&EPA
United States
Environmental Protection
Agency
Office of Research
and Development
Washington DC 20460
EPA/600/R-96/074
July 1996
Analysis of the National
Human Activity Pattern
Survey (NHAPS)
Respondents from a
Standpoint of Exposure
Assessment
EJBD
V ARCHIVE
EPA
600-
R-
96-
074
-------
EPA/600/R-96/074
Ju|v1996
Analysis of the National Human Activity
Pattern Survey (NHAPS) Respondents from
a Standpoint of Exposure Assessment
Percentage of Time Spent, Duration, and
Frequency of Occurrence for Selected
Microenvironments by Gender, Age,
Time-of-day, Day-of-week, Season,
and U.S. Census Region
Final Report RepOSltOiy M&\&\
^ Permanent Collection
Neil E. Klepeis and Andy M. Tsang
Information Systems and Services Incorporated
4220 South Maryland Parkway, Suite 31 1
Las Vegas, Nevada 89119
CO
Q
=!§ i^- Joseph V.Behar
§ Characterization Research Division
4E g § ^ c\i o 5 ° co Post Off jce Box 93478
2? c <5 o I o <° Las Vegas, NV 89193-3478
^^J _— »*«* ,. . rrj Si. 1 **
f ..-, i; r^ O)rvi
>r ™ 4.., —. i\j
W ,? 03 O — O
o^SO f ^ U.S. EPA Contract No. 68-01-7325
r"> "^
_ " i— to
TO tf O ,3»
§ 2i ^ ^* Work Assignment, Project Officer: Joseph V. Behar
"ro Characterization Research Division
X Las Vegas, Nevada 89193
National Exposure Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Las Vegas, Nv 89193-3478
-------
Notice
The U.S. Environmental Protection Agency (EPA), through its Office of Research and Development
(ORD), partially funded and collaborated in the research described here. It has been peered reviewed by
the Agency and approved as an EPA publication.
-------
Acknowledgments
The National Human Activity Pattern Survey (NHAPS) was conducted for the U.S. Environmental
Protection Agency (EPA) between October 1992 and September 1994 by the Survey Research Center at
the University of Maryland, College Park. The current report was funded by the Office of Research and
Development (ORD) of the U.S. Environmental Protection Agency (EPA), National Exposure Research
Laboratory, Characterization Research Division (CRD-LV) in Las Vegas, Nevada. Timothy Triplett and
John Robinson of the University of Maryland were very helpful in answering our questions, sending us
updated versions of the NHAPS database, and guiding us in some of our analyses including the use of
sub-population weights. William Nelson, Wayne Ott, Lance Wallace (all of the U.S. Environmental
Protection Agency), and John Robinson provided guidance for determining what analyses and
breakdowns were to be included in the final report.
in
-------
Abstract
The National Human Activity Pattern Survey (NHAPS) 24-hour diary data collected from 9,386
respondents interviewed between October 1992 and September 1994 were receded and regrouped into
ten locations, seven exposure-relevant activities, and two smoker-present categories. For all the NHAPS
respondents, the mean number of occurrences and durations in selected microenvironments (location x
activity x smoker present category), and the percentage of time spent in the microenvironments were
calculated over the 24-hour diary day and by time-of-day (midnight to midnight of the diary day in one
minute and 3-hour time segments). In addition to overall calculations, the data were divided by gender
(male/female), age (adult/child), day-of-week (weekend/weekday), season (fall/winter and
spring/summer), and the four census regions. The largest overall percentages of tune were spent in the
Residential-Indoors location (69%) with 5.5% percent of the time spent in vehicles and almost 8% of the
time spent in outdoor locations. The largest percentages of time in any location x exposure activity
microenvironment were for the Eating/Drinking, Housekeeping, Food Preparation, and Bathing exposure
activities — all taking place in the Residential-Indoor location.
Many of the NHAPS follow-up questions on personal exposure to air and water pollutants - asked
after the 24-hour diaries were collected - contained many missing, "refused to answer," or "don't know"
responses, or were coded in an unwieldy mixed-type format (categorical and numeric data). Only the
yes/no categorical follow-up questions with an adequate sample size were analyzed (percentage of "yes"
responses). Overall percentages of "yes" responses that were greater than 50% occurred for: using an air
freshener, using a dish washer; washing dishes by hand; giving or taking a bath; taking a shower;
children swimming over the last month; having a door leading directly outdoors; using a microwave;
running/walking outdoors; walking to a car in the driveway; and having a welcome mat.
All analyses of the NHAPS sample were weighted against the 1990 US Census to account for the:
(1) oversampling done on weekends; (2) probability of sampling adults (18 and up) vs. children; (3)
probability of a household's selection; (4) disproportionate weekday ratios; (5) unrepresentative
male/female ratio; (6) disproportionate sampling by season; and (7) unrepresentative ratios among ten
age groups. The NHAPS data base is large and rich enough to provide material for very detailed
exposure studies targeted on specific populations in the United States. The results presented here are a
fraction of the possible analyses that can be used in total human exposure assessment. Since minute-by-
minute 24-hour diaries provide the most complete and accurate human activity pattern data, future human
activity pattern studies should further focus the exposure-relevance of the diary activity categories.
IV
-------
Table of Contents
Notice ii
Acknowledgments iii
Abstract iv
Notice to the Reader vii
List of Tables ix
List of Figures xiv
Section 1 - Introduction 1-1
1.1 Objective and Scope of This Report 1-1
1.2 Background 1-2
1.3 The Analyses Presented in This Report 1-3
1.4 Calculation Methodology 1-6
1.5 Using the Results: Exposure Assessments 1-7
Section 2 - Summary of the Data-Collection Methodology 2-1
Section 3 - Geographic, Socioeconomic, and Temporal Divisions of the Respondents 3-1
3.1 Geographic Divisions 3-1
3.2 Socioeconomic Divisions 3-7
3.3 Temporal Divisions 3-14
Section 4 - Weighting of the Data 4-1
Section 5 - Microenvironments in the 24-hour Diaries: Selected Location x Activity
x Smoker-Present Categories 5-1
5.1 Locations, Activities, and Smoker-Present Categories 5-2
5.2 Location x Activity and Location x Smoker-Present Microenvironments 5-10
Section 6, Part I - 24-hour Diary Results by Day-of-week, Season, and Background Factors ... 6-1
6.1 Locations 6-2
6.2 Exposure Activities 6-14
6.3 Smoker-Present Categories 6-24
6.4 Locations x Smoker-Present Categories : 6-26
6.5 Locations x Activities 6-38
6.6 Summary and Discussion 6-47
-------
Section 6, Part n - 24-hour Diary Results by Time-of-day in Minute Time Segments
by Day-of-week, Season, and Background Factors 6-48
6.7 Locations 6^48
6.8 Exposure Activities 6-56
6.9 Smoker-Present Categories 6-63
Section 6, Part HI - 24-hour Diary Results by Time-of-day in 3-hour Time Segments 6-65
Section 7 - Results of Follow-up Questions on Personal Exposure 7-1
Section 8 - Conclusions, Recommendations, and Future Work 8-1
References ^"1
Appendix A - Conceptual Methodology and Background A-l
Appendix B - Detailed Examples of the Methodology Used for Calculations in this Report
(With SAS31 Computer Code) B-l
VI
-------
Notice to the Reader
To help avoid any confusion, we have compiled the following eleven items concerning our analyses of the
National Human Activity Pattern Survey (NHAPS) data:
1. The analyses in this report are focused on the minute-by-minute NHAPS 24-hour diaries with
limited analyses of the NHAPS follow-up questions (see Section 7).
2. Table 1-1 in the introduction gives a complete index of the figures and tables in this report
containing the results of the 24-hour diaries - broken down by the type of human activity pattern
calculation (percentage of time spent, 24-hour duration, etc.) and the subgroups covered (age,
gender, time-of-day, day-of-week, and season).
3. The final ten location groupings that were used for diary calculations in this report were:
Residential-Indoor, Residential-Outdoor, In Vehicle, Near Vehicle(s), Other Outdoor,
Office/Factory, Mall/Store, School/Church/Public Bldg., Bar/Restaurant, and Other Indoor.
4. The final eight activity groupings that were used for diary calculations in this report were: Food
Preparation, Laundry/Dishes, Housekeeping, Bathing, Yardwork/Maintenance, Sports/Exercise,
Eating/Drinking, and No Exposure (sleeping, watching TV, etc.). The No Exposure activity group
contains activity categories that contribute little or no additional information on exposure beyond
what the locations contribute. For example, if the location is In Vehicle and the activity is
Traveling in Vehicle, then the location is sufficient to designate that the respondent may be
exposed to automobile exhaust while driving on roadways. Thus, the Traveling in Vehicle activity
was assigned to the No Exposure group.
5. In this report both the doers and the non-doers were analyzed for diary calculations of the
percentage of time spent in each location and activity. Only the doers were analyzed for the
percentage of time spent exposed to environmental tobacco smoke (ETS). For all calculations of
mean microenvironment duration and frequency of occurrence, only the doers were considered,
although tables are included giving both the doer and the total sample sizes.
6. No frequency distributions or statistical comparisons of distributions of time spent (e.g., percentage
or duration) are presented in this report.
7. In exposure research, microenvironments are defined as combinations of locations and activities
(see Appendix A), although in this report we also present analyses according to locations alone and
activities alone.
8. Another report (J. Robinson; J. Blaire (1995) "Estimating Exposure to Pollutants Through Human
Activity Pattern Data: The National Microenvironmental Activity Pattern Survey", Annual Report,
Survey Research Center, U. of Maryland) provides the detailed National Human Activity Pattern
Survey (NHAPS) data-collection methodology.
Vll
-------
9. This report is meant only as a presentation of selected examples of results that can be obtained
from the NHAPS data, and not a comprehensive treatment that will be of use to all human
exposure assessors.
10. Appendix A contains a development of exposure concepts — including the use of activity patterns
in total human exposure assessment.
11. Appendix B contains detailed examples of the computer methodology used in this report including
the SAS computer code, the computer hardware used, and the amount of computer time required.
VUl
-------
List of Tables
Page
No. Description No.
1-1 Index of the statistical calculations presented in this report 1-5
1-2 How different percentages of time spent and numbers of people experiencing a 1-8
microenvironment (doers) contribute to the overall significance of the
microenvironment
1-3 How each microenvironmental quantity (HAP calculation or experiment) is 1-9
used to estimate exposure
2-1 Summary of the NHAPS features 2-1
2-2 Number of NHAPS respondents at the time the interview began 2-3
2-3 Number of NHAPS respondents by type of interview 2-3
2-4 Number of NHAPS respondents by number of non-business phones in each 2-3
household
2-5 Description of the two NHAPS data files 2-4
3-1 The NHAPS geographic variables for each respondent 3-1
3-2 List of NHAPS frequencies grouped by EPA Region, Census Region, and state 3-2
3-3 Comparison of Census with NHAPS by EPA Region 3-6
3-4 Comparison of Census with NHAPS by Census Region 3-6
3-5 The number of NHAPS respondents hi each area code by state 3-8
3-6 The 20 most populated U.S. cities in 1990 3-11
3-7 The NHAPS demographic variables for each respondent 3-11
3-8 Number of NHAPS respondents by gender , 3-13
3-9 Number of NHAPS respondents by age 3-13
3-10 Number of NHAPS respondents by race 3-13
3-11 Number of NHAPS respondents by Hispanic origin 3-13
IX
-------
3-12 Number of NHAPS respondents by education level 3-13
3-13 Number of NHAPS respondents by employment status 3-13
3-14 Number of NHAPS respondents by number of adults in household 3-14
3-15 Number of NHAPS respondents by number of children under 18 3-14
3-16 List of the NHAPS temporal variables 3-14
3-17 Number of NHAPS respondents by seasonal quarter 3-14
3-18 Number of NHAPS respondents by season 3-14
3-19 Number of NHAPS respondents by year 3-15
3-20 Number of NHAPS respondents by weekend vs. weekday 3-15
3-21 Number of NHAPS respondents by day-of-week 3-15
4-1 The twenty gender and age subgroups for the calculation of weights 4-3
4-2 The 1990 U.S. Census population joint frequency and proportions for each age 4-4
and gender subpopulation
4-3 The NHAPS joint frequency and proportions for each age and gender subgroup 4-4
4-4 The sample weights Aft calculated for each subgroup h and assigned to each 4-4
individual i
4-5 Comparison of the weighted sample with the 1990 U.S. Census by marginal 4-7
grouping factors
4-6 Consistency check of the weighted joint sample proportion for gender and age 4-8
across season and weekend vs. weekday
5-1 The NHAPS microenvironmental variables 5-1
5-2 The original WHR location codes matched with NEWLOC groupings 5-2
5-3 The NEWLOC locations with overall frequency of occurrence 5-5
5-4 The NHAPS activity codes for RACT with the REGACT regroupings 5-5
5-5 The NEWACT activities regrouped from the FACT and RACT variables 5-8
5-6 The eight REGACT activities regrouped from NEWACT 5-8
5-7 The five largest non-exposure RACT categories 5-9
5-8 The NHAPS smoker-present categories 5-10
5-9 Frequency table for SMKEXP (time exposed to ETS) with SMKB assignment 5-10
(number of people exposed)
-------
5-10 The location x activity rnicroenvironmental matrix 5-11
5-11 The location x smoker-present microenvironmental matrix 5-12
6-1 The weighted percentages of time spent in each location on the diary day across 6-2
each subgroup
6-2 The mean weighted 24-hour duration (minutes) in each location (for doers) on 6-4
the diary day for each subgroup
6-3 The weighted number of NHAPS respondents (sample size N) in each location 6-4
(the doers) on the diary day by each subgroup
6-4 The weighted percentages of time spent in each exposure activity on the diary 6-15
day across each subgroup
6-5 The mean weighted 24-hour duration (minutes) in each exposure activity (for 6-16
doers) on the diary day for each subgroup
6-6 The weighted number of NHAPS respondents (doer sample size N) in each 6-17
exposure activity on the diary day by each subgroup
6-7 The weighted percentages of time spent in each location x smoker-present 6-29
microenvironment on the diary day across each subgroup
6-8 The mean weighted 24-hour duration (minutes) in each location x smoker- 6-29
present microenvironment (for doers) on the diary day for each subgroup
6-9 The weighted mean frequency of occurrence of each location x smoker-present 6-30
microenvironment (for doers) on the diary day by each subgroup
6-10 The weighted number of NHAPS respondents (doer sample size N) in each 6-30
location x smoker-present microenvironment on the diary day by each subgroup
6-11 The overall weighted percentage of time spent in 21 out of 70 location x activity 6-39
microenvironments
6-12 The overall weighted percentage of people doing (the doers) 21 out of 70 6-40
location x activity microenvironments
6-13 The overall weighted mean 24-hr durations of 21 out of 70 location x activity 6-41
microenvironments
6-14 The overall weighted mean 24-hr number of occurrences of 21 out of 70 6-42
location x activity microenvironments
6-15 Comparison across subgroups of percentage of time spent in nine selected 6-44
location x activity microenvironments
6-16a Comparison across subgroups of 24-hour duration in nine selected location x 6-46
activity microenvironments
XI
-------
6-16b The weighted doer sample sizes of nine selected location x activity 6-46
microenvironments across each subgroup
6-17 Percentage of respondents who entered a location beginning in each of eight 6-65
consecutive 3-hour time segments
6-18 Cumulative durations (minutes) of locations originating in the eight consecutive 6-67
3-hour time segments
6-19 Percentage of respondents who entered an exposure activity beginning in each 6-69
of eight consecutive 3-hour time segments
6-20 Cumulative durations (minutes) of exposure activities originating in the eight 6-71
consecutive 3-hour time segments
6-21 Cumulative duration and percentage of respondents for ETS exposures 6-73
originating in each of eight 3-hour time segments
7-1 The ten NHAPS follow-up question categories with example activities and 7-2
associated chemicals
7-2 The NHAPS follow-up variables with category, A or B questionnaire, and the 7-3
number of missing, refused, and "don't know" responses
7-3 Percentage of "yes" responses for follow-up questions on chemical exposure 7-12
(CE)
7-4 Percentage of "yes" responses for follow-up questions on combustion sources 7-14
(CS)
7-5 Percentage of "yes" responses for follow-up questions on washing and bathing 7-15
(WA and BA)
7-6 Percentage of "yes" responses for follow-up questions on medical background 7-16
(MB)
7-7 Percentage of "yes" responses for follow-up questions on housing 7-17
characteristics (HC)
7-8 Percentage of "yes" responses for follow-up questions for other categories 7-18
8-1 Comparison of the utility of 24-hour diaries vs. follow-up questions 8-2
8-2 Suggested specific exposure activity categories for microenvironments in the 8-3
minute-by-minute 24-hour diaries for future HAP studies
A-l An example episodic formulation of ten persons in indoor and outdoor A-8
microenvironments over a 12-hour period
A-2 An example time-of-day formulation of ten persons with frequencies of indoor A-10
and outdoor microenvironments over 12-hours in half-hour increments
xu
-------
A-3 Example breakdown of the time-of-day formulation M, into microenvironmental A-12
factor terms m
A-4 Example breakdown of the 12-hour total exposure array E(ij) into exposure A-13
arrays e for each combination of microenvironmental factors (indoors/outdoors
andsmoker-present/no-smoker-present)
B-l Example 24-hour diary file in the episodic format for person ID# 1.1 B-2
B-2 Example microenvironment calculations of percentage of time spent, 24-hour B-3
duration, and 24-hour frequency of occurrence using person E># 1.1
B-3 Example time-of-day format for person ID# 1.1 B-7
B-4 Example output data file containing the overall weighted fraction of B-11
respondents in each location (NEWLOC category) for every minute of the diary
day
Note: Tables and figures are numbered in the order they were referenced in the text with their number preceded by the
section number and a hyphen (-). For example, the 5th table in Section 3 is labeled Table 3-5'.
Kill
-------
List of Figures
Page
No. Description No.
1-1 The four levels of human activity pattern analysis used in this report 1-4
2-1 Histogram of the number of minutes each interview lasted 2-2
3-1 Map of the number of NHAPS respondents in each state and EPA region 3-3
3-2 Map of the number of people in each state and EPA region from the 1990 3-3
Census
3-3 Bar chart comparison of NHAPS respondents vs. the Census by EPA Region 3-4
3-4 Bar chart comparison of NHAPS respondents vs. the Census by Census Region 3-5
3-5 Plots of males and females by four age groups and ten age groups in 1990 3-12
4-1 SAS program for calculation of weights 4-6
6-1 The overall weighted percentage of time spent in each location 6-2
6-2 The overall weighted and unweighted mean 24-hr durations in each location 6-3
6-3 The weighted percentage of tune spent in each location for males vs. females 6-5
6-4 The weighted and unweighted mean 24-hr durations in each location for males 6-5
vs. females
6-5 The weighted percentage of time spent in each location for each age group 6-6
6-6 The weighted and unweighted mean 24-hr durations in each location for each 6-7
age group
6-7 The weighted percentage of time spent in each location for each census region 6-9
6-8 The weighted and unweighted mean 24-hr durations in each location for each 6-10
census region
6-9 The weighted percentage of time spent in each location for weekdays vs. 6-11
weekends
6-10 The weighted and unweighted mean 24-hr durations in each location for 6-11
weekdays vs. weekends
nv
-------
6-11 The weighted percentage of time spent in each location for each season 6-12
6-12 The weighted and unweighted mean 24-hr durations in each location for each 6-13
season
6-13 The overall weighted percentage of time spent in each exposure activity 6-15
6-14 The overall weighted and unweighted mean 24-hr durations of each exposure 6-16
activity
6-15 The weighted percentage of time spent in each exposure activity for males vs. 6-17
females
6-16 The weighted and unweighted mean 24-hr durations in each exposure activity 6-18
for males vs. females
6-17 The weighted percentage of time spent in each exposure activity for each age 6-18
group
6-18 The weighted and unweighted mean 24-hr durations in each exposure activity 6-19
for each age group
6-19 The weighted percentage of time spent in each exposure activity for each census 6-20
region
6-20 The weighted and unweighted mean 24-hr durations in each exposure activity 6-20
for each census region
6-21 The weighted percentage of time spent in each exposure activity for weekdays 6-21
vs. weekends
6-22 The weighted and unweighted mean 24-hr durations in each exposure activity 6-22
for weekdays vs. weekends
6-23 The weighted percentage of time spent in each exposure activity for each season 6-22
6-24 The weighted and unweighted mean 24-hr durations in each exposure activity 6-23
for each season
6-25 The weighted percentage of people exposed to ETS for each subgroup 6-25
6-26 The weighted percentage of time that those who were exposed to ETS on the 6-25
diary day were exposed for each subgroup
6-27 The weighted mean 24-hr duration of ETS exposure by each subgroup 6-25
6-28 The overall weighted percentage of time spent, mean number of occurrences, 6-28
and mean 24-hr duration of ETS exposure in each location
6-29 The weighted percentage of time spent exposed to ETS in each location for 6-31
males vs. females
6-30 Weighted mean 24-hr number of occurrences and mean 24-hr duration of ETS 6-32
exposure in each location for males vs. females
XV
-------
6-31 Weighted percentage of time spent exposed to ETS in each location by age 6-33
group
6-32 Weighted mean 24-hr duration of ETS episodes in each location by age group 6-33
6-33 Weighted mean 24-hr number of occurrences of ETS exposure in each location 6-33
by age group
6-34 Weighted percentage of time spent exposed to ETS in each location by census 6-34
region
6-35 Weighted mean 24-hr duration of ETS episodes in each location by census 6-34
region
6-36 Weighted mean 24-hr number of occurrences of ETS exposure in each location 6-34
by census region
6-37 Weighted percentage of time spent exposed to ETS in each location by weekend 6-35
vs. weekday
6-38 Weighted mean 24-hr number of occurrences and mean 24-hr duration of ETS 6-36
episodes in each location by weekends vs. weekdays
6-39 Weighted percentage of time spent exposed to ETS in each location by season 6-37
6-40 Weighted mean 24-hr duration of ETS episodes in each location by season 6-37
6-41 Weighted mean 24-hr number of occurrences of ETS exposure in each location 6-37
by season
6-42 3-D plot of the overall weighted percentage of time spent in 21 out of 70 6-38
location x activity microenvironments
6-43 3-D plot of the overall weighted number of people doing (the doers) 21 out of 6-39
70 location x activity microenvironments
6-44 3-D plot of the overall weighted mean 24-hr duration in 21 out of 70 location x 6-40
activity microenvironments
6-45 3-D plot of the overall weighted mean 24-hr number of occurrences of 21 out of 6-41
70 location x activity microenvironments
6-46 3-D plot of the overall weighted percentage of time spent in nine location x 6-43
activity microenvironments across each subgroup
6-47 3-D plot of the overall weighted mean 24-hour duration in nine location x 6-45
activity microenvironments across each subgroup
648 Stacked plot of overall fraction of respondents in each location by time-of-day 6-50
6-49 Individual plots of the overall fraction of respondents in each exposure activity 6-51
by time-of-day
6-50 Individual plots of the fraction of respondents in each location by time-of-day 6-52
for each age group
XVI
-------
6-51 Individual plots of the fraction of respondents in each location by time-of-day 6-54
by weekday vs. weekend
6-52 Plots of the fraction of respondents in the Residential-Outdoor and Other 6-55
Outdoor locations by season
6-53 Stacked plot of overall fraction of respondents in each exposure activity by 6-57
time-of-day
6-54 Individual plots of the overall fraction of respondents in each exposure activity 6-58
by time-of-day
6-55 Individual plots of the fraction of respondents in each exposure activity by time- 6-60
of-day for each age group
6-56 Individual plots of the fraction of respondents in each exposure activity by time- 6-61
of-day for weekdays vs. weekends
6-57 Individual plots of the fraction of respondents in the Yardwork/Maintenance 6-62
and Sports/Exercise exposure activities by time-of-day for each season
6-58 Plots of the fraction of respondents exposed to environmental tobacco smoke 6-64
(ETS) by time-of-day for gender, age, weekday vs. weekend, and season
6-59 3-D plot of the percentage of respondents initiating a microenvironment in each 6-66
of the eight 3-hour time segments for each location
6-60 3-D plot of the mean total duration in each location for each of the eight 3-hour 6-68
time segments
6-61 3-D plot of the percentage of respondents initiating a microenvironment in each 6-70
of the eight 3-hour time segments by exposure activity
6-62 3-D plot of the mean total duration in each exposure activity for each of the 6-72
eight 3-hour time segments
6-63 Bar charts of the percentage of respondents exposed to ETS and the duration of 6-74
exposure to ETS (minutes) in each of the eight 3-hour time segments
7-1 The percentage of "yes" responses to NHAPS follow-up variables for chemical 7-10
exposure (CE) #1
7-2 The percentage of "yes" responses to NHAPS follow-up variables for chemical 7-11
exposure (CE) #2
7-3 The percentage of "yes" responses to NHAPS follow-up variables for 7-13
combustion sources (CS)
7-4 The percentage of "yes" responses to NHAPS follow-up variables for washing 7-15
and bathing (WA/B A)
7-5 The percentage of "yes" responses to NHAPS follow-up variables for medical 7-16
background (MB)
xvu
-------
7-6 The percentage of "yes" responses to NHAPS follow-up variables for housing 7-17
characteristics (HC)
7-7 The percentage of "yes" responses to NHAPS follow-up variables for other 7-18
categories (DS, ED, OQ
A-l The basic questions posed in human activity pattern analysis A-2
A-2 Example percentage of people in indoor vs. outdoor microenvironments by A-10
time-of-day in half-hour increments over a 12-hour period
B-l The SAS code for calculating total time spent, 24-hour durations, and 24-hour B-4
frequencies of occurrence of microenvironments using the episodic formulation
B-2 The SAS code for converting the NHAPS data from the episodic to the time-of- B-5
day formulation
B-3 The SAS code used to concatenate the 24 files containing the time-of-day B-9
formulation for each hour of the diary day
B-4 The SAS code used to calculate the unweighted and weighted numbers of B-10
respondents in each location for every minute of the diary day
B-5 The SAS code used to calculate weighted and unweighted overall mean B-12
durations in each of the eight 3-hour time segments over the 24-hour diary day
Note: Figures and tables are numbered in the order they were referenced in the text with their number preceded by the
section number and a hyphen (-). For example, the 5th figure in Section 3 is labeled 'Figure 3-5*.
XVlll
-------
Section 1
Introduction
1.1 Objective and Scope of This Report
The U.S. Environmental Protection Agency (EPA) has a continuing interest in the estimation of
human exposure to toxic chemicals using the microenvironmental data (locations and activities) collected
in human activity pattern (HAP) surveys. Since the Total Exposure Assessment Methodology (TEAM)
studies have shown that indoor sources of pollution outnumber outdoor sources, and the 1988-1990
California Air Resource Board (GARB) California Activity Pattern (CAP) HAP study1'3 and the 1985
University of Maryland Americans' Use of Time (AUT) HAP study4 show that Americans spend most of
their time in a variety of indoor locations, information on where and how people are exposed to indoor
pollutants is crucial in total human exposure (THE) assessment.5"7 This report analyzes the EPA's
National Human Activity Pattern Survey (NHAPS)8'9 by duration and frequency of occurrence of
microenvironments (combinations of locations, activities, and smoker-present categories) for selected
time periods (time-of-day, day-of-week, season) and background factors (age, gender, census region).
Completed in December 1994, NHAPS is the largest and most current HAP survey available. It will soon
be included in the THERDbASE modeling and relational data base computer environment,10 which has
been developed by the EPA. For this report, durations and frequencies of occurrence for an exposure-
relevant subset of all possible microenvironments are given in the form of means, percentages of time
spent, and percentage of respondent occurrences, and can be readily included in the EPA's Exposure
Factors Handbook. The contents of this report will also help to focus the exposure emphasis, follow-up
questions, and location and activity categories for future HAP studies. A comprehensive description of
the NHAPS data-collection methodology — including example questionnaires - is presented in another
work.8
In addition to a cross section of the kinds of analyses that are useful in exposure research (Sections 6
and 7), we have included the major NHAPS features (Section 2), time and background factor divisions
with comparisons to the 1990 U.S. Census (Section 3), weighting factors (Section 4), and
microenvironment designations (Section 5) - including lists of variables, samples sizes, and codes - so
that this report will be a useful reference to all human exposure modelers, exposure/risk assessors, and
environmental health professionals as they use the NHAPS data to conduct future exposure studies across
the United States. The analyses in this report are not a comprehensive summary of every possible
categorical division of the data and many researchers will desire more specific breakdowns. However,
when combined with the detailed descriptions of the analysis methodology in Appendices A and B, the
analyses that are presented here provide a guide for future work. By reading this report researchers can
find: (1) useful breakdowns and examples of ways to present the NHAPS data by location, exposure
activity, smoker-present category and selected combinations; (2) which NHAPS variables and/or
microenvironments are relevant for their work; (3) whether NHAPS contains sufficient data for the
breakdowns of interest; and (4) examples of appropriate ways to weight the NHAPS data.
1-1
-------
We only considered those exposure activities that could be related to explicit exposures such as
cooking (particles), cleaning (detergents), showering (water contaminants or chloroform), eating (food
contaminants), or exercising (accelerated inhalation of air contaminants). Activities such as traveling,
sleeping, dressing, talking, or watching TV were placed into a "No Exposure" category since no explicit
exposure to contaminants corresponds to these categories. Exposure may still be occurring during these
activities if there is contact with (i.e., proximity to) pollutant sources such as engine exhaust in traffic,
carpet fumes in the bedroom, or fabric irritants, but this information is not contained in the NHAPS data.
After locations such as In Vehicle or In Residence are analyzed, the activities in the No Exposure
category give very little or no additional information that can be used to infer exposure. In addition,
there are too many different kinds of activities in the No Exposure category to warrant analyzing each of
them. Consequently, separate analyses were not done for these activities.
The results in this report may be used directly in point estimates (i.e., long-term averages) of human
exposure. However, estimates of frequency distributions of exposure (i.e., probabilistic modeling) will
require researchers to use the NHAPS data to calculate frequency distributions of microenvironment
duration and frequency of occurrence. These statistics are not presented in this report.
1.2 Background
NHAPS and CAP are part of recent efforts to estimate human exposure to pollutants by interviewing
a sample of individuals in a target population, and categorizing their 24-hour routines (diaries) and/or
answers to follow-up exposure questions according to exposure-related events such as: (1) occupying a
location; or (2) an activity. These events are stored as human activity patterns which provide answers to
the basic "who", '*what", "where", "how long", "how often", and "when" questions about the target
population that are critical in efforts to model personal exposure to pollutants for specific populations.
The 24-hour diaries in HAP surveys are particularly useful in probabilistic models (using Monte-Carlo
sampling techniques), which provide frequency distributions of exposure.
For CAP, California adults and children were interviewed between October 1987 and February
19901*3 and follow-up questions were focused on personal exposure to contaminants in air (tobacco
smoke, solvents, engine exhaust, etc.), while NHAPS has collected information on both air and water
exposure (showering, swimming, tap water consumption, etc.). Both studies contain essentially
equivalent 24-hour diary formats (although the location and activity categories are somewhat different)
including information on the presence of smokers. The CAP data have been used in exposure
modeling11"13 and other exposure assessments14*18 to estimate what portions of the target population are
being exposed, how much they are exposed, and for how long. Likewise, NHAPS will also provide a rich
database for estimates of exposure for various subpopulations across the entire United States.
There are two approaches to total human exposure (THE) assessment: the "direct" and the
"indirect" approaches.19'21 In the direct approach, multiple individuals or households in the target
population are equipped with monitoring devices that measure concentrations of pollutant for later
analysis. This approach provides immediate data for the estimation of frequency distributions of
exposure across an entire population. Examples of the direct approach are the Total Exposure
Assessment Methodology (TEAM) studies.22"26 In contrast, the indirect approach combines HAP data
and pollutant concentration data for the same population — usually from different studies - and uses an
exposure model to calculate frequency distributions of exposure across the target population.11"13'27"29 The
direct and indirect approaches produce similar results, but with the use of a large HAP survey such as
1-2
-------
NHAPS, the indirect approach is a more cost-effective way to study many different populations across
the country.
In the evolution of exposure modeling and THE assessment, the concept of the microenvironment
has proved useful in classifying different kinds of exposure. The microenvironment is the basic building
block of HAP survey diaries and is defined by the time period that some combination of exposure events
(an activity occurring in a particular location) — called an episode — occurs. In this report we split the
microenvironmental concept into a microenvironmental-factor component (location, activity, and
smoker-present categories by themselves or in combination) and a time component (beginning and
ending times over the 24-hour diary day or - for time-of-day analysis — equal time intervals such as a
minute or 3-hours). See Appendix A for a discussion of the microenvironmental concept and its use in
human exposure modeling.
Before undergoing a new monitoring study or modeling effort, investigators can discover — using
the indirect exposure assessment approach and a data base such as NHAPS - for which time periods and
geographic and socioeconomic subpopulations a specific microenvironment occurs long enough and
frequently enough to pose a potentially significant risk to human health. For example, estimates of the
exposure to carbon monoxide (CO) from environmental tobacco smoke (ETS) in office settings (as
determined by monitoring studies using actual smokers in real offices) can be weighted by the total
duration of all office setting locations obtained from NHAPS to determine the seriousness of office CO
exposures from ETS, and which city, state, gender, age group, etc., is connected with the most risk.
13 The Analyses Presented in This Report
In this report, 24-hour statistical calculations on each respondent's 24-hour diary are based on two
derived variables that quantify how long and how often microenvironments are occurring in the NHAPS
sample: (1) the duration of microenvironments D; and (2) the frequency of occurrence of
microenvironments O. The number of respondents experiencing each microenvironment - the doer
sample size N — was also calculated. The six calculations presented in this report are the: (1) sample
proportion (percent) of time spent over 24-hours - based on D alone; (2) sample proportion of people
over 24-hours — based on N; (3) descriptive statistics (means) of both D and O; (4) sample proportion of
occurrences by minute - based on O alone; (5) descriptive statistics of D by 3-hour segments; and (6) the
sample proportion of people by each of the eight 3-hour segments in the 24-hour diary day (see Table 1-1
for an index of results figures and tables corresponding to each of the six analyses). The descriptive
statistics of D and O were only performed on respondents for which each microenvironment occurred -
referred to as the "doers". The sample proportions of time spent were performed over all respondents,
i.e., both the doers and non-doers, except for calculations on exposure to ETS where only the doers were
considered.
Since many of the follow-up questions on frequency of occurrence and duration of exposure events
(e.g., number of minutes spent exposed to glue) had missing data or were recoded in an unwieldy format
that contained both categorical (for example, "under one minute" or "over two hours") and interval (an
exact number of minutes) data, they were not analyzed. The follow-up variables concerning the
occurrence or non-occurrence of exposure events on the diary day or in some other time period (e.g., the
last month or the last six months) that had enough valid responses were analyzed by the percentage of
"yes" answers (Section 7). A "yes" response can be used to represent an exposure to the corresponding
1-3
-------
chemical source, with higher percentages of "yes" responses in each subgroup representing higher
exposures.
For the 24-hour diaries, the original NHAPS microenvironmental factors (locations, activities, and
smoker-present categories) were receded into new categories according to (Section 4): (1) their
relevance to personal exposure assessment; and (2) adequate sample size. The resulting
microenvironments were broken down according to the background factors of gender (male/female), four
age groups (0-4,5-17,18-64,65 and over), and the four U.S. census regions (northeast, midwest, south,
and west), and the time factors of day-of-week (weekend/weekday), season (fall, winter, spring, summer),
and time-of-day. The follow-up questions were also analyzed by each background and time factor except
time-of-day since the responses to each follow-up question did not contain time-of-day information. The
development of appropriate weighting factors for all calculations is discussed in Section 4.
In theory, the total number of possible analyses (of the 24-hour diaries) is equal to the number of
microenvironmental factors times the number of time factors times the number of background factors
times the number of calculations (Figure 1-1). For example, in this report we use 10 locations, 7
activities, two smoker-present categories, time-of-day by minute and 3-hour segments, day-of-week by
weekend and weekday, spring/summer and fall/winter seasons, four age groups, gender, the four census
regions, and six D,O,N-calculations, which gives 10x7x2x2x2x2x4x2x4x6 = 215,040 possible
analyses. Of these, 150,528 are based on the 49 microenvironments that had fewer than 100 occurrences.
Since for obvious reasons we cannot present all the possible analyses, we have chosen to omit location x
activity and location x smoker-present analyses for calculations involving time-of-day (calculation #'s 4,
5, and 6 in Table 1-1). The proportion of microenvironment occurrences for time-of-day at minute
resolution (calculation #4) were done for selected background factors, and the 3-hour segmented
descriptive statistics and sample proportions of people (calculations #5 and #6) were only done over all
the respondents. Besides time-of-day by day-of-week and season, no calculations were done for mixed
time factors or mixed background factors.
Microenvironmental Time
Factors Factors
Background
Factors
Statistical
Calculations
* Activity
• Location
* Presence of a Smoker
I
Location x Activity x
Smoker-Present
Durations, D
(How Long?)
- Descriptive Statistics
- Proportion of Time Spent
Frequency of Occurrence, O
(How Often?)
- Descriptive Statistics
- Proportion of Occurrences
Rgure 1-1. The four levels of analysis used in this report for the NHAPS 24-hour human activity pattern diaries: (1)
three microenvironmental factors; (2) three time factors; (3) three background factors; and (4) two
calculations for the "How Long?" category and two calculations for the "How Often?" category.
1-4
-------
Table 1-1. Index of Tables and Figures in Section 6 Containing the Results of D,O,N Statistical Calculations for Selected Microenvironments Over the
Respondents' 24-hour Diaries by Time and Background Grouping Factors
1.
2.
3.
4.
5.
6.
•Derived
Varlable(s)
D
N
D,0
O
D
N
"Statistical
Calculations
Sample
Proportion of
Time Spent
Sample
Proportion (or
number) of
Respondent
Occurrences
Descriptive
Statistics
Sample
Proportion of
Microenvironmen
t Occurrences
Descriptive
Statistics
Sample
Proportion of
Respondent
Time Unit of
Study
24-Hour Diary Day
24-Hour Diary Day
24-Hour Diary Day
Minute
3-Hour Segment
3-Hour Segment .
'Grouping Factors
Time
Day of Week
Season
Day of Week
Season
Day of Week
Season
Time of Day
Time of Day
Time of Day
Time of Day
Time of Day"
Day of Week
Time of Day"
Season
Time of Day
Time of Day
Backgroun
d
Gender
Age
Region
Gender
Age
Region
Gender
Age
Region
Gender
Age
Region
-
*Mlcroenvlronments
Location
(10)
f 6-1, 1 6-1
« 6-3, 1 6-1
f 6-5, 1 6-1
f 6-7, 1 6-1
f 6-9, 1 6-1
f6-11,t6-1
16-3
16-3
t6-3
16-3
16-3
t6-3
f 6-2, 1 6-2
f 6-4, 1 6-2
f 6-6, t 6-2
f 6-8, 16-2
f 6-10,16-2
f 6-12,16-2
f 6-48,49
f6-50
f6-51
f 6-52
f 6-60,
16-18
f 6-59,
t6-17
Activity
(7)
f 6-13,16-4
f 6-15. 16-4
f 6-17, 1 6-4
f 6-19,16-4
f 6-21, 16-4
f 6-23, 1 6-4
t6-6
t6-6
t6-6
16-6
t6-6
16-6
f 6-14,16-5
f 6-16, 1 6-5
f 6-18,16-5
f 6-20, 1 6-5
f 6-22, 1 6-5
f 6-24, 1 6-5
f 6-53,54
f6-55
f6-56
f6-57
f 6-62,
t6-20
f6-61,
t6-19
Smoker-
Present
(2)
f6-26
f6-26
f6-26
f6-26
f6-26
f6-26
f6-25
f6-25
f6-25
f6-25
f6-25
f6-25
f6-27
f6-27
f6-27
f6-27
f6-27
f6-27
f6-58
f6-58
f6-58
f6-58
f6-58
f6-58
f 6-63,
t6-21
f6-63,
16-21
Location x
Smoker-Present
(20)
f 6-28, 16-7
f 6-29, 1 6-7
f 6-31, 16-7
f 6-34, 1 6-7
f 6-37, 16-7
f 6-39, 1 6-7
16-10
t6-10
t6-10
16-10
t6-10
t6-10
f 6-28, 1 6-8, 9
f 6-30, 1 6-8,9
f 6-32,33, 1 6-8,9
f 6-35,36. 1 6-8,9
f 6-38, 1 6-8,9
f 6-40,41.1 6-8,9
Location x
Activity
(70)
f 6-42, 1 6-1 1
t6-15
16-15
te-15
16-15
te-is
f 6-43, 16-12
t6-16b
t6-16b
t6-16b
t6-16b
t6-16b
f 6-44,45, 16-13,14
t6-16
16-16
t6-16
16-16
te-16
(J\
• D = duration of microenvironments, 0 = frequency of occurrence of microenvironments, N = sample size of respondents experiencing a microenvironment (the doers) on the 24-hour
diary day or some other time unit (e.g., 3 hours).b Proportion of Time Spent and Proportion of Occurrences are usually reported in percentages or counts for each microenvironment.
Descriptive statistics are reported as means only. "The time unit used for each statistical calculation is an absolute fixed-interval time such as 24-hours, minute or 3-hour segment.
" Hyphens designate analyses over ajl respondents. ' Parentheses contain the number of categories in each microenvironment. Cells contain the figure (f) or table (t) in Section 6 where
the results of corresponding calculations are presented. Blank cells indicate no analyses were done for the corresponding breakdown. Note: This table contains only analyses of the
24-hour time diaries and no analyses of the follow-up variables. Sample proportions are over all respondents except for smoker-present calculations which are for doers only. All
calculations were weighted by time and background factor selection probabilities. Weighting is discussed in Section 4.
-------
1.4 Calculation Methodology
The 24-hour time unit (the diary day) can be used to calculate proportions of time spent and
frequency distributions of D and O by day-of-week and seasonal time factors (calculations #1 and #3 in
Table 1-1). But for calculations by time-of-day (calculation #'s 4, 5 & 6), we need to use a series of
much smaller fixed-length time units such as minutes or 3-hour segments. While D-calculations
(proportion of time spent and frequency distributions) require durations obtained from the beginning and
ending times of each microenvironment, O-calculations are based on the simple occurrence or non-
occurrence of a microenvironment in a specified time interval, and can be easily applied over these
smaller time units. But small time units destroy duration information within the time unit since the
beginning and ending times of some microenvironments extend beyond each absolute time boundary, and
when the length of the time unit is reduced to the finest resolution of the microenvironment's beginning
and ending times (one minute for NHAPS), duration information is completely destroyed. Thus, no D-
calculations could be done for time-of-day at minute resolution (calculation #4 in Table 1-1).
For 24-hour calculations the NHAPS respondents' 24-hour diaries were first sorted by
microenvironment (including their beginning times, ending times, and corresponding durations), and then
the doers were selected and grouped according to the time and background factors. D and O were
obtained by adding up all the time each doer spent in each microenvironment over the entire 24-hour
diary day and counting the number of times each doer entered each microenvironment over the entire 24-
hour diary day, respectively.
In contrast, when the time unit was less than 24-hours (for one minute and 3-hour time-of-day
analyses), the beginning and ending times in the multiple-record NHAPS respondents' 24-hour diaries
(grouped by time and background factors) were first converted into a constant number of records (1440
minutes or eight 3-hour segments) specifying the microenvironments occurring during each time unit.
After the records were sorted by time unit and microenvironment, O was obtained by counting the
number of microenvironments within each time and microenvironment grouping. Technically, D cannot
be obtained within time units that are smaller than the microenvironment durations, but for 3-hour
segment time units we assigned D to the 24-hour time unit durations of microenvironments that began
within each 3-hour time unit. For example, as we count all the respondents that are located inside their
home beginning in the 2nd 3-hour time segment (out of 8 total), the ending time of some of the episodes
may fall outside the 2nd 3-hour time unit boundary. Instead of truncating the episode so that it fits in this
3-hour time unit, we used the full duration — with untruncated ending times - regardless of the 3-hour
segment in which it terminates (for example, see Appendix B).
1-6
-------
1.5 Using the Results: Exposure Assessments
Total human exposure assessments are concerned with the confluence of a pollutant concentration at
a certain place (x, y, z coordinate) and time and a human being at the same place and time. In this way,
as chemical toxins come into contact with a person via air (lungs, skin), water (gut, skin), soil (gut, skin),
or food (gut) media (i.e., the inhalation, ingestion, or dermal pathways) at time t and spatial coordinates
x, y, and z, the person is said to be exposed. The results of the HAP calculations described above can be
used to determine the time (time-of-day, season, day-of- week) and spatial coordinates (location) for
different exposure events (exposure activities such as cooking or having a smoker present) and
subgroups. The concept of a microenvironment is used to describe all the exposure-relevant time and
space factors that are contained in the HAP data (see Appendix A).
In this report we present the time and space aspects of exposure in terms of: (1) which
microenvironments occupy the largest fraction of time over the entire sample; (2) the fraction of
respondents that is experiencing each microenvironment on the 24-hour diary day - the doers; (3) the
length of time (24-hour duration) that the doers are spending in the microenvironments; and (4) the
relative fractions of time, fractions of respondents, and durations across background (gender, age, region)
and time (time of day, day of week, season) factors. These quantities obtained purely from HAP data can
be used to estimate exposure across different subgroups (background and time factors); however, one
must assume that pollutant concentrations have approximately the same magnitude across the subgroups.
Accurate assessments of exposure for populations depend on accurate knowledge of the pollutant
concentrations and their movement to the lungs, gut, or skin of the population of human subjects
(obtained from measurement, air monitoring, or modeling). Ideally, HAP data for a population and
microenvironmental pollutant data for the same population are combined to give point estimates (using
mean concentrations and times spent in microenvironments) or frequency distributions of population
exposure (distributions of time spent and pollutant concentration are Monte-Carlo sampled). For dermal
and ingestion exposures, in addition to measurements of toxin concentrations, we must measure the
volume of material eaten or applied and the surface area for the person being exposed over the duration
of the microenvironment. Air exposure estimates require: (1) the pollutant concentrations (assumed to
be homogeneous) as determined from monitoring surveys or deterministic models that in turn require
pollutant source strengths and pollutant movement mechanisms (air exchange rates, removal rates, etc.);
and (2) inhalation rates.
Note: The conceptual methodology employed in this report is given in Appendix A. Detailed examples of the
calculation methodology for both the 24-hour and time-of-day analyses in this report - including SAS
computer code — are given in Appendix B.
Technical Note: The work presented in this report was completed using the SPSS {© SPSS Inc,) and SAS (© SAS
Inc.) statistical software packages. All graphics were done using SAS, Freelance Graphics for
Windows™ (©Lotus), and Microsoft Excel for Windows™ (©Microsoft).
Technical Disclaimer The data summaries and methodology in this report are not comprehensive, and are meant
as examples for researchers that will be conducting more specific (and thorough) analyses
of the NHAPS data for use in their own work. Sample weights and selected time and
background weights have been incorporated in reported statistics, but these may not be
appropriate for all uses of the data.
1-7
-------
Given HAP data only, the estimation of population exposures across different subgroups is probably
most accurate for the air-inhalation pathway since the amount of exposure is approximately proportional
to the duration of time spent in a microenvironment (assuming pollutant magnitudes and inhalation rates
are about the same for different demographic groups at different times). Conversely, dermal or ingestion
exposures can vary greatly from person-to-person for the same microenvironment duration.
In this report, we don't use the HAP results to assign explicit exposures (i.e., we are not conducting
an exposure assessment), but we do use the results to compare potential exposure (i.e., exposure that may
occur in a microenvironment) across subgroups for each kind of microenvironment (location, activity,
and smoker-present), which is probably most accurate for exposure to air pollutants. The "yes or no"
follow-up questions (in Section 7) are also used to assign potential exposures for different subgroups.
The percentage of time spent in microenvironments during the 24-hour diary day over the entire
population is used to assign the significance of a microenvironment to the population. Likewise, as more
people are experiencing a microenvironment during the 24-hour diary day, that microenvironment is
assigned more exposure significance for the population. Microenvironments may appear significant
based on the overall amount of time being spent, but not significant based on the number of people
experiencing them. In this case, the relatively small segment of the population experiencing them - the
doers — are spending large amounts of time (durations) in these microenvironments (see Table 1-2, Case
#2). Alternatively, microenvironments that do not appear significant based on the overall percentage of
time spent may be experienced by a large segment of the population. Here, the large number of doers are
not spending much time in these microenvironments (see Table 1-2, Case #3). When using amounts of
time spent in microenvironments or frequencies of occurrence of microenvironments to assign relative
exposures for the population as a whole and for the doers, it should be noted that small durations may
result in large exposures if the magnitude of exposure is large. Table 1-3 gives a summary of each
microenvironmental quantity (for HAP data and experimental data) and describes how they can be used
to estimate population exposures.
Table 1-2. How Different Percentages of Time Spent and Numbers of People Experiencing a
Microenvironment (Doers) Contribute to the Overall Significance of the Microenvironment With the
Corresponding Doer Durations
Case Percentage of Time Spent Number of People (Doers) Duration for Doers
1 Significant Significant Either Short or Long
2 Significant Not Significant Long
3 Not Significant Significant Short
4 Not Significant Not Significant Either Short or Long
1-8
-------
Table 1-3. How Each Microenvironmental Quantity (HAP Calculation or Experiment) Is Used to Estimate Exposure
Microenvironmental
Quantity
HAP Variable
Doers or
Non-Doers + Doers
Purpose/Comment
1 Percentage of Time
Spent
2 Percentage of
Respondents
3 Percentage of Time
Spent
4 Mean 24-Hour
Durations
5 Percentage of
Microenvironment
Occurrences
6 Mean 24-Hour
Frequency of
Occurrence
7 'Exposure Magnitude
(Means, Standard
Deviations,
Percentiles)
Duration, D
Doers + Non-Doers
Sample Size, N Doers + Non-Doers
Duration, D
Duration, D
Doers
Doers
Mlcroenviron. Doers + Non-Doers
Occurrences, O
Mlcroenviron.
Occurrences, O
from
Monitoring,
Measurement,
or Modeling
Doers
Doers
Assigns significance of microenvironments based on the number of people being
exposed and the length of the exposures; large percentages could result from long
exposure durations or from large numbers of people experiencing the exposures
Assigns significance of microenvironments based on the number of people being
exposed; confirms that large percentages of time spent are resulting from large numbers
of people experiencing a microenvironment and not a small number of people
experiencing long exposure durations; also used to analyze the movement of
respondents through microenvironments over the diary day in fixed time frames (1
minute, 3-hours, etc.)
Assigns significance of microenvironments based on the length of exposures of those
exposed; can be compared with the percentage of time spent for doers and non-doers
combined to estimate number of people experiencing each microenvironment as
provided by the percentage of respondents (#2)
Assigns significance of microenvironments based on the length of exposures of those
exposed
Assigns relative significance of microenvironments during a given time frame (1 minute,
3-hours, 24-hours, etc.); equal to the proportion of respondents at the time resolution of
the study
Indicates significance of microenvironments based on how often exposures occur during
the day for those that are exposed
Determines microenvironments that may pose a significant exposure risk based on
magnitudes of exposures, e.g., average pollutant concentrations; cannot assign
significance of exposure across a population since this requires population exposure
durations; can be combined with HAP studies to estimate exposure magnitudes across
populations
Note: Magnitudes of exposures are determined from factors "unknown" in human activity pattern (HAP) studies alone such as air exchange rates, source strengths,
or from actual monitoring studies. The significance of a microenvironment for a given population is defined by how much it poses a serious exposure risk for
that population. HAP studies can assign significance based on the time spent in microenvironments as determined from: the duration of microenvironments
D, the number of respondents in each microenvironment N, and the number of times the microenvironment occurs O. A microenvironment may not be
significant for the population as a whole but can pose a very serious exposure risk for the members of the population that experience it (small numbers of
people with large durations). Alternatively, microenvironments that appear to be significant for the whole population may be experienced by large numbers of
respondents, but they may not have very large durations. * = Not obtained from HAP studies.
-------
Section 2
Summary of the Data-Collection Methodology
The NHAPS study was designed to be used in the assessment of personal exposure to pollutants in
air and water systems with which people in the United States come into contact throughout their typical
daily routine. The complete data-collection methodology (including example questionnaires) is
presented in another work.8 Carried out between October 1992 and September 1994, NHAPS is an
extensive data resource, containing geographic (EPA region, U.S. census region, state, zip code, etc.)
socioeconomic (gender, age, race, education, etc.) and time/season (quarter, month, day of week, etc.)
information on 9,386 different respondents (i.e., respondents were never re-interviewed in the study)
distributed over 8 seasonal quarters throughout the entire nation (see Table 2-1 for a summary of the
NHAPS features).
Table 2-1. Summary of the National Human Activity Pattern Survey (NHAPS) Features
Dates of Data Collection
Data Collection Instrument
Data Types
Number of Total Respondents
Response Rate
Cooperation Rate
Geographic Coverage of
Respondents
Socio-economic Coverage of
Respondents
Emphasis of 175 Follow-up
Questions
24-Hour Diary Location Categories
24-Hour Diary Activity Categories
October 1992 through September 1994 (8 three-month seasonal quarters)
Telephone interviews using a Random-Digit Dial (ROD) method and
Computer Assisted Telephone Interviewing (CATI)
(1) 24-hour diaries with beginning and ending times at minute resolution, (2)
demographic questions, and (3) 175 follow-up questions on medical
background, housing characteristics, and exposure to chemicals
9,386
63% (65% for last seven quarters; first quarter was 46% from difficulties in
procedure and training schedules)
Excluding those respondents not contacted (or because they were not
interviewed for other factors) the cooperation rate was over 75%
The 48 contiguous United States, i.e., excluding Alaska and Hawaii, by state,
EPA region, U.S. census region, telephone area code, working postal zip
code, and residential postal zip code
Ages 0 to 93, gender, race, education, employment status, etc.
Personal exposure to contaminants in air and water from household sources
82 categories arranged by Own House, Friend's/Other's House, Traveling,
Other Indoor, and Other Outdoor
91 categories arranged by Non-Free Time (Paid Work, Household Work,
Child Care, Obtaining Goods/Services, Personal Care) and Free Time
(Educational, Organizational, Social, Recreational, Communications)
2-1
-------
Table 2-1. (Continued)
Diary Variables for each
Microenvironment
Total Diary Records
(Microenvironments)
Number of Different
Microenvironments Per Person
date (month, day and year), starting time, ending time, elapsed time
(duration), presence of a smoker, heavy breathing activity (yes or no),
location category, and activity category
157,234
Range = 1 to 59; Mean = 17
Detailed minute-by-minute 24-hour diaries were collected for each respondent containing 82
different possible locations (Residence-Kitchen, Residence-Garage, Office, School, Bar-Restaurant,
Automobile, etc.), 91 different activities (Cleaning, Food Preparation, Bathing, etc.), and whether a
smoker was ever present or not. Additionally, respondents were asked some fraction of 175 exposure-
related follow-up questions (focused on air and water pathways) on specific pollutant sources (paint,
glue, etc.) or prolonged background activities (gas heaters, wood smoke, etc.).
The interviews began with random selection of a respondent from the selected household. Saturdays
and Sundays were over-sampled to insure an adequate weekend sample size. The interviews lasted an
average of 23 minutes with most beginning between 6:00 and 9:00pm (Figure 2-1 and Table 2-2).
3000
2000-
1000-
Std. Dev. = 24.14
Mean = 23.2
N = 8490.00
Amount of Time for Interview
Figure 2-1. Histogram of the number of minutes that each interview lasted.
2-2
-------
Table 2-2. Number of NHAPS Respondents at the Time the Interview Began
TIME
Code Frequency
MIDNIGHT
8:00 AM
9:00 AM
10:00 AM
11:00 AM
NOON
1:00 PM
2:00 PM
3:00 PM
0
8
9
10
11
12
13
14
15
2
1
67
215
308
417
452
503
565
0
0
0.7
2.3
3.3
4.4
4.8
5.4
6
TIME
4:00 PM
5:00 PM
6:00 PM
7:00 PM
8:00 PM
9:00 PM
10:OOPM
11:OOPM
Missing
TOTAL
Code
16
17
18
19
20
21
22
23
Frequency
598
914
1,329
1,347
1,302
1,002
305
46
13
9,386
%
6.4
9.7
14.2
14.4
13.9
10.7
3.2
0.5
0.1
100
Note: The variable used to produce the frequency distribution is in the upper left hand comer of the table.
If the respondent chosen was a child too young to provide responses, an adult (18 and older) in the
household gave a proxy interview (Table 2-3). Since either an adult or a child could have been chosen
from the household, and their probability of selection depends on the number of adults/children in the
household and the criteria used for selection, a compensating weight was created. This weight was
combined with the probability of household selection based on the number of non-business phones in
each household (Table 2-4) to give the weighting variable WEIGHT. In this report WEIGHT was used to
calculate a joint frequency table across various subgroups (age, gender, day of week, season), which was
then used to create an overall weighting variable WEIGHT4 that improved the representativeness of the
NHAPS sample with respect to these subgroups using 1990 Census data (see Sections 3 and 4).
Table 2-3. Number of NHAPS
Respondents by Type of
Interview
TYPE
ADULT
CHILD
PROXY
PROXY
TOTAL
Code
1
2
3
4
Frequency
7,514
837
338
697
9,386
%
80.1
8.9
3.6
7.4
100
Table 2-4. The Number of NHAPS Respondents
by the Number of Non-Business
Phones in Each Household
PHONES
Missing
TOTAL
Code
1
2
3
4
5
6
9
Frequency
8,394
840
83
8
4
6
51
9,386
%
89.4
8.9
0.9
0.1
0
0.1
0.5
100
Note: The variable used to produce the frequency distribution is in the upper left hand comer of each table.
2-3
-------
Each of the 9,386 persons interviewed were asked to recount their entire daily routine from midnight
to midnight on the day preceding the day that they were interviewed. The beginning and ending times of
each microenvironment in these diaries were recorded with a time resolution of one minute in the "diary"
data file (Table 2-5). Together, the set of beginning and ending times for each microenvironment
spanning one day comprise a comprehensive, sequential account of a person's locations, activities, and
proximity to smokers (for example, see Table B-l in Appendix B). The qualification of presence-of-a-
smoker was expanded in the last two quarters of the study (April 1994 to September 1994) from the
simple "smoker-present" category to the fraction of time a smoker was present: "the entire time"; "more
than half the time"; "half of the time"; "less than half the time"; or "hardly any of the time".
Table 2-5. Description of the two NHAPS Data Files
File Name Description
MAIN.XXX Main file containing household and identification information (respondent ID #, length of
455 Variables interview, etc.), demographic background information (region, gender, age, etc.), temporal
9,386 records information (day of week, month, year, etc.), responses to the 175 exposure follow-up
questions, and frequencies in each location and activity diary category for all 9,386
respondents.
DIARY.XXX Diary file containing multiple-record 24-hour diaries for each of the 9,386 respondents. The
16 variables variable set of records correspond to all the different microenvironments each respondent
157,234 records visited and include the location, activity, and smoker-present codes, and the beginning and
ending times for each microenvironment
Note: Each file has a variable containing a unique identification code for each respondent that was used to link
the information in the two files. The files were supplied in SPSS-format (XXX = SYS) but were converted
into SAS-format (XXX = SSD) for most of the analyses.
The NHAPS study did not collect any minute-by-minute diary data on the respondents' proximity to
specific pollutant sources (besides a smoker). For example, there are gaps in the source type such as
cleaning agent, pesticide, solvent, or stove, and in their method of use. Minute-by-minute categories for
housing characteristics such as windows open or heat on, and types of exercise such as running or hiking
are also missing.
After the 24-hour diaries were collected each respondent was asked follow-up questions on personal
exposure, household characteristics, and medical background, which were stored in the "main" data file
(Table 2-5). Each of the follow-up questions were placed on either an A questionnaire or a B question-
naire, or both. A nationally representative number of respondents were given each questionnaire (4,723
for A only and 4663 for B only). The main file also contains all the demographic, respondent
identification, and time variables (respondent ID #, interview date, time interview began, duration of
interview, number of phones in household, type of interview, age, race, etc.). In addition to "yes or no"
questions on specific exposure issues (do you use a humidifier?, does your house have a basement?, etc.),
many of the follow-up questions concerned frequencies of occurrence and durations of exposure events
(how many cigarettes do you smoke?, for how many minutes did you take a shower?, etc.). Table 7-2 in
Section 7 lists each variable and its descriptive label.
2-4
-------
Section 3
Geographic, Socioeconomic, and Temporal Divisions of the Respondents
To conduct exposure assessments for different populations in the U.S., the NHAPS data must be
broken down into specific socioeconomic and geographic subgroups. For example, scientists may wish
to determine the effects of tobacco smoke in the home on females in the state of Pennsylvania. Before
the NHAPS data can be used for such a specific subgroup, there must first be enough NHAPS
respondents for the statistical analysis. Researchers should examine the NHAPS data to determine
whether the subgroups they are interested in: (1) have adequate sample size; and (2) occur in
representative proportions when compared to the population. Here, the NHAPS sample sizes are
compared to the 1990 U.S. Census30 for EPA region, U.S. census region, gender, and age to determine
their representativeness. The proportions of respondents interviewed on each quarter and day of the
week are also compared to their desired values (1/7 for each day of the week and 1/8 for each seasonal
quarter). The gender, age, day of week, and season subgroups were used in Section 4 to develop
appropriate weighting factors based on the 1990 U.S. Census.
3.1 Geographic Divisions
The NHAPS geographic variables are zip code, state, area code, EPA region, and U.S. census region
(Table 3-1). Although the zip codes variables may be useful for the study of subpopulations in several
densely-populated urban regions, the usefulness of zip codes may be limited because: (1) more than 50%
of the work zip codes (ZIPWORK) are missing; and (2) the number of respondents in most zip codes may
not be adequate to conduct a complete analysis. Divisions by area code will probably be more
appropriate in the study of a larger number of metropolitan areas.
Table 3-1. The NHAPS Geographic Variables for Each Respondent
•Variable Name
AREA
EREGION
CREGION
RACD
ZIPWORK
ZIPHOME
Variable Label
State
EPA Region
Census Region
Telephone Area Code
U.S. Postal Zip code of Main Workplace
U.S. Postal Zip code of Home
Example
Alabama, Arizona, ...
1 10
1,2,3,4
702,415,617,301,...
•57.8% of the ZIPWORK values are missing; 6.3% of the ZIPHOME values are missing.
A relative comparison of NHAPS and Census proportions is defined as the ratio of the percentages
in each state of the Census data to NHAPS. A comparison of the number of NHAPS respondents in each
state vs. the 1990 U.S. Census (Table 3-2) shows generally good agreement with the relative comparisons
of most states close to 1 (see Figures 3-1 to 3-3). The only state that was oversampled in NHAPS
3-1
-------
Table 3-2. List of States Grouped Alphabetically Within Each EPA Region, Ranked by the 1990 U.S. Census Population with
Comparison to the Number of NHAPS Respondents That Were Interviewed
EPA
Region
1
1
1
1
1
1
2
2
3
3
3
3
3
3
4
4
4
4
4
4
4
4
5
5
5
5
5
5
6
6
6
6
6
7
7
7
7
8
8
8
8
8
8
9
9
9
9
10
10
10
10
Census
Region
1
1
1
1
1
1
1
1
3
3
3
1
3
3
3
3
3
3
3
3
3
3
2
2
2
2
2
2
3
3
4
3
3
2
2
2
2
4
4
2
2
4
4
4
4
4
4
4
4
4
4
Rank
27
38
13
40
43
49
9
2
46
48
19
5
12
34
22
4
11
23
31
10
25
17
6
14
8
20
7
16
33
21
37
28
3
30
32
15
36
26
44
47
45
35
51
24
1
41
39
50
42
29
18
State
Code
(STF1PS)
9
23
25
33
44
50
34
36
10
11
24
42
51
54
•f
12
13
21
28
37
45
47
17
18
26
27
39
55
5
22
35
40
48
19
20
29
31
8
30
38
46
49
56
4
6
15
32
2
16
41
53
State Name
CONNECTICUT
MAINE
MASSACHUSETTS
NEW HAMPSHIRE
RHODE ISLAND
VERMONT
NEW JERSEY
NEWYORK
DELAWARE
DIST. OF COLUMBIA
MARYLAND
PENNSYLVANIA
VIRGINIA
WEST VIRGINIA
ALABAMA
FLORIDA
GEORGIA
KENTUCKY
MISSISSIPPI
NORTH CAROLINA
SOUTH CAROLINA
TENNESSEE
ILLINOIS
INDIANA
MICHIGAN
MINNESOTA
OHIO
WISCONSIN
ARKANSAS
LOUISIANA
NEW MEXICO
OKLAHOMA
TEXAS
IOWA
KANSAS
MISSOURI
NEBRASKA
COLORADO
MONTANA
NORTH DAKOTA
SOUTH DAKOTA
UTAH
WYOMING
ARIZONA
CALIFORNIA
HAWAII
NEVADA
ALASKA
IDAHO
OREGON
WASHINGTON
State
Abbrev.
CT
ME
MA
NH
Rl
VT
NJ
NY
DE
DC
MD
PA
VA
WV
AL
FL
GA
KY
MS
NC
SC
TN
IL
IN
Ml
MN
OH
Wl
AR
LA
NM
OK
TX
IA
KS
MO
NE
CO
MT
ND
SD
UT
WY
AZ
CA
HI
NV
AK
ID
OR
WA
TOTAL
No.
NHAPS
Respond.
129
54
275
39
63
12
278
687
22
21
176
538
270
62
157
534
266
99
63
290
124
180
455
176
327
186
373
134
108
180
40
126
565
82
94
174
68
153
63
14
19
77
14
188
988
0
63
0
18
154
208
9,386
%
1.37
0.58
2.93
0.42
0.67
0.13
2.96
732.
0.23
0.22
1.88
5.73
2.88
0.66
1.67
5.69
2.83
1.05
0.67
3.09
1.32
1.92
4.85
1.88
3.48
1.98
3.97
1.43
1.15
1.92
0.43
1.34
6.02
0.87
1.00
1.85
0.72
1.63
0.67
0.15
0.20
0.82
0.15
2.00
10.53
0.00
0.67
0.00
0.19
1.64
2.22
100
1990 U.S.
Census
Population
3,287,116
1,227,928
6,016,425
1,109,252
1,003,464
562,758
7,730,188
17,990,455
666,168
606,900
4,781,468
11,881,643
6,187,358
1,793,477
4,040,587
12^37,926
6,478,216
3,685,296
2,573,216
6,628,637
3,486,703
4,877,185
11,430,602
5,544.159
9,295,297
4,375,099
10,847,115
4,891,769
2,350,725
4,219,973
1,515,069
3,145,585
16,986,510
2,776,755
2,477,574
5,117,073
1,578,385
3,294,394
799,065
638,800
696,004
1,722,850
453,588
3,665,228
29,760,021
1,108,229
1,201,833
550,043
1,006,749
2,842,321
4.866.692
248,709,873
%
1.32
0.49
2.42
0.45
0.40
0.23
3.11
7.23
0.27
0.24
1.92
4.78
2.49
0.72
1.62
5.20
2.60
1.48
1.03
2.67
1.40
1.96
4.60
2.23
3.74
1.76
4.36
1.97
0.95
1.70
0.61
1.26
6.83
1.12
1.00
2.06
0.63
1.32
0.32
0.26
028
0.69
0.18
1.47
11.97
0.45
0.48
022
0.40
1.14
1.96
100
%
Drff.
0.05
0.08
0.51
0.03
0.27
0.10
0.15
0.09
0.03
0.02
0.05
0.95
0.39
0.06
0.05
0.49
0.23
0.43
0.36
0.42
0.08
0.04
0.25
0.35
0.25
0.22
0.39
0.54
0.21
022
0.18
0.08
0.81
0.24
0.01
0.20
0.09
0.31
0.35
0.11
0.08
0.13
0.03
0.53
1.44
0.45
0.19
022
021
0.50
0.26
%
Ratio
0.96
0.86
0.83
1.07
0.60
1.77
1.05
0.99
1.14
1.09
1.03
0.83
0.86
1.09
0.97
0.91
0.92
1.40
1.54
0.86
1.06
1.02
0.95
1.19
1.07
0.89
1.10
1.38
0.82
0.88
1.43
0.94
1.13
1.28
0.99
1.11
0.88
0.81
0.48
1.72
1.38
0.84
1.22
0.74
1.14
0.00
0.72
0.00
2.11
0.70
0.88
Note: The last two columns (% Dffi and % Ratio) are the absolute difference between the NHAPS vs. Census percentages in each state (absolute
comparison) and the ratio of the Census percentage to the NHAPS percentage in each state (relative comparison), respectively.
3-2
-------
Fi»qu«icy Count
o
SK
too
SX
7W
300 ^M 400
800 000
Figure 3-1. Map of the U. S. showing the number of NHAPS respondents in each state grouped by
the ten labeled EPA Regions. There were no NHAPS respondents in either Alaska or
Hawaii.
PafSC 0*313
21000000
9000000
24000000
12000000
2700000C
15000000
30000000
Rgure 3-2. Map of the U.S. showing the population of each state as reported in the 1990 U.S.
Census and grouped into the ten labeled EPA Regions.
3-3
-------
12-
10-
8-
6-
u>
4-
2-
I
a
NHAP8
(N- 9,386)
1990 CENSUS
(Nn 248,709,873)
07
0.4 |o.4
°107
CT ME MA NH Rl VT
1
U NY
2
DE DC MO PA VA WV
3
AL FL GA KY MS NC SC TN
4
IN Ml MM OH Wl
5
R LA NM OK TX
6
IA KS MO NE
7
CO MT NO SO UT WY
8
State Grouped by EPA Region
Figure 3-3. Percentage comparison bar chart of the 9,386 NHAPS respondents vs. the 248,709,873 1990 Census participants broken down by state and EPA Region.
-------
(assuming the Census is an accurate measure of the number of inhabitants) with a relative comparison
under 0.5 was Montana. States that were undersampled at a relative comparison over 1.5 were Vermont,
Mississippi, North Dakota, and Idaho. No NHAPS respondents were interviewed in either Alaska or
Hawaii.
Because statistical analysis will require further subdivisions by demography and/or time, there
should probably be at least 100 people in each area code grouping. The 20 states (including Washington
D.C.) that do not have at least 100 NHAPS respondents are Delaware, the District of Columbia (D.C.),
Idaho, Iowa, Kansas, Kentucky, Maine, Mississippi, Montana, Nebraska, Nevada, New Hampshire, New
Mexico, North Dakota, Rhode Island, South Dakota, Utah, Vermont, West Virginia, and Wyoming (see
Table 3-2). The states with over 500 NHAPS respondents were California, Florida, New York,
Pennsylvania, and Texas.
The percentage of NHAPS respondents sampled in each of the 10 EPA Regions is comparable to the
population observed in the 1990 U.S. Census (Figure 3-4, Table 3-3) with all relative comparisons of the
Census vs. NHAPS clustered from 0.9 to 1.2. There is a sufficient NHAPS sample size in each EPA
Region to perform a detailed statistical analysis with a low of 340 NHAPS respondents in EPA Region 8.
Six of the other regions have between 900 and 1800 respondents, with the remaining three having
between 350 and 600 respondents.
8 9 10
Figure 3-4. Side-by-side percent comparison of the number of NHAPS respondents vs. the 1990 U.S.
Census in each EPA Region.
3-5
-------
Table 3-3. Comparison of 1990 U.S. Census Population with Number of NHAPS Respondents in Each EPA Region
EPAREG
1
2
3
4
5
6
7
8
9
10
EPA
Region Name
New England
North Atlantic
Mid Atlantic
South Atlantic
Midwest
South Central
Central
North Central
Pacific
Mountain
Composite States
CT, ME, MA, NH, Rl, VT
NJ.NY
PA, DE, DC, MD, VA, WV
AL, FL, GA, KY, MS, NC, SC, TN
IL, IN, Ml, MN, OH, Wl
AR, LA, OK, TX, NM
IA, KS, MO, NE
ND, SD, CO, MT, UT, WY
AZ, CA, HI, NV
AK, ID, OR, WA
TOTAL
No. NHAPS
Respondents
572
965
1,089
1,713
1,651
1,019
418
340
1,239
380
9,386
%
6.1
10.3
11.6
18.3
17.6
10.9
4.5
3.6
13.2
4.0
100
1990 U.S.
Census
Population
13,206,943
25,720,643
25,917,014
44,707,766
46,384,041
28,217,862
11,949,787
7,604,701
35,735,311
9,265,805
248,709,873
%
5.3
10.3
10.4
18.0
18.6
11.3
4.8
3.1
14.4
3.7
100
%
Ratio
1.2
1.0
1.1
1.0
0.9
1.0
0.9
1.2
0.9
1.1
Note: See Table 3-2 for STRIPS and state names. The NHAPS data do not include any respondents for Alaska or Hawaii. The last
column (% Ratio) is the ratio of the percentages in each state (relative comparison).
Each of the four U.S. Census regions (Table 3-4) has approximately 2000 to 3000 NHAPS respondents each.
The relative comparisons between the 1990 Census and NHAPS are all near 1.
Table 3-4. Comparison of 1990 U.S. Census Population with Number of NHAPS Respondents in Each Census Region
CENSREG
1
2
3
4
Census
Region Name
Northeast
Midwest
South
West
States
CT, ME, MA, NH, Rl, VT, NJ, NY,
PA
IL, IN, Ml, MN, OH, Wl, IA, KS,
MO, NE, ND, SD
DE, DC, MD, VA, WV, AL, FL,
GA, KY, MS, NC, SC, TN, AR, LA,
OK.TX
NM, CO, MT, UT, WY, AZ, CA,
HI, NV, AK, ID, OR, WA
TOTAL
No. NHAPS
Respondents
2,075
2,102
3,243
1,966
9,386
%
22.1
22.4
34.6
20.9
100
1990 U.S.
Census
Population
50,809,229
59,668,632
85,445,930
52,786,082
248,709,873
%
20.4
24.0
34.4
21.2
100
%
Ratio
1.1
0.9
1.0
1.0
Note: See Table 3-2 for STFIPS and state names. The NHAPS data do not include any respondents for Alaska or
Hawaii. The last column (% Ratio) is the ratio of the percentages in each state (relative comparison).
3-6
-------
The NHAPS sample sizes by area code within each state are listed in Table 3-5. The 37 area code
subregions within each state that have more than 100 NHAPS respondents are: (1) Alabama; (2) Arizona;
(3) Arkansas; (4) the Los Angeles, California area; (5) the San Francisco, California Bay area; (6) eastern
California; (7) northeastern California; (8) northwest California; (9) the northwest section of Colorado;
(10) Connecticut; (11-14) all four area codes in Florida; (15) the Atlanta, Georgia area; (16) north
central, Illinois; (17) the Baltimore, Maryland area; (18) the Boston, Massachusetts area; (19) central
Massachusetts; (20) the Detroit, Michigan area; (21) central Minnesota; (22) the Newark, New Jersey
area; (23) central New Jersey; (24) the New York City area - 212, 516, 718,914; (25) western North
Carolina; (26) eastern North Carolina; (27) northeastern Ohio; (28) Oregon; (29) the Philadelphia,
Pennsylvania area; (30) east central Pennsylvania; (31) the Pittsburgh, Pennsylvania area; (32) south
Carolina; (33) eastern Tennessee; (34) south central Texas; (35) eastern Virginia; (36) western Virginia;
and (37) western Washington.
The 20 most-populated cities according to the 1990 U.S. Census are listed in Table 3-6 with the
surrounding metropolitan NHAPS sample size. In future analyses of the NHAPS data, it may be useful
to study these metropolitan regions. Notice that the 20 most populated cities in 1990 tend to fall into the
most populated states (see Table 3-2).
3.2 Socioeconomic Divisions
The major NHAPS socioeconomic divisions are gender, age, race, education, employment status,
and number of children and adults in the household (Table 3-7). Other variables that relate to housing
characteristics or living options such as the number of windows kept open or when the respondent moved
into their house are summarized in Section 7 (the NHAPS follow-up questions).
Joint age and gender frequencies in the 1990 U.S. Census for two different age groupings are given
in Figure 3-5. Notice how males occur in higher numbers under age 25, whereas females occur in higher
numbers for age 35 and up. The age group with the largest discrepancy between numbers of males and
females is the 75 and up group where the number of males is 54% of the number of females.
Fifty-four percent of the NHAPS respondents were female as compared to 51% for the 1990 U.S.
Census (Table 3-8). The largest percentages of the NHAPS respondents fell in the 25 to 34 (17%) and
35 to 44 (16%) age groups (Table 3-9). About 28% of the respondents were under age 25 and 37% were
age 45 and older. For the groupings shown, the largest discrepancy between NHAPS and the Census
occurred for the 5 to 17 age group (14% vs. 18%).
Over 80% of the NHAPS respondents were white with 10% black (Table 3-10). About 4% were
Hispanic and less than 2% were Asian. The remainder were either some other race or refused to answer.
Ninety-one percent of respondents said they were not of Hispanic origin (Table 3-11). About 28% of the
NHAPS respondents graduated from high school only, with 13% finishing college and 10% with some
kind of post-graduate work. For 21% of the respondents no information (either missing or "refused to
answer") is available on level of education (Table 3-12). Forty-four percent of the NHAPS respondents
were employed full time with 8.5% employed part time (Table 3-13). About 28% of the respondents said
they were not employed and a substantial number of respondents have no employment data available
(20%). The majority of the NHAPS respondents' households had two adults (56%) with one adult as the
next largest percentage (26%). Above four adults (4%) the number of respondents trailed off rapidly
(Table 3-14). Sixty-one percent of the respondents' households had no children under age 18 (Table 3-
15).
3-7
-------
Table 3-5. The Number of NHAPS Respondents in Each Area Code Arranged by State
No.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
STFIPS
1
2
4
5
6
8
9
10
11
12
13
15
16
17
18
19
Census
Region
3
4
4
3
4
4
1
3
3
3
3
4
4
2
2
2
EPA
Region
4
10
9
6
9
8
1
3
3
4
4
9
10
5
5
7
State
Abbrev.
AL
AK
AZ
AR
CA
CO
CT
DE
DC
FL
GA
HI
ID
IL
IN
IA
State
Alabama
Alaska
Arizona
Arkansas
California
California
California
California
California
California
California
California
California
California
California
California
California
Colorado
Colorado
Connecticut
Delaware
D.C.
Florida
Florida
Florida
Florida
Georgia
Georgia
Georgia
Georgia
Hawaii
Idaho
Illinois
Illinois
IBinois
IBinois
IHinois
Illinois
Indiana
Indiana
Indiana
Iowa
Iowa
Area Code
Sub Region
All
All
All
All
Fresno Area
Los Angeles
Santa Monica Area
South Bay Area
West Bay Area
East Bay Area
Eastern
Northwestern
Santa Ana Area
San Luis Obispo Area
San Fernando Area
Southeastern
Northeastern
Northwest
Southeast
All
All
All
Miami Area
Eastern
Southwestern
Northern
Atlanta Area
Northern
Southern
All
AH
Springfield Area
Peoria Area
Chicago Area
Southern
North Central
Rockford Area
Northern
Central
Southern
Eastern
Central
Area
Code
205
907
602
501
209
213
310
408
415
510
619
707
714
805
818
909
916
303
719
203
302
202
305
407
813
904
212
404
706
912
808
208
217
309
312
618
708
815
219
317
812
319
515
Area Code
Frequency
157
0
188
108
92
70
148
38
58
50
105
44
77
61
76
49
120
115
38
129
22
21
127
128
150
129
1
130
86
49
0
18
91
21
78
38
165
62
61
80
35
33
23
State
Frequency
157
0
188
108
988
153
129
22
21
534
266
0
18
455
176
82
State
%
1.7
0
2
1.2
10.5
1.6
1.4
0.2
0.2
5.7
2.8
0
0.2
4.8
1.9
0.9
3-8
-------
Table 3-5. (Continued)
No.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
STRIPS
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Census
Region
2
3
3
1
3
1
2
2
3
2
4
2
4
1
1
4
1
3
EPA
Region
7
4
6
1
3
1
5
5
4
7
8
7
9
1
2
6
2
4
State
Abbrev.
KS
KY
LA
ME
MD
MA
Ml
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
State
Iowa
Kansas
Kansas
Kentucky
Kentucky
Louisiana
Louisiana
Maine
Maryland
Maryland
Massachusetts
Massachusetts
Massachusetts
Michigan
Michigan
Michigan
Michigan
Minnesota
Minnesota
Minnesota
Mississippi
Missouri
Missouri
Missouri
Montana
Nebraska
Nebraska
Nevada
New Hampshire
New Jersey
New Jersey
New Jersey
New Mexico
New York
New York
New York
New York
New York
New York
New York
New York
North Carolina
North Carolina
North Carolina
Area Code
Sub Region
Western
Southern
Northern
Western
Eastern
Western
Eastern
All
Western
Baltimore Area
Western
Central
Boston Area
Detroit Area
Central
Western
Northeastern
Northern
Southern
Central
All
Eastern
Southwestern
Northwestern
All
Western
Eastern
All
All
Newark Area
Trenton Area
Central
All
Manhattan Area
North Central
Hempstead Area
Eastern
South Central
Western
Bronx, Brook, Queens
White Plains Area
Western
Central
Eastern
Area
Code
712
316
913
502
606
318
504
207
301
410
413
508
617
313
517
616
810
218
507
612
601
314
417
816
406
308
402
702
603
201
609
908
505
212
315
516
518
607
716
718
914
704
910
919
Area Code
Frequency
26
34
60
54
45
91
89
54
62
114
32
137
106
184
46
82
15
6
36
144
63
73
38
63
63
30
38
63
39
101
54
123
40
64
78
111
87
43
90
158
56
100
30
160
State
Frequency
94
99
180
54
176
275
327
186
63
174
63
68
63
39
278
40
687
290
State
%
1
1.1
1.9
0.6
1.9
2.9
3.5
2
0.7
1.9
0.7
0.7
0.7
0.4
3
0.4
7.3
3.1
3-9
-------
Table 3-5. (Continued)
No.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
STF1PS
38
39
40
41
42
44
45
46
47
48
49
50
51
53
54
55
56
Census
Region
2
2
3
4
1
1
3
2
3
3
4
1
3
4
3
2
4
EPA
Region
8
5
6
10
3
1
4
8
4
6
8
1
3
10
3
5
8
State
Abbrev.
NO
OH
OK
OR
PA
Rl
SC
SD
TN
TX
UT
VT
VA
WA
WV
Wl
WY
State
North Dakota
Ohio
Ohio
Ohio
Ohio
Oklahoma
Oklahoma
Oregon
Pennsylvania
Pennsylvania
Pennsylvania
Pennsylvania
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Tennessee
Texas
Texas
Texas
Texas
Texas
Texas
Texas
Texas
Texas
Utah
Vermont
Virginia
Virginia
Washington
Washington
West Virginia
Wisconsin
Wisconsin
Wisconsin
Wyoming
Area Code
Sub Region
All
Northeast
Northwest
Southwest
Southeast
Western
Eastern
All
Philadelphia Area
Pittsburgh Area
Reading Area
East Central
West Central
All
All
All
East
West
Southwestern
Dallas Area
Southeastern
South Central
Houston Area
Northern
North Central
Northeastern
Western
AN
AH
Eastern
Western
Western
Eastern
All
Eastern
Southwestern
Northern
All
Area
Code
701
216
419
513
614
405
918
503
215
412
610
717
814
401
803
605
615
901
210
214
409
512
713
806
817
903
915
801
802
703
804
206
509
304
414
608
715
307
Area Code
Frequency
14
154
50
95
74
62
64
154
163
111
46
142
76
63
124
19
157
23
62
72
62
103
91
38
70
33
34
77
12
153
117
166
42
62
86
22
26
14
TOTAL
State
Frequency
14
373
126
154
538
63
124
19
180
565
77
12
270
208
62
134
14
9,386
State
%
0.1
4
1.3
1.6
5.7
0.7
1.3
0.2
1.9
6
0.8
0.1
2.9
2.2
0.7
1.4
0.1
100
3-10
-------
Table 3-6. The 20 Most Populated U.S. Cities in 1990 with the Surrounding NHAPS Sample Size
Rank
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
City
New York, NY
Los Angeles, CA
Chicago, IL
Houston, TX
Philadelphia, PA
San Diego, CA
Detroit, Ml
Dallas, TX
Phoenix, AZ
San Antonio, TX
San Jose, CA
Indianapolis, IN
Baltimore, MD
San Francisco, CA
Jacksonville, FL
Columbus, OH
Milwaukee, Wl
Memphis, TN
Washington, DC
Boston, MA
Area Code(s)
212,718
213
312
713
215
619
313
214
602
210
408
317
410
415
904
614
414
901
202
617
1990 U.S. Census
Population
7,322,564
3,485,398
2,783,726
1,630,553
1,585,577
1,110,549
1,027,974
1,006,877
983,403
935,933
782,248
741,952
736,014
723,959
672,971
632,910
628,088
610,337
606,900
574,283
Surrounding NHAPS
Sample Size
222
70
78
91
163
105
184
72
188
62
38
80
114
58
129
74
86
23
21
106
Note: Area codes in less populated regions have a larger coverage so the NHAPS sample sizes are larger than
they would be for just the immediate metropolitan region.
Table 3-7. The NHAPS Demographic Variables for Each Respondent
Variable
Label
Example
ADULT # of Adults in Household
AGEKID* Age of Youngest Child
EDUC* Res.'s Last Grade or Year of School
EMP* Employment Status of Respondent
HISP Is Res. of Hispanic Origin-Descent
KID1 * # Children Under 18 in Household
KIDS Any Children Under 18 Living in HH
KREL* Relationship to Child
KSEX Gender of Child with Proxy Interviews
RACE Race or Ethnic Group of Respondent
RECAGE Age of Respondent
REDUC* Education Level of Respondent
RSEX Respondent Sex
WORKPLCE* Main Work Place
WOUT* # of Hours per Week Worked Outdoors
WKHRS* # of Hours Worked for Pay Last Week
WNGHT* # of Hours Worked Between 6 PM and 6 AM
UNEMP* Reason for Not Employed
YOB Respondent Year of Birth
1,2,3.... 12, ...18,...
Yes/No
Yes/No
White, Black, Asian,...
High School, College, Adv. Deg.,,
Male/Female
Note: Demographic variables denoted by an asterisk (*) have a substantial number of missing or 'refused to
answer* values. Additional variables that may have some demographic interest are described in Section 7
(the NHAPS follow-up questions).
3-11
-------
Very Young
(0-4)
School Age
(5-17)
Working
(18-64)
Retired
(65&up)
Age Groups
0-4 5-11 12-17 18-24 25-34 35-44 45-54 55-64 65-74 75&UD
Age Groups
Figure 3-5. Plots of the number of male and female Americans in each of four age groups (top) and ten age
groups (bottom) obtained from the 1990 U.S. Census.
3-12
-------
Note: In the following tables the NHAPS variable is given in the upper left-hand corner of the table.
JTable 3-8. Number of NHAPS Respondents by Gender
RSEX
Males
Females
Refused
TOTAL
Code
1
2
9
No. NHAPS
Respondents
4,294
5,088
4
9,386
%
45.75
54.21
0
0.0004
1990
U.S. Census
121,239,418
127,470,455
248,709,873
%
48.75
51.25
100
Table 3-9. Number of NHAPS Respondents by two Different Age Groupings
RECAGE
Under 5
5 TO 11
12 TO 17
18 TO 24
25 TO 34
35 TO 44
45 TO 54
55 TO 64
65 TO 74
75 and Older
Missing
TOTAL
Codel
1
2
3
4
5
6
7
8
9
10
99
Code 2
1
2
-
3
-
-
-
-
4
-
No. NHAPS
Respondents
499
703
589
799
1,622
1,489
1,187
962
816
533
187
9,386
%
1
5.3
7.5
6.3
8.5
17.3
15.9
12.6
10.2
8.7
5.7
2.0
% 1990
2 U.S. Census
5.3 18,354,443
13.8 25,207,871
20,042,118
64.7 26,737,766
43,175,932
37,578,903
25,223,086
21,147,923
14.4 18,106,558
13,135,273
248,709,873
%
1
7.4
10.1
8.1
10.8
17.4
15.1
10.1
8.5
7.3
5.2
%
2
7.4
18.2
61.9
12.5
Table 3-10. Number of
RACE
White
Black
Asian
Some Other Race
Hispanic
Refused
TOTAL
NHAPS
Code
1
2
3
4
5
9
Respondents by
No. NHAPS
Respondents
7,591
945
157
182
385
126
9,386
Race
%
80.9
10.1
1.7
1.9
4.1
1.3
100
Table 3-11. Of Hispanic Origin?
Table 3-12. Number of NHAPS Respondents by
Education Level
REDUC
Less than High School
High School Grad
Less than College
College Grad
Post Grad
No Data
Missing
TOTAL
Code
1
2
3
4
5
-
9
Frequency
834
2,612
1,801
1,247
924
1,872
96
9,386
%
8.9
27.8
19.2
13.3
9.8
19.9
1
100
HISP
No
Yes
DK
REF
TOTAL
Value
0
1
8
9
Frequency
8,534
702
47
103
9,386
%
90.9
7.5
.5
1.1
100.0
Table 3-13. Number of NHAPS Respondents by
Employment Status
EMP
Full time
Part time
Not Employed
Refused
Missing
TOTAL
Code
1
2
3
9
Frequency
4,096
802
2,644
71
1,773
9,386
%
43.6
8.5
28.2
0.8
18.9
100
3-13
-------
Table 3-14. Number of NHAPS Respondents by
Number of Adults in Household
ADULTS
1
2
3
4
5
6
7
8
10
More than 10
Refused
TOTAL
Code
1
2
3
4
5
6
7
8
10
11
99
Frequency
2,479
5,296
1,057
406
71
13
5
1
2
1
55
9,386
%
26.4
56.4
11.3
4.3
0.8
0.1
0.1
0
0
0
0.6
100
Table 3-15. Kids Under 18?
KIDS
Value Frequency %
No
Yes
Refused
0
1
9
5,691
3,566
129
60.6
38.0
1.4
TOTAL
9,386 100
33 Temporal Divisions
Variables related to the date of the interview (DATE, QUARTER, WEND) give information for day of
week, day of month, month, year and seasonal quarter (Table 3-16). The number of respondents in each quarter
of the NHAPS study were fairly uniform (approximately 13%) except for the first quarter when only 7.8% of
the respondents were interviewed (Table 3-17). The proportion of respondents interviewed during each season
(winter, spring, summer, fall) ranged from 20 to 27% (Table 3-18). Most of the respondents were interviewed
on a weekday (67%). This value is somewhat smaller than the ideal figure (5/7 = 71%) since the weekends
were intentionally oversampled (Tables 3-20 and 3-21).
Table 3-16. List of the NHAPS Temporal Variables
Variable
Date
Quarter
Wend
Season*
Day
Year
Label
Date of Interview
Study Quarter
Day of Sample
Season of Sample
Day of Week
Year of Sample
Example
010192 (January 1, 1992)
1,2,3,4,5,6,7,8
Weekday, Weekend
Win, Spr., Sum., Fall
M,T,W,Th,Fr,Sat,Sun
1992-1994
* The data were receded from the original NHAPS variables.
Note: In the following tables the % Desired is the desired proportion and % Ratio is the ratio of the desired
proportion with the sample proportion.
Table 3-17. Number of NHAPS Respondents in Each
Seasonal Quarter
Quarter
9/92 to 12/92
1/93 to 3/93
4/93 to 6/93
7/93 to 9/93
10/93 to 12/93
1/94 to 3/94
4/94 to 6/94
7/94 to 10/94
TOTAL
Code
1
2
3
4
5
6
7
8
Frequency
731
1,288
1,228
1,256
1,157
1,236
1,210
1.280
9,386
%
7.8
13.7
13.1
13.4
12.3
13.2
12.9
13.6
100
Desired
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
Ratio
1.6
0.91
0.95
0.93
1.02
0.95
0.97
0.92
Table 3-18. Number of NHAPS Respondents in Each
Season
SEASON
Winter
Spring
Summer
Fall
TOTAL
Code
1
2
3
4
Frequency % Desired
2,524
2,438
2,536
1,888
26.9
26.0
27.0
20.1
100
25
25
25
25
Ratio
0.93
0.96
0.93
1.24
3-14
-------
Table 3-19. Number of NHAPS Respondents by Year of Table 3-20. Number of NHAPS Respondents by
Interview Weekend vs. Weekday
% % % %
YEAR Code Frequency % Desired Ratio WEND Code Frequency % Desired Ratio
Weekday 1 6,316 67.3 71.43 1.06
Weekend 2 3,070 32.7 28.57 0.87
TOTAL ~
Missing
TOTAL
92
93
94
-
356
2,475
1,885
7
4,723
7.5
52.4
39.9
.1
100.0
12.5
50.0
37.5
1.66
0.95
0.94
Table 3-21. Number of NHAPS Respondents by Dav-of-Week
DAY Code Frequency % Desired Ratio
Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sundav
1
2
3
4
5
6
7
1,457
1,457
1,391
1,179
832
1,267
1,803
15.5
15.5
14.8
12.6
8.9
13.5
19.2
14.3
14.3
14.3
14.3
14.3
14.3
14.3
0.92
0.92
0.97
1.13
1.61
1.06
0.74
TOTAL 9,386 100
3-15
-------
Section 4
Weighting of the Data
As shown in the previous section, the NHAPS data are not ideally representative of the larger U.S.
population in terms of geographic, socioeconomic, and temporal variables. The data have fairly similar
proportions as the 1990 U.S. Census for states, EPA regions, and Census regions. But since they do not
adequately match the Census for age, gender, day of week, or season, we must derive weighting factors
for these subdivisions. Other divisions of the data such as employment status, educational status, race,
etc., were not considered because of one or more of the following: (1) proportions of these subgroups
vary greatly across other geographic and socioeconomic variables, and they can be considered by future
investigators for specific divisions of the country; (2) responses to status-related questions may not be
very reliable when obtained from telephone-interview surveys; (3) a substantial number of values (>20%)
for education and employment status variables were missing or "refused to answer"; and (4) it is simpler
to obtain a two-way frequency distribution across age and gender only, rather than for three or more
socio-economic subdivisions. Only age and gender were obtained from the Census since day-of-week
and season proportions are absolute quantities (e.g., 1/4 for each season) that are easily calculated for the
NHAPS study and that are constant across the Census. Although the U.S. 1990 Census is the most recent
nationwide survey of population characteristics that is complete and readily available, other researchers
may choose to weight the NHAPS data with more up-to-date population surveys available for
subdivisions of the population of interest.
Each subdivision of the population (e.g., the Census) is called a subpopulation, and each subdivision
of the sample data (e.g., NHAPS) is called a subgroup. When weighting data for subgroups that appear
to misrepresent the "true" population, one must consider the following eight points: (1) weighting can be
used to compensate for oversampling in different regions or time periods, e.g., day of the week; (2) the
data must have a sufficient sample size in each subgroup to weight their proportions so the overall
weighted statistics (e.g., mean) give a good estimate; (3) in general, an n-way population proportion (the
joint proportion) should be applied to define the weight for each subgroup when we don't know if the n-
ways (age x gender, etc.) are independent; (4) by using joint proportions, one universal weighting factor
variable can be obtained for weighting across all subgroups; (5) if joint proportions are not used, then the
universal weighting factor must be determined by repeatedly weighting by the marginal proportions until
every subgroup proportion matches the true population proportion; (6) if the weighted joint sample
proportions match the population joint proportions, then the weighted marginal sample proportions match
the population marginal proportions, but not vice-versa; (7) it should be noted that not every possible
unrepresentative variable can be identified or measured, and that only the variables that are thought to be
the most important should be used to construct the /i-way population proportion; and (8) if weighted
results are significantly different from unweighted results, then the data are probably severely
undersampled and cannot be corrected by weighting.
S ' ~~ N
Technical Disclaimer The sample weights and selected time and background weights that have been incorporated in
the statistics given in this report may not be appropriate for all uses of the data, and further
analysis may be required.
4-1
-------
The Weighting Methodology Used in This Report
In this report we consider the NHAPS data to be a stratified random sample with subgroups formed
from background factors (gender, age, region) and time factors (day-of-week, season). The sample of n
total observations can be classified into L subpopulations with sample sizes n,,n2,...,nLand means
JEj, x2,..., XL. As Cochran32 notes, in stratified sampling the sample mean can be defined as:
(1)
where
y. = the value obtained for each individual i
yh = the sample mean for all y, in subgroup h
y = the sample mean
nk = the number of samples in subgroup h
n — the total number of samples in the stratified sample
L = the number of different subgroups
If some of the sample subgroups are misrepresented with respect to the true population, i.e., some of
the subgroups were either oversampled or undersampled in the survey, then the sample mean y is not the
best estimator of the population mean. As Cochran points out, a suitable estimator of the population
mean for use in stratified sampling is:
N
where
Nh = the number of individuals in the A* subpopulation
... + WL = the total number of individuals
L = the number of different subdivisions
The number of individuals Nk in each subpopulation is obtained from the true population, e.g., the
U.S. Census.
To obtain ya in computerized calculations, the concept of a weighting variable X is introduced.
First, we define the sample proportion pk for each subgroup /t, which — for an unrepresentative stratified
random sample — is not significantly equal to the population proportion Ph:
_ n*
* = T
4-2
-------
Taking the ratio of the sample proportion and the population proportion gives the subgroup h sample
weight XA:
1 - 1
Afc ~ _
where \h > 0 and
(4)
In general, each subgroup mean yh should be weighted by XA so that the sample proportion ph is
rescaled to match the population proportion Ph. However, in our computer methodology, we assign a
sample weight 5l, = XA to each individual that falls into the A* subgroup so that the weighted sample mean
yw( = yst is calculated as follows:31
yw{ =
= the weighted sample mean
(5)
where
Vi -
the sample weight assigned to individual i
the value obtained for each individual i
Although Equation 5 is equivalent to Equation 2, it is formulated in terms of weighting variables X,
for each individual / rather than subgroup means yh for each subgroup h. When the sample and
population proportions are obtained from n-way frequency tables for the given grouping factors (age,
gender, etc.), the individual weights A., can be used to calculate the weighted sample mean ywtfor any
combination of subgroups.
Example for Age and Gender Subgroups
In this report the major background factors for analysis of microenvironments were two gender
groups (male/female) and ten age groups - giving twenty total subgroups (Table 4-1). The population
reference was the 1990 U.S. Census with 248,709,873 total individuals. The sample weights Xh were
calculated for each subgroup h by forming joint frequency tables - containing the population proportion
Ph and the sample proportion ph for each subgroup h - for both the Census population (Table 4-2) and the
NHAPS survey (Table 4-3) and taking the ratio PJph for each subgroup. The individuals were assigned a
weight A.. = Phlph if they fell into the A* subgroup (Table 4-4).
Table 4-1. The Twenty Gender and Age Subgroups for the Calculation of Weights
Gender
Males
Females
0-4
1
11
5-11
2
12
12-17
3
13
18-24
4
14
Age
25-34
5
15
Groups
35-44
6
16
45-54
7
17
55-64
8
18
65-74
9
19
>74
10
20
4-3
-------
Table 4-2. The 1990 U.S. Census Population Joint Frequency and Proportions for Each Age and Gender
Subpopulation (N= 248,709,873)
Age Group
0-4
5-11
12-17
18-24
25-34
35-44
45-54
55-64
65-74
>74
Males
9,392,409
12,908,033
10,283,836
13,616,119
21,572,869
18,594,227
12,325,335
9,981,417
7,941,613
4,623,560
P, = 3.78
P2 = 5.19
P3 = 4.13
P4 = 5.47
PS = 8.67
P6 = 7.48
P7 = 4.96
P8 = 4.01
P9 = 3.19
P,o = 1.86
A/,
8,962,034
12,299,838
9,758,282
13,121,647
21,603,063
18,984,676
12,897,751
11,166,506
10,164,945
8,511,713
Females
% (Pft)
P,,=3.60
P12 = 4.95
P, 3 = 3.92
P14 = 5.28
P15 = 8.69
P16 = 7.63
P17 = 5.19
P18 = 4.49
P19 = 4.09
Pa, = 3.42
Table 4-3. The NHAPS Sample Joint Frequency and Proportions for Each Age and Gender Subgroup (n -
9,196)
Age Group
0-4
5-11
12-17
18-24
25-34
35-44
45-54
55-64
65-74
>74
"ft
270
369
285
414
781
679
528
435
303
177
Males
p,=2.88
p2 = 3.93
Pa = 3.04
p4 = 4.41
ps = 8.32
Ps = 723
p7 = 5.63
p. = 4.63
P9 = 3.23
p,0=1.89
Females
228
333
303
385
841
810
659
527
513
356
p,,=2.43
p 12 = 3.55
P, 3 = 3.23
Pu = 4.10
P, s = 8.96
p 16 = 8.63
P, 7 = 7-02
P, 8 = 5.61
p,9 = 5.47
Pa, = 3.79
Note: Respondents with missing age or gender values were discarded.
Table 4-4. The Sample Weights A,, Calculated for Each Subgroup h and Assigned to Each Individual / (Sorted by Age
and Gender)
WeWit
\
^•h
Subgroup
A, "-"270 "27V""839 "«40—"S24 •" ^7801"'^*327
P P P P
PI P^ Pz PI*
= 1.31 =1-32 =1.36 '" =0.80
n
^8328—^8*40
Pl9
= 0.78
"»
"ftftt'l * *"01 96
JP
^20
= 0.90
Note: The /subscripts for each individual weight A, run over all the respondents sorted by subgroup from individual 1
to individual 9,196.
4-4
-------
The Final Weight
Given the individual sample weights A,, there are currently many statistical software packages -
SAS, SPSS, and BMDP — that have built-in procedures to compute the weighted mean to estimate the
population mean. In this report we have used a SAS program (Figure 4-1) to: (1) calculate the n-way
sample proportion tables for gender (2 levels), age (10 levels), day-of-week (7 levels), and season (2
levels) - giving 2x10x7x2x2 = 560 different subgroups; (2) assign the sample weights to each
individual based on the 1990 Census data; and (3) check the representativeness of the final weighted joint
sample proportion over combinations of each subgroup. The initial joint sample frequency tables were
generated based on the sample likelihood of a child versus an adult being chosen from a household and
for the likelihood of a household being chosen based on the number of non-business telephones -
obtained using the WEIGHT variable given in the main file. In the Census data, the population
proportions for the factors of age and gender were assumed to have the same distribution across season
and day of week, hi other words, season and day of week are independent from the rest of the n-way
factors. However, in the NHAPS sample data the same assumption should not be made, and the season
and day-of-week factors must be included with age and gender factors in the n-way proportion table. In
other words, we cannot simply use the n-way sample proportion of age and gender and multiply them by
the marginal proportions for day-of-week (1/7 per day) and season (1/4 for Winter, Spring, Summer, and
Fall).
The weighted sample proportion significantly matches the 1990 U.S. Census proportion for each
gender and age group (Table 4-5). The geographic factors (Census regions) were not considered in the
weighting process, but there is still good agreement with the Census. The day-of-week and season
divisions match their required proportions of 1/7 and 1/4, respectively, after weighting, hi addition,
since we have used the n-way proportions for all subdivisions, the final weighted n-way sample
proportions contain statistically equal joint proportions of males, females, and respondents in each age
group across each day-of-week and season division (Table 4-6), i.e., the final weighted sample
proportions ph are now equal to the corresponding population proportion Ph with respect to each
subdivision. Thus, our final weight (WEIGHT4), which allows for the interdependence of each grouping
factor and is based on the joint proportion of 560 subgroups and the household sample weight, can be
used for analysis by any combination of subgroups.
4-5
-------
Note: In the following SAS source code, demo2 is the main file containing all background factors of the respondents,
weight is "the likelihood of kids+adults being selected", and weight4 is the final weight that we used in all
estimations of population means.
Part I: Create the sample proportion given the weight variable in the main file
proc freg data=demo2;
table rsex*age*day*season/out=resl noprint;weight weight;run;
Part II: Create the overall weight by using the joint frequency of gender*age, day of week, season and weight due to
kids+adults
data res2;set resl;
if rsex=l then do;
if age=l then wtsad=(3.776452/28)/percent;
else if age=2 then wtsad=(5.189996/28)/percent;
else if age=3 then wtsad=(4.134872/28)/percent;
else if age=4 then wtsad= (5.4747/28) /percent/-
else if age=5 then wtsad=(8.673909/28)/percent;
else if age=6 then wtsad=(7.476272/28)/percent;
else if age=7 then wtsad=(4.955708/28)/percent;
else if age=8 then wtsad=(4.013277/28)/percent/-
else if age=9 then wtsad=(3.193123/28)/percent;
else if age=10 then wtsad=(1.859017/28)/percent;
else if age=. or age=99 then wtsad=.;
end;
else if rsex=2 then do;
if age=l then wtsad=(3.603409/28)/percent;
else if age=2 then wtsad=(4.945456/28)/percent;
else if age=3 then wtsad=(3.923560/28)/percent;
else if age=4 then wtsad=(5.275885/28)/percent;
else if age=5 then wtsad=(8.686050/28)/percent;
else if age=6 then wtsad=(7.633262/28)/percent;
else if age=7 then wtsad=(5.185862/28)/percent;
else if age=8 then wtsad=(4.489772/28)/percent;
else if age=9 then wtsad=(4.087069/28)/percent;
else if age=10 then wtsad=(3.422346/28)/percent;
else if age=. or age=99 then wtsad=.;
end;
else wtsad=. ;
run;
Part ffl: Assign the overall weight from each subgroup to each individual respondent
proc sort data=demo2; by rsex age day season;
proc sort data=res2; by rsex age day season;
data demosad; merge demo2 res2; by rsex age day season;
drop count percent;
we ight3=weight*wtsad;
weight4=weight3*9196/9379.539021;
run;
Part IV: Check the weighted sample proportion and unweighted sample proportion. We want the weighted sample
proportion = population proportion
proc freq data=demosad;
table day wend age rsex rsex*age4 rsex*age;weight weight4;run;
proc freg data=demosad;
table day wend age rsex rsex*age4 rsex*age,-run;
proc freg data=demosad;
table regionc regione;run;
proc freg data=dentosad;
table regionc regione;weight weight4,-run;
proc freg data=demosad;
table quarter season;weight weight4;run;
proc means data=demosad n sum;
var weight wtsad weights weight4;run;
Rgure 4-1. The SAS code used in this report to generate the final weight (WEIGHT4).
4-6
-------
Table 4-5. Comparison Check of the Weighted Sample with the U.S. 1990 Census by Marginal Grouping
Factors
Grouping Factor
Male
Female
Under 5
5-11
12-17
18-24
25-34
35-44
45-54
55-64
65-74
75+
Refused
Northeast
Midwest
South
West
Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday
Weekday
Weekend
Winter
Spring
Summer
Fall
Unweighted
Sample Proportion
n = 9,386
45.7
54.2
5.3
7.5
6.3
8.5
17.3
15.9
12.6
10.2
8.7
5.7
2.0
22.1
22.4
34.6
20.9
15.5
15.5
14.8
12.6
8.9
13.5
19.2
67.3
32.7
26.9
26.0
27.0
20.1
Weighted
Sample Proportion
n = 9,196
48.71
51.29
7.38
10.14
8.06
10.76
17.37
15.12
10.15
8.51
7.29
5.22
20.3
24.7
34.4
20.7
14.30=(1/7)*100
14.30
14.30
14.30
14.30
14.30
14.30
71.41 =(5/7)*100
28.59 = (2/7)100
25.02 = (1/4)*1 00
25.02
25.02
24.95
1990 U.S. Census
Proportion
N= 248,709,873
48.7
51.3
7.37
10.14
8.06
10.75
17.36
15.11
10.14
8.50
7.28
5.28
20.4
24.0
34.4
21.2
-
-
-
-
-
-
-
-
-
-
-
-
-
Note: Individuals with missing age or gender values were dropped in the weighting procedure.
4-7
-------
Table 4-6. Consistency Check of the Joint Sample Proportion for Gender and Age Across Season and
Weekend vs. Weekday After Weighting
Time
Overall
Census
Overall
Winter
Spring
Summer
Fall
Weekday
Weekend
Overall
Census
Overall
Winter
Spring
Summer
Fall
Weekday
Weekend
Gender
Male
Male
Male
Male
Male
Male
Male
Male
Female
Female
Female
Female
Female
Female
Female
Female
0-4
3.77
3.78
3.78
3.78
3.78
3.79
3.78
3.78
3.60
3.61
3.60
3.60
3.60
3.61
3.61
3.60
5-11
5.19
5.19
5.19
5.19
5.19
5.20
5.19
5.19
4.95
4.95
4.95
4.95
4.95
4.96
4.95
4.95
12-17
4.13
4.14
4.13
4.13
4.13
4.15
4.14
4.13
3.92
3.93
3.92
3.92
3.92
3.93
3.93
3.92
18-24
5.47
5.48
5.47
5.47
5.47
5.49
5.48
5.47
5.28
5.28
5.28
5.28
5.28
5.29
5.28
5.28
Age
25-34
8.67
8.68
8.67
8.67
8.67
8.70
8.68
8.67
8.69
8.69
8.69
8.69
8.69
8.71
8.69
8.69
Group
35-44
7.48
7.48
7.48
7.48
7.48
7.50
7.48
7.48
7.63
7.64
7.63
7.63
7.63
7.65
7.64
7.63
45-54
4.96
4.96
4.96
4.96
4.96
4.97
4.96
4.96
5.19
5.19
5.19
5.19
5.19
5.20
5.19
5.19
55-64
4.01
4.02
4.01
4.01
4.01
4.02
4.02
4.01
4.49
4.49
4.49
4.49
4.49
4.50
4.49
4.49
65-74
3.19
3.20
3.19
3.19
3.19
3.20
3.20
3.19
4.09
4.09
4.09
4.09
4.09
4.10
4.09
4.09
75+
1.86
1.79
1.86
1.86
1.86
1.60
1.77
1.86
3.42
3.42
3.42
3.42
3.42
3.43
3.43
3.42
Note: This table shows that after weighting (by WEIGHT4) the joint sample proportions for age and gender are
almost all statistically identical across the time factors of season and day-of-week.
4-8
-------
Section 5
Microenvironments in the 24-hour Diaries:
Selected Location x Activity x Smoker-Present Categories
The NHAPS data contain three variables - WHR, RACT, and SMK - corresponding to locations,
activities, and smoker-present categories (Table 5-1). For this report, WHR was regrouped into a new
variable NEWLOC. RACT and the adjunct activity variable FACT that contains verbatim accounts of
the respondent's activity were used to create the regrouped exposure activity variables NEWACT and
REGACT. Locations and activities were regrouped from the original NHAPS WHR, RACT, and FACT
variables based on two criteria: (1) differentiation of types of human exposure; and (2) adequate sample
size. For this report, SMK was simplified from categories that specified a broad smoker-present duration
into categories for smoker-present and no-smoker-present only. Additional variables collected for each
person give the diary month (DD), diary year (DYR), day of week (DAY), day of month (DMO), and the
starting time (STRT), ending time (END), and elapsed time (ETOA) of each microenvkonment (Table 5-
1). As discussed in Section 2 the NHAPS diary data for each respondent consist of a variable number of
multiple records corresponding to each microenvironment that the respondent visited for the entire day
immediately before the day that the respondent was interviewed.
The microenvironments resulting from the regrouped locations and activities are relevant to different
kinds of chemical exposure via the air and water pathway, and they can be used to ascertain exposure to
chemicals via statistical analysis and modeling. The NEWACT and REGACT groupings were used for
all analyses presented in this report. Since this report is meant as a summary of NHAPS with an
emphasis on scientific exposure assessment, the location and activity groupings are not sociologically
based.
Note: When the data were receded into the new variables (NEWLOC, NEWACT, REGACT) with fewer location, activity, and
smoker-present categories, the original WHR, RACT, FACT, and SMK variables were retained in the data base.
Tabte 5-1. The NHAPS Microenvironmental Variables for Each Respondent Diary
Variable Variable Label Example
DAY Day of Diary (Day of Week) Monday, Tuesday Sunday
DD Diary DE (Day of Month) 1, 2, 3,4 30 (31) (29) (28)
DMO Diary Month 1,2,3,4 12
DYR Diary Year 92,93,94
STRT Start Time
END End Time
ETOA Elapsed Time of Microenvironment
FACT Activity Text "Bathing or Showering'
RACT Activity Code
NEWACT* Primary Regrouped Activity Code
REGACT Secondary Regrouped Activity Code
SMK Smoker Present (excluding the respondent) Yes: "Entire'; ">Half; "Half"; -
-------
5.1 Locations, Activities, and Smoker-Present Categories
Locations
The 83 original NHAPS location codes WHR (Table 5-2) are broken down into five areas: Own
House, Friend's/Other's House, Traveling, Other Indoor, and Other Outdoor. Since they are too
numerous for convenient analysis, they were regrouped into ten NEWLOC codes (Table 5-3). Whereas
in the NHAPS study the residential location was treated the most thoroughly with specific codes for each
kind of room indoors and each outdoor residential setting, the NEWLOC scheme groups all the
residential locations into either an indoor or an outdoor residential location. Residential locations at
one's own home were not differentiated from Residential locations at someone else's home (i.e.,
respondent locations were grouped into a "Residential" NEW ACT group even if the original NHAPS
code states that they were at someone else's house). The remaining non-residential NHAPS locations
were grouped into In Vehicle, Near Vehicle, Other Outdoors, Office/Factory, Mall/Other Store,
School/Church/Hospital/Public Bldg., Bar/Restaurant, and Other Indoors. The In Vehicle locations
include travel inside all cars, trucks, buses, trains, airplanes, boats, and public transit Travel outdoors
via motorcycle, bicycle, walking, stroller, etc., or waiting for transit outdoors were all grouped into Near
Vehicle. The relatively small Other Outdoors location grouping (2,149 out of 154,234 occurrences)
mostly includes various outdoor places associated with recreation such as sports stadiums, parks, and
playgrounds. The Other Indoor grouping (2,056 occurrences) included all the remaining indoor locations
such as auto-repair shops, health clubs, laundromats, salons, and parking garages. These locations may
be associated with very different kinds of exposures, but there are too few occurrences of each of them to
warrant separate categories.
The frequencies for each NEWLOC grouping over all those reported are all at least 2000. A
compromise was established between the creation of as many location groupings with different exposure
characteristics as possible within the limitation of number of occurrences of that location over all 9,386
respondents. Additionally, we did not attempt to distinguish locations that are not explicitly exposure-
related. For example, locations were not divided according to work-related activities. The only
NEWLOC group that can be associated with work-related activities is Office/Factory. It is not possible
to determine - based on location alone - whether work-related activities were occurring in any of the
other locations. Respondents that are in stores, restaurants, bars, hospitals, etc. could be present either
as patrons or as staff.
Table 5-2. The Original NHAPS Locations Codes WHR with the NEWLOC Regroupings
WHR
Value WHR Label Frequency
NEWLOC
10
20
30
40
SO
60
70
80
90
100
100 Other, Home 1
101 Home Kitchen 19,001
102 Home Uving Room/ 21,292
Family Room/Den
103 Home Dtrtng Room 3£53
104 Home Bathroom 10,964
10S Home Bedroom 30,243
106 Home Stuoybffice 669
107 Home Garage 250
108 Home Basement 417
110 Home UtBly Room/ 572
Laundry Room
10
10
10
10
10
10
10
10
10
10
5-2
-------
Table 5-2. (Continued)
uyub
Value WHR Label Frequency
Own House IftrifteiMff ;>>
111 Home Pool, Spa (Outdoors) 98
112 Home Yard/ 3,731
Other Outside House
113 Home Moving from 6,106
Room to Room
114 Home Moving 664
In/Out of the House
120 Other Verified 227
199 Ref 39
NEWLOC
10
~, -,„;",--- •
10
10
10
20
30
20
20
20
40
SO
60
70
80
80
100
Friend'sJOttere'House
200 Other, Other's House
201 Other's Kitchen 760
202 Other's Living Room/ 1395
Family Room/Den
203 Other's Dining Room 256
204 Other's Bathroom 169
205 Other's Bedroom 648
206 Other's Study/Office 24
207 Othert Garage . 35
208 Other's Basement 61
210 Other's Utaty Room/ 11
Laundry Room
211 Othef s Pool, Spa (Outdoors) 31
212 Other's Yard/ 726
Other Outside House
213 OthefsMoving 849
From Room to Room
214 Other's Moving In/Out 209
of the House
220 Other Verified 34
299 Ref 8
10
10
10
10
10
10
10
10
10
10
10
10
10
20
20
20
Traveltefl
300 Other, Travel
301 Car 23.562
302 Truck (Pick-up/Van) 3.663
303 Truck (Others) 287
304 Motorcycle/Moped/Scooter 78
305 Bus 864
306 Walking 3.683
307 Bicycle/Skateboard/ 249
Roller Skates
308 Ina Stroller/ 17
Carried by an Adult
310 Train/Subway/Rapid Transit 195
311 Airplane 62
312 Boat 38
313 Waiting for Bus, Train, Ride 177
(At Stop)
314 Waiting for Travel, Indoors 41
320 OtherVerified 27
399 Ref 1
30
30
30
30
30
30
30
30
30
30
40
40
40
40
40
100
Other Indoor
400 Other, Indoor 1
401 Office Building/Bank/ 2.904
Post Office
402 Industrial Plant/ 786
Factory/Warehouse
60
60
100
5-3
-------
TabteS-2. (Continued)
MIUB
WfWI
VMM WHRUM Fraqwncy
NEWLOC
10
20
30
40
SO
60
70
80
90
100
403 Grocery Store/ 1546
Conwentenc* Store
404 ShopptagliMV 2,179
Non-grocary Store
405 BKMghtClub/ 442
Bowing Atoy
406 Autoftap«rShop/ 222
GasSMion
407 Indoor GynV 454
Sports or Hailtti Club
408 Pubfc BiACnoAJbnry/ 628
MuaaunVThaater
409 Laundrerat 48
410 ttMptaHtaaMiCara 1,092
FtcttypDoctof% OKc0
411 BMuh/Parin/ 153
BtttwShop/
HarDraniri
412 At Wok/ 197
No SpaoCc M«in Location
413 School 2.108
414 nulaurrt 2£33
415 Chun* 976
416 HoMMoM 475
417 OfyOMrar 39
418 OhtrMiMirShap 40
419 IndDorPMWngQcnQa 6
420 OfMrVMtod 364
499 M 16
70
70
80
80
80
80
90
90
100
100
100
100
100
100
100
100
100
100
100
500 Otvr Outdoor
501 72
(Outdooi^
511 Frnm 213
520 OMwrVMM 210
599 M 6
ToW 157234
97^84
5.459
2&699
40
40
40
40
64)93
50
50
50
SO
50
50
50
50
2.149
3,690
3.725
4404
2475
2.056
NotK tht NEWLOC lc««toncMlM (torn Tabto 5^
(OiMoon). 50. OlwOultecv,
. m
Timing Mde Vehfcte, 40. NMT VaHde
rS«or^80.Scr»cVO*n*VHo»pit^>ubicBldg.,90.Barrttas»aurant,
•nd 100. OtMr Indoor. Alc«cunwx«c<«»chcWlca«toi«r»m^jprtor»«»ino>»r*ewhxatiorL
5-4
-------
Table 5-3. Receded NEWLOC Locations with Overall Frequency of Occurrence
NEWLOC
Residence (Indoors)
Residence (Outdoors)
Traveling Inside Vehicle
Traveling/Near Vehicle (Outdoors)
Other Outdoor
Office or Factory
Man/Grocery Store/Other Store
SchooWSiurchMospitalT'ublicBMg.
Bar or Restaurant
Other Indoor
Total
Abbreviations
Res. Indoors
Res. Outdoors
InVeh.
NearVeh.
Other Outdoors
Office/Fact
Mai/Other Store
School/Pub. BMg.
Bar/Rest
Other md.
Code
10
20
30
40
50
60
70
80
90
100
Original NHAPS Codes
100, HJ1-110, 113, 115-199, 201-210, 213, 21 5-299
111,112,114,211,212,214
300, 301-303. 305, 310-31 2, 320, 399
304, 306-308, 313, 501-504
505-599
401,402
403,404
408,410,413,415
405,414
314, 400, 406, 407, 409, 41 1 , 412,
416-499
Frequ.
97,584
5,459
28.699
6,093
2,149
3,690
3,725
4,804
2,975
2,056
157,234
%
62.1
3.5
18.3
3.9
1.4
2.3
2.4
3.1
1.9
1.3
100.0
Note: The Frequency and Percent columns give the number of times and the percentage of the numbwrf times that a rricroer^^
microenviium rents for a» 9,386 NHAPS respondents.
Activities
The original 91 NHAPS activity codes are broken down into NON-FREE time and FREE time
groups that are each further divided into the five additional areas of Paid Work, Household Work, Child
Care, Obtaining Goods/Services, and Personal Needs/Care for NON-FREE time, and Educational,
Organizational, Entertainment/Social, Recreational, and Communications for FREE time (Table 5-4).
These RACT categories are not the best groupings for exposure studies. Thus, they were used with the
FACT variable to create more exposure-relevant categories in the NEWACT and REGACT variables
(Tables 5-5 and 5-6).
Table 5-4. The Original NHAPS Activity Codes for RACT with the REGACT Regroupings
RACT
Value RACT Label Frequency
REGACT
0
10
20
30
40
50
60
Non-free Time/Paid Work
01 Main Job 5,461
02 Unemployment 56
03 Travel During Work 172
05 Second Job 41
08 Breaks 241
09 Travel To/From Work 6,421
0
0
0
0
0
0
Household Work
10 Food Preparation 6,670
11 Food Cleanup 1>348
12 Cleaning House 2.465
13 Outdoor Cleaning 825
14 Clothes Care 1.177
15 Car Repair/Maintenance 169
16 Other Repairs 386
17 PtentCare 299
18 Animal Care 1.189
19 Other Household Work 1.786
0
0
0
0
0
0
0
0
10
10
20
20
20
30
30
30
30
50
50
50
50
50
50
70
5-5
-------
JaMeS-4 (Continued)
RACT
Value RACT Label Frequency
REGACT
0
10
20
30
40
50
60
70
CHUCam
20 Baby Care 397
21 ChHCare 1.065
22 HatpingrTeaching 113
23 TflNn^RBftdnQ 86
24 Moor Playing 226
25 Outdoor Ptaying 68
26 Merited Care -CNd 11
27 ChUCare 662
28 Dry Cleaning 18
29 Travel, ChMCare 901
nfcl • fiihi •• l^fLftrf* **-• — * • '
ummng uuuuvy OWICM - ; ••• ~- .: - : .
30 Shopping for Food 1,466
31 Shopping for CMhesMousehoU Hems 1.931
32 Personal Care Services 121
33 Merited Appointments 287
34 Government/Financial Sarvtoes 410
35 Car Repair Services 297
36 Other Rap* Service* 12
37 OtherServices 225
38 Euan* 282
30 Travel, Goods and Service* 7320
l 0
r o
0
0
0
0
0
0
0
0
0
0
0
0
40
40
40
50
50
50
40 W«et*)8.etc. ' 7,188
41 Miacel Cam 197
42 HetpandCiM 985
43 Ejttng 18.102
44 Petsonel Hygiene 1,071
45 CleepingftlapplnB 19329
47 Dreetng,e(c. 11^75
48 NAAdivtie* 461
49 TnMl.PannnalCan> 5,161
0
0
0
0
0
0
0
0
40
70
FraeTtaeJEducaHonel
50 TenoVgFUl Time School 1^52
51 OhBfCllliii 21
S* Homewoik 803
55 LWngUxwy 49
56 OtarEducatfon 52
59 Other Travel, Education 2,127
0
0
0
0
0
0
Orgeabattonel
60 PiotmionaVOnioo 44
61 Special Meraet 16
82 PoMcaVCMc 13
63 Vokrteer Helping 42
64 Reigiou* Qmup* 52
65 Reigtou»Prtoe« 981
66 Fmtemal 4
67 CHUfYoOHftnff 21
68 Other Otoantzafion 161
69 Travel Qroaraeonal 1,758
0
0
0
0
0
0
0
0
0
0
5-6
-------
Table 5-4 (Continued)
RACT
Value RACT Label Frequency
REGACT
0
10
20
30
40
50
60
70
EnttrttmentfSocW
70 Sports Event 208
71 Entertainment 118
72 Movies/Videos 287
73 Theater 47
74 Museums 25
75 Visiting 3,071
76 Parties 278
77 Bars/Lounges 206
78 Other Social 116
79 Travel/Social 5,033
0
0
0
0
0
0
0
0
0
0
Recreation
80 Active Sports 1,807
81 Outdoor Recreation 303
82 Exercise 631
83 Hobbies 44
84 Domestic Crafts 301
85 Art 100
86 Music/Drama/Dance 292
87 Games 2,903
88 ComputerUse 256
69 Travel/Recreation 2,663
0
0
0
0
0
0
0
0
50
50
60
60
60
Communication*
90 Radra 229
91 TV 13,289
92 Records/Tapes 86
93 Readrig Books 527
94 Magazines, etc. 1,600
95 Rearing Newspaper 1,345
96 Conversations 1,988
97 Utters, Writing Paperwork 764
98 ThMdncyRelaxing 2,314
99 Travel Related/Passive Leisure 234
Total 157,234
0
0
0
0
0
0
0
0
0
0
114,344
6,771
2,730
2,452
8.577
1,919
2,439
18,102
Note: Some of the RACT coded activities were augmented wrtomc«detaS from the FACT variable. In addrSon. by using FACT some RACT codes were found
to have been miscoded. Thus, some occurrences lor same value of the RACT vanabte were catedmtodrrferentNEWACT or REGACT categories. If
ariy commences in RACT/FACTrr*tohed a NEWACT or REGACT categxxytr^ Different
cases of the same RACT code were sometimes mapped to one of several REGACT categories.
5-7
-------
Table 5-5. The NEWACT Activities Regrouped From the FACT and RACT Variables
NEWACT
No Exposure Activity
Food Preparation/Cooking
Eating Of Drinking
Ctearing Dishes or Kitchen
Laundry
Housekeeping
Sweeping/Vacuuming/Dusting
Mowing Grass/Yardwork
Taking/Giving a Bath/Shower
Hygiene/Using Bathroom
Washing Car
C?ariFcmHyi^tf U^^MIWS
Wwf d^U^ffm H VMPWHKMH^
Pajnting/Remodeling/Finishing
Carpertry/House Maintenance
Starling a Fire
Sports or Exercise
Total
New
Co*
0
10
15
20
21
22
23
25
30
31
32
40
41
42
50
60
Original
NHAPSCodn
—
10,19
43
11,12,16
12, 14, 18
12, 16. 19
12
13,17
20,21.32,40,42
44
15,16,19,35
15, 16, 19, 35
15,16,19,84.85
36,37
10, 16, 19
80.82.83
Frequency
114,344
6,671
18,102
1,527
1203
2,362
90
1,122
7,504
1,073
72
307
106
274
38
2,439
157,234
%
72.7
42
115
1.0
£
15
.1
.7
4.8
.7
.0
2
.1
2
.0
1.6
100.0
Note: TheFrequency and Percent f» columns give the number of times and the percentage of the number of times that
a microemironment occurred over al microenvironments lor al 9,396 NHAPS respondents.
Table 5-6. The Eight REGACT Activities Regrouped from NEWACT
Non Exposure Delated Activity
CooHnryFood Preparation
UundryOiiheeOeaning Wchen
Housekeeping
gfnnjnj^fr(>rfir'"fj**>"**'i'fj/
Using Bathroom
YardworWSardanoig/
Sports or Exercise
No Exposure
Food Preparation
Dishes/Laundry
Housekeeping
Bathing
Yardwork/
Maintenance
SpOCtS/EXBfCQG
EaSng/DrtnWng
Numeric
Cod*
0
10
20
30
40
50
60
70
NEWACT Codec
0
10
20,21
22,23
30,31
25,32,40,41,42,50
60
15
Frequency
114,344
6,671
2,730
2,452
8,577
1,919
2,439
18,102
Note Trie Frequency and Percert(%)columr*grvB me ra^^
-».^..^-~j.~»i«»n»«far»iagaBH>i*PSf«>mi»idaas. The nmanQB-of-a-Bnoker activttv is coded in the SMK variable.
%
72.7
42
1.7
1.6
S£
12
1.6
115
curred
The NEWACT categories were created by using as much detail as was available in the FACT and
RACT variables. Whereas RACT contains very broad exposure-relevant groupings such as "Food
Preparation" and "Cleaning House", the FACT variable contains more information on the specific kind
of activity such as "Baking Cookies" or "Vacuuming the Carpet." However, the data appeared to be
recorded inconsistently for different respondents with some of the activities coded with more detail than
others. For example, sometimes the FACT variable was very specific in the kind of activity ("Baking",
"Vacuuming", "Waxing the Floor", etc.) and sometimes it simply contained the general value label of the
RACT variable or text of a similarly general nature. Thus, it does not seem appropriate to use the
specific FACT variable for our final activity groupings, especially since the specific categories that were
coded using FACT have a small sample size (Table 5-5). For example, Car/Equipment Maintenance,
Building a Fire, Painting/Remodeling, etc. all have less than 500 occurrences, which is probably too
small to conduct analyses across subgroups.
The NEWACT categories were regrouped into eight broader REGACT categories (Table 5-6) that
each contain nearly 2,000 occurrences or more. These categories are not the only possible exposure-
5-8
-------
relevant groupings, but they represent our attempt to create useful exposure activities from the available
data. The most frequent activities in the No Exposure exposure activity category (Table 5-7) - into
which 73% of all microenvironments were grouped — were Sleeping/Napping (17% of
microenvironments), Watching TV (12%), and Dressing, etc. (10%). The next two largest No Exposure
activities were Traveling for Goods and Services (7%) and Traveling To and From Work (6%).
Although it is possible for exposure to occur during activities in the No Exposure category, little or no
information is added beyond that which the location category contributes. For example, it is only
necessary to know that a person is inside a vehicle in estimating their exposure from automobile exhaust
on the roadwav.
Table 5-7. The Five Largest Non-Exposure RACT Categories (REGACT =
NEWACT = 0)
NEWACT/REGACT
RACT Code Frequency %
Sleeping/Napping
Watching TV
Dressing, etc.
Travel-Goods/Services
Travel-To and From Work
0
0
0
0
0
19,329
13,289
11,375
7,620
6,921
16.9
11.6
9.9
6.7
5.6
Note: Percent (%) is the percentage of the number of occurrences over all 157,234 micro-
environments.
The Cooking/Food Preparation exposure activity encompasses any activity where the respondent
was preparing a snack or meal, regardless of whether cooking was involved. The
Laundry/Dishes/Cleaning Kitchen exposure activity includes activities related to cleaning up in the
kitchen and any activity related to washing clothes. The Housekeeping exposure activity includes
general tidying-up, cleaning of floors or carpets, and any other activity related to house cleaning. The
Bathing/Showering/Washing/Using the Bathroom exposure activity includes all activities related to
giving or taking a bath or shower, and any other activity related to hygiene, toiletry, or other personal
matters that may occur in the bathroom. The Yardwork/Gardening/Car or House Maintenance exposure
activity covers any type of automobile, house, or yard maintenance, including remodeling, painting,
washing the car, car repair, carpentry, fire-making, gardening, and mowing the lawn. The
Sports/Exercise exposure activity includes all team or individual sports and any other exercise-related
activity. The Eating/Drinking exposure activity includes any activity where the respondent was
consuming food.
Smoker-Present Categories
For the first 6 quarters of the NHAPS study the SMK microenvironment variable had only two
Smoker-Present possibilities: 0 for no smoker present and 1 for a smoker present (not including times
when the respondents themselves were smoking). The last two quarters contain seven possible Smoker-
Present categories corresponding to different intensities of environmental tobacco smoke (ETS) exposure
in each microenvironment, ranging from no exposure to exposure during the entire microenvironment
(Table 5-8). The vast majority of microenvironments (90%) do not involve any ETS exposure, and of the
10% that do involve exposure, almost 8% indicate exposure the entire time. Thus, the six categories for
incremental amounts of exposure — including the "don't know" (SMK = 8) category of exposure — were
grouped into one SMKEXP ETS exposure category, giving three possible Smoker-Present categories: 0
5-9
-------
for No Smoker Present, 1 for Smoker Present, and 8 for "don't know" responses. The presence of a
smoker can occur simultaneously with any other type of activity and in any location.
Table 5-8. The NHAPS Smoker-Present Categories
SMK
No
Yes - The Entire Time
Yes - More Than Half
Yes - About Half
Yes - Less Than Half
Yes - Hardly Any
Don't Know
Total
Code
0
1
2
3
4
5
8
Regrouped
Code
0
1
1
1
1
1
1
Frequency
141,507
12,146
180
734
1,073
734
860
157,234
%
90.0
7.7
.1
.5
.7
.5
.5
100.0
Note: SMK codes 2-5 were only used in the last two quartets of the study. The Frequency arid Percent columns give the number of times and the
ntmmtaMntlhaiMT*>«KrflgMislnMaii»uoenvinCTieniocci»iBdovafa^ The Regrouped
Code is 0 for no smoker present and 1 if there is a smoker present tor any amount of time. The'eJonl know responses (SMK = 8) mean that me
respondent oW not know how long the exposure lasted. These responses were considered as SMK = 0 in al analyses.
The number of people exposed to ETS was determined from the SMKEXP microenvironmental
variable by defining a variable SMKB for each respondent with values of 0 for people that were never
exposed on the diary day, and 1 for people that were exposed for at least one minute on the diary day
(Table 5-9). For example, if respondents had SMKEXP values of 0 all day (never exposed), then SMKB
= 0 for those people. Respondents with all 1's were assigned SMKB = 1, respondents with both 1's and
8's were assigned SMKB = 1, and respondents with only O's and 8's were omitted from calculations
using SMKB.
Table 5-9. The Frequency Table for SMKEXP (time exposed) with SMKB (people exposed) Assignment
SMKEXP
{0},{1},or{8}
{Oand1}or{0and8}
(0,1, and 8}
SMKB
Case*
1
2
3
9,150 =
r^r+JnmiKf*** rtf rt flO/l Or
O'S
5,145
0
0
0
5,145
nrdrart 1 f Cmnlrai
1's
11
3,623
0
1
+3,634
• Pnxmrm mrt 8 n>
8's and O's
2
234
0
+0
ml Know) in lt» SMKB '
8's and 1's
0
0
371
1
+371
variable ocves the
SMKEXP assignments of (0) no exposure over the entire alary day and (1) exposure ofa Person at sometime on the diary day. The combinations
corresponding to the shaded regions (al donl know 9 and no exposure 0 responses) were omitled in SMKB calculations. The samples sizes are
afferent for the weighted data as presented in Section 6.
5.2 Location x Activity and Location x Smoker-Present Microenvironments
Location x Activity
In this report location x activity microenvironments are defined by a combination of: (1) a
NEWACT location such as Residence-Indoors, Office/Factory, In Vehicle, Bar/Restaurant, or
School/Public Bldg.; and (2) a REGACT exposure activity such as Yardwork/Maintenance, Bathing,
Housekeeping, or Dishes/Laundry. A microenvironment matrix can be devised that contains rows for
every possible location and columns for every possible activity (Table 5-10). The intersection of each
row and column defines a unique microenvironment. For example, following the row for "Residential
(Indoor)" to the column for "Bathing/Showering" results in a cell for the "Bathing or Showering at
Home" microenvironment (01040).
5-10
-------
The actual number of real microenvironments is typically less than the total number of locations
times the total number of activities. It is possible that an intersection of a row and column results in a
microenvironment that does not exist. For example, the "Doing Dishes/Laundry in Vehicle"
microenvironment (03020) does not occur for any of the NHAPS interview respondents (Table 5-10). In
addition, many of the microenvironments (cells) may not contain enough NHAPS respondents to do a
complete analysis. For example, there were very few people interviewed who were preparing food at
places other than the Residential (Indoor) or Residential (Outdoor) locations. Only 21 out of the 70
possible microenvironments for exposure activities 10-70 contain more than 100 occurrences. The most
occurrences of any microenvironment were those in the Residential (Indoors) location with 22,605 total
occurrences for microenvironments with exposure activities 10-70.
Table 5-10. The Location x Activity Microenvironmental Matrix with the 70 Resulting 5-Digit Numerical
Microenvironmental Codes Obtained by Merging the Regrouped Location and Activity Codes
Sample Sizes in Parentheses)
(with
NEWLOC
Regrouped
Locations Codes
Residential Ind.
Residential Out.
In Vehicle
Near Vehicle
Other Outdoors
Office/Factory
Mall/Store
School/Public Bldg.
Bar/Restaurant
Other Indoors
010
020
030
040
050
060
070
080
090
100
REGACT Regrouped Activities
NoExp.
00
01000
(9339)
02000
(2028)
03000
(7742)
04000
(2430)
05000
(887)
06000
(1959)
07000
(2661)
08000
(2887)
09000
(630)
10000
(818)
Food
Prep.
10
01010
(4204)
02010
(94)
03010
(2)
04010
d)
05010
(8)
06010
(3)
07010
(2)
08010
(2)
09010
(6)
10010
(5)
Dishes/
Laundry
20
01020
(1814)
02020
(49)
03020
(0)
04020
(0)
05020
(2)
06020
(0)
07020
(0)
08020
(3)
09020
(0)
10020
(56)
House-
keeping
30
01030
(1842)
02030
(122)
03030
(10)
04030
(D
05030
(8)
06030
(D
07030
(1)
08030
(3)
09030
(D
10030
(1)
Bathing/
Shower
40
01040
(6309}
02040
(14)
03040
(9)
04040
(3)
05040
(7)
06040
(13)
07040
(10)
08040
(21)
09040
(7)
10040
(181)
Yardwork/
Maintenance
50
01050
(281)
02050
(1008)
03050
(4)
04050
(150)
05050
(44)
06050
(2)
07050
(3)
08050
(3)
09050
(D
10050
(48)
Sports/
Exercise
60
01060
(308)
02060
(446)
03060
(7)
04060
(450)
05060
(528)
06060
(2)
07060
(11)
08060
(75)
09060
(46)
10060
(256)
Eating/
Drinking
70
01070
(7847)
02070
(225)
03070
(40)
04070
(74)
05070
(143)
06070
(332)
07070
(90)
08070
(312)
09070
(1833)
10070
(132)
Note: Shaded cells designate the 21 location x activity microenvironments with frequencies greater than 100 that were analyzed
in this report (see Section 6-I).
Location x Smoker-Present
A location x smoker-present microenvironmental matrix can also be devised containing location
codes in each row and the two smoker-present categories in each column (one for "Smoker-Present" and
one for "No-Smoker-Present"). The sample sizes in Table 5-11 show that each microenvironment exists
and has a substantial number of occurrences in the NHAPS data because it is possible for a smoker to be
present in any of the regrouped locations. The location with the fewest number of smoker-present
occurrences (code 1) is School/Church/Public Bldg. (80) with 264. Residential Outdoors, Near Vehicle,
Other Outdoors, Office/Factory, Mall/Store, and Other Indoors all have fewer than 550 occurrences. The
largest sample size falls in the Residential (Indoors) location at 2937 — probably since the majority of
time is spent indoors at home (see Section 6) - with In Vehicle next at 1559, and Bar/Restaurant third at
1074.
5-11
-------
Table 5-11. The Location x Smoker-Present Microenvironmental Matrix with the 20 Resulting 4-
Digit Numerical Microenvironmental Codes Obtained by Merging the Location and
Smoker-Present Codes (with Sample Sizes in Parentheses)
NEWACT
Regrouped
Locations
Residential Ind.
Residential Out.
In Vehicle
Near Vehicle
Other Outdoors
Office/Factory
Mall/Store
School/Public Bldg.
Bar/Restaurant
Other Indoors
Codes
010
020
030
040
050
060
070
080
090
100
SMKEXP Smoker-Present Categories
No Smoker Present
0
0100 (9327)
0200 (2834)
0300 (7217)
0400 (2587)
0500(1126)
0600(1713)
0700 (2439)
0800 (2790)
0900(1442)
1000(1027)
Smoker-Present
1
0101 (2937)
0201 (517)
0301 (1559)
0401 ( 475)
0501 ( 348)
0601 ( 424)
0701 ( 374)
0801 ( 264)
0901 (1074)
1001 ( 287)
Note: SMKEXP codes with values of 8 (dont know) were assumed to equal 0 (no smoker) in this report
5-12
-------
Section 6, Part I
24-hour Diary Results by
Day-of-week, Season, and Background Factors
Each location, exposure activity, and presence of a smoker category was analyzed by: (1) the
percentage of time spent; and (2) the mean doer 24-hour duration. Location x activity and location x
smoker present microenvironments were also analyzed by the mean frequency of their occurrence (for
doers) during the diary day. Comparisons between subgroups are based only on these results; no
statistical comparisons of distributions were performed. Percentage of time spent was calculated over all
respondents (both doers and non-doers) except for microenvironments when exposure to environmental
tobacco smoke (ETS) occurred, where only the doers were considered. All mean 24-hour durations were
calculated over the pool of doers. Only the time spent actually experiencing each microenvironment was
considered. For example, percentages of time spent exposed to ETS were only done for the respondents
that were exposed for at least one minute on the diary day, and the mean 24-hour durations of exposures
were calculated only from the microenvironments where these respondents reported a smoker present.
Section 1 and Appendix A contain the methodology used to calculate durations and number of
occurrences, and Appendix B contains detailed examples of each type of calculation.
The number of occurrences of microenvironments were only determined for each unique location x
activity and location x smoker-present microenvironment. Since there are different entries in the diary
file for each possible location x activity microenvironment, it is difficult to obtain number of occurrences
for each location and activity by themselves. For example, a count of the number of microenvironments
in the Residential-Indoors location would give new occurrences every time a new activity was initiated
even though the respondent had not left the location. Counts of this sort can greatly overestimate the
number of occurrences of locations and activities by themselves. Similarly, 24-hr cumulative durations
were tabulated instead of the durations of individual microenvironments (with the same activity and
location) since the individual durations would be smaller than the actual time spent in one location across
several activities — or one activity across several locations.
All results described below are for weighted calculations (see Section 4) unless stated otherwise;
however, the bar charts for 24-hr mean durations also contain the unweighted results. The development
of the microenvironment categories (location, activity, smoker-present, and combinations) is described in
Section 5.
Technical Note: The percentage of time that respondents spend in each microenvironment (location, activity, smoker-present,
and combinations) was determined by summing the time spent in each microenvironment and dividing by
the total amount of time over all the respondents: 1440 minutes/day x 9,386 respondents = 13,515,840 total
minutes. The overall frequency of occurrence and duration were determined by: (1) sorting the 157,234
NHAPS diary records for all 9,386 respondents by person and exposure activity; (2) compressing the
records for each exposure activity into one record per person containing the number of times the exposure
occurred during the diary day and the total time summed over each occurrence; and (3) calculating means of
frequency of occurrence and duration for each exposure activity for each person. The doer sample size N is
the number of people -- out of the 9,386 total respondents -- that were engaged in a given exposure activity
at any time during the diary day.
6-1
-------
6.1 Locations
The largest percentage of time spent over all the respondents was in the Residential-Indoors location
(69%) (see Figure 6-1; Table 6-1 contains percentages of time spent across each subgroup). The next
largest percentages were for the School/Public Bldg (7%), the In Vehicle (6%), and the Office/Factory
(5%) locations. The percentage of time spent in the Residential-Outdoor location was about 4% with the
Residential-Indoors 68.73
Residential-Outdoor 3.69
In Vehicle 5.52
Other Indoor 2.07
Bar/Restaurant 1.84
School/Public Bldg. 6.61
Mall/Other Store 2.26
Office/Factory 5.39
Other Outdoor 2.19
Near Vehicle 1.7
Figure 6-1. The overall weighted percentage of time spent by the respondents in each location. The total
amount of time is 1,440 min x 9,196 respondents = 13,242,240 minutes.
Table 6-1. The Weighted Percentages of Time Spent in Each Location on the Diary Day Across Each Subgroup
Residential
Percentage : - Indoors
Overall
Males
Females
0-4
5-7
17-64
65+
Northwest
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
68.73
64.79
72.47
84.08
67.81
64.71
80.84
68.77
68.22
68.75
69.24
66.91
73.26
71.2
66.76
67.27
6968
Residential
- Outdoor
3.69
4.49
2.93
5.38
5.05
2.93
4.48
3.25
3.41
4.03
3.89
3.12
5.12
1.71
5.33
5.18
254
In
Vehicle
5.52
5.94
5.12
3.14
4.29
6.43
4.17
5.57
5.62
5.57
5.25
5.35
5.93
5.34
5.76
5.39
5.57
Near
Vehicle
1.7
2.49
0.94
0.56
1.41
2.06
0.99
1.78
1.51
1.64
1.93
1.89
1.2
1.45
1.67
1.89
1.77
Other
Outdoor
2.19
2.96
1.45
0.96
2.83
2.33
1.27
2.02
2.26
2.05
2.48
1.87
2.97
1.27
2.33
3.42
1.72
Office/
Factory
5.39
6.46
4.38
0.05
0.18
8.42
1.18
5.9
5.27
5.49
4.88
7.02
1.34
5.37
5.5
5.44
5.26
Mall/Other
Store
2.26
1.85
2.65
1.39
1.15
2.77
1.89
2.34
2.05
2.29
2.4
2.15
2.54
1.95
2.29
2.4
2.41
School/
Public Bldg.
6.61
6.53
6.68
3.45
15.33
5.19
2.83
6.66
7.24
6.63
5.77
7.86
3.5
7.48
6.77
4.98
7.21
Bar/
Restaurant
1.84
1.94
1.75
0.57
0.76
2.43
1.27
1.65
2.18
1.64
1.96
1.65
2.31
1.97
1.77
1.8
1.83
Other
Indoor
2.07
2.55
1.63
0.42
1.18
2.74
1.07
2.05
2.24
1.91
2.19
2.17
1.84
2.24
1.85
2.22
1.99
Note: Subgroups are for gender, age, U.S. census region, day of week, and season. See Section 4 for a discussion of weighting.
6-2
-------
Near Vehicle, Other Outdoor, Mall/Store, Bar/Restaurant, and Other Indoor locations all at about 2%.
The Residential-Indoors location also has the largest 24-hour cumulative mean duration (996 min) over
all the respondents (Figure 6-2; Table 6-2 contains durations across each subgroup). The next largest
duration was for the Office-Factory location (388 min) followed by the School/Public Bldg (284 min)
and the Other Indoor (226 min) locations. All the other locations had mean 24-hr durations of less than
200 min (3.3 hours). The smallest 24-hr durations were for the In Vehicle and Near Vehicle locations
(78-95 min). There is only a 1-4% difference between the overall unweighted and weighted mean
durations. Samples sizes for the doers across each subgroup are given in Table 6-3.
Males spent more time than females in every location except for Residential-Indoor (72.47% F vs.
64.79% M) and Mall/Other Store (2.65% F and 1.85% M) (Figure 6-3). Differences also occurred for
the Office/Factory (6.45% M and 4.38% F), Other Outdoor (2.96% M and 1.45% F), Near Vehicle
(2.49% M and 0.94% F), and Residential-Outdoor (4.49% M and 2.93% F). The 24-hr mean durations
for males are larger or roughly equal in every category except Residential-Indoors (1,048 min M and 940
min F) (Figure 6-4). The only category where males durations are more than twice as large as female
durations are for the Near Vehicle location (111 min M and 45 min F).
There are striking similarities between the Very Young (ages 0-4) and the Retired (65+) age groups
for both percentage of time spent and 24-hour mean durations (Figures 6-5 and 6-6). Both groups spend
the majority of their time in the Residential-Indoor location (68% VY and 65% R) for about the same
Residential-Indoors
Residential-Outdoor
In Vehicle
Near Vehicle
g Other Outdoor
° Office/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant-
Other Indoor-
Weighted (N=9196)
Unweighted (N=9386)
995.6
1,001.39
226.46
225.75
200 400 600 800 1,000
Mean 24-hour Cumulative Duration (minutes)
1,200
Figure 6-2. The overall weighted and unweighted mean 24-hour cumulative durations in each
location. In the weighted analyses, 190 respondents with missing age or gender
values were excluded. See Section 4 for a discussion of the weighting
methodology.
6-3
-------
durations (980 rain VY and 940 min R). The percentage of time spent and mean durations are also
similar for every other location category except for Office/Factory (0.05%, 33 min VY; 1.18%, 166 min
R). As expected, the Working age group (ages 18-64) spent the largest percentage of time in the
Office/Factory location (8.42%) and for the longest mean duration (414 min); and the School Age age
group (ages 5-17) spent the most time in the School/Public Bldg location (15.33%, 360 min).
Table 6-2. The Mean Weighted 24-Hour Duration (Minutes) in Each Location on the Diary Day by Each Subarouo
Duration D
Overall
Males
Females
0-4
5-7
17-64
65+
Northwest
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fal
Residential
-Indoors
995.6
940.33
1047.91
1213.14
980.04
939.55
1165.41
993.65
991.43
995.46
1002.74
968.91
1062.38
1031.57
967.46
976
1007.34
Residential In
-Outdoor
155.84
176.41
13321
190.73
167.46
142.36
166.34
149.79
158.37
16224
147.82
141.62
183.99
116.36
170.98
169.15
139.41
Vehicle
95.48
101.15
89.93
66.96
74.89
104.95
87.87
98.12
96.92
9521
91.66
91.74
105.18
95.08
97.15
93.75
95.9
Note: Subgroups are for gender, age, U.S. census
Table 6-3. The Weighted Number
Near
Vehicle
78.38
111.08
44.99
40.97
44.69
99.8
58.42
69.71
7121
87.54
82.89
82.87
64.6
72.16
76.04
83.4
81.3
Other
Outdoor
198.85
225.78
161.35
148.19
149.17
231.05
192.91
205.35
201.88
191.08
202.16
17925
240.09
181.35
185.76
228.57
182.15
Office/
Factory
387.96
411.54
359.16
32.68
105.86
413.73
165.97
39122
38421
405.75
358.37
401.56
269.06
379.55
394.8
390
387.6
Mall/Other
Store
118.5
121.4
116.66
93.07
82.92
133.14
91.32
118.49
104.51
131.36
115.71
120.14
115.19
109.48
124.66
117.01
122.48
School/
Public Bldg.
283.65
292.09
27625
238.8
360.3
265.3
148.07
280.1
295.72
283.44
271.35
312.92
186.08
303.01
284.05
264.36
278.85
Bar/
Restaurant
111.74
111.15
112.36
62.4
6627
125.64
902
115.44
11328
110.02
109.31
110.16
114.66
126.8
105.18
109.51
106.67
Other
226.46
260.78
189.41
106.12
153.77
258.3
147.42
216.6
251.02
22225
215.83
222.1
240.49
240.9
209.87
248.76
206.86
region, day of week, and season. See Section 4 for a discussion of weighting.
of NHAPS Respondents (Sample Size
N) in Each
Location on the Diary Day
Each Subgroup
Doer
Sample
SizeN
Overall
Males
Females
0-4
5-7
17-64
65+
Northwest
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fal
Residential
-Indoors
9141
44445
4696.5
677.8
16682
5646.5
1148.6
1836.4
2222.1
3219.7
1862.8
6530.3
2610.7
2286.5
2285.8
22832
2285.5
Residential
-Outdoor
3138
16435
1494.5
275.6
727.7
1689
445.8
574.9
694.6
1157.7
710.9
2084.7
1053.3
487.4
1033.3
1014.3
603.1
In
Vehicle
7652.4
3786.1
38662
458.7
13825
5024.6
7865
1506
1873.3
2728.7
1544.4
5518.9
2133.4
1861.8
1963.4
1906.4
1920.7
Near
Vehicle
2863.6
1446.8
1416.7
134.7
759
16902
279.7
678
686
8715
628
2160.1
703.4
667.8
725.8
749.7
720.3
Other
Outdoor
1455.4
847
608.4
632
457.3
825.4
109.4
261.3
3622
500.8
331.1
986.4
469
232.7
414.7
495.4
312.7
Office/
Factory
1841
1012.1
828.9
15.6
40.1
1667.6
117.7
400.1
442.6
630.7
367.6
1651.9
189.1
468.9
4612
462.3
448.7
Mall/Other
Store
2529.6
984.1
1545.6
145.6
334.9
1705.6
3435
5242
633
811.7
560.7
1695.3
834.3
590.1
608.4
680.6
650.5
School/
Public Bldg.
3085.8
1442.6
16432
141.4
1025.8
1602.4
3162
631.1
7902
1090.6
573.9
2373.8
712
8182
789
624.4
854.1
Bar/
Restaurant
2182.6
1124.1
1058.6
89.7
276.8
1583.8
232.4
3802
621.7
696.3
484.4
1419
763.6
513.8
556.5
545.3
567.1
by
Other
Indoor
1213
629.8
5832
38.7
1852
868.5
120.6
251.5
288
399.9
273.7
925
288
308.5
291.3
296
317.3
Note: Subgroups are for gender, age, U.S. census region, day of week, and season. See Section 4 for a discussion of weighting. The
doer sample sizes are equal to the number of respondents entering each location on the diary day. The maximum number in
each location is the total weighted sample size (N=9196).
6-4
-------
80
40
20
72.47
M.79
I !
Males I I Females
Resldential-lndoors
Residential Outdoor
In Vehicle
Other Indoor
Bar/Restaurant
School/Public Bldg.
Mall/Other Store
Office/Factory
Other Outdoor
Near Vehicle
I r 12.93 • 2.49
I] | _
6.46
6.53 6.68
2.96
4.38
1.85
2.65
Tit3
Residenlial-lndoors In Vehicle Other Outdoor : Mall/Other Store Bar/Restaurant
Residential-Outdoor Near Vehicle Office/Factory School/Public Bldg. Other Indoor
Location
Residential-lndoors
Residential-Outdoor
In Vehicle
Near Vehicle
Other Outdoor
0
a
j Office/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
I 260.78
] 260.32
I 189.41
| 190.6
Weighted
Unweighted
Females
H Weighted
_j Unweighted
940.33
945.9
1047.91
] 1048.07
i 1 1 1 r-
1000 1200 1400
200 400 600 800
Mean 24-hour Cumulative Duration (minutes)
Figure 6-3. The weighted percentage of time spent in each location for males
vs. females.
Figure 6-4. The weighted and unweighted mean 24-hour cumulative
durations in each location for males vs. females.
-------
Residential-lndoors
• 67.81
64.71
I 64.08
|80.84
Residential-Outdoor
In Vehicle
1
5.38
5.05
2.93
4.48
3.14
6.43
|4.17
| Very Young
2 Working
School Aged
Retired
Near Vehicle
0.56
1.41
0.96
Other Outdoor • f*
rd.Ai
1.27
r
M
Office/Factory
8.42
Residential (Indoors)
...- _____________
r
1.39
1.15
2.77
1.89
13.45
.83
School/Public Bldg.
Bar/Restaurant 2.43
• 1J7
5.19
Other Indoor
Bar/Restaurant
School/Public Bldg.
Mall/Other Store
5 33 T^/iSS^s • Office/Factory
Residential (Outdoor) ^^1»
In Vehide Other Outdoor
Near Vechicle
Other/Indoor
10.42
lJ.18
l~1.07
20
40 60
Percentage
80
100
Figure 6-5. The weighted percentage of time spent in each location for each age group.
6-6
-------
1400
.E 1200
E,
o
'•g 1000
Q
0>
,g 800
E
3 600
k.
P
4 400
^^"""^81 — ^
j] w W *C o>H~i
= rd
-------
There were not large differences between the percent time spent in each location between each of
the four Census regions (Northeast, Midwest, South, and West) (Figure 6-7). The largest differences in
percentage of time spent were only about 1% and occurred for: (1) the Residential-Indoors location
between the Northeast (68%) and the other regions (69%); (2) the Residential-Outdoor location for the
Northeast and Midwest (3%) vs. the South and West (4%); and (3) the School/Public Bldg location for
the West (6%) vs. the other regions (7%). Differences between the mean 24-hr durations across
geographic regions (Figure 6-8) were also small with the largest absolute differences occurring for: (1)
the Office/Factory location between the West (358 min) and the other regions (384-406 min); (2) the
Mall/Other Store location between the South (131 min) and the other regions (104-118 min); and (3) the
Other Indoor location between the Midwest (251 min) and the other regions (216-222 min).
On weekends (WE) a higher percentage of time was spent and durations were longer than on
weekdays (WD) in the Residential-Indoors (73%, 1062 min WE; 67%, 969 min WD), Residential-
Outdoors (5%, 184 min WE; 3%, 142 min WD), and Other Outdoor (3%, 240 min WE; 2%, 179 min
WD) locations (Figures 6-9 and 6-10). As expected, on weekdays more time was spent in the
Office/Factory (7%, 402 min WD, 1%, 269 min WE) and School/Public Bldg (8%, 313 min WD, 4%,
186 min WE) locations.
There is not much variation across different seasons for percentage of time spent and mean duration
in locations (Figures 6-11 and 6-12) except that: (1) slightly less time is spent in die Residential-Indoors
location in the Spring and Summer seasons (67%, 967-977 min) than in the Winter and Fall seasons (69-
71%, 1,007-1,032 min); (2) more time is spent in the Residential-Outdoor location in the Spring and
Summer seasons (5%, 170-171 min) than in the Winter and Fall seasons (2-3%, 116-140 min); and (3)
slightly more time is spent in the Other Outdoor location in the Summer (3%, 229 min) than for the
Winter, Spring, and Fall seasons (1-2%, 181-186 min). In addition, in the summer season about 2% less
time is spent in the School/Public Bldg location than in the other seasons, although the 24-hr durations
are approximately the same (264-303 min).
6-8
-------
Residential-lndoors
Residential-Outdoor
In Vehicle
Near Vehicle
Other Outdoor
Office/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other/Indoor
68.77
68.22
68.75
69.24
Residential (Indoors)
. Northeast
Other Indoor
Bar/Restaurant
School/Public Bldg.
Mall/Other Store
Office/Factory
Other Outdoor
Near Vechicle
Residential (Outdoor)
In Vehicle
2. Midwest
40
Percentage
Figure 6-7. The weighted percentage of time spent in each location for each census region.
6-9
-------
1200
1000-
800-
600-
400-
200-
?* — "" 55 «<»<,, i$.
^ (vi mi r *-. «t s> ^;
• X >^ $SS°!
•HHiii
8 sS
SIS
L
8 ?
ri?
Northeast
• Weighted
D Unweighted
Midwest
YA Weighted
L] Unweighted
South
ION Weighted
CD Unweighted
West
E9 Weighted
L I Unweighted
I
Location
Figure 6-8. The weighted and unweighted mean 24-hour cumulative durations in each location for each census region.
-------
80
20
I Weekday • Weekend
6691
Residential- Indoors
Other Indoor
Bar/Restaurant
School/Public Bldg.
Mall/Other Store
. Other Outdoor
Residential-Outdoor Office/Factory
In Vehicle Near Vehicle
. -.
u.OB
Resldentlal-lndoors ; In Vehicle Other Outdoor Mall/Other Store Bar/Restaurant !
Residential-Outdoor Near Vehicle Office/Factory School/Public Bldg. Other Indoor
Location
Figure 6-9. The weighted percentage of time spent in each location for
weekdays vs. weekends.
Residential-lndoors
Residential-Outdoor
In Vehicle
Near Vehicle
Other Outdoor
Office/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
141.62
I 141.16
] 193.99
J174.92
I 120.14
117.45
115.19
110.61
312.92
309.84
968 91
I 965.69
•• 1062.38
. ] 1074.81
Weekday
H| Weighted
[ ] Unweighted
Weekend
H| Weighted
3] Unweighted
200 400 600 800 1000 1200 1400
Mean 24-hour Cumulative Duration (minutes)
Figure 6-10. The weighted and unweighted mean 24-hour cumulative
durations in each location for weekdays vs. weekends.
-------
Residential-lndoors
Residential-Outdoor j
In Vehicle
Near Vehicle
Other Outdoor
Office/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other/Indoor
Winter
I Summer
Spring
]FaH
Residential (Indoors)
Other Indoor
/ Bar/Restaurant
School/Public Bldg.
Mall/Other Store
5.37
! 5.26
Residential (Outdoors)
In Vehicle
Office/Factory
Other Outdoor
Near Vechide
0
20 40
Percentage
60
80
Figure 6-11. The weighted percentage of time spent in each location for each season.
6-12
-------
U)
1200
. 1000
(A
i
!
§ 80°
0>
i 6°°
a
I
| 400
e
I 200
all
r • Q£
*\ |J
' O
i>
>:
>
c
i j
y el
ij
c
^s
Ci
• ij
^, C1
)
>
s
?
<
>
'
ff
•
1
-
/
f
f,
•
[
'•
fj
f.
f.
f,
\
v S
X ' \
,
ls"«
Et885s(
P*01r^? S^Z.KK-V
—~% >St"l ssfeRsi&seS;
HJLUniu
•
/
Winter
• Weighted
O Unweighted
Spring
^ Weighted
CD Unweighted
Summer
HI Weighted
CD Unweighted
Fall
IS1 Weighted
71 Unweighted
^ass-ssi
s^Sstcgg
i
i
/
Location
Figure 6-12. The weighted and unweighted mean 24-hour cumulative durations in each location for each season.
-------
6.2 Exposure Activities
Below, we have only considered the percentage of the total time over all respondents that was spent
in one of the exposure activities, i.e. the "No Exposure" category was excluded (see Section 3 for an
explanation of die construction of exposure activities). Overall the percentage of time spent in all
exposure activities was 13%. Males (12%) had less time in exposure activities than females (14%) and
the Very Young and School Aged had less (10%) than the Working and Retired (13% and 17%,
respectively). The Northeast and West had more time in exposure activities (14%) than the Midwest and
South (13%). More time in exposure activities occurred on weekends (15%) than on weekdays (12%),
and the Spring and Summer had more time in exposure activities (14%) than the Winter and Fall seasons
(12%).
The largest percentage of any of the exposure activities (over the time spent being exposed only)
was for Eating/Thinking (35%) (Figure 6-13; Table 6-4 contains percentages of time spent across all
subgroups). The percentages of time spent in the Food Preparation, Housekeeping, Yardwork/
Maintenance, and Sports/Exercise exposure activity categories were all 11-13%. The Bathing category
took up about 9% and the Dishes/Laundry/Clean Kitchen category took up 6% of the time. In contrast,
the largest mean 24-hr duration (Figure 6-14; Table 6-5 contains 24-hour durations across all subgroups)
occurred for Yardwork/Maintenance (147 min), followed by Housekeeping (116 min) and
Sports/Exercise (122 min). Eating/Drinking had the fourth largest mean 24-hr duration at 74 min. The
durations for Food Preparation and Dishes/Laundry/Clean Kitchen were comparable to Eating/Drinking
at 53 and 52 min, respectively. The smallest duration was for Bathing at 26 minutes. Unlike for
locations, the ranking of exposure activities by duration does not closely match their ranking by
percentage of time spent. This phenomenon occurs because percentages of time spent depend on the
length of events and the number of people experiencing them, whereas the mean 24-hour durations
presented here depend only on the average length of events for just the people that experience them - the
doers. Table 6-6 contains samples sizes of the doers in each exposure activity across all subgroups.
The differences between the percentage of time spent in exposure activities by males vs. females are
striking for every category except Bathing (9-10% for both) (Figure 6-15). Females are engaged in
higher proportions for Food Preparation (6% M, 16% F), Dishes/Laundry/Clean Kitch (2% M, 9% F),
and Housekeeping (6% M, 18% F), while males are occupied in higher proportions for
Yardwork/Maintenance (18% M, 6% F), Sports/Exercise (18% M, 9% F), and Eating/Drinking (40% M,
32% F). The 24-hr durations (Figure 6-16) follow the same pattern except for: (1) Housekeeping and
Eating/Drinking where the durations are approximately the same for both males and females (115-116
min and 73-74 min for each exposure activity, respectively); and (2) Bathing/Hygiene, where male
durations are slightly smaller than those for females (24 min M, 28 min F).
The Very Young/School Age age groups, when compared to the Working/Retired age groups, have
smaller percentages of time spent (Figure 6-17) in the Food Preparation (1-2% vs. 14%),
Dishes/Laundry/ Clean Kitch (0-1% vs 7-8%), Housekeeping (1-5% vs. 14-17%), and
Yardwork/Maintenance (1-4% vs 12-14%) exposure activities. Of all age groups, the Very Young have
the highest percentages of time spent for the Bathing (14% vs 6-11% for all the others) and
Eating/Drinking (63% vs. 32-40% for all the others) exposure activities. The Sports/Exercise exposure
activity has the highest percentage of time spent for the School Age age group (37%), followed by the
Very Young (18%), the Working (10%), and the Retired (6%). The mean 24-hr durations (Figure 6-18)
for the Dishes/Laundry/Clean Kitch, Housekeeping, Sport/Exercise, and Yardwork/Maintenance
6-14
-------
exposure activities follow the pattern of the percentages of time spent. However, the durations of the
Very Young for Food Preparation are nearly equal to those for the Working, probably since many small
children are assisting their parents. The Very Young/Retired durations are very similar for Bathing and
Eating/Drinking (30 min and 92-94 min, respectively) as they are for the School Age/Working durations
(24-27 min and 62-70 min, respectively).
Bathing 9.38
Yard/Mainten. 11.28
Housekeeping 12.42
Dishes/Clean Kitchen 6.02
Food Preparation 11.8
Sports/Exercise 13.26
Eating/Drinking 35.82
Figure 6-13. The overall weighted percentage of time spent by the respondents in each exposure activity
excluding time spent in the No Exposure category. The total amount of time is 1,737,104
minutes = 1,440 min x 9,196 respondents (13,242,240 min) minus 11,505,136 min (the time
spent in the No Exposure category, REGACT = 0 -- 86.88% of 1,440 x 9,196).
Table 6-4. The Weighted Percentages of Time Spent in Each Exposure Activity on the Diary Day
Across Each Subgroup
Percentage
of Time
Overall
Males
Females
0-4
5-17
17-64
65+
Northwest
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
No
Exposure
86.88
88.1
85.73
89.82
89.91
86.52
82.53
86.38
87.48
87.17
86.16
87.54
85.23
87.6
86.02
86.32
87.59
Cooking
1.55
0.73
2.33
0.14
0.22
1.92
2.47
1.57
1.52
1.52
1.6
1.56
1.51
1.54
1.63
1.48
1.54
Dishes/
Laundry
0.79
0.22
1.34
0.06
0.12
0.96
1.4
0.86
0.76
0.75
0.83
0.76
0.88
0.86
0.76
0.81
0.74
House-
keeping
1.63
0.69
2.51
0.16
0.45
1.87
3.01
1.68
1.51
1.66
1.66
1.53
1.88
1.5
1.56
1.58
1.86
Bathing
1.23
1.14
1.32
1.43
1.11
1.29
1.03
1.19
1.11
1.33
1.25
1.22
1.26
1.32
1.37
1.13
1.11
Yardwork/
Maintenance
1.48
2.18
0.82
0.15
0.4
1.75
2.5
1.36
1.4
1.45
1.77
1.29
1.95
0.98
2.04
1.68
1.23
Sports/
Exercise
1.74
2.16
1.34
1.84
3.74
1.29
0.97
2.07
1.35
1.71
1.94
1.58
2.14
1.39
1.97
2.22
1.38
Eating/
Drinking
4.7
4.78
4.62
6.4
4.06
4.4
6.09
4.9
4.88
4.41
4.79
4.51
5.15
4.81
4.66
4.77
4.55
Note: Subgroups are for gender, age, U.S. census region, day of week, and season
of weighting. Percentage of time is over all respondents (doers + non-doers).
See Section 4 for a discussion
6-15
-------
Weighted
Unweighted
50 100 150
Mean 24-hour Cumulative Duration (minutes)
200
Figure 6-14. The overall weighted (N=9196) and unweighted (N=9386) mean 24-hour cumulative durations
in each exposure activity. In the weighted analyses, 190 respondents with missing age or
gender values were excluded. See Section 4 for a discussion of the weighting methodology.
Table 6-5. The Mean Weighted 24-Hour Duration (for Doers) in Each Exposure Activity on the Diary
Day by Each Subgroup
Doer
Duration D
Overal
Mates
Females
0-4
5-17
17-64
65+
Northwest
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fal
No
Exposure
1251.1
1268.59
1234.49
1293.47
1294.75
1245.88
1188.4
1243.89
1259.66
1255.3
1240.7
1260.64
122728
1261.48
1238.72
1242.96
1261.27
Cooking
53.1
36.97
61.03
52.08
21 2.
53.71
62.37
53.52
50.82
54.97
52.44
51.85
56.61
51.09
5759
5124
52.64
Dishes/
Laundry
61.48
47.83
6428
31.78
29.95
62.57
67.68
58.57
60.52
63.3
62.92
5826
69.77
6326
63.95
62.88
56.05
House-
keeping
116.06
11525
11628
65.4
62.44
120.41
128.83
109.64
11727
121.52
112.64
107.47
138.52
116.85
120.44
108.99
118.34
Bathing
26
2426
27.64
3024
23.96
25.65
29.03
25.16
24.12
27.13
27.03
25.31
27.85
27
28.4
23.94
24.53
Yardwork/
Maintenance
14724
17026
109.69
8024
100.61
150.99
155.11
138.85
146.32
158.16
140.78
138.77
163.79
129.73
160.37
139.83
154.04
Sports/
Exercise
121.72
135.78
105.08
11251
142.13
115.6
89.19
136.07
110.34
118.4
123.97
110.44
149.97
115.56
125.66
129.14
11228
Eating/
Drinking
73.58
74.32
72.86
94.38
62.37
70.42
91.72
75.78
75.99
69.8
74.93
70.52
8128
74.85
73.45
74.56
71.43
Note: Subgroups are for gender, age, U.S. census region, day of week, and season. See Section 4 for a discussion
6-16
-------
Table 6-6. The Weighted Number of NHAPS Respondents (Doer Sample Size N) in Each Exposure
Activity on the Diary Day by Each Subgroup
Doer Sample
SizeN
Overall
Males
Females
0-4
5-17
17-64
65+
Northwest
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
No
Exposure
9196
4479.7
4716.3
679.1
1674.2
5692.9
1150
1842.6
2242.5
3237.4
1873.5
6567
2629.2
2300.5
2300.5
2301
22944
Cooking
3860.1
1272.8
2587.3
25.6
246.6
2930.9
657.1
776.8
967
1291.1
825.2
2850
1009.8
1001.3
938.4
954.7
965.7
Dishes/
Laundry
1705.9
290.2
1415.7
17.2
94.1
1252.9
341.8
388.7
405.9
555.4
355.9
1229
476.9
449.4
392.5
428.2
435.8
House-
keeping
1857
388.7
1468.3
24.2
172.6
1273.1
387.1
407.2
416.4
635.6
397.7
1343
514
426.4
429
480.9
520.6
Bathing
6270.7
3037.6
3233.1
461.9
1112.3
4108.2
588.3
1257.4
1481
2284.7
1247.6
4560
1711.2
1613.9
1597.6
1565
1493.8
Yardwork/
Maintenance
1332.3
826
506.3
18.6
94.7
952.6
266.3
259.1
3082
426
338.9
881.2
451
249
420.5
398.5
264.2
Sports/
Exercise
1893.8
1026.2
867.6
159.7
635.1
918.1
180.9
402.8
395.4
673.9
421.6
1354
540.1
398.6
518.6
570.7
405.9
Eating/
Drinking
8453.8
4151.2
4302.6
663.5
1569.6
5121.7
1099
1714,5
2071 .7
2943.1
1724.6
6053
2400.4
2128.9
2009.9
2121
2104.1
Note: Subgroups are for gender, age, U.S. census region, day of week, and season. See Section 4 for a discussion
of weighting. The doer sample sizes are equal to the number of respondents entering each location on the
diary day. The maximum number in each location is the total weighted sample size (N = 9196).
I
u
Housekeeping
Dishes/
Clean Kitch
Food
Prepartion
Food Preparation
Housekeeping
Yard/Mainten.
Eating/Drinking
Dishes/Clean Kitch.
Bathing
Exposure Activity
Sports/Excercise
Figure 6-15. The weighted percentage of time spent in each exposure activity for males vs.
females.
6-17
-------
Food Preparation
Dishes/Clean Kltch.
Housekeeping
| 36.97
37.81
61.03
59.02
Sports/Exercise
Eating/Drinking
Males
| Weighted
[ ] Unweighted
Females
H Weighted
| | Unweighted
115.25
109.42
116.26
121.05
170.26
" [174.84
135.78
130.67
74.32
| 75.83
72.86
I 74.08
Food Preparation
Dishes/Clean Kitch.
Housekeeping
Bathing
Yard/Mairiten.
Sports/Exercise
Eating/Drinking
Very Young ^ Working
School Age Q Retired
Yard/Malnten.
Bathing
Housekeeping
Dishes/Clean Kltch
Food Preparation
0 50 100 150 200
Mean 24-hour Cumulative Duration (minutes)
Figure 6-16. The weighted and unweighted mean 24-hour cumulative
durations in each exposure activity for males vs. females.
30 40
Percent
Figure 6-17. The weighted percentage of time spent in each exposure activity
for each age group.
-------
Food
21.2
2279
! 56.46
3 53.71
52.09
62.37
60.5
' 49.3
Kitch.
| 6257
63.24
HimiHMII Illllllllllll 67.68
63.44
• Weighted
D Unweighted
V, Weighted
D Unweighted
18-64 H Wei9hted
D Unweighted
65+
We*9hted
Unweighted
74.09-
Housekeeping
////S////////// 62.44
63
120.41
120.42
128.83
12822
Location
Bathing i
024
1 93.17
Yard/Mainten.
'109.55
150.99
15021
155.11
: 149.3
112.51
118.98
Sports/Exercise
144.58
IIIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIE
! 115.6
109.69
82.05
.19
Eating/Drinking
iiiiiiiiiiimiiiiiiimimiiiiiiiiiiiiiiii
91.72
.191.7
0 50 100 150 200
Mean 24-hour Cumulative Duration (min.)
Figure 6-18. The weighted and unweighted mean 24-hour cumulative
durations in each exposure activity for each age group.
Differences in percentages of time spent in each exposure activity are less than 1% across the
Census regions (Figure 6-19) except that: (1) 10.4% of the time is spent Bathing in the South in
comparison to 8-9% elsewhere; (2) 10% of the time is spent in Yardwork/Maintenance in the Northeast
with 1 1% in the Midwest and South and 13% in the West; (3) 1 1% of the time is spent in Sports/Exercise
in the Midwest with 13-15% spent elsewhere; and (4) 39% of the time is spent Eating/Drinking in the
Midwest with 34-36% spent elsewhere. The only differences in mean 24-hr duration (Figure 6-20) th<
exceed 10-11 min are for: (1) Yardwork/Maintenance between the South (158 min) and the other regions
(139-141 min); and (2) Sports/Exercise between the Northeast (136 min) and the other regions (110-124
min).
6-19
-------
Food Preparation
Dishes/Clean Kltch.
Housekeeping
Bathing
Yard/Mainten.
Sports/Exercise
Eating/Drinking
Figure 6-19. The weighted percentage of time spent in each exposure
activity for each census region.
Food Preparation
Dishes/Clean Kilch.
Housekeeping
Bathing
Yard/Mainten.
Sports/Exercise
Eating/Drinking
M.»r
iiimiiiiiiimiiiimiiw ",™
19.9\
Northeast
• Weighted
I I Unwalghled
Midwest
V) Weighted
! I UnwdlgMod
South
n VKHghlod
I I Unwetghlod
West
III Wolghtod
i I Unwolghlod
iiiiiiiiiiimiiiiiiiiimmiiiiiX'
'••I HI
'
~| 119.23
,
1 117.85
'-] 121.5!
Illlllllllllllllllllllllllllllllllllllllllllllllllllllllll!
iiiiiiiiiiiiimmmiiiiiiiiiiiiiiimiiiiiimiiimmuiiiiiiiiiiiii
T M».
I flttST "
105.69
11111 1
IlllllllllUlllllllllllimilllllllllllllllllllllllllllllllllllll 12307
J 116.95
50 100 150
Mean 24-hour Cumulative Duration (minutes)
200
Figure 6-20. The weighted and unweighted mean 24-hour cumulative
durations In each exposure activity for each census region.
-------
The exposure activities that had percentage of time differences that were greater than 2% between
days of the week falling on Weekdays vs. Weekends (Figure 6-21) were Food Preparation (12.5% WD,
10% WE) and Yardwork/Maintenance (10% WD, 12% WE). The 24-hr durations of all exposure
activities (Figure 6-22) were smaller on the Weekdays than on the Weekends. The greatest differences
occurred for Housekeeping (107 min vs. 139 min), Yardwork/Maintenance (139 min vs. 164 min), and
Sports/Exercise (110 min vs 150 min).
Differences in percentage of time spent tend to occur between the Winter/Fall (WF) and
Spring/Summer (SS) season groupings (Figure 6-23). The largest differences were for Yardwork/
Maintenance (8-10% WF, 12-15% SS) and Sports/Exercise (11% WF, 14-16% SS). The most House-
keeping occurred in the Fall (15% vs 11-12% for the other seasons) and the highest percentage of time
spent Eating/Drinking occurred in the Winter (39% vs. 33-37% for the other seasons). The largest mean
24-hr durations (Figure 6-24) for Yardwork/Maintenance were for the Spring (160 min) and the Fall (154
min) seasons (130-140 min for Winter/Summer). The largest durations for Sports/Exercise were for the
Spring/Summer season (126-129 min vs. 112-116 min for Winter/Fall). Differences in mean duration
between the four seasons did not exceed 10-11 min for the other exposure activities.
Food Preparation Housekeeping Yard/Mainten. Eating/Drinking
Dishes/Clean Kitch. Bathing Sports/Exercise
Exposure Activity
Figure 6-21. The weighted percentage of time spent in each exposure activity for weekdays vs. weekends.
6-21
-------
Food Preparation
Dishes/Clean Kilch.
Housekeeping
Bathing
Weekday
j§ Weighted
Unweighted
Weekend
Weighted
Unweighted
Sports/Exercise
Eating/Drinking
10379
149.97
136.03
70.52
71.21
S3 81.28
82.47
0 50 100 150
Mean 24-Hour Cumulative Duration (minute*)
Figure 6-22. The weighted and unweighted mean 24-hour cumulative
durations in each exposure activity for weekdays vs.
weekends.
200
Food Preparation
Dishes/Clean Kitch.
Housekeeping
Bathing
12.42
"11.66
IOB2
) 12.41
I Winter ^ Spring
I Summer [ | Fall
Housekeeping
Bathing
Dishes/Clean Kllch
Food
Preparation
Percent
Figure 6-23. The weighted percentage of time spent in each exposure
activity for each season.
-------
Food Preparation
Winter
• Weighted
_J Unweighted
Spring
Weighted
d Unweighted
Summer
1 Weighted
J Unweighted
Fall
I Weighted
LJ Unweighted
Dishes/Clean Kitch.
Housekeeping
.0
Bathing
//////////////////////}. 57.59
i^^X^^^^fri^/'^^
Yard/Mainten.
y//////////////////W/////y/////////^^^ 160.37
I 162.23
J 139.83
137.92
| 154.04
149.97
Sports/Exercise
115.56
104.09
125.66
123.45
j 129.14
Eating/Drinking
50 100 150
Mean 24-hour Cumulative Duration (minutes)
200
Figure 6-24. The weighted and unweighted mean 24-hour cumulative durations in each exposure activity
for each season.
6-23
-------
6.3 Smoker-Present Categories
The overall percentage of people exposed to ETS for at least one minute on the diary day was 45%
(Figure 6-25). More male respondents were exposed (46%) than females (44%), and individuals in the
School Age (5-17) and Working (18-64) age groups were more exposed (41% and 51%, respectively)
than the Very Young (0-4) and Retired (65+) age groups (34% and 28% exposed, respectively). Those in
the Northeast had the highest percentage of people exposed (49%), followed closely by the Midwest
(48%) and the South (45%) with the West having the lowest exposure (38%). There were slightly more
people exposed on the weekends (46%) than on weekdays (44%). The Winter, Spring, and Summer
seasons all had similar numbers of people exposed (43-44%) while in the Fall season substantially more
people were exposed (49%).
The overall percentage of time over the diary day that respondents were exposed to ETS — including
only those that were exposed at least one minute on the diary day - was 26% (Figure 6-26). Males were
exposed for a larger percentage of time (28%) than females (24%). The age group with the largest
percentage of time exposed was the Working (28%), followed by the Very Young (26%), the Retired
(24%), and the School Aged (18.5%). The South had the largest percentage of time exposed (28%),
while the other regions had about the same exposure (24-25%). On weekends the percentage of time was
greater (28%) than on weekdays (25%), while each season had about the same percentage of time
exposed to ETS (25-26.5%).
The mean 24-hr cumulative durations of ETS exposure (371 min overall) follow a similar pattern as
the percentage of time respondents were exposed (Figure 6-27). Males durations were longer than
females (397 min vs. 345 min); and of any age group, the Working were exposed the longest (402 min),
but they were followed closely by the Retired (382 min) and the Very Young (360 min) with the School
Ages having the least exposure durations (261 min). The South had larger durations (404 min) than the
other regions (345-363 min), weekend durations were longer (395 min) than on weekdays (362 min), and
each of the seasons all had similar durations (359-378 min).
6-24
-------
I I
' >,
50.5
10 20 30
-HI
50 60
Percentage of People Exposed to
ETS on the Diary Day
Figure 6-25. The weighted percentage of people that
were exposed to ETS for at least one
minute on the diary day according to
each subgroup. The weighted sample
size was 4,951.5 for those who were
exposed and 4025.9 for those who were
never exposed on the diary day.
Overall
Males
Females
0-4
5-17
18-64
65+
ET
e
j»> Midwest
w South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
18.49
| 26 02
| 27.86
24.16
25.45
| 28.06
| 23.64
24.44
25.36
• 28.35
24.23
25.33
• 27.67
| 26.45
| 26.08
| 26.51
25.15
Overall
Males
Females
0-4
5-17
18-64
65+
§• Northeast
£> Midwest
m South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
0 5 10 15 20 25 30 35
Percent Time Spent Exposed to ETS on
the Diary Day
Figure 6-26. The weighted percentage of time that
those who were exposed to ETS on the
diary day were exposed -- for each
subgroup. Total Time = 5,747,716 min.
100 200 300 400 500
Mean 24-Hour Cumulative Duration of
ETS Exposure (minutes)
Figure 6-27. The weighted mean 24-hour cumulative
durations of ETS exposure for those who
were exposed for at least one minute on
the diary day -- for each subgroup.
-------
6.4 Locations x Smoker-Present Categories
Percentage of time spent exposed to ETS is dependent on both the length of exposure
microenvironments and the number of people being exposed, whereas 24-hr durations are dependent on
the length of each exposure microenvironrnent and the number of times it occurs on the diary day. Of
those that were exposed to ETS for at least one minute on the diary day, the location with the largest
overall percentage of time spent being exposed (based on total time spent exposed, i.e., not including
time spent not being exposed) was Residential-Indoors (48%) (Figure 6-28; Table 6-7 contains the
percentage of time spent across all subgroups). The next largest percentages of time spent exposed were
for the Office/Factory (9.7%), Bar/Restaurant (8.8%), and In Vehicle (7%) locations. The two locations
with the smallest percentage of time spent exposed to ETS were Mall/Other Store (3.5%) and
School/Public Bldg (3.9%). In other locations respondents were exposed between 4% and 5.5% of the
time. The largest overall 24-hr duration of ETS exposure (Figure 6-28; Table 6-8 contains 24-hr
durations across all subgroups) occurred in the Office-Factory location (363 rain), followed by
Residential-Indoors (305 min), Other Indoor (256 min), School/Public Bldg (249 min), Other Outdoor
(247 min), Mall/Other Store (198 min), Residential-Outdoor (178 min), Near Vehicle (160 min),
Bar/Restaurant (143 min), and In Vehicle (79 min). The largest mean number of occurrences (Figure 6-
28; Table 6-9 contains the frequency of occurrence across all subgroups) was in the Residential-Indoors
location (3) with the next closest location being In Vehicle (2). The next two highest mean number of
occurrences were for the Office/Factory (1.6) and the Near Vehicle (1.5) locations. Number of
occurrences in all the other locations fell between 1.1 and 1.4. Table 6-10 contains the doer sample size
across all subgroups.
Females were exposed for a larger percentage of the time than males (Figure 6-29) in the
Residential-Indoors (56% F, 41% M), Mall/Other Store (4% F, 3% M), and School/Public Bldg (5% F,
3% M) locations. But males had substantially larger percentages in the Near Vehicle (7% M, 1% F),
Other Outdoor (8% M, 3% F), Office/Factory (12% M, 7% F), and Other Indoor (5% M, 3% F)
locations. The percentages of time spent were about the same for males and females for the Residential-
Outdoor (5%), ID Vehicle (7%), and Bar/Restaurant (9%) locations. The largest differences in mean 24-
hr durations of ETS exposure between males and females (Figure 6-30) occurred for the Other Outdoor
(299 min M, 163 min F), Near Vehicle (215 min M, 65 min F), and Residential-Outdoor (207 min M, 150
min F). There were also differences for Other Indoor (271 min M, 227 min F) and School/Public Bldg
(222 min M, 273 min F). Differences in 24-hr duration were within 20 min for all the other locations.
The mean 24-hr number of occurrences of ETS exposure for each location were about the same for
females and males (Figure 6-30).
The Very Young age group had the largest percentage of time exposed to ETS in the Residential-
Indoor (82%) and Residential-Outdoor (9%) locations, while the Working age group had the least (41%
and 4.5%, respectively) (Figure 6-31). The School Aged and Retired had nearly the same percentages for
Residential-Indoors (63-67%). The Very Young had the least percentage of time exposed to ETS in the
In Vehicle (4% vs. 6-7% for the other age groups), Near Vehicle (1%), Other Outdoor (1% vs. 5-6%),
Office Factory (0%), Mall/Other Store (1% vs. 2-4%), School/Public Bldg (0% vs. 2-5%), Bar Restaurant
(2.5%), and Other Indoor (0.2%) locations. As expected, the Working and Retired age groups had the
largest percentage of time for Office/Factory (12% and 6%, respectively) and Bar/Restaurant (8-10%).
24-hr durations of ETS exposure (Figure 6-32) were approximately the same for the Very Young and the
Retired in the Residential-Indoor location (365-381 min vs. 257-303 for School and Working), whereas
the Very Young had the longest 24-hr durations in the Residential-Outdoors location (222 min vs. 166-
6-26
-------
177 min for the others). Durations were substantially larger for the Working/Retired than the Very
Young/School Aged in the Near Vehicle (125-189 min vs. 35-60 min for Very Young/School Aged),
Other Outdoor (277-281 min vs. 94-150 min), and Office/Factory (356-363 min vs. 0 min) locations. The
Working had the longest exposures in the Mall/Other Store (232 min vs. 101-128 for the others),
Bar/Restaurant (154 min vs. 98 min for the others), and Other Indoor (293 min vs. 96-150 min for the
others) locations. The duration of exposures in the School/Public Bldg locations were small for the Very
Young (5 min) with the other age groups being fairly closely spaced (169-265 min). Most of the 24-hr
mean number of occurrences of exposure to ETS (Figure 6-33) were closely spaced for each age group.
However, the Working/Retired had a slightly higher mean number of occurrences in the Residential-
Indoors (3.1-3.3 vs. 2.7 for the Very Young/School Aged) and In Vehicle (2.3-2.4 vs. 1.94-1.97)
locations, the Very Young had the highest number of occurrences in the Near Vehicle location (1.9 vs. 1-
1.5 for the others), and, as expected, the Working had the highest number of occurrences in the
Office/Factory location (1.6 vs. 1.1 for Retired and Of or Very Young/School Aged).
The percentage time spent being exposed to ETS for the Residential-Indoors locations is smallest for
the West (45%) with the others regions having approximately the same percentage (49%) (Figure 6-34).
The West also has the lowest percentage for the In Vehicle (5% vs. 6-8% for the other regions) and
Office/Factory (8% vs. 10-11% for the other regions) locations. The Midwest and West have the largest
percentage of time exposed to ETS in the Bar/Restaurant location (10-11% vs. 7-8% for the
Northeast/South), and the Northeast had the largest percentage in the School/Public Bldg location (6%
vs. 3-4% for the other regions). The mean 24-hr durations of ETS exposure in the Near Vehicle location
was substantially larger for the South and West (212-237 min) than for the Northeast and Midwest (99
min) (Figure 6-35). The West had the smallest durations for the Residential-Outdoor, Other Outdoor,
Office-Factory, and Other Indoor locations, but the largest durations in the Mall/Other Store location.
The Northeast had the largest durations in the Other Outdoor and School/Public Bldg locations. The
mean 24-hr number of occurrences were very closely clustered across all regions (Figure 6-36).
The percentage of time exposed to ETS and exposure durations were substantially larger on
weekdays than on weekends (Figure 6-37 and 6-38) for the Office/Factory (13%, 268 min vs. 2%, 302
min), School/Public Bldg (5%, 368 min vs. 2%, 178 min), and Near Vehicle (6%, 179 min vs. 2%, 92
min) locations, whereas the percentages and durations were larger on weekends for the Residential-
Indoors (54%, 352 min vs. 45%, 284 min), Residential-Outdoors (8%, 212 min vs. 3.5%, 153 min), Other
Outdoor (8%, 284 min vs. 4%, 224 min), and Bar/Restaurant (10%, 150 min vs. 8%, 139 min) locations.
The number of occurrences of ETS exposure on weekends and weekdays were about the same for each
location (Figure 6-38).
The Spring/Summer seasons have lower percentages of time spent exposed to ETS than Winter/Fall
in the Residential-Indoors location (40-46% vs. 51-55%), whereas Spring/Summer have higher
percentages than Winter/Fall in the Residential-Outdoor (7.5% vs. 3%) and Other Outdoor (6-10% vs. 2-
4%) locations (Figure 6-39). Other locations do not have percentages of exposure to ETS that are very
different across the seasons. Winter has the largest mean 24-hr duration of exposure to ETS (Figure 6-
40) in the Residential-Indoors location (341 min), while summer has the smallest duration (263 min).
Winter also has the largest duration in the Residential-Outdoor location (219 min vs. 166-180 for the
other seasons), and the largest mean number of occurrences of ETS exposure (Figure 6-41) for the
Residential-Indoors (3.5) and Office/Factory (1.9) locations.
6-27
-------
Percent Time Spent/No. Occurrences
0 10
Residential-Indoors
48.06
5.01
Residential-Outdoor 1.39
305.11
In Vehicle 2.32
Near Vehicle
178.22
7.01
78.65
4.45
5.49
Other Outdoor 1.34
159.93
I % Time
I I No. Occurr.
24-Hr Duration
Office/Factory
I 3 54
Mall/Other Store 1.14
I 3 9
School/Public Bldg.
247.15
9.68
362.98
198.35
Bar/Restaurant
Other Indoor
248.96
8.77
142.95
255.46
100 200 300 400
24-Hour Duration
500
Figure 6-28. The overall weighted percentage of time spent, mean number of occurrences, and mean 24-hour
cumulative durations (in minutes) of ETS exposure in each location.
6-28
-------
Table 6-7. The Weighted Percentages of Time Spent in Each Location x Smoker-Present Microenvironment on the
Diary Day Across Each Subgroup
Percentage
of Time
Overall
Males
Females
0-4
5-17
17-64
65+
Northwest
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fa"
Residential
- Indoors
48.06
40.87
56.44
81.88
67.3
41.37
62.68
48.53
48.97
48.57
44.74
45.23
54.23
55.29
45.69
39.74
51.16
Residential
- Outdoor
5.01
5.38
4.58
8.68
7.01
4.47
4.73
4.18
3.98
6.27
4.68
3.53
8.24
2.5
7.51
7.46
2.76
In
Vehicle
7.01
7.27
6.7
3.67
6.62
7.43
5.66
6.41
7.59
7.7
5.16
6.97
7.09
5.84
7.82
7.02
7.32
Near
Vehicle
4.45
7.02
1.44
0.64
1.36
5.47
1.38
3.8
2.64
5.14
6.5
5.68
1.75
4.39
3.21
4.72
5.41
Other
Outdoor
5.49
7.64
2.48
1.26
5.53
5.85
4.79
5.63
5.5
4.99
6.52
4.4
7.88
2.43
6.35
9.53
3.85
Office/
Factory
9.68
11.99
7
0
0
12.23
5.45
10.74
9.69
9.87
7.87
13.14
2.12
9.48
9.86
10.35
9.09
Mall/
Other Store
3.54
3.01
4.16
1.24
2.51
4
2.08
3.14
3.13
3.73
4.27
3.72
3.15
2.9
3.96
3.75
3.55
School/
Public Bldg.
3.9
3.06
4.87
0.004
4.57
4.22
2.37
5.7
3.68
3.01
4.07
4.82
1.88
4.13
4.61
3.58
3.3
Bar/
Restaurant
8.77
8.5
9.07
2.46
4.1
9.99
8.03
7.58
10.35
7.39
11.09
8.05
10.33
9.13
7.98
8.79
9.14
Other
Indoor
4.1
5.25
2.76
0.17
0.99
4.97
2.83
4.28
4.48
3.34
5.1
4.45
3.33
3.91
3.01
5.06
4.41
Note: Subgroups are for gender, age, U.S. census region, day of week, and season.
The percentage of time spent is over all respondents (doers + non-doers).
See Section 4 for a discussion of weighting.
Table 6-8. The Mean Weighted 24-Hour Duration (Minutes) in Each Location x Smoker-Present Microenvironment
Doer
Duration
D
Overall
Males
Females
(M
5-17
17-64
65+
Northwest
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
Residential
305.11
301.64
308.11
38157
257.49
30356
365.47
292.72
304.82
313.88
301.37
284.17
352.51
341.61
300.59
263.27
310.75
Residential
-Outdoor
178.22
206.99
149.69
221.56
165.52
176.75
17757
184.77
17755
187.62
149.06
152.59
211.53
21952
170.91
179.91
165.86
In
Vehicle
78.65
87.3
69.87
57.03
6252
82.8
77.3
74.84
81.37
82
68.7
77.26
81.82
69.23
90.31
77.49
77.62
Near
Vehicle
159.93
214.94
65.18
60.1
35.3
188.87
125.12
98.7
98.92
237.5
212.13
178.69
91.66
166.49
139.85
155.92
171.98
Other
Outdoor
247.15
29952
162.61
94.59
149.71
27756
280.92
288.45
237.04
255.78
21254
223.57
283.66
271.09
229.37
268.52
221-59
Office/
Factory
362.98
369.82
350.04
0
0
363.29
355.78
365.54
370.86
371.59
322.8
368.43
302.4
360.19
421.85
33655
345.79
Mall/
Other Store
198.35
200.35
196.69
128.32
101.68
232.19
105.03
202.35
171.63
198.91
235.56
209.62
1745
167.51
209.05
250.11
178.05
School/
Public Bldg.
248.96
221.76
273.54
5
206.98
2655
169.02
297.39
237.91
217.95
256.01
268.18
177.58
204.42
274.8
334.59
221.76
Bar/
Restaurant
142.95
142.24
143.74
98.22
98.09
153.79
98.43
134.94
141.38
14853
144.63
138.88
150.48
153.67
131.75
145.8
141.03
Other
Indoor
255.46
271.07
226.45
150
96.36
293.41
104.18
257.09
265.96
258.38
236.47
257.53
249.59
259.17
211.69
333.96
226.08
Note: Subgroups are for gender, age. U.S. census region, day of week, and season. See Section 4 for a discussion of weighting.
6-29
-------
Table 6-9. The Weighted Number of Occurrences of Each Location x Smoker-Present Microenvironment (for Doers) on
the Diary Day by Each Subgroup
Doer
Frequency
O
Overall
Males
Females
0-4
5-17
17-64
65+
Northwest
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
faR
Residential
- Indoors
3.02
2.69
3.3
2.74
2.74
3.1
3.34
2.94
2.95
3,16
2.91
2.98
3.12
3.48
2.8
2.68
3.09
Residential
- Outdoor
1.39
1.47
1.32
1.37
1.43
1.38
1.48
1.39
1.38
1.44
1.29
1.39
1.39
1.3
1.38
1.47
1.31
Note: Subgroups are for gender, age, U.S
In
Vehicle
2.32
2.31
2.33
1.97
1.94
2.41
2.3
2.09
2.58
2.32
2.14
2.36
2.24
2.4
2.28
2.21
2.39
Near
Vehicle
1.5
1.5
1.51
1.91
1.36
1.54
1.07
1.42
1.34
1.53
1.82
1.56
1.-3
1.6
1.43
1.4
1.56
Other
Outdoor
1.34
1.36
1.3
1
1.32
1.36
1.33
1.53
1.33
125
1.32
1.36
1.3
1.51
1.19
1.37
1.39
Office/
Factory
1.61
1.59
1.65
0
0
1.63
1.13
1.56
1.67
1.56
1.71
1.61
1.56
1.92
1.35
1.59
1.55
. census region, day of week, and season.
Mall/
Other Store
1.14
1.18
1.11
1
1
1.18
1.1
1.12
1.19
1.09
1.21
1.18
1.05
1.06
1.2
1.16
1.14
School/
Public Bldg.
1.21
1.16
1.26
1
1.07
1.26
1.02
1.36
1.11
1.19
1.18
1.25
1.06
1.24
1.2
1.24
1.16
Bar/
Restaurant
1.26
1.28
1.23
1.27
1.04
1.27
1.34
1.2
1.27
1.31
1.22
1.23
1.31
1.27
1.18
1.35
1.24
Other
Indoor
1.31
1.3
.34
.05
.38
.05
.29
1.38
1.37
1.16
1.34
1.24
1.32
1.26
1.49
1.21
See Section 4 for a discussion of weighting.
Table 6-10. The Weighted Number of NHAPS Respondents (Doer Sample Size N) in Each Location x Smoker-Present
Microenvironment on the Diary Day by Each Subgroup
Dow
Sample
SizeN
Overal
Males
Females
0-4
5-17
17-64
65+
Northwest
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fal
Residential
-Indoors
2355.6
1091.3
1264.3
173.7
449.2
1549.8
182.9
507
609.7
887.3
351.7
1633.6
721.9
599.8
557.9
550
647.9
Residential
-Outdoor
420.3
2092
211
31.7
72.8
287.3
28.4
69.1
85.3
191.5
74.4
237.6
182.7
42.3
161.3
151.1
65.5
In
Vehicle
13322
670.6
661.6
52
182.9
1019.1
78.1
261.8
354
538.5
177.9
925.7
406.5
312.6
318
330.3
371.3
Near
Vehicle
416
2632
152.8
8.6
66.3
329.3
11.8
117.9
101.4
124.1
72.6
326.3
89.7
97.7
842
110.4
123.7
Other
Outdoor
332.4
205.7
126.7
10.8
63.5
239.9
182
59.7
88
111.9
72.8
202
130.4
332
101.6
129.3
68.3
Office/
Factory
399
261
137.9
0
0
382.6
16.3
89.9
99.1
1522
57.8
366.1
32.9
97.5
85.8
1122
103.5
Mall/
Other Store
266.9
120.9
146
7.8
42.4
195.6
21.1
47.4
69.1
107.5
42.9
182
84.9
64.3
69.6
54.7
78.4
School/
Public Bldg.
234.1
111.1
123
0.6
37.9
180.6
14.9
58.6
58.7
792
37.6
184
49.6
74.9
61.6
39
58.6
Bar/
Restaurant
9172
481.5
435.7
20.3
71.7
7382
87
171.8
277.7
286
181.7
5952
322.1
220.1
222.4
219.6
255.2
Other
Indoor
240
156
84
0.9
17.7
192.4
29
50.9
63.9
74.1
51.1
177.6
62.6
55.9
522
552
76.8
Note: Subgroups are for gender, age, U.S. census region, day of week, and season. See Section 4 for a discussion of weighting.
The doer sample sizes are equal to the number of respondents entering each location on the diary day. The maximum
number in each location is the weighted sample size (N = 9196).
6-30
-------
70-
60
Ill
o 50
TJ
-------
Percent Time Spent/No. Occurrences
Residential-lndoors
Residential-Outdoor
In Vehicle >
Near Vehicle
5
"re
o
o
Other Outdoor
Office/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
1.36
Males
I No. Occurrences
j 24-Hr Duration
Females
1 No. Occurrences
24-Hr Duration
299.22
369.82
350.04
100 200 300
24-Hour Duration
400
500
Figure 6-30. Weighted mean 24-hr number of occurrences and weighted mean 24-hr durations of ETS
exposure in each location for males vs. females.
6-32
-------
Resldentlal-lndoors
Residential-Outdoor
In Vehicle p
Near Vehicle
I1
1.28
~1 5.53
Other Outdoor IS! 5.85
o
S Office/Factory
Very Young 0-4
School Agod 5 17
Wording 18-54
Retired 65*
1 24
2.51
Mall/Other Store U 4
School/Public Bldg
Bar/Restaurant
Other Indoor
20 40 60
Percent Time Spent
Figure 6-31. Weighted percentage of time spent
exposed to ETS in each location by age
group.
Residential-lndoors
Residential-Outdoor
In Vehicle
Near Vehicle
C Other Outdoor
3S1.27
, 30326
] 365.47
Very Young 0-4
• School Agod 5-17
D Working 18-64
I ] Rellrcd 65*
14971
277.26
280.92
363 29
[355.78
O Office/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
24-Hour Duration
Figure 6-32. Weighted mean 24-hour duration of ETS
episodes in each location by age group.
Residential-lndoors
Residential-Outdoor
In Vehicle
Near Vehicle
c Other Outdoor
O
1
O Office/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
01234
No. Occurrences
Figure 6-33. Weighted mean 24-hour number of
occurrences of ETS exposure in each
location by age group.
-------
Resldentlal-lndoore
Residential-Outdoor
In Vehicle
Near Vehicle
Other Outdoor
Office/Factory
w Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
0 10 20 30 40 50
Percent Time Spent
Figure 6-34. Weighted percentage of time spent
exposed to ETS in each location by
census region.
Resldential-lndoors
Rosldentlal-Ouldoor
In Vehicle
Near Vehicle
School/Public Bldg.
385.W
370.8*
371.69
322.8
2IT.9S
258.01
Other Indoor
9EHHHRBBBHHHH9HI
25709
20596
25838
_ I 23«.47
Resldentlal-lndoors
Residential-Outdoor
In Vehicle
Near Vehicle
Other Outdoor
OHice/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
100 200 300 400
24-Hour Duration
500
Figure 6-35. Weighted mean 24-hour duration of ETS
episodes in each location by census
region.
o 0.5 1
No. Occurrences
Figure 6-36. Weighted mean 24-hour number of
occurrences of ETS exposure in each
location by census.
-------
60
,>
in
• K
50H
LU
O
Q)
(f)
O
CL
X
LU
4->
CD
a
(f)
Q)
40
30-
20
a>
10
54.23
45.23
Weekday
Residential-lndoors
Residential-Outdoor
In Vehicle
Near Vehicle
Other Outdoor
Office/Factory
Other Indoor
Bar/Restaurant
School/Public Bldg
Mall/Other Store
13.14
8.24
3.51
3.72 3.15
4.82
Weekend
10.33
4.45
3.33
Residential-lndoors I In Vehicle Other Outdoor Mall/Other Store Bar/Restaurant
Residential-Outdoor Near Vehicle Office/Factory School/Public Bldg. Other Indoor
Location
Figure 6-37. Weighted percentage of time spent exposed to ETS in each location by weekend vs. weekday.
-------
Residential-lndoors !
Residential-Outdoor
In Vehicle
Near Vehicle
.2 Other Outdoor
s
c
Office/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
No. Occurrences
0.5 01 1.5 2 2.5
3.5
J 352.51
2.36
2.24
233.57
1.3
283.66
Weekday
• No. Occur.
• 24-Hour Dur.
Weekend
• No. Occur.
I 24-Hour Dur.
368.43
302.4
1.18
209.62
268.18
1.24
257.53
249.59
100
200 300
24-Hour Duration
400
500
Figure 6-38. Weighted mean 24-hr number of occurrences and weighted mean 24-hr durations of ETS
exposure in each location for weekends vs. weekdays.
6-36
-------
Residenlial-lndoors
Residential-Outdoor
In Vehicle
Near Vehicle
Other Outdoor
Office/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
0 10 20 30 40 50 60 70
Percent Time Spent
Figure 6-39. Weighted percentage of time spent
exposed to ETS in each location by
season.
Residential-lndoors
Residential-Outdoor
In Vehicle
Near Vehicle
Other Outdoor
Office/Factory F
Other Indoor
(66.49
139.85
ii.JlS5.92
1 171.98
Mall/Other Store B ^3'250.11
~T 178.05
School/Public Bldg.
Bar/Restaurant
100 200 300 400
24-Hour Duration
500
Figure 6-40. Weighted mean 24-hour duration of ETS
episodes in each location by season.
Residential-lndoors
Residential-Outdoor
In Vehicle
Near Vehicle
c
0
Other Outdoor
° Oflice/Factory
Mall/Other Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
01234
No. of Occurrences
Figure 6-41. Weighted mean 24-hour number of
occurrences of ETS exposure in each
location by season.
-------
6.5 Locations x Activities
Only the 21 location x activity microenvironments that had more than 100 occurrences over all the
respondents' diary days were analyzed. Of the total time spent by all the weighted respondents on the
diary days (9,196 x 1440 min = 13,242,240 min), the location x exposure activity microenvironment with
the largest percentage of time (Figure 6^*2 and Table 6-11) is Eating/Drinking-Residential-Indoors
(3.4%) - ignoring the time spent in the No Exposure category (Sleeping, Watching TV, Reading, etc.)
which accounts for approximately 87% of the total time. The next largest percentages are for
Housekeeping-Residential-Indoors (1.50%), Food Preparation-Residential-Indoors (1.49%),
Bathing/Hygiene-Residential-Indoor (1.15%), Yardwork/Maintenance-Residential-Outdoor (1.08%),
Eating/Drinking-Bar/Restaurant (0.87%), Dishes/Laundry/Clean Kitchen-Residential-Indoor (0.74%),
Sports/Exercise-Other Outdoor (0.6%), and Sports/Exercise-Residential Outdoor (0.47%). The
respondents spent 0,29% or less in each of the remaining microenvironments.
10. Food Preparation
20. Dishes/Clean Kitch.
30. Housekeeping
40. Bathing
50. Yard/Mainten. Activity
60. Sports/Exercise
70. Eating/Drinking
Location
Figure 6-42. 3-D plot of the overall weighted percentage of time spent in 21 out of 70 location x
activity microenvironments. The missing microenvironments had less than 100
occurrences over the diary days of all the respondents.
6-38
-------
Table 6-11. The Overall Weighted Percentage of Time Spent in 21 Location x Activity Microenvironments
Activity Location
10. Residential-Indoor
20. Residential-Outdoor
30. In Vehicle
40. Near Vehicle
50. Other Outdoor
60. Office/Factory
70. Mall/Other Store
80. School/Public Bldg.
90. Bar/Restaurant
100. Other Indoor
10. 20. 30.
Food Dishes/Clean House-
Preparation Kitchen keeping
1 .49 0.74 1 .5
0.1
40. 50.
Yard/
Bathing Maintenance
1.15 0.29
1.08
0.04
0.06
60.
Sports/
Exercise
0.13
0.47
0.21
0.6
0.2
70.
Eating/
Drinking
3.42
0.07
0.05
0.08
0.11
0.87
0.05
Note: The missing percentages were for the 49 microenvironments with less than 100 occurrences overall.
The location x activity microenvironments that had the largest percentage of people experiencing
them at least once on the diary day were Residential-Indoor-Eating-Drinking (84%), Residential-Indoor-
Bathing (67%), and Residential-Indoor-Food Preparation (41%) (Figure 6-43 and Table 6-12). Notice
that while more people were Bathing than Preparing Food, the overall percentage of time spent was not
as great.
70. Eating/Drinking
60. Sports/Exercise
50. Yard/Mainten
40. Bathing ^
Activity 30. Housekeeping
20. Dishes/Clean Kitch"
10. Food Preparation
§ 8
Location
Figure 6-43. 3-D plot of the overall weighted (N = 9,196) percentage of doers (people
experiencing a microenvironment on the diary day) in 21 out of 70 location
x activity microenvironments. The missing 49 microenvironments had less
than 100 occurrences over the diary days of all the respondents.
6-39
-------
Table 6-12. The Overall Weighted Percentage of People Doing (the Doers) Each of 21 Location x Activity
Microenvironments on the Diary Day
Activity Location
10. Residential-Indoor
20. Residential-Outdoor
30. In Vehicle
40. Near Vehicle
50. Other Outdoor
60. Office/Factory
70. Mall/Other Store
80. School/Public Bldg.
90. Bar/Restaurant
100. Other Indoor
10. 20. 30.
Food Dishes/Clean House-
Preparation Kitchen keeping
41.24 18.07 19.06
1.34
40. 50.
Yard/
Bathing Maintenance
66.91 2.82
10.23
1.57
1.91
60.
Sports/
Exercise
3.21
5.94
4.81
5.93
2.63
70.
Eating/
Drinking
83.89
2.23
1.62
3.10
4.10
18.70
1.47
Note: The missing percentages were for the 49 microenvironments with less than 100 occurrences overall.
The location x activity microenvironment with the largest mean 24-hr duration (Figure 6-44 and
Table 6-13) is Yardwork/Maintenance-Residential-Outdoor (152 min), followed closely by
Yardwork/Maintenance-Residential-Indoor (149 min), and Sports/Exercise-Other Outdoor (146 min).
The four next largest durations were for Sports/Exercise-Residential-Outdoor (113 min),
Housekeeping/Residential-Outdoor (113 min), Sports/Exercise-Other Indoor (112 min), and
Housekeeping-Residential-Outdoor (108 min). The microenvironment with the smallest duration was
Bathing-Residential-Indoor (25 min). The remaining microenvironments had 24-hr durations that were
all between 30 and 70 min. Although more people experienced Bathing-Residential-Indoor on the diary
day than any other microenvironment except Eating/Drinking-Residential-Indoor, since its duration is so
small, its overall percentage of time spent is only the fourth largest.
10. Food Preparation
20. DishesOean Kitch.
30. Housekeeping
40. Baling
Activity SO. Yanl/Mainten.
60.
70. Eating Drinking
100. Other Indoor
90. Bar/Restaurant
80. School/Public Bid.
70. Mall/Other Store
60. Office/Factory
50. Other Outdoor
40. Near Vehicle Location
30. In Vehicle
20. Residential-Outdoor
10. Residential/Indoor
Figure 6-44. 3-D plot of the overall weighted mean 24-hour duration in 21 out of 70 location
x activity microenvironments. The missing 49 microenvironments had less
than 100 occurrences over the diary days of all the respondents.
6-40
-------
Table 6-13. The Overall Weighted Mean 24-Hr Duration in 21 Location x Activity Microenvironments
Activity Location
10. Residential-Indoor
20. Residential-Outdoor
30. In Vehicle
40. Near Vehicle
50. Other Outdoor
60. Office/Factory
70. Mall/Other Store
80. School/Public Bldg.
90. Bar/Restaurant
100. Other Indoor
10. 20. 30.
Food Dishes/Clean House-
Preparation Kitchen keeping
52.19 58.84 112.96
108.07
40. 50.
Yard/
Bathing Maintenance
24.82 149.06
151.92
32.59
41.65
60.
Sports/
Exercise
58.86
113.44
62.4
145.69
111.91
70.
Eating/
Drinking
58.71
47.67
42.5
37.51
38.1
66.96
45.77
Note: The missing percentages were for the 49 microenvironments with less than 100 occurrences overall.
The location x activity microenvironment with the largest mean 24-hr number of occurrences
(Figure 6-45 and Table 6-14) for all respondents over the 24-hr diary day was Eating/Drinking-
Residential-Indoor (1.9). The other two microenvironments with mean number of occurrences of 1.5 or
greater were Food Preparation-Other Indoor (1.6) and Dishes/Laundry/Clean Kitch-Residential-Indoor
(1.5). The microenvironments with mean durations between 1.2 and 1.3 were Housekeeping-Residential-
Indoor, Bathing-Residential-Indoor, Yardwork/Maintenance-Residential-Indoor, Housekeeping-
Residential-Outdoor, Yardwork/Maintenance-Residential-Outdoor, Sports/Exercise-Residential-Outdoor,
Sports/Exercise-Other Outdoor, and Eating/Drinking-Bar/Restaurant. The remaining ten
microenvironments had mean number of occurrences between 1 and 1.2.
i
10. Food Preparation
20. Dishes/Clean Kteh.
30. Housekeeping
40. Bathing
50. YaroyMainten
Activity ^ Sports/Exercise
70. Eating Drinking
100. Other Indoor
90. Bar/Restaurant
80. SchooWublic BW.
70. Mall/Other Store
60. OfficeyFactory
SO. Other Outdoor
40. Near Vehicle
30. In Vehicle
20. Residential-Outdoor
10. Residential/Indoor
Location
Figure 6-45. 3-D plot of the overall weighted mean 24-hour number of
occurrences in 21 out of 70 location x activity microenvironments.
The missing 49 microenvironments had less than 100 occurrences
over the diary days of all the respondents.
6-41
-------
Table 6-14. The Overall Weighted Mean 24-Hr No. Of Occurrences in 21 Location x Activity
Microenvironments
Activity Location
10. Residential-Indoor
20. Residential-Outdoor
30. In Vehicle
40. Near Vehicle
50. Other Outdoor
60. Office/Factory
70. Mall/Other Store
80. School/Public BMg.
90. Bar/Restaurant
100. Other Indoor
10. 20. 30.
Food Dishes/Clean House-
Preparation Kitchen keeping
1.57 1.46 1.24
1.08
40. 50.
Yard/
Bathing Maintenance
1.32 1.23
1.3
1.02
1.08
60.
Sports/
Exercise
1.08
1.24
1.09
1.23
1.03
70.
Eating/
Drinking
1.89
1.11
1.17
1.06
1.1
1.15
1.09
Note: The missing percentages were for the 49 microenvironments with less than 100 occurrences overall.
The largest overall percentages of time spent in the Residential-Indoor-Eat/Drink microenvironment
were for females respondents (7%), respondents aged 0-4 (6%), and respondents aged 65 and over (5%)
(Figure 6-46 and Table 6-15). The percentages of time spent for the other subgroups (males, ages 5-7,
ages 18-64, seasons, Census regions, and weekend vs. weekday) ranged between 2.9 and 4%. Of the nine
selected microenvironments with the largest percentages of time spent, female respondents and
respondents aged 65 and over had the largest percentages of time spent in the Housekeeping-Residential-
Indoor (2.4-2.8% vs. 0.4 to 1.8 for the other subgroups), Dishes/Laundry-Residential-Indoor (1.3% vs. 0
to 0.9%), and Residential-Outdoor-Yardwork/Maintenance microenvironments (2.1% vs. 0.1 to 1.6%).
Female respondents had the highest percentage of time spent in the Bar/Restamant-Eatmg/Drinking
(1.7% vs. 0.5 to 1.1% for the other subgroups) and Residential-Indoor-Bathing (2.2% vs. 1 to 1.4%)
microenvironments.
Of the nine selected microenvironments with the largest percentages of time spent, the Residential-
Indoor-Food-Preparation location x activity microenvironment had the smallest 24-hour mean duration
(Figure 6-47 and Table 6-16a) for males and respondents aged 5-17 (20-35 rain vs. 50-61 min for the
other subgroups). Those aged 0-17 had the smallest durations in the Residential-Indoor-Dishes/Laundry
(27-28 min vs. 42-64 min), Residential-Indoor-Housekeeping (61-65 min vs. 106-118 min), and
Residential-Outdoor-Yardwork/Maintenance (100-111 min vs. 123-175 min for the other subgroups)
microenvironments. Those aged over 65 had the smallest duration in the Residential-Outdoor-
Sports/Exercise microenvironment (63 min vs. 94-128 min). Table 6-16b contains the weighted doer
sample sizes for the nine microenvironments.
642
-------
T:
Res. Ind.-Eat/Drink
Res. Ind.-Housekeep.
Res. Ind.-Food Prep.
Res. Ind.-Bathing
Res. Out-Yard. Main.
Bar/Rest.-Eat/Drink
Res. Ind.-Dishes
Other Out.-Sports/Exer.
Res. Out.-Sports/Exercise
>,
•fs
o c
S o
8.2
Subgroup
Figure 6-46. 3-D plot of the overall percentage of time spent in nine location x activity microenvironments across each subgroup.
-------
Table 6-15. The Percentage of Time Spent in Nine Selected Location x Exposure Activity Microenvironments
Across Each Subgroup
Location
X
Activity
Males
Females
0-4
5-17
18-64
65+
Northeast
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
Food
Pr&psration
0.66
0.33
0.14
022
1.84
2.40
1.50
1.48
1.48
1.57
1.53
1.43
1.52
1.56
1.40
1.52
Residential
-Indoor
Dishes
0.18
1.27
0.00
0.11
0.89
1.28
0.80
0.70
0.71
0.74
0.73
0.77
0.80
0.72
0.72
0.68
House-
keeping
0.53
2.40
0.14
0.38
1.71
2.80
1.60
1.39
1.45
1.57
1.43
1.68
1.36
1.40
1.48
1.76
Bathing
1.09
2.24
1.35
1.10
1.18
0.96
1.15
1.07
1.25
1.13
1.15
1.15
1.24
1.28
1.08
1.04
Eat/
Drink
3.47
6.67
5.69
3.13
2.92
4.88
3.69
3.40
3.27
3.44
3.30
3.71
3.60
3.36
3.44
3.33
Residential
-Outdoor
Yard/
Mainten.
1.54
2.11
0.14
0.27
1.23
2.08
0.95
0.90
1.14
1.37
0.94
1.43
0.52
1.60
1.28
0.88
Sports/
Exercise
0.60
0.35
0.95
1.54
0.16
0.08
0.45
0.29
0.62
0.44
0.38
0.70
0.28
0.64
0.64
0.32
Other
Outdoor
Sports/
Exercise
0.80
0.41
0.27
1.15
0.52
0.40
0.85
0.45
0.51
0.69
0.53
0.77
0.36
0.72
0.92
040
Bar/
Restaurant
Eat/
Drink
0.88
1.70
0.54
0.49
1.03
0.88
0.75
1.07
0.77
0.93
0.77
1.08
0.84
0.88
0.84
092
Note: The percentage of time spent is over the time spent in all 70 possible kxatkxi x activity microenvironments. Results
are over all respondents (doers + non-doers).
6-44
-------
180
l"
Res. Out-Yard. Main.
Other Out.-Sports/Exer.
Res. Ind.-Housekeeping
Res. Out.-Sports/Exercise
L Bar/Rest.-Eat/Drink
Res. Ind.-Eat/Drink
Res. Ind.-Dishes
Res. Ind.-Food Prep.
Res. Ind.-Bathing
.
"I
<0
x \
§1
is
Subgroup
Figure 6-47. 3-D plot of the mean 24-hour duration of nine location x activity microenvironments across each subgroup.
-------
Table 6-16a. The 24-Hour Duration (Minutes) of Nine Selected Location x Exposure Activity
Microenvironments (the Doers) Across Each Subgroup
Location
X
Activity
Males
Females
0-4
5-17
18-64
65+
Northeast
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
FaR
Residential
-Indoor
Food
Preparation
34.91
60.47
53.94
20.13
52.71
61.6
51.32
49.98
54.34
52.25
51.46
54.24
50.17
56.27
50.03
52.48
Dishes
41.92
6228
27.47
28.38
59.87
6424
56.46
58.74
60.99
58.15
56.94
63.82
60.96
62.38
59.48
52.86
House-
keeping
101.63
115.67
65.4
60.82
11722
124.79
108.18
114.44
11524
112.89
106.05
13129
11124
113.78
108.44
117.82
Bathing
23.58
25.98
30.13
23.78
24.17
27.09
24.13
23.3
26.04
25.08
2424
26.36
26.18
26.83
23.11
22.99
Eat/
Drink
59.32
58.12
84.49
50.83
5326
77.15
61.16
59.45
56.19
59.69
56.33
64.81
59.73
58.64
59.1
57.32
Residential
-Outdoor
Yard/
Mainten.
169.67
122.69
100.13
110.7
155.53
154.73
141.07
147.92
163.75
147.47
145.37
163.81
133.09
159.75
139.19
174.63
Sports/
Exercise
123.08
100.96
119.38
125.73
93.71
62.63
114.44
102.54
117.44
112.97
105.33
127.1
103.19
120.56
128.3
89.82
Other
Outdoor
Sports/
Exercise
152.14
134.98
122,48
128.55
158.4
167.65
168.02
129.7
133.63
153.67
131.73
17923
155.79
13924
151.14
137.23
Bar/
Restaurant
Eat/
Drink
63.75
70.41
58.91
50.01
69.98
71.16
69.13
68.91
62.82
68.95
64.17
72.5
69.48
63.98
66.07
68.62
Note: 24-hour durations were calculated for the doers only.
Table 6-16b. The Weighted Doer Sample Sizes of Nine Selected Location x Exposure Activity
Microenvironments Across Each Subgroup
Location
x
Activity
Males
Females
0-4
5-17
18-64
65+
Northeast
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
FaR
Residential
-Indoor
Food
Prspfiration
1229
2563
24.6
240.3
2871.7
655.3
765.9
950.5
1264.1
811.4
2801.8
990.1
990.8
911.7
930.9
958.5
Dishes
281.1
1380.6
12.4
883
1223.5
3372
377.6
3912
543.4
349.5
12032
458.5
439
383.5
411
4282
House-
keeping
339
1413.7
242
159.5
1202.9
3662
397.5
395
585.5
374.8
12732
479.6
4062
407.1
448.5
491
Bathing
2978.9
3173.9
457.8
10955
4023.5
575.9
1235.1
1453.6
2244.1
1220.1
4478.7
1674.1
1592
1558.6
1527.6
1474.6
Eat/
Drink
37722
3942.4
6555
1492.9
4512.3
1053.9
1603
18442
2712.3
15552
5549.6
2165
1989.8
1895
1922.1
1907.7
Residential
-Outdoor
Yard/
Mainten.
585.1
355.5
11.1
58.1
646.5
224.8
177.8
192.9
321.3
248.6
606.7
333.8
133.8
334.6
3055
166.7
Sports/
Exercise
308.4
2382
82
2952
144.3
25
108.5
94.7
242.5
100.9
343.1
203.5
91.3
179.1
160.7
115.5
Other
Outdoor
Sports/
Exercise
340.4
205.3
18.6
221.5
269
36.7
132
108.8
181.8
123
385.4
160.3
78.3
170.8
202.7
94
Bar/
Restaurant
Eat/
Drink
892.3
826.9
80.8
227.7
1202
208.7
288.1
497.1
564.4
369.6
1144.4
574.9
402.7
463.9
412.7
440
6-46
-------
6.6 Summary and Discussion
The percentage of time spent in microenvironments considers both the length of time spent in
microenvironments and the number of people entering the microenvironments; and thus, it is a good
indicator of the significance of a microenvironment to the entire population and its relative significance
between subgroups. However, the duration and frequency of occurrence of microenvironments for the
doers - as calculated in this report — are more useful in the estimation (modeling) of exposure
magnitudes; and they are a better indicator of the potential exposure associated with microenvironments
for the specific subgroups that are being exposed. Calculations of human exposure usually give time-
averages over minutes, hours, days, etc., and require knowledge of how long and how often
microenvironments occur. Assuming characteristics of the locations and the mechanisms of pollutant
emission and transport are approximately the same across a population, we can assign exposures based
on durations in microenvironments without necessarily knowing the magnitude of exposures.
The Residential-Indoor location is by far the most important overall location with the
Eating/Drinking, Food Preparation, and Housekeeping exposure activities taking up the most time. The
most time spent being exposed to ETS also occurs in the Residential-Indoor location (females/weekends)
- followed by the Office/Factory (older males/weekdays) and Bar/Restaurant locations. The next most
significant location was Residential-Outdoor which has the most amount of time being spent in the
Yard/Maintenance exposure activity followed by Sports/Exercise. Although much time was spent in the
In Vehicle location, most of it did not involve specific exposure activities (other than exposure to other
vehicles' exhaust fumes). However, after Bar/Restaurant, the In Vehicle location has the next largest
amount of time spent exposed to ETS.
Of the microenvironments in the Residential-Indoor location, those with Yardwork/Maintenance and
Housekeeping exposure activities have the largest 24-hr durations. Thus, since Housekeeping occurs for
a large overall proportion of individuals, it may imply larger potential exposures when considering the
entire population; however, those members of the population that spend time in the
Yardwork/Maintenance exposure activity may also face significant exposures. Two of the exposure
activities in the Residential-Outdoor location with large durations — Yardwork/Maintenance and
Sports/Exercise — suggest larger exposures to the entire population than Housekeeping, which is more
important for specific subgroups — such as females over 18. The largest 24-hour durations of exposure to
ETS occur in the Office/Factory and the Residential-Indoor locations; and these microenvironments
occur over large proportions of time for the entire population. Females have more potential exposure in
Residential-Indoors while males have more potential exposure in Office/Factory. Persons over 18 have
more potential exposure in Office/Factory and those under 5 have more potential exposure in
Residential-Home.
6-47
-------
Section 6, Part n
24-hour Diary Results by Time-of-day in Minute Time Segments
by Day-of-week, Season, and Background Factors
Besides frequency of occurrence and duration of microenvironments over the entire diary day, the
NHAPS human activity pattern data base can be examined by the time-of-day that each
microenvironment occurs. Here, we present the fraction of respondents that experienced each location,
exposure activity, or smoker-present category over the 24-hour (1440-minute) diary day - with selected
subgroup breakdowns by gender, age, Census region, weekend vs. weekday, or season. There are two
kinds of plots presented: (1) stacked area charts of fraction of respondents vs. time of day for direct
time-of-day comparisons between each location and exposure activity; and (2) independent plots for
precise peak determinations and comparison across subgroups. Sometimes the peaks for different
subgroups fall at the approximately the same time so that analysis by time-of-day does not provide any
more information than the percentage of time spent over the entire 24-hour day - as reported in Part I of
Section 6. Thus, we have only included those plots where there were significant variations in peak
occurrences between subgroups. Since the 24-hour diaries were collected at one-minute time resolution
(i.e., every respondent was matched with a microenvironment for every minute of the day) and the plots
contain 1440 minutes of data corresponding to every minute of the day, the calculated fraction of
microenvironments being experienced for every minute is equal to the fraction of respondents
experiencing the microenvironment.
6.7 Locations
By following a straight vertical line in Figure 6-48 corresponding to a particular minute of the day,
we are able to examine the relative fraction of the NHAPS respondents that spent the current minute in
each of nine locations (see Section 5) - Residential (Indoor), which takes up the largest fraction
throughout the day, accounts for the remaining time: (1) Residential-Outdoors; (2) In Vehicle; (3) Near
Vehicle (walking near roadway, bicycling, motorcycling, etc.); (4) Other Outdoor (park, campground,
etc.); (5) Office/Factory; (6) Mall/Other Store; (7) SchooVChurch/Other Public Building; (8)
Bar/Restaurant; and (9) Other Indoor (hotel/motel, health club, repair shop, etc.). Except for the middle
of the day between 9:00 AM and 5:00 PM, the majority of the respondents (more than 50%) are in the
Residential-Indoor location. Between 9:00 AM and 5:00 PM between 50% and 65% of respondents are
Technical Note: The time of day plots were produced by creating 1440 activity pattern records for each of the 9,386 NHAPS
respondents corresponding to each minute of the day, sorting these records by minute and location, activity,
and smoker-present category, and calculating the frequency of respondents in each category for each minute
of the day. Lastly, the fraction of respondents was obtained by dividing each frequency by the total number
of activities (= respondents) for the current minute. The total number of microenvironments for each
minute is equal to the total number of respondents (9386) since the time resolution of the study was one
minute. All results presented are weighted.
6-48
-------
outside of the Residential-Indoor location. The four locations with the highest fraction between 9:00 AM
and 5:00 PM are the Residential-Outdoor, In Vehicle, Office/Factory, and School/Church/Public
Building locations with the Office/Factory and School/Church/Public Building being the larger of the
four until about 4:00 PM. Towards 5:00 PM most people are returning from school and work by car so
the frequency of the In Vehicle location becomes larger and more people may be spending time outside at
their homes so the frequency in the Residential (Outdoors) location also becomes larger. The
Bar/Restaurant location replaces Office/Factory in the top four locations after 5:00 PM.
The Office/Factory and School/Public Bldg. locations each have double peaks of occurrence in the
middle of the day corresponding to the before and after lunch waves (Figure 6-49). The first broad peak
falls between 9:00 AM and 12:00 PM (about 13% of respondents) for Office Factory and between 10:00
and 11:00 AM (almost 20%) for School/Public Bldg. The second peak occurs at about 2:30 PM Oust
over 12%) for Office/Factory and about 1:00 PM (about 17%) for School/Public Bldg. The
School/Public Bldg. location also has a third peak at about 7:30 PM probably corresponding to night
school (about 5% of respondents). The Residential-Outdoor location has two peaks, one just before noon
(about 7%) and another slightly higher one between 4:00 and 5:00 PM (about 9%). Single broad peaks
occur between morning and evening for the Other Outdoor (6:00 AM to midnight, high of 5%),
Mall/Other Store (6:00 AM to 10:00 PM, high of about 6%), and Other Indoor (6:00 AM to 10:00 PM,
about 3.5%). About 1% of respondents are in the Other Indoor location between 11:00 PM and 6:00
AM. The most striking peaks occur for the Bar/Restaurant location at about 12-12:30 PM for lunch time
(4.5%) and 7:00 PM for dinnertime (just over 4%). The occurrences of Bar/Restaurant extend into the
night - trailing off at about 3:00 AM. The In Vehicle and Near Vehicle locations display a jagged
pattern that probably arises because their durations are fairly brief and respondents tend to round their
diary beginning and ending times to the nearest hour or half-hour. The smoothest, broadest peak for In
Vehicle occurs during morning traffic between 7:00 AM and 8:00 AM (about 10%) with the highest peak
occurring just after 5:00 PM for the rush-hour commute home (17-18%). The number of respondents in
the Near Vehicle location is a fairly even plateau between 7:30 AM and 3:00 PM (3-4%).
The Very Young (0-4) and Retired (65+) age groups have two distinct peaks each about five hours
long - one in the late morning and one in the afternoon - for the Residential-Outdoor location (Figure 6-
50). The Very Young peaks are delayed about an hour after those for the Retired. The Working (18-64)
and School Aged (5-17) age groups each have steadily rising occurrences with peaks at about 5:00 PM.
The School Aged and Working account for the In Vehicle peak at 7:00-8:00 AM. The peak for School
Age in the afternoon occurs at 3:00 PM while peaks for the Working are at noon and 5:00 PM. While the
peak for the Working hi the Near Vehicle location is a broad plateau during the day, the School Aged
have one peak near 8:00 AM and another near 3:00 PM. The School Aged have the largest fraction of
respondents in the Other Outdoor location after 5:00 PM. As expected from our study of percentages of
time spent over the entire diary day of all respondents, the Working has the largest fraction of
respondents in the Office/Factory location for the entire day and the School Aged has the largest fraction
in the School/Public Bldg. location for the entire day. The Working also have the largest fraction of
respondents in the Other Indoor location from midnight until about 7:30 PM. The School Aged and Very
Young age groups have the most respondents in the Bar/Restaurant location at about 11:30 AM and 6:30-
7:00 PM, while the Working have the most respondents in the Bar/Restaurant location at about 12:30 PM
and 7:30-8:00 PM. The Retired have the most respondents in Bar/Restaurant at 12:30 PM to match the
Working, but their peak at just after 6:00 PM is close to that for the School Aged and the Very Young.
6-49
-------
In Vch.
Office/Fact.
Res. Out.
Other Iml.
Bar/Rest.
Sch./Pub.Bldg
Mall/Store
Other Out.
D Near Vch.
^1 ^H ^M ^H ^N ^N ^N ^N ^N ^N ^S ^% ^^l
8§888888888§||||||||||||8
Time of Day
Figure 6-48. Stacked overall fraction of respondents in each location by time-of-day. The Residential-Indoor location accounts for the fraction not shown
(white space). The time scale is from midnight to midnight on the diary day. Since all respondents are in some location at any given minute
the fraction of microenvironments is equal to the fraction of respondents.
-------
Figure 6-49. Overall fraction of respondents in each location by time-of-day in separate plots. The stacked plot is in Figure 6-48.
-------
20. Residential-Outdoor
80. School/
Public Building
Figure 6-50. Proportion of respondents in each location by time-of-day by each of four age groups: Very Young (0-4), School Aged (5-17), Working
(18-64), and Retired (65+). The time scales run from midnight to midnight on the diary day.
-------
On weekends more respondents were in the Residential-Indoor location between 6:00 AM and 5:30
PM than on weekdays (Figure 6-51). There were slightly fewer respondents on weekends in Residential-
Indoor from midnight to about 4:00 AM. The peaks of occurrence of In Vehicle are distinct for the
morning and evening commutes on weekdays, but on weekends the fraction of respondents in In Vehicle
is fairly even (2.5% to 10%) between 8:00 AM and 10:00 PM. The fraction of respondents is greater on
weekends in Other Outdoor between 8:30 AM and 6:00 PM, in Mall/Other Store between 11:00 AM and
6:00 PM, in Bar/Restaurant from 10:00 PM to 5:00 AM, and in Other Indoor between 10:30 PM and 7:00
AM. In contrast to the double peaks (before and after lunch) on weekdays, on weekends there is a small,
broad peak at about 10:30 AM in Office/Factory and a peak at about 11:00 AM for School/Public Bldg.,
which decays gradually until 8:00 PM.
The fraction of respondents were similar during the Winter and Fall seasons and during the Spring
and Summer seasons (Figure 6-52). The fractions of respondents in the Residential-Outdoors and Other
Outdoor locations were greater during most of the diary day in Spring/Summer with the peak in summer
extending until about 7:00 PM.
6-53
-------
20. Residential-Outdoor
80. School/
Public Building
Figure 6-51. Proportion of respondents In each location by time-of-day by weekday vs. weekend. The time scales run from midnight to midnight on the
diary day.
-------
0.16
0.14
0.12
0.1
0.08
e 0.06
i
20.04
0.02
0
20. Residential-Outdoors
giiiiigggfigiggg
Time of Day
50. Other Outdoor
Time of Day
Figure 6-52. The fraction of respondents in the Residential-Outdoor and Other Outdoor locations by time-of-day for
the four seasons. The time scales run from midnight to midnight on the diary day. Since all respondents
are in some location at any given minute, the fraction of microenvironments is equal to the fraction of
respondents.
6-55
-------
6.8 Exposure Activities
At any time during the diary day, between 0% and about 35% of the NHAPS respondents (Figure 6-
53) were occupied in seven exposure activities that are related to exposure to chemical pollutants by the
air, water, dermal, or food pathway (see Section 5): (1) Food Preparation, (2) Do Dishes/Do
Laundry/Clean Kitchen, (3) Housekeeping, (4) Giving or Taking a Bath or Shower, (5)
Yardwork/Gardening/Maintenance, (6) Sports/Exercise, and (7) Eating/Drinking. The peak fraction of
respondents in exposure activities (about 65%) occurs at about 6:00 PM when people are returning home
to make dinner. Close to 100% of the respondents are involved in non-exposure activities from midnight
to about 5:00 AM when most people are sleeping. Examples of other non-exposure related activities are
watching TV, dressing, traveling for goods and services, or traveling to and from work (see Table 5-7 in
Section 5). By following a straight vertical line in Figure 6-53 corresponding to a particular minute of
the day, we are able to examine the relative fractions of respondents in the seven defined exposure
activities. At midday and the early evening, Eating/Drinking takes up the major portion of time, with
Cooking/Food Preparation at night also taking up a substantial fraction. From 2:30 to 4:30 PM
Yardwork and Sports/Exercise seem to be the predominant activities, while between 9:30 and 11:30 AM
Housekeeping occurs in the largest proportion.
The individual time of day plots in Figures 6-54 illustrate how the number of respondents in each
exposure activity fluctuates during the day. The Food Preparation and Eating/Drinking exposure
activities display three distinct peaks in frequency corresponding to the three typical meals of the day
(breakfast, lunch, and dinner). The majority of Food Preparation takes place in the evening at
approximately 5:30 PM (about 7% of respondents). The peaks for breakfast at 7:00 AM and lunch at
noon (both about 3%) are substantially smaller, probably because more people are eating out or skipping
those meals. Likewise, the breakfast peak (about 9% of respondents) for Eating/Drinking is much
smaller than the peaks for lunch at 12:30 PM (15%) and dinner at 6:00 PM (about 19%). Notice how the
dinner and breakfast peaks are more spread out than the lunch peak suggesting that Americans adhere to
the noon lunch hour more than any specific breakfast or dinner hour. Additionally, the Eating/Drinking
peaks are about a half-hour delayed from the Food Preparation peaks, and the frequency of
Eating/Drinking is much higher during the entire day than the frequency for Cooking/Food Preparation
since more people are (1) eating only, or (2) both eating and cooking, rather than (3) cooking only. The
Dishes/Clean Kitchen/Laundry exposure activity has lower frequencies than both the Cooking/Food
Preparation or Eating/Drinking exposure activities, and the peaks are spread out more evenly throughout
the entire day — except for the peak at about 6:30 PM which is delayed about a half hour after the
Eating/Drinking dinner peak. The Housekeeping exposure activity fraction of respondents rises steadily
from 5:00 AM until its peak at about 9:30 AM (about 5.5% of respondents); it then falls abruptly at noon
before decreasing steadily until midnight. Like the In Vehicle location the Giving/Taking a Bath/Shower
exposure activity displays sharp peaks every hour and half-hour. The two major areas of high frequency
are in the rooming at about 6:00-8:00 AM (a high of 7% of respondents), and the other in the evening
centered at approximately 8:00 PM (about 3%). Both peaks span 5-7 hours and jagged sub-peaks occur
throughout each of them. For the Yardwork/Maintenance exposure activity there is a peak in the
morning at about 10:30 AM (3.5%), an abrupt drop-off in the fraction of respondents at noon, and
another peak at about 3:00 PM (3.5%). The fraction of respondents in the Sports/Exercise exposure
activity rises from 5:00 AM, levels off between 10:00 AM and 1:00 PM (about 2.5% of respondents),
and reaches a peak at about 4:30 PM (4.5%).
6-56
-------
Bathing
• Yanl./Main.
] Housekeep.
• I-at/Drink
I Dishes/Laun.
Food Prep.
Sports/Excer
O »H
8|88888i8§8§§§8888888
Time of Day
Figure 6-53. Stacked overall fraction of respondents in each exposure activity by time-of-day. The No Exposure category accounts for the fraction
not shown (white space). The time scale is from midnight to midnight on the diary day. Since all respondents are in some exposure
activity at any given minute, the fraction of microenvironments that include a given exposure activity is equal to the number of
respondents in that exposure activity.
-------
A
t?^3 ^jj^ Jl ^
Ib.
^^77777777777777777777777^
Figure 6-54. Overall fraction of respondents In each exposure activity by time-of-day In separate plots. The stacked plot is in Figure 6-53.
-------
For respondents in the Retired age group, the dinner peak for Food Preparation is about one half-
hour ahead (5:00 PM) of the same peak for those in the Working age group (5:30 PM) (Figure 6-55). In
the Food Preparation, Dishes/Laundry, Housekeeping, and Yardwork/Maintenance exposure activities,
the fraction of respondents in the Retired age group is larger than the other age groups for most of the
day. However, for the Sports/Exercise exposure activity, the School Aged take up the largest fraction for
most of the day, and for Eating/Drinking, the Very Young have approximately the same fraction as the
Retired throughout the day.
For Food Preparation, Dishes/Laundry, Housekeeping, Bathing, and Eating/Drinking, the peaks in
the morning are consistently earlier on weekdays than on weekends; but the peaks later in the day occur
at approximately the same times for weekdays and weekends (Figure 6-56). On weekends
Sports/Exercise has the largest fraction of respondents in the middle of the day between 10:00 AM and
5:00 PM, whereas on weekdays the largest fraction occurs at about 6:00 PM.
Throughout the morning and early afternoon, the highest fraction of respondents in the
Yardwork/Maintenance exposure activity occurs in the Spring (Figure 6-57). For Spring/Summer the
fractions are larger than those in Winter/Fall from 5:00 PM to 10:00 PM. Summer has the largest
fraction of respondents (3-6% of respondents) in the middle of the day (noon to 3:00 PM) for
Sports/Exercise, while in Spring the fraction of respondents becomes large (5%) from about 3:30 to 6:00
PM. In both Spring and Summer the fraction of respondents in Sports/Exercise remains larger than
Winter/Fall until 9:00 PM.
6-59
-------
10. Food Preparation
50. Yardwork/
Maintenance
70. Eating/Drinking
j>J
20. Dishes/Laundry
60. Sports/Exercise
TUMidta*
Figure 6-55. Proporttonof respondents in each exposure activity by time-of-day by each of four age groups: Very Young (0-4); SchoolAged (5-17); Working
(18-64); and Retired (65+). The time scales run from midnight to midnight on the diary day.
-------
10. Food Preparation
30. Housekeeping
»~ •,* «^ *^ ^ ^? ^ ^ ,p* ^p *j* j& A* j
^^ ^ v *\r "T *^ S^ »^ "r v v ^r ^f* ^?
50. Yardwork/
Maintenance
70. Eating/Drinkin
20. Dishes/Laundry
60. Sports/Exercise
Figure 6-56. Proportion of respondents in each exposure activity by time-of-day for weekdays vs. weekends. The time scales run from midnight to midnight
on the diary day.
-------
0.06
aos
50. YardworK/Maintenance
Time of Day
0.07
0.06
0.05
0.04
OJJ3
OQ2
O01
60. Sports/Exercise
Winter
Spring
Summer
Fall
f 9 * t 9 9 9 9 i t f £ £
Time of Day
Figure 647. The fraction of respondents in the Yardwork/Maintenance and Sports/Exercise exposure activities by
time-of-day for the four seasons. The time scales run from midnight to midnight on the diary day. Since
all respondents are in one exposure activity at any given minute, the fraction of microenvironments is
equal to the fraction of respondents.
6-62
-------
6.9 Smoker-Present Categories
The fraction of respondents exposed to environmental tobacco smoke (ETS) has its largest peak
(Figure 6-58) just before 9:00 PM (about 18% of respondents). The fraction falls towards midnight but
remains between 5 and 8% until it reaches a low of 2.5% at 4:00 AM. For most of the work day (11:00
AM to 7:00 PM) about 15-16% of the respondents are exposed.
Males are exposed in larger fractions than females for the entire day (a high of 20% vs. a high of
15%), and the Working are substantially more exposed than any other age group between 7:00 AM and
11:00 PM (a peak difference of more than 10% at about 2:00 PM). Between noon and 9:00 PM a smaller
fraction of respondents in the West are exposed to ETS than any other region (10-15% vs. 15-20%).
Between noon and 4:30 AM the fraction is larger on weekends than on weekdays, and in the Fall there
are slightly more respondents exposed to ETS between about 8:00 AM and 5:00 PM than in the other
seasons.
6-63
-------
SSSSg-ggSSSSSiSSeSeSSeSlS
S2as?S2S2SB22-s
-------
Section 6, Part m
24-hour Diary Results by
Time-of-day in 3-hour Time Segments
To study the duration of microenvironments at different times of the day, we broke the diary day
into eight 3-hour time segments and calculated the mean total duration - across all respondents - of
microenvironments (location, exposure activity, and smoker-present categories) whose beginning times
originated in each time segment. The entire length of the microenvironment was considered regardless of
whether the ending time extended beyond the rightmost boundary of the 3-hour segment; and for each
respondent, durations of all occurrences of a microenvironment in a time segment were summed to obtain
the total duration. Thus, the mean total duration can exceed three hours (=180 minutes). No
comparisons were made across subgroups.
The percentage of respondents initiating a microenvironment in each location for each of the 3-hour
segments (Table 6-17 and Figure 6-59) was greatest for the Residential-Indoor location for all eight time
segments. The location with the second highest percentage of respondents initiating a microenvironment
over all time segments was Residential-Outdoor. The longest mean duration (over all occurrences) of
any location was 408 minutes between 12:00 AM and 3:00 AM in Residential-Indoors (Table 6-18 and
Figure 6-60). The next five longest durations were Mall/Store between 3:00 AM and 6:00 AM (400
min), Other Indoor between 12:00 AM and 3:00 AM (392 min), Office/Factory between 3:00 AM and
6:00 AM (382 min), Office/Factory between 6:00 AM and 9:00 AM (368 min), and School/Public Bldg
between 3:00 AM and 6:00 AM (343 min). The peak duration for In Vehicle occurred between 3:00 AM
and 6:00 AM (54 min).
Table 6-17. Weighted Percentage of Respondents Who Entered a Location Beginning in Each of Eight
Consecutive 3-Hour Time Segments
Sample Size
Residential - Indoor
In Vehicle
Residential - Outdoor
Near Vehicle
Other Outdoor
Office/Factory
Mall/Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
12am-3am
9685.2
91.33
0.37
3.22
0.43
0.36
0.88
0.23
0.90
1.15
1.13
3am-6am
2833.7
68.50
2.88
16.84
2.71
1.55
2.89
0.80
1.32
0.96
1.54
6am-9am
15875.2
43.19
4.21
23.63
6.26
2.01
6.58
1.34
9.32
1.31
2.15
9am-12pm
11631.1
35.88
7.02
22.96
5.78
3.37
3.88
6.31
8.57
3.67
2.57
12pm-3pm
13793.5
32.74
7.17
25.22
6.52
3.57
5.46
6.15
6.26
4.62
2.30
3pm-6pm
16080.2
39.68
7.74
29.03
6.74
2.46
1.65
4.50
3.43
2.87
1.90
6pm-9pm
13785.3
52.96
5.41
22.41
4.44
1.87
0.60
3.11
3.16
4.46
1.58
9pm-12am
9835.2
73.38
1.67
16.54
2.78
0.47
0.52
0.87
0.87
2.19
0.71
Note: The total weighted sample size is 9,196 (see Section 4). The total sample size of respondents entering locations in
each 3-hour can exceed 9,196 since respondents may enter more than one location in any given 3-hour segment.
6-65
-------
12am-3am
3am-6am
6am-9am
9am-12pm
12pm-3pm
3pm-6pm
6pm-9pm
9pm-12am
3-Hour
Segment
Location
•
-
:•
'
Figure 6-59. 3-D plot of the weighted percentage of respondents initiating a microenvironment in each of the eight 3-hour time segment for each location
throughout the 24-hour diary day (see Table 6-17). Notice that the fractions in each time segment can add up to greater than 1 since a
respondent can enter more than 1 location in any given 3-hour time segment.
-------
Table 6-18. Weighted Cumulative Durations (Minutes) of Locations Originating in Eight Consecutive 3-Hour
Time Segments
Residential - Indoor
Residential - Outdoor
In Vehicle
Near Vehicle
Other Outdoor
Office/Factory
Mall/Store
School/Public Bldg.
Bar/Restaurant
Other Indoor
12am-3am
408.54
140.14
69.4
217.13
338.5
308.46
335.53
281.16
124.69
391.65
3am-6am
135.3
100.69
54.41
122.94
231.85
382.45
400.22
343.1
165.07
285.62
6am-9am
111.7
134.85
36.66
81.34
211.92
368.44
217.88
303.17
144.26
233.99
9am-12pm
141.62
116.17
39.86
44.58
151.74
178.46
111.66
147.71
85.07
120.35
12pm-3pm
154.06
126.03
37.77
47.86
126.58
194.44
93.36
143.72
63.38
135.83
3pm-6pm
132.59
84.96
35.33
29.43
118.15
118.52
68.91
105.53
106.91
105.83
6pm-9pm
191.32
72.16
32.33
26.63
108.92
120.67
53.87
120.73
85.34
100.77
9pm-1 2am
128.73
39.93
26.39
14.39
63.63
65.69
29.26
88.64
78.88
93.26
6-67
-------
450 -
12am-3am
3am-6am
6am-9am
9am-12pm
12pm-3pm *
3pm-6pm Segment
6pm-9pm
9pm-12am
Location
Figure 6-60. 3-D plot of the weighted mean duration in each location for each of the eight 3-hour time segments. The beginning times of the
microenvironments originate in each time segment the durations are assigned to, but the ending times of the microenvironments may occur
in a later time segment.
-------
The percentage of respondents initiating a microenvironment in each exposure activity - excluding
the No Exposure category - for each of the 3-hour segments (Table 6-19 and Figure 6-61) was greatest
for the Eat/Drink exposure activity between 6:00 AM and 9:00 PM (23% to 26%). The Bathing exposure
activity had the largest percentage between 12:00 AM and 3:00 AM (1.7%), between 3:00 AM to 6:00
AM (24%), and between 9:00 PM and 12:00 AM (9.3%). Bathing also had the largest percentage
besides Eat/Drink between 6:00 AM and 9:00 AM (17%), 9:00 AM and 12:00 PM (9%), and 6:00 PM to
9:00 PM (9%). Besides the No Exposure category, the longest mean total duration occurred for
Yardwork/Maintenance between 6:00 AM and 9:00 AM (161 min) (Table 6-20 and Figure 6-62). The
two next largest mean durations occurred for Sports/Exercise between 3:00 AM and 6:00 AM (141 min)
and Housekeeping between 6:00 AM and 9:00 AM (120 min). During the 12:00 AM to 3:00 AM time
segment, the greatest mean total duration occurred for the Eat/Drink (52 min), Dishes Laundry (70 min),
and Food Preparation (67 min) exposure activities. The 12:00 AM to 3:00 AM time segments also have
the smallest percentages of respondents initiating these exposure activities, so that only a small portion of
the population is contributing to the long durations.
Table 6-19. Weighted Percentage of Respondents Who Entered an Exposure Activity Beginning in Each of
Eight Consecutive 3-Hour Time Segments
Sample Size
No Exposure
Food Preparation
Dishes/Laundry
Housekeeping
Bathing
Yardwork/Mainten.
Sports/Exercise
Eat/Drink
12am-3am
9507.4
96.56
0.24
0.05
0.14
1.70
0.11
0.10
1.10
3am-6am
3588.3
51.56
8.19
0.84
0.98
23.83
0.46
1.48
12.66
6am-9am
16530.5
41.63
9.03
2.02
3.70
16.75
1.92
1.75
23.19
9am-12pm
10883.8
49.26
6.10
3.08
5.24
9.08
3.06
3.54
20.63
12pm-3pm
12122.6
53.35
5.52
2.64
2.82
2.69
3.26
3.77
25.95
3pm-6pm
14312.2
50.79
11.54
2.67
2.21
4.66
2.68
4.80
20.67
6pm-9pm
16039.3
49.25
5.59
4.73
1.77
9.20
1.26
2.66
25.53
9pm-12am
9392.7
79.47
1.39
1.44
0.80
9.28
0.41
0.89
6.32
Note: The total weighted sample size is 9,196 (see Section 4). The total sample size of respondents entering exposure
activity in each 3-hour can exceed 9,196 since respondents may enter more than one exposure activity in any given 3-
hour segment.
6-69
-------
30
0
Eat/Drink
Bathing r
Food Prep.
Exposure Sports/Exerc. i
Activity Yard./Maint.
Housekeeping
Dishes/Laund.
9pm-12am
6pm-9pm
3pm-6pm
12pm-3pm
9am-12pm
6am-9am
3-Hour Segment
3am-6am
12am-3am
Figure 6-61. 3-D plot of the weighted percentage of respondents initiating a mtcroenvironment in each of the eight 3-hour time segments by exposure activity
throughout the 24-hour diary day (see Table 6-19). Notice that the percentages can sum to greater than 100 since a respondent can be in more
than one microenvironment for any given 3-hour time segment.
-------
Table 6-20. Weighted Cumulative Durations (Minutes) of Exposure Activities Originating in Each of Eight
Consecutive 3-Hour Time Segments
No Exposure
Food Preparation
Dishes/Laundry
Housekeeping
Bathing
Yardwork/Mainten.
Sports/Exercise
Eat/Drink
12am-3am
410.02
66.98
70.17
55.19
16.07
97.95
62.02
51.66
3am-6am
184.58
19.93
31.23
96.22
15.33
109.13
140.92
27.87
6am-9am
256.12
23.77
59.9
120.08
17.05
161.08
98.21
29.61
9am-12pm
181.57
34.07
54.54
111.17
22.83
126.2
109.77
35.12
12pm-3pm
182.95
38.29
62.64
105.47
28.12
132.53
112.44
35.72
3pm-6pm
147.18
48.61
39.47
67.35
24.04
79.43
91.41
40.84
6pm-9pm
182.1
34.5
33.91
50.05
24.22
78.39
77.24
39.3
9pm-12am
130.26
20.14
33.35
39.36
19.1
36.57
53.01
29.44
6-71
-------
180
160
W
fj
c
o
I
Q
YardVMaint
Sport s/Exerc
Housekeeping
Dishes/Laund.
Eat/Drink
Exposure Activity Food Prep.
12am-3am
3am-6am
6am-9am
9am-12pm
12pm-3pm
3pm-6pm
- 6pm-9pm 3-Hour Segment
Bathing 9pm-12am
Figure 6-62. 3-D plot ot the weighted mean duration in each exposure activity for each of the eight 3-hour time segments. The beginning times of the
microenvironments originate in each time segment the durations are assigned to, but the ending times of the microenvironments may occur
in a later time segment.
-------
The largest percentage of respondents initiating a microenvironment where they were exposed to
environmental tobacco smoke (ETS) occurred in the 3:00 PM to 6:00 PM (20%) and 6:00 PM to 9:00 PM
time segments (21%) (Table 6-21 and Figure 6-63). The longest mean total duration of
microenvironments where ETS exposure was initiated was in the 12:00 AM to 3:00 AM time segment
(235 min), followed by the 6:00 AM to 9:00 AM (206 min) and 3:00 AM to 6:00 AM (183 min) time
segments. The long mean durations between 12:00 AM and 6:00 AM are due to a small segment of the
population since only 3-6% of the population is starting to experience ETS exposure during these times
versus 10 to 21% for the other time segments.
Table 6-21. Cumulative Weighted Duration and Weighted Percentage of Respondents for ETS Exposures
Originating in Each of Eight 3-hour Time Segments
12am-3am 3am-€am 6am-9am 9am-12pm 12pm-3pm 3pm-6pm 6pm-9pm 9pm-12am
Percentage 5.94 3.15 16.86 13.83 17.30 19.89 20.81 9.87
Duration 235.6 182.86 205.92 159.35 149.43 127.76 144.31 88.86
Note: The Percentage column is the number of respondents who began an ETS exposure episode in each 3-hour segment
divided by the weighted total sample size of persons who were exposed vs. persons who were never exposed (4025.9 +
4951.5 = 8977.4).
6-73
-------
3-Hour Segment
250 235.6
3-Hour Segment
Figure 6-63. Bar charts of the weighted percentage of respondents exposed to ETS and the weighted mean
duration of exposure to ETS (in minutes) for each of the eight 3-hour segments in the 24-hour
diary day (see Table 6-21). The beginning times of the microenvironments originate in each time
segment the durations are assigned to, but the ending times of the microenvironments may occur
in a later time segment.
6-74
-------
Section 7
Results of Follow-up Questions
On Personal Exposure
Although the 24-hour diary data - analyzed in Section 6 - provides the most accurate frequencies
and durations of specific exposure-related microenvironments, specific exposure-related background
activities are not represented. For example, the general areas of Food Preparation and Housekeeping
were not broken down into smaller classifications such as "Frying" or "Waxing the Floor". Similarly,
while the diary data contain variables for the presence of a smoker in every microenvironment, direct
smoking or the type of smoke source is not addressed. The follow-up questions presented here were
designed to retrieve specific exposure information for typical household pollutant sources such as dish
detergent, floor wax, paints, glue, pesticides, engine exhaust, wood smoke, gas ovens, air fresheners, etc.,
and indirect and direct exposure to tobacco smoke. Unfortunately, however, many of the follow-up
questions contain a substantial number of missing values or were coded in a way that makes them
difficult to analyze.
To shorten the length of each interview, the follow-up questions were split into two groups. Some
were placed on the A questionnaire, some on the B questionnaire, and some on both. Of the 9,386 total
respondents, 4,723 were given the A questionnaire and 4,663 were given the B questionnaire. Both
questionnaires were asked of respondents that were evenly distributed throughout the United States.
The TEAM studies22"26 give the higher-than-outdoor kinds and levels of pollutants that are found in
typical indoor microenvironments. Since the NHAPS follow-up questions mostly queried respondents on
the activity portion of microenvironments, they have been grouped into ten categories of exposure based
on the type of activity and assigned a unique two-letter code (Table 7-1). All tobacco smoke-related
questions were placed into the TB group. Questions related to water exposure pathways were placed in
the groups for Taking (and/or giving) a Bath, Showering, and Swimming (BA), Eating or Drinking (ED),
and Washing (WA). Many of the air exposure pathway questions concerning a variety of combustion
sources (other than tobacco smoke) such as fires, heating systems, and engine exhaust were placed in the
CS group. Any question on chemical or solvent exposure via either air or dermal contact was placed in
the CE group. The general questions on housing characteristics such as renovations, number of carpets,
number of windows left open, water sources, etc. were put in the HC group. Any question not easily
placed in one of the above categories was put into the OC group.
The follow-up variables that contain the NHAPS respondents' answers to each of the follow-up
exposure questions are listed in Table 7-2. The variables are listed alphabetically under each of the
groups described above. Thirty-four of the follow-up questions were asked of all 9,386 NHAPS
respondents, 77 of them were asked of 4,723 respondents on the A Questionnaire, and 60 of them were
asked of 4,663 respondents on the B Questionnaire. The columns in Table 7-2 contain: the variable
number, the variable name, the category name (see Table 7-1), the variable label, the questionnaire (A or
7-1
-------
B), the variable type (either categorical or numeric), either the overall % answered yes or no (categorical)
or the overall mean (numeric), the number of missing values, the number of refused values, and the
number of "don't know" answers. Categorical variables can have two values (yes/no) or multiple values
(e.g., for housing type). Either the percent answered yes/no or the mean is included for variables with
enough actual data.
Unfortunately, many of the NHAPS follow-up variables contain substantial numbers of missing,
"refused to answer", or "don't know" responses, especially the variables containing numeric or interval
data (an exact number) that attempt to measure frequencies of occurrence ("how many times...") or
durations ("how many minutes...") of exposure events. Some of these missing values arise because the
variables are dependent on yes/no questions where most of the responses were "no". However, in
addition, most of these variables were coded with both interval and categorical data types, which makes
them difficult to analyze.
Table 7-1. The Ten NHAPS Follow-Up Question Categories with Example Activities and Associated Chemicals
Category Name/Description Code Examples of Activities
"Associated Chemicals
1. BatrtJnoTSrwweringtewirnrning BA
2. Misc. Chemical Exposure CE
3. Misc. Combustion Sources CS
Giving/Taking a Bath or Shower
Pesticides. Solvents,...
Heating. Engine Exhaust,...
4. Dust/Soil
5. Eating/Drinking
6. Misc. Home Characteristics
7. Medical Background
8. Some Other Category
DS On Grass, Sand, In Garden,...
ED Water Drank, Seafood Eaten,...
HC Carpets, Windows Open....
MB Angina, Asthma....
OC Microwave Use, Walking Outside,
Chloroform,...
Pesticides. VOCs....
Particles, CO, PAH, benzene, VOCs, NO2,
Particles, Heavy Metals,...
Pesticides, Heavy Metals,...
9. Tobacco Smoke Exposure TB
10. Washing WA
Indirect/Direct Smoking
Laundry. Dishes,...
Particles, PAH, CO, benzene, VOCs,
Chloroform,...
"CO = Carbon Monoxide
VOC = Volatte Organic Chemical
NOj = Nitrogen Dioxide
PAH = Porycydic Aromatic Hydrocarbons
Heavy Metals include Lead, Mercury and Cadmium.
Measured parbctes typicaJy consist of matter with diameters of 10 microns or less (PM10).
The yes/no variables with less man 10% overall invalid responses were analyzed by the percentage
of "yes" responses out of the total number of "yes" and "no" answers (missing values, "don't know", or
"refused" responses were omitted). The results were broken down by the exposure categories in Table 7-
1. The ten yes/no follow-up variables asked of 9,386 combined respondents on both questionnaire A
and B that contain enough useful information are ANGIN, ASTHMA, BASEMNT, CARP, EMPHYS,
GASCAN, HUML, LAWN, PAINT, and STOVE. There are 28 yes/no variables on questionnaire A and
21 on questionnaire B that had adequate amounts of valid data. There were 14 variables analyzed for the
Chemical Exposure (CE) category (Figures 7-1 and 7-2; Table 7-3), 10 for the Combustion Sources (CS)
category (Figure 7-3; Table 7-4), 8 for the combined Washing and Bathing (WA/BA) categories (Figure
7-4; Table 7-5), 4 for the Medical Background (MB) category (Figure 7-5; Table 7-6), 9 for the Housing
Characteristics (HQ category (Figure 7-6; Table 7-7), and 8 for other categories (OC, DS, ED) (Figure 7-
7; Table 7-8). The percentage of "yes" answers to exposure questions were used to assign relative
amounts of exposure across the different subgroups.
7-2
-------
Table 7-2. The Categorized NHAPS Follow-up Variables, If They Are on the A or the B Questionnaire, If They Are a Categorical Variable, and the Number of
Missing, Refused, and "Don't Know" Responses (9383 total for A+B with 4723 for A and 4663 for B)
Variable
No. Name
Cat. Variable Label
A B Type
%Y/N
or Mean
Missing
Refused
DK
1. BATH BA
2. BATH# BA
3. BATHP BA
4. KBASTAY BA
5. KBATH1 BA
6. KBATH1# BA
7. KBATH1M BA
8. KBAWIN BA
9. KBDOOR BA
10. KIDSWIM BA
11. KSWIM BA
12. KSWM BA
13. SHBAST# BA
14. 8HDOOR BA
15. SHOWBA BA
16. SHOWBA# BA
17. SHOWER BA
18. SHOWER* BA
19. SHROOM BA
20. SHWIND BA
21. AEROSAL CE
22. AEROSAL# CE
23. AGENT CE
24. AGENT* CE
25. DEODORT CE
26. FLOORWAX CE
27. FIOQRWAX# CE
28. FRESHNER CE
29. GLUE CE
30, GLUE# CE
31. GLUEUSE CE
DID YOU TAKE OR GIVE ANY BATH YESTERDAY.
MINUTES RES SPEND TAKING SHOWERS 1
DID YOU TAKE BATH YESTERDAY
MINUTES RES STAY B-RM RIGHT AFTER BATH
DID RES TAKE OR BE GIVEN BATH YESTERDAY 1
# OF BATH RES TOOK OR GAVE 1
# OF MINUTES RES SPEND IN TAKE-GIVEN BATH
WAS WINDOW OPEN OR EXHAUST FAN ON
WAS BATHROOM DOOR CLOSED
# OF TIMES SWIMMING 1
DID KID SWIM IN SWIMMING POOL LAST MONTH 1
MINUTES RES IN WATER WHEN SWIMMING
HOW LONG KES STAY B-RM RIGHT AFTER BATH
WAS B.RM DOOR CLOSED WHEN RES SHOWERED
DID RES TAKE SHOWER OR BATH YESTERDAY 1
HOW LONG FOR SHOWER OR BATH 1
DID RES TAKE A SHOWER YESTERDAY 1
# OF TIMES RES TOOK SHOWER 1
MINUTE RES STAYB-RM RIGHT AFTER SHOWER
WAS B-RM WINDOW OPEN OR EXHAUST FAN ON
DID SM USE AEROSOL SPRAY PRODUCTS 1
ABOUT HOW MANY TIMES 1
EXPOSED TO CLEANING AGENTS-POWDERS-AMMONIA 1
ABOUT HOW MANY MINUTES
TOILET-BOWL DEODORIZERS USED AT HOME 1
EXPOSED TO FLOOR/FURNITURE WAX 1
ABOUT HOW MANY MINUTES
ARE AIR FRESHENERS USED AT HOME 1
EXPOSED TO GLUES/ADHESIVES 1
ABOUT HOW MANY MINUTES
B C
B
B C
B C
B C
B
B
B
B C
B
B C
B
A
B C
A C
A
B C
B
B
B C
A C
A
A C
A
A
A
A
A
A
A
C
C
C
C
B C
14/86
14786
91/8
77/22
43
32/67
20/80
44/56
7/92
64/36
6/93
1021 (21.9)
44
4213 (90.3)
8
4137 (88.7)
6
7
, 7 ,
• 7 v
6
3
0
6
43
32
4
38
40
44
1030(22.1)
36
39
26
7
24
.7 •'
22
7
5
51
3617 (77.6)
3617 (77.6)
3617 (77.6)
3617 (77,6)
4004(85.9)
436(9.1);
1030(22.1)
430(9.1)
1030(22.1)
20
3795 (80.4)
1
4385 (92.8)
1
4416(93.5)
4219 (90.5)
17
1
28
5
69
20
8
15
5
22
13
101 (2.1)
70
17
37
10
15
55
32
40
18
16
35
14
5
14
31
5
35
-------
Table 7-2. (Continued)
No.
r/32,:
33.
34.
35.
36.
37.
, 38.
39.
40.
41.
42.
43.
44.
45.
46.
;47.
48.
49,
50.
51.
52,
53.
54.
55,
58,
57.
58.
59.
60.
61.
62.
63.
Variable
Nama
/;d8TAT|0#
GSTATION
IPAINT
LAWN
MOTHBALL
NPOLISH
OTHPUMP
PAINT
PAINTAP*
PAINTAPP
PERFUME
PERFUME*
PERSPEST
PEST
PEST1
PEST1#
PE8TLOC
PROPEST :
PUMPGAS
SOLV
8OLV#
SOLVENT
STAIN
STAIN*
AQAR#
A ra AP^;Jt "
^Vwl^^nVrTr '
AUTOGAR
ESPAHEAT
PlREHEAT
FIREPLAC
FLAMES
FUEL
Cat.
CE
CE
CE
CE
CE
CE
CE
CE
CB
CE
CE
CE
CE
CE
CE
CE
CE
OE
CE
CE
OE
CE
CE
CE
cs
08
CS
OS
CS
CS
CS
CS
Variable Label
\ ' ' '^ *" * ' '
ABOUT HOW MANY MINUTES ALTOGETHER
INDOOR PAINTING
LAWNMOWERS-OTHER DEVICES STORED AT HOME
MOTHBALLS/CRYSTALS/CAKES USED AT HOME
USE NAIL POLISH YESTERDAY
DID 8M PUMP GAS WHILE RES IN CAR
PAINTS/VARNISHES STORED AT HOME
ABOUT ftdWMAN?y'MINUfE8|! 1 • .-
EXPOSED TO APPLIED PAINT/OPEN PAINT CAN
USE COLOGNE/PERFUME/AFTERSHAVE/FRAGRANCE
HOW MANY TIMES
# OP TIMES P-CIDE6 APPLIED PERSONALLY
ANY PESTICIDES USED TO GET RID OF PESTS
EXPOSED TO PESTICIDES/BUG SPRAYS
ABOUT HOW MANY MINUTES
WHERE WERE PESTICIDES USED
# f IMES P-CIDES APPLIED BY PROFESSIONAL
DID YOU PUMP GAS AT GAS STATION
EXPOSED TO SOLVENTS/FUMES
ABOUT HOW MANY MINUTES
WOODWORKING SOLVENTS/ADHESIVES AT HOME
EXPOSED TO STAIN OR SPOT REMOVERS
ABOUT HOW MANY MINUTES
# OF TIMES CAR WAS STARTED
# OF TIMES OAR STARTED W/G-DOOR CLOSED
ANY AUTO STARTED IN GARAGE YESTERDAY
ANY RM HEATED WITH ELECTRIC SPACE HEATER
ANY RM HEATED WITH FIREPLACE
DO YOU USE A FIREPLACE
EXPOSED TO OPEN FLAMES
MAIN TYPE OF FUEL USED IN THIS SYSTEM
%Y/N
A B Typa or Maan
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
A
A
A
A
A
A
A
A
A
A
A
A
A
<
A
A
A
A
A
A
A
A
A
?
B
B
B
B
?
B
B
B
B
B
B
B
B
.
21/78
C 25/74
C 26/73
C 14/85
C 5/87
C
C 44/55
, ' , /
C 6/93
C 47/45
C 42/58
C 6/94
C
C
C
C
C
C
C
C
C 10/89
C 10/89
C
Missing
7 ,
29
4
364 (7.7)
3746(79,3),
. ,. i *"< - •
17
364 (7.7)
369 (7.8)
'B'.'
24
'7- '
'• 6- • '
c ••
142
22
7
27
7
. • .. 7
48
8
3
3
4
19
1
Refused
,3739(79,2) ,
. '
•'•< 4436,
2712(68.2)
1
4451 (94.2)
2712(68.2)
2712(88,2)
3739
1
4217 (89.3)
4605 (97,5)
285960.5)
5643(60.1)
5643(60.1)
2813(59.6)
2813 (59.6)
1
DK
10
15
15
31
48
8
5
82
3 «.
19
18
22
29
37
15
•7
•; 7 ,
, 23,
8
25
3
87
12
2
27
33
15
4
1
43
9
296
-------
-1
tin
Table 7-2. (Continued)
No.
64,
65,
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91-,
92.
93.
94.
95.
Variable
Name
FUELY
FURNY
QASCAN
GASQUIP
GASQUIPtf
HEAT
HEATY
HEAWT#
HEAVYTRF
INDQAR
INDQAR#
KERHEAT2
KEROHEAT
OFUEL
OTHEAT
OTHEATD
OTHHEAT
OVENHEAT
OVENON
OVENON*
OVHEAT#
RUNHEAV
RUNHEAV*
SPILOT
STOVE
WOODSTOV
WSTOVE
:DIRT#
DUST
DUST*
GRASS*
GROUND
Cat.
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
DS
DS
DS
DS
DS
Variable Label
WHAT KIND FUEL USED IN CENTRAL FURNACE
WAS CENTRAL FURNACE TURNED ON
GASOLINE/KEROSENE STORED AT HOME
EXPOSED TO GAS/DIESEL POWERED EQUIPMENT
ABOUT HOW MANY MINUTES
MAIN TYPE OF HEATING SYSTEM USED AT HOME
WAS ANY HEAT TURNED ON AT HOME YESTERDAY
FOR HOW MANY MINUTES ALTOGETHER
WAS RES IN CAR/VAN/TRUCK IN HEAVY TRAFFIC
WAS RES IN GARAGE/INDOOR PARKING LOT
FOR HOW MANY MJNUTES ALTOGETHER
ANY RM HEATED WITH KEROSENE SPACE HEATER
DO YOU USE A KEROSENE SPACE HEATER
ANY OTHERS
ANY OTHER SOURCES OF HEATING
ANY OTHER HEAT SOURCES
ANY RM HEATED WITH ANY OTHER HEAT SOURCE
DID YOU USE GAS RANGBOVEN TO HEAT HOME
WAS RES AT HOME WHEN GAS OVEN USED
HOW LONG RES AT HOME WHEN OVEN WAS ON
ABOUT HOW MANY MINUTES
DID RES RUN/WALK ON HEAVY TRAFFIC RD
FOR HOW MANY MINUTES ALTOGETHER
DOES RANGE/OVEN HAVE BURNING PILOT LIGHT
DO YOU HAVE A GAS RANGE OR OVEN
DO YOU USE A WOOD STOVE
ANY ROOM HEATED WITH WOOD STOVE
H0W MANY MINUTES ON DIRT
EXPOSED TO EXCESSIVE DUST IN THE AIR
ABOUT HOW MANY MINUTES
HOW MANY MINUTES ON GRASS
DID KID PLAY ON SAND/GRAVEL/DIRT/GRASS
A
A
A
1 A
1 A
A
1
1 A
A
1 A
1 A
A
A
1
A
1
1
A
A
A
A
A
1 A
A
A
1 A
1
A
A
1 A
A
A
A
BType
B
B
B
B
B
•
B
B
B
B
B
B
C
c
C
C
C
C
C
C
C
C
C
C
C
C
c
c
c
c
c
c
c
c
c
%Y/N
or Mean
19/80
8/91
40/58
26/73
6/92
2/97
22/77
3/96
9/90
27/72
38/61
6/93
15/83
Missing
386
3
24
6
1
34
6
43
44
7
3
4
452 (9.6)
1
4
3
8
6
. ., 7 , .
' ' 7
43
7
11
28
4
3
5699 (60.7)
26
6
5699(60.7)
5699(60.7)
Refused
2813 (59.6)
2813 (59.6)
4322(91.5)
18
3503 (74.2)
4418 (93.5)
2813 (59.6)
4103 (86.9)
2813 (59.6)
3668 (77.7)
2934(62.1)
3668 (77.7)
4687(99.2)
4310(91.3)
5782(61.6)
2813(59.6)
2987(31.8)
57
4015(85)
2987(31.8)
2800 (29.8)
DK
65
25
57
15
5
160
31
17
24
17
4
• 2
44
12
49
43
2
1
10
22
6
21
5
103
61
43
:: 1
53
23
43
12
-------
Table 7-2. (Continued)
Variable
No. Nam*
Cat. Variable Label
A B Type
%Y/N
or Mean
Missing
Refused
DK
100. SOILHR
101. VACUUM
DS ANY WELCOME MATS FOR WIPING FEET
08 -H6W:Wy,MWi8SP|Nf^8ANt),6rtQRA
DS HOURS RES WORKED IN GARDEN • WITH SOIL
DS FREQ OF SWEEPING/VACUUMING FLOORS
103. BOTTWAT
104. CHAR*
;Tf,WfJpJpOTi • *f Wil^Wfc* -9 t W^ >|f^P*iSP*'lr*;»
ED IS BOTTLED WATER EVER USED IN YOUR HOME
ED # OF SERVINGS OF CHARRED FOOD
106. FRIED
108. GLASS*
109. JUICE*
1tO. 0EACP
111. SEAFOOD
112. SEAFOOD*
113. SODA
115, BASEMNT
116. CARP
117. CARPORT
118. DOOROUT
;f,i. .pq^INf!
ED EXPOSED TO FOOD BEING FRIED/GRILLED
ED
ED
ID
ED
ED
ED
HC
HC
HC
HC
HC
HC
GLASSES TAP WATER RES DRANK YESTERDAY
GLASSES JUICE WITH TAP WATER RES DRANK
WAS SEAFOOD PURCHASED OR CAUGHT
DID RES EAT SEAFOOD LAST MONTH
OF SERVINGS RES ATE SEADFOOD
CAN/BOTTLE CARBONATED DRINKS RES DRANK
DO YOU HAVE AN ATTACHED GARAGE OR CARPORT
IS THERE A BASEMENT IN YOUR BUILDING
# OF ROOMS BEING CARPETED
LAYING CARPET
ANY HOME DOORS OPEN DIRECTLY TO OUTSIDE
HOJtf fltytf TiMiS AN $JTf IDE! b00R OPiNIO - "•''
122. ELECHEAT
12& FtAME# .
124. FLOOR
126. QAL#
127. HOUSING
HC
HO
HC
HO
HC
HC
DO YOU USE AN ELECTRONIC HEATER
ABOUT HOW MANY MINUTES
REFINISHING FLOORS
IS tHAf FLOOR AR6A CARPETED
# OF GALLONS OF, WATER USED PER WEEK
HOUSING TYPE
1
1
1
1
1
1
y
1
1
1
1
1
1
1
1
1
1
1
A
A
A
A
A
A
A
A
A
A
A
A
A
A
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
C
• ,
,
C
c
c
«
c
c
c
c
c
c
p
,
c
c
c
1 A
B
BC
88/12
43/57
23/76
60/39
43/55
9/90
84/15
8/91
5/95
4723(50.3)
19
> '"e ;;r
31
34
1
35
!4
2
4
4
36
7
4
2
1+1
5707(60.8) 2987(31.8)
3
5699(60.7) 2413(25.7)
*•• , -1- ;;
1
3641(77.1)
1884(40.4)
18
3530 (74,7)
t68 (1CI.3)
$829{74J)
4227(89.5)
2655(56.9)
00 ,;
5
38 J
88
25
a
7+1
100
83 !
22
20
138
66
39
72
136(2.9)
27
1
125
161 (1.7)
15
12
17
45
10
16
70
98
-------
Table 7-2. (Continued)
No.
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
153,
:154
155.
156.
157.
158.
159.
Variable
Name
HUM!
HUMID
HUMIDR
RENOVA
ROOM
ROOMS
STORIES
STORWORK
WATER
WINDMIN
WINDY
WINDY#
YEARHOME
ANGIN
ASTHMA
EMPHYS
PREGNANT
MICRO
MICRO*
OTHOUT
OTHOUT*
SLEEP
WALKCAR
WALKCAR#
CIGAR
CIGARMIN
;OTHSMO
OTHSMO#
SMOAWAY*
SMOHO
SMOHO#
SMOKE*
Cat.
HC
HC
HC
HC
HC
HC
HC
HC
HC
HC
HC
HC
HC
MB
MB
MB
MB
OC
OC
OC
OC
OC
OC
OC
TB
TB
TB
TB
TB
TB
TB
TB
Variable Label
IS A PORTABLE HUMIDIFIER USED AT HOME
FREQ OF USING HUMIDIFIER
TYPES OF HUMIDIFIER
WAS HOUSE RENOVATED
ADDING ROOMS TO THE HOUSE
# OF ROOMS IN HOME
# OF FLOORS IN HOME OR BUILDING
# OF STORIES WORK BUILDING
WHERE DO YOU GET WATER FOR HOUSEHOLD USE
AMOUNT OF TIME WINDOWS LEFT OPEN
WERE WINDOWS LEFT OPEN WHEN RES AT HOME
# OF WINDOWS LEFT OPEN
YEAR OF MOVING INTO HOME
DOES RES HAVE ANGINA
DOES RES HAVE ASTHMA
DOES RES HAVE BRONCHITIS OR EMPHYSEMA
IS RES PREGNANT
DID RES USE A MICROWAVE YESTERDAY
ABOUT HOW MANY MINUTES
DID RES SPEND OTHER TIME RUN/WALK OUTSIDE
FOR HOW MANY MINUTES ALTOGETHER
WHERE DID YOU SLEEP WHILE RENOVATING
DID RES WALK OUTSIDE TO CAR IN DRIVEWAY
FOR HOW MANY MINUTES ALTOGETHER
DID YOU SMOKE CIGARStfOBACCO YESTERDAY
FOR ABOUT HOW MANY MINUTES
DID SB SMOKE CIGARETTES AT R's HOME
# OF CIGARETTES THEY: SMOKED
# OF CIGARETTES RES SMOKED OUTSIDE HOME
IS SMOKING ALLOWED IN YOUR HOME
# OF HOUSEHOLD MEMBERS WHO SMOKE AT HOME
# OF CIGARETTES SMOKED AT HOME TOTAL
%Y/N
A B Type or Mean
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
8
B
C 21/79
C
C 31/69
C 4/96
C
C 43/56
1984
C 3/97
C 7/92
C 5/95
C
C 49/42
C 27/70
C
C
C
c
C
Missing
1
5
5
4
893 (18.9)
25
35
40
366 (7.7)
7
89
?
4
44
6
1045(11.1)
527 (S<6)
897(19)
901(19.1)
519(11)
1382 (14.7)
2782(59.7)
3826(81)
Refused
3610 (77.4)
3785(80.1)
1373(29.1)
2712(57.4)
18
18
1
2384 (50.5)
3210 (68.8)
1363(28.9)
1
4619(49.2)
18
DK
22
24
84
13
16
164
46
26
35
55
262
75
60
71
45
32
28
5
22
51
4
7
17
85
37
28
3+14-1
17
-------
Table 7-2. (Continued)
Variable
No. Name
Cat. Variable Label
A B Type
%Y/N
or Mean
Missing
Refused
DK
160. SMOKE1
MINUTES SPENT SMOKING
ATR'SHOME
166.
DHAND
DISH
168. DISHHAND
169. DISHWASH
170. LAUNDRY
171. WASHAH
172. WASHK
173. WASHLOAD
174. WASHMACH
175. WMACH#
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
SY.HAND
DID YOU SMOKE Cl(
FREQRESWASHmiNSE
DO YOU HAVE A DISHWASHER AT HOME
PRlO RE8 USED DISHWASHER'* ,">
DOES RES WASH/RINSE DISHES BY HAND
WAS DISHWASHER USED WHEN RES AT HOME
ARE CLOTHES DONE AT HOME OR AT LAUNDROMAT
WAS WASHING MACHINE USED WHEN RES AT HOME
# OF TIMES KID WASHED HANDS YESTERDAY
# OF LOADS OF LAUNDRY DONE WHEN RES HOME
WHERE IS YOUR WASHING MACHINE LOCATED
FREQ RES WASH CLOTHES IN MACHINE
1
A
A"
A
B
B
B
B
B
B
B
B
B
B
B
C
C
C
C
382
57/43
79/17
21/77
0 = ?
3824 (81)
810 (41)
! 4-
206 (4.4)
33
33
38
39
1
368
998(21,4)
2025(43.4)
18
21
18
18
746(16)
35
,,'10 .
/ 12
14
40
16
32
49
189(4.1)
87
" 6 '
37
Note: The numbers In parentheses are percentages of the total number: 9386 for A+B, 4723 for A, and 4663 for B. Shaded rows designate variables that have more than 10%
missing, refused or "don't know" responses.
-------
The highest absolute percentages of "yes" answers can be used to determine which exposures are the
most significant both overall and within each subgroup. Overall percentages of "yes" answers over 50%
occurred for exposure to air fresheners, the use of dish washers, taking a shower, washing dishes by
hand, giving/taking a bath, children swimming, having a door leading directly outdoors, use of a
microwave, running/walking outdoors, walking to a car in the driveway, and the use of a welcome mat.
Females had more exposures to nail polish, floor wax, spot/stain removers, cleaning agents,
aerosols, and air fresheners, while males had more exposures to applied paint, glue, solvents, stored lawn
mowers, and stored paint (Figures 7-1 and 7-2; Table 7-3). Respondents aged 5-17 had the most
exposure to glue, stored paint, stored lawn mowers, and air fresheners, while those aged 18-64 had the
most exposure to every other item in the CE category except mothballs and deodorant, which were
highest for those aged 65 and older. There was more exposure to pesticides in the summer than in any
other season, and exposure to glue occurred the most in the fall.
The only three exposures in the CS category that were greater than 15% were the use of heat
yesterday, and the use of other heat sources besides fireplace, kerosene, wood stove, etc. (Figure 7-3;
Table 7-4). As expected, the Northeast and the Midwest had the most use of heat and the Winter/Fall
seasons had much more exposure than Spring/Summer.
Those respondents aged 0-4 had the lowest number of exposures to washing dishes by hand and
taking a shower, but they had the highest exposures for giving/taking a bath and using a clothes-washing
machine or dish-washing machine (Figure 7-4; Table 7-5). Respondents aged 18-64 had the highest
number of exposures from taking a shower.
The most respondents with angina and bronchitis or emphysema were by far in the 65+ age group
(Figure 7-5; Table 7-6). Females had slightly more cases of these ailments and asthma than males. The
most occurrences of asthma were for the 5-17 age group. Curiously, angina and emphysema also occur
most frequently in the South.
Humidifiers were used most often by respondents aged 0-4 and by respondents in the Midwest
(Figure 7-6; Table 7-7). Respondents in the Northeast had both the largest number of basements and the
fewest number of doors leading directly outdoors, while the West had the largest number of doors left
open.
7-9
-------
14
12
10
Chemical Exposure (CE) #1
i i
I I
Overall Females 5-17 65+ Midwest West Weekend Spring Fall
Males 0-4 18-64 Northeast South Weekday Winter Summer
Subgroup
FIJOORWAX GLUE NPOIJSH PAINTAPP PEST1 SOLV STAIN
Figure 7-1. The weighted percentage of respondents that answered *yes" to questions on chemical exposure: floor wax, glue, nail polish, applied paint
or open paint cans, pesticides/bug sprays, solvents/fumes, stain/spot removers.
-------
80
60
s
a
20
0
Chemical Exposure (CE) #2
\ /
/
s „ - -
^
OveraD Females 5-17 65+ Midwest West Weekend Spring Fall
Males 0-4 18-64 Northeast South Weekday Winter Summer
Subgroup
AGENT AEROSOL DEODORT FRESHNER MOTHBALL LAWN PAINT
Rgure 7-2. The weighted percentage of respondents that answered "yes' to questions on chemical exposure:
cleaning agents, aerosols, deodorants, air fresheners, mothballs, stored lawn mowers, or stored
paints/varnishes.
7-11
-------
Table 7-3. Percentage of 'Yes" Responses (or Follow-Up Questions on Chemical Exposure (CE)
T4
t«™*
to
Variable
Overall
Males
Females
0-4
5-17
18-64
65+
Northeast
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
AGENT
18.99
13.01
24.63
10.29
10.18
22.01
20.97
20.85
16.12
18.51
21.60
18.56
20.08
19.60
17.07
19.29
19.98
AEROSOL
33.33
26.53
39.76
17.42
31.06
37.35
24.93
30.66
33.88
37.10
28.67
33.03
34.07
31.99
33.51
32.67
35.05
DEODORT
45.62
48.82
42.63
42.12
43.15
45.87
49.83
45.88
41.86
51.99
38.85
45.23
46.61
45.24
43.01
46.26
47.89
FLOORWAX
7.21
4.34
9.93
6.05
5.83
8.14
5.16
6.99
8.23
7.80
5.18
6.78
8.31
8.80
6.57
6.08
7.42
FRE3HNER
65.01
63.26
68.66
60.44
67.15
66.47
57.34
63.33
63.50
69.07
61.41
64.04
67.44
63.40
64.39
65.04
67.09
GLUE
6.28
9.88
6.77
4.46
13.46
8.58
1.78
8.59
8.58
7.27
9.37
9.37
5.54
7.59
6.85
6.79
11.70
MOTHBALL
12.87
12.85
12.88
4.57
11.43
11.96
24.09
17.51
10.57
15.61
6.55
12.98
12.59
12.97
12.24
13.82
12.46
NPOLI8H
4.85
0.66
8.79
0.00
4.71
5.25
3.04
3.97
4.13
4.71
6.83
4.47
5.82
4.40
5.97
4.53
4.51
PAINTAPP
6.06
7.19
4.99
2.74
4.68
7.21
4.08
5.95
6.18
5.54
6.87
6.99
3.72
4.84
6.36
6.79
6.21
PEST)
5.54
5.75
5.34
2.37
5.07
6.10
5.16
3.71
5.57
6.54
5.49
5.85
4.77
2.59
6.54
9.12
3.95
30LV
10.92
12.94
9.00
2.40
8.27
14.16
3.14
13.06
11.40
9.20
11.29
11.73
8.87
12.04
9.93
12.22
9.59
STAIN
2.65
2.21
3.05
0.62
1.26
3.52
1.33
2.56
2.12
2.75
3.19
2.61
2.73
2.13
3.07
3.67
1.73
LAWN
29.73
32.67
26.94
29.47
32.36
30.61
21.73
26.02
32.88
30.75
27.85
29.71
29.79
29.66
29.64
28.16
31.49
PAINT
46.70
48.58
44.90
46.06
52.69
46.98
36.82
51.81
54.59
37.81
48.02
46.77
46.52
46.53
48.10
43.09
49.06
-------
80
|60
at
I
e
&40
20
Combustion Source (CS)
__—-—•..
Overall Females 5-17 65+ Midwest West Weekend Spring Fall
Males 0-4 18-64 Northeast South Weekday Winter Summer
Subgroup
FIREPLAC KEROHEAT OTHEAT OTHEATD WOODSTOV
FLAMES HEATY HEAVYTRF INDGAR RUNHEAV
Figure 7-3. The weighted percentage of respondents that answered "yes" to questions on exposure to combustion
sources: a fireplace, kerosene, wood stove, open flames, other heat sources, if heat was turned on
(HEATY), heavy traffic, indoor garages, or running in heavy traffic.
7-13
-------
Table 7-4. Percentage o( "Yes' Responses for Follow-Up Questions on Combustion Sources (CS)
Variable*
Overall
Males
Females
0-4
5-17
18-64
65+
Northeast
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
FIREPLAC
11.25
11.97
10.56
8.67
12.91
11.52
9.06
8.81
9.43
12.15
14.34
11.33
11.07
11.71
12.45
8.87
12.03
KEROHEAT
2.03
1.94
2.10
1.93
2.94
1.93
1.14
1.81
2.06
2.91
0.67
1.79
2.62
1.83
1.97
1.54
2.80
OTHEAT
24.11
24.38
23.86
22.11
28.35
23.43
22.24
23.14
21.91
22.20
30.98
24.50
23.16
28.14
23.93
19.68
24.68
OTHEATD
3.21
3.06
3.35
2.16
4.73
3.04
2.37
3.65
2.97
2.49
4.28
2.96
3.80
3.41
2.69
3.25
3.50
WOODSTOV
6.42
6.89
5.98
6.88
7.60
6.23
5.32
7.88
6.18
3.83
9.72
6.38
6.52
8.09
6.90
4.84
5.82
FLAMES
10.08
11.54
8.70
5.18
7.39
12.29
5.45
12.88
11.12
8.54
8.83
9.27
12.10
12.11
9.89
9.69
8.71
HEATY
41.74
42.11
41.39
41.52
42.71
40.55
46.53
48.71
48.70
34.64
38.81
42.07
40.92
81.50
22.64
2.59
60.08
HEAVYTRF
26.08
26.15
26.01
17.12
18.97
29.48
23.86
22.33
24.44
27.09
29.86
28.77
19.26
24.11
27.62
23.79
28.60
INDQAR
6.05
6.63
5.49
4.05
4.95
7.25
2.63
5.40
5.12
4.93
9.70
6.17
5.73
5.47
6.07
6.49
6.15
RUNHEAV
8.48
8.92
8.06
4.26
7.93
9.67
5.62
11.71
7.52
6.84
9.45
9.72
5.34
8.06
9.66
7.94
8.24
-------
100
a
§, 8°
ee
I
60
O
0)
I40
B
ft* 20
Washing and Bathing (WA and BA)
Overall Females 5-17 65+ Midwest West Weekend Spring Fall
Males 0-4 18-64 Northeast South Weekday Winter Summer
Subgroup
DISH DISHHAND KBATH KSWIM SHOWER WASHAH SHOWBA DISHWASH
WA WA BA BA BA WA BA WA
Figure 7-4. The weig hted percentage of respondents that answered "yes" to
questions on exposure to washing and bathing: dishwashing
machine present (DISH), dishwasher used (DISHWASH), dishes
hand-washed (DISHHAND), clothes-washing machine used
(WASHAH), a bath given or taken (KBATH), child swam
(KSWIM), a shower/bath taken (SHOWBA).
Table 7-5. Percentage of "Yes" Responses for Follow-Up Questions on Washing and Bathing (WA and BA)
Variable
Category
Overall
Males
Females
0-4
5-17
18-64
65+
Northeast
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
DISH
WA
58.07
59.23
56.97
54.15
60.35
59.58
49.59
53.80
52.82
60.37
64.62
57.50
59.48
57.61
59.20
58.32
57.13
DISHHAND
WA
79.49
72.58
86.04
22.53
56.25
87.92
86.80
78.47
78.82
77.94
84.03
79.76
78.82
81.09
79.71
78.71
76.32
WASHAH
WA
42.43
37.16
47.35
50.81
43.94
43.06
32.49
42.11
42.77
42.77
41.73
42.33
42.69
41.43
43.69
42.52
42.08
DISHWASH
WA
23.66
24.78
22.60
28.89
23.39
23.85
20.09
21.30
23.63
24.25
24.88
22.20
27.36
23.62
23.12
21.66
25.90
KBATH
BA
61.93
60.03
63.94
73.46
49.82
53.13
51.67
73.90
61.29
58.33
70.59
55.97
65.75
64.14
62.49
KSWIM
BA
15.51
14.69
16.30
25.60
24.53
13.55
5.38
15.14
12.35
17.18
16.72
15.35
15.90
5.20
17.81
32.14
6.53
SHOWER
BA
75.46
77.43
73.58
16.98
58.50
89.99
66.18
78.51
71.03
76.80
75.20
76.33
73.31
72.86
76.08
76.53
76.48
SHOWBA
BA
91.79
91.59
91.98
84.50
83.19
95.84
87.57
91.17
90.87
93.49
90.57
91.94
91.41
90.23
93.11
92.77
91.07
7-15
-------
Medical Background (MB)
Overall Females 5-17 65+ Midwest West Weekend Spring Fall
Males 0-4 18-64 Northeast South Weekday Winter Summer
Subgroup
ANGINA ASTHMA EMPHYS PREGNANT
Figure 7-5. The weighted percentage of respondents that answered "yes"
to questions on exposure to medical background: does the
respondent have angina, asthma, bronchitis/emphysema
(EMPHYS), or is the respondent pregnant.
Table 7-6. Percentage of 'Yes' Responses for Foltow-Up Questions on Medical Background
(MB)
Variable
Overall
Males
Females
0-4
5-17
18-64
65+
Northeast
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
ANGINA
2.26
2.17
2.35
0.00
0.66
1.53
9.59
1.98
2.37
2.57
1.87
2.12
2.60
2.45
2.06
1.96
2.57
ASTHMA
7.41
6.67
8.11
8.72
10.40
6.90
4.81
7.10
7.48
7.41
7.65
6.97
8.51
6.60
7.70
7.63
7.73
EMPHYS
4.27
4.11
4.41
1.62
2.84
3.68
9.85
3.63
3.41
5.69
3.45
3.85
5.31
3.73
4.20
4.25
4.89
PREGNANT
2.94
0.00
2.95
0.00
1.18
3.04
0.00
3.69
3.03
2.49
2.88
2.66
3.63
2.33
3.57
3.60
2 28
7-16
-------
100
as
i
i
30
c
09
60
20
Housing Characteristics (HC)
Overall Femalea 5-17 65+ Midwte Wat Weekend Spring Fill
Mala 0-4 18-64 Nattbau Son* Weekday Winter Summer
Subgroup
BASEMNT HUMI DOOROUT WINDY CARPORT
ELECHEAT FLOOR RENOVA ROOM
Figure 7-6. The weighted percentage of respondents that answered
"yes" to questions on the characteristics of their housing:
a basement, humidifier, electric heater, windows left open
(WINDY), rooms being added, house being renovated,
floors being refinished, carpet being laid, or doors open
directly to the outside.
Table 7-7. Percentage of "Yes1
Variable
Overall
Males
Females
0-4
5-17
18-64
88*
Northeast
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
BASEMNT
44.41
44.46
44.35
43.95
45.81
44.26
43.40
79.84
71.87
18.80
21.14
44.14
45.10
45.31
44.90
43.50
4396
Responses for Follow-Up Questions on Housing Characteristics
HUMI DOOROUT
24.06
22.63
25.42
38.92
25.51
22.75
19.65
27.38
33.34
19.03
18.39
23.70
24.97
26.84
22.27
22.11
2504
84.93
85.29
84.60
84.37
85.39
85.00
84.25
79.51
82.42
87.42
88.82
84.52
85.97
86,82
86.49
84.26
8232
WINDY
41.90
41.27
42.50
37.69
41.96
42.13
43.02
48.46
35.47
32.71
59.50
41.13
43.85
25.49
54.60
54.31
33.44
CARPORT
9.92
9.01
10.79
8.91
10.56
10.61
6.16
9.99
12.64
8.38
9.31
9.89
10.00
10.04
9.32
8.26
12.18
ELECHEAT
8.98
9.38
8.59
11.57
9.00
8.27
10.78
8.46
9.02
7.20
12.52
9.31
8.15
10.88
8.26
6.82
9.95
FLOOR
4.99
5.28
4.71
5.45
6.61
4.90
2.68
6.53
3.56
5.54
4.12
5.26
4.32
5.99
4.17
4.93
4.82
(HC)
RENOVA ROOM
34.21
34.38
34.05
37.80
40.16
34.92
19.38
41.13
37.19
29.88
31.08
34.13
34.40
31.82
32.02
34.49
38.78
3.94
3.91
3.96
6.38
5.03
3.71
1.92
3.97
3.86
3.97
3.96
3.71
4.52
4.41
3.31
3.07
5.02
7-17
-------
100
80
40
20
Other Categories (DS, ED and OQ
'~ "
Overall Females 5-17 65* Midwest West Weekend Spring Fall
Males 0-4 18-64 Northeast South Weekday Winter Summer
Subgroup
DUST
DS
FRIED
ED
MICRO
OC
OTHOUT
OC
WALKCAR
OC
BOTTWAT
ED
MAT
DS
SEAFOOD
ED
Figure 7-7. The weighted percentage of respondents that answered "yes"
to questions concerning other exposure categories: dust/dirt,
welcome mat used, foods eaten (SEAFOOD), food fried or
grilled (FRIED), microwave used, running/walking outdoors
(OTHOUT), or walking to car in the driveway (WALKCAR).
Table 7-8. Percentage of 'Yes' Responses for Follow-Up Questions on Other Categories (DS, ED, OC)
Variable
Category
Overall
Males
Females
0-4
5-17
18-64
65+
Northeast
Midwest
South
West
Weekday
Weekend
Winter
Spring
Summer
Fall
DUST
DS
16.23
17.31
15.21
10.67
18.99
18.17
5.64
14.69
16.09
15.17
19.63
16.72
14.99
14.60
15.97
17.62
16.69
FRIED
ED
23.99
23.33
24.62
12.80
27.03
25.73
17.27
20.95
25.02
25.86
22.34
22.92
26.69
24.11
25.37
23.70
22.84
MICRO
OC
55.06
51.36
58.52
100.00
38.87
58.30
58.09
49.03
55.10
57.11
57.05
55.77
53.28
59.01
53.45
50.51
57.23
OTHOUT
OC
28.46
29.93
27.07
39.52
42.72
24.97
21,18
25.63
28.77
27.90
31.68
28.21
29.11
26.17
33.57
34.08
20.45
WALKCAR
OC
72.47
73.35
71.64
57.29
71.87
76.92
59.37
71.09
73.50
71.38
74.37
72.09
73.42
70.95
74.38
74.94
69.69
BOTTWAT
ED
42.96
41.61
44.25
50.93
39.79
46.89
23.35
46.40
36.58
42.79
47.20
43.32
42.08
43.82
43.43
45.54
38.80
MAT
DS
89.33
89.59
89.09
86.13
90.24
89.83
87.52
88.72
88.51
90,28
89.28
89.37
89,22
88.39
88.46
89.70
90 87
SEASFOD
ED
59.25
60.97
57.62
41.53
45.50
65.57
59.87
61.63
52.83
62,82
58.21
58.60
60.86
61.88
58.98
57.52
58 57
7-18
-------
Section 8
Conclusions, Recommendations, and Future Work
This report has presented analyses of time spent (duration D) and number of occurrences (of people
N and microenvironments O) for selected location, activity, smoker-present, location x activity, and
location x smoker-present microenvironments by selected geographic, socioeconomic, and time factors.
These analyses were based on the minute-by-minute 24-hour diary day data collected in the National
Human Activity Pattern Survey (NHAPS). Analyses of exposure-related follow-up questions were also
presented but these took the form of the percentage of "yes" responses concerning exposure events
occurring on the day before the diary day or some other time frame (such as the previous month or 6
months) for each subgroup. The 24-hour diary results are more desirable for probabilistic modeling of
population exposure and, unlike the follow up questions, they provide detailed time information (time-of-
day, day-of-week) and allow for accurate determinations of frequency of occurrence and duration of
microenvironments. See Table 8-1 for a comparison of diary data and follow-up question data.
The follow-up questions are, in general, not conducive to classification by specific
microenvironments (beginning and ending time with a specified location, activity and/or background
activity), but since they reveal the occurrence or non-occurrence of selected exposure events for different
subgroups, some follow-up questions will provide more accurate information than the diaries. For
example, some respondents might forget they went to the gas station on the way to work and could be
reminded by a follow-up question. However, these responses could possibly be incorporated back into the
minute-by-minute diary format.
While the location categories in the NHAPS diaries were very detailed and acceptable for use in
exposure modeling, most of the exposure activity categories in the 24-hour diaries were very general (the
FACT diary variable contained more specific, verbatim descriptions of the exposure activity, but was
recorded inconsistently) and sociologically-based; and so their usefulness in exposure modeling may be
limited. For example, there were very few specific breakdowns for exposure activities involving smoking,
housing characteristics, food preparation, housekeeping, maintenance (house or automobile), and
yardwork. The use of more detailed exposure activity categories in the 24-hour diaries instead of in the
follow-up questions would probably not require any more interview time and would allow for the detail in
the limited format of the follow-up questions to be used in the more flexible diary format. Future human
activity pattern (HAP) studies should collect 24-hour diaries that explicitly categorize exposure categories
such as baking, frying, cold food preparation, vacuuming, waxing the floor, and general tidying (Table 8-
2), which correspond to different sources (or a lack of sources) of chemical pollutants. Studies that target
more specific kinds of exposure could utilize even more finely split categories such as baking with a gas
stove versus an electric stove. For these finer categories to be useful, the sample size of the study must be
large enough to provide a sufficient number of occurrences for each exposure activity. These studies will
probably be focused on much smaller geographic areas than the entire nation such as states, counties, or
cities. In addition, future HAP studies that sample the same respondents on more than one day would give
more accurate estimates of the how the duration and frequencies of occurrence of microenvironments vary
by time factors such as day-of-week, month, season, and year.
8-1
-------
Table 8-1. Comparison of the NHAPS 24-Hour Mlnute-by-Minute Diaries and the NHAPS Follow-Up Questions
Data Type
Pros
Cons
24-Hour Diaries
1. Complete time-of-day information
2. The most accurate way to obtain exposure Information
3. Complete human activity pattern data (the beginning and
ending times of locations, activities, background activities,
etc.)
4. Flexible data format for analysis by any combination of
mlcroenvlronmental factors (location, activity, smoker-
present) and any combination of demographic factors
5. Useful In probabilistic modefing of human exposure
1. Bulky data base
2. Can sometimes miss exposure events caught by the follow-
up questions, but these can be added to the minute-by-
minute diaries afterward
Follow-Up Questions
1. Obtain answers to questions on personal exposure in
specific mlcroenvlronments for different demographic
subgroups
2. Catch exposure events that might be missed in diaries
3. Relatively small data base
1. No time of day Information
2. Responses may not be as accurate as 24-hour diaries with
many missing values since the questions may be difficult for
respondents to answer; many questions on the duration or
frequency of occurrence of specific mlcroenvironments
were not answered by many respondents or were coded in
a mixed-type format, making the data unusable or difficult to
use.
3. Inflexible and/or incomplete and/or imprecise data that is
difficult or impossible to analyze across different
microenvironments (both location and activity); usually only
contain information on one specific exposure activity (e.g.,
the use of an appliance or other household/personal
products) with no specified location.
4. May not be useful in modeling of human exposure
-------
Table 8-2. Suggested Specific Exposure Activity Categories for Microenvironments in the Minute-by-Minute 24-
hour Diaries for Future HAP Studies
General Exposure Activity Examples of More Detailed 24-Hour Categories
Smoking Number of Cigarettes Smoked, Smoking Time of Each Cigarette
Housing Characteristics Was Window Open, Was Fireplace Going, Was Gas Heat On
Food Preparation Baking, Frying, Grilling, Cold Food Preparation, Using a Microwave
Housekeeping Sweeping Bare Floor, Vacuuming Carpet, Dusting, Waxing the Floor, General
Tidying, Using Air Freshener
Car Maintenance Changing Oil, Cleaning Engine, Washing the Car, Waxing the Car, Tune-up
Home Maintenance Painting, Varnishing, Removing Paint or Varnish, Sanding, Carpentry, Electrical
Repair
Yarclwork Mowing the Lawn, Gardening, Trimming Trees, Applying Pesticide
Sports/Exercise Running, Biking, Hiking, Swimming, Football, Basketball
Note: Many of these more detailed 24-hour diary exposure activity categories are the same as those contained in the follow-up
question part of the interview (see Section 7).
The NHAPS 24-hour diary data should be used in the future for many different kinds of exposure
assessments - perhaps most importantly for the probabilistic modeling of human exposures for different
populations in the United States. The raw data can be used to obtain minute-by-minute 24-hour diaries
for many subgroups, e.g., female respondents aged 5-11 in Texas. The NHAPS data should also be
systematically compared to previous national studies (e.g., the American's Use of Time Study4-18 and the
California Activity Pattern Survey (CAP) l~*) by the amounts of time people spend in different
microenvironments. Additionally, NHAPS can be used to determine commuting patterns and transition
probabilities between microenvironments.
8-3
-------
References
1. Jenkins, P. L., Phillips, T. J., Mulberg E. J., and Hui, S.P., (1992) "Activity Patterns of Califomians:
Use of and Proximity to Indoor Pollutant Sources," Atmospheric Environment, Vol 26A, No. 12, pp.
2141-2148.
2. Wiley, J., Robinson, J., Piazza, T., Garrett, K., Cirksena, K., U. Cheng and G. Martin, (1991)
"Activity Patterns of California Residents," Final Report Under Contract No. A6-177-33, California
Air Resources Board, Sacramento, CA.
3. Wiley, J., Robinson, J., Piazza, T., Stork, L. and Pladsen, K., (1991) "Study of Children's Activity
Patterns," Final Report Under Contract No. A733-149, California Air Resources Board, Sacramento,
CA.
4. Robinson, J.P., (1987) 'Time Diary Evidence About the Social Psychology of Everyday Life," J.
McGrath, ed., Sage Publications, Newberry Park, CA.
5. Ott, W., (1985), "Total Human Exposure: An emerging science focuses on humans as receptors of
environmental pollution," Feature Article, Environmental Science and Technology, Vol. 19, pp. 880-
885.
6. Ott, W., (1990) "Total Human Exposure: Basic Concepts, EPA Field Studies, and Future Research
Needs," Journal of Air & Waste Management Association, Vol. 40, No. 7, pp. 966-975.
7. Wallace, L., (1993) "A Decade of Studies of Human Exposure: What Have We Learned?," Risk
Analysis, Vol. 13, No. 2, pp. 135-143.
8. Robinson, J.P. and Blake, 3., (1995) "Estimating Exposure to Pollutants Through Human Activity
Pattern Data: The National Microenvironmental Activity Pattern Survey," Annual Report, Survey
Research Center, University of Maryland.
9. Nelson, W., Ott, W., and Robinson, J. (1994) "The National Human Activity Pattern Survey
(NHAPS): Use of Nationwide Activity Data for Human Exposure Assessment," paper no. 94-
WA75A.01, presented at the 87th Annual Meeting and Exhibition of the Air & Waste Management
Association.
10. Pandian, M.D., Bradford, J. and Behar, J.V., (1990) "THERDbASE: Total Human Exposure
Relational Database," in Proceedings of the EPA/AWMA Symposium on Total Exposure
Assessment Methodology - New Horizons, Las Vegas.
R-l
-------
11. Lurmann, F. W. and Korc, M. E. (1994) "Characterization of Human Exposure to Ozone and PM-10
in the San Francisco Bay Area," Final Report STI-93150-1416 FR, for the BAAQMD, San
Francisco, CA.
12. Behar, J.V., Thomas, J., and Pandian, M.D., "Estimation of the Exposure to Benzene of Selected
Populations in the State of Texas Using the Benzene Exposure Assessment Model (BEAM)," EPA
600/X-93/002, Environmental Monitoring Systems Laboratory, U.S. Environmental Protection
Agency, Las Vegas, NV, January 1993.
13. Klepeis N. E., Ott W., and Switzer P., (1994) "A Total Human Exposure Model (THEM) for
Respirable Suspended Particles (RSP)," National Technical Information Service (NTIS) No. PB94-
197415, Presented at the 87th annual meeting of the A&WMA meeting in Cincinnati, OH.
14. Robinson, J., Switzer, P., Ott, W., (1994) "Smoking Activities and Exposure to Environmental
Tobacco Smoke (ETS) in California: A Multivariate Analysis," Report No. 1 for the California
Activity Pattern Survey, Department of Statistics, Stanford University, Stanford, CA.
15. Robinson, J., Switzer, P., Ott, W., (1994) "Exposure to Environmental Tobacco Smoke (ETS)
Among Smokers and Nonsmokers," Report No. 2 for the California Activity Pattern Survey,
Department of Statistics, Stanford University, Stanford, CA.
16. Robinson J., Switzer, P., Ott, W., (1994) "Microenvironmental Factors Related to Califomians'
Potential Exposures to Environmental Tobacco Smoke (ETS)," Report No. 3 for the California
Activity Pattern Survey, Department of Statistics, Stanford University, Stanford, CA.
17. Ott, W., Switzer, P., Robinson, J., (1994) "Exposures of Califomians to Environmental Tobacco
Smoke (ETS) by Time-of-Day: A Computer Methodology for Analyzing Activity Pattern Data,"
Report No. 4 for the California Activity Pattern Survey, Department of Statistics, Stanford
University, Stanford, CA.
18. Robinson, J.P. and Thomas J., (1991) 'Time Spent in Activities, Locations, and Microenvironments:
A California-National Comparison," Las Vegas, Nevada: Environmental Monitoring Systems
Laboratory, U.S. EPA.
19. Fugas, M., (1975) "Assessment of Total Exposure to Air Pollution," in Proceedings of the
International Conference on Environmental Sensing and Assessment, Las Vegas, NV, Paper No. 38-
5, Vol. 2, IEEE #75-CH1004-l ICESA.
20. Duan, N., (1982) "Microenvironment Types: A Model for Human Exposure to Air Pollution,"
Environment International, Vol. 8, pp. 305-309.
21. Ott, W., (1982), "Concepts of Human Exposure to Air Pollution," Environment International, Vol.
7, pp. 179-196.
22. Wallace, L.A., (1987) "The Total Exposure Assessment Methodology (TEAM) Study: Summary
and Analysis: Volume L" U.S. EPA, Washington, D.C.
R-2
-------
23. Pellizzari, E.D., K. Perritt, T.D. Hartwell, L.C. Michael, R. Whitmore, R.W. Handy, D. Smith, H.
Zelon. (1987) "Total Exposure Assessment Methodology (TEAM) Study: Elizabeth and Bayonne,
New Jersey; Devils Lake, North Dakota; and Greensboro, North Carolina, Vol. O," U.S. EPA,
Washington, DC.
24. Pellizzari, E.D., K. Perritt, T.D. Hartwell, L.C. Michael, R. Whitmore, R.W. Handy, D. Smith, H.
Zelon, (1987) "Total Exposure Assessment Methodology (TEAM) Study: Selected Communities in
Northern and Southern California, Vol. m," U.S. EPA, Washington, DC.
25. Pellizzari, E.D., Thomas, K.W., Clayton, C.A., Whitmore, R.W., Shores, R.C., Zelon, H.S., and
Perritt, R.L. (1992) "Particle Total Exposure Assessment Methodology (PTEAM): Riverside,
California Pilot Study," Report No. RTI/4948/108-02F prepared for the U.S. Environmental
Protection Agency by Research Triangle Institute, Research Triangle Park, NC.
26. Ozkaynak, H., Xue, J., Weker, Butler, and Spengler, J. (1994) "The Particle Team (PTEAM) Study:
Analysis of the Data," Draft Final Report, Volume m, prepared under Contract No. 68-02-4544.
27. Ott, W., (1984) "Exposure Estimates Based on Computer Generated Activity Patterns," Journal of
Toxicology: Clinical Toxicology, Vol. 21, pp. 97-128.
28. Ott W., J. Thomas, D. Mage, and L. Wallace, (1988) "Validation of the Simulation of Human
Activity and Pollutant Exposure (SHAPE) Model Using Paired Days from the Denver, CO, Carbon
Monoxide Field Study," Atmospheric Environment, Vol. 22, No. 10, pp. 2101-2113.
29. Ott W., Mage, D., and Thomas, J., (1992) "Comparison of Microenvironmental CO Concentrations
in Two Cities for Human Exposure Modeling," Journal of Exposure Analysis and Environmental
Epidemiology, Vol. 2, No. 2, pp. 249-267.
30. "1990 Census of Population and Housing," (1992) Summary Social, Economic, Housing
Characteristics, United States, U.S. Department of Commerce.
31. "SAS Procedures Guide," Release 6.03 Edition, (1988) SAS Institute, Inc., SAS Circle, Box 8000,
Gary, NC 27512, p. 317.
32. Cochran, W.G., Sampling Techniques, 3rd edition, John Wiley & Sons, New York, 1977, p. 91.
33. Switzer, P., and Ott, W. (1992) "Derivation of an Indoor Air Averaging Time Model from the Mass
Balance Equation for the Case of Independent Source Inputs and Fixed Air Exchange Rates,"
Journal of Exposure Analysis and Environmental Epidemiology, Vol. 2, Suppl. 2, pp. 113-135.
34. Ott, W., Langan, L., and Switzer, P. (1992) "A Time Series Model for Cigarette Smoking Activity
Patterns: Model Validation for Carbon Monoxide and Respirable Particles in an Chamber and an
Automobile," Journal of Exposure Analysis and Environmental Epidemiology, Vol. 2, Suppl. 2, pp.
175-200.
R-3
-------
35. Klepeis, N., Ott, W., Switzer, P, (1995) "Modeling the Time Series of Carbon Monoxide and
Respirable Suspended Particles from Multiple Smokers: Validation in Two Public Smoking
Lounges," presented at the 88th Annual Meeting and Exhibition of the A&WMA, San Antonio, TX,
June 1995.
36. Ott, W., Klepeis, N., and Switzer, P., (1995) "Modeling Environmental Tobacco Smoke in the Home
Using Transfer Functions," presented at the 88th Annual Meeting of the A&WMA, San Antonio,
TX, June 1995.
R-4
-------
Appendix A
Conceptual Methodology and Background
Al. Total Human Exposure, Human Activity Patterns, and the Generalized Microenvironment
The study of human exposure to pollutants has recently been focused away from the traditional
sources (industrial waste, factory emissions, etc.) and onto the non-traditional sources (consumer
products, building materials, etc.) with which a person typically comes into personal contact during their
daily routine. With the recognition that most exposure sources are associated with personal activities and
consumer products,7l22'26 the concept of total human exposure (THE) was introduced to consider all
possible routes of exposure by which a pollutant reaches a human target. Ott21 defines exposure as the
confluence of two events in space and time: (1) person i is present at location x,y,z at time t; and (2)
pollutant concentration c is present at location x,y^. at time t.
With the individual human being - and his or her movement though space and time - placed at the
center of attention, the most important part of THE studies has been the human activity pattern (HAP),
which includes all exposure-relevant information about a person's behavior (locations, activities, etc.)
over time. Human activity patterns do not describe the mechanisms of exposure or the flow of pollutants
from sources; these topics are left to mass-flow modeling or monitoring surveys. Instead, HAP's
describe a person's activities or activities occurring nearby that person, and the type of location in which
those activities are occurring. The six most basic exposure-related questions one might ask in human
activity pattern studies — along with their measured or calculated NHAPS quantities in parentheses — are
(Figure A-l): "What are people doing (activities, background activities)?"; "Where are they doing it
(locations)?"; "Who are they (demographics)?"; "How long were they doing it (duration)?"; "How often
were they doing it (frequency of occurrence)?"; and "When were they doing it (time of day, season, day
of week)?"
Most generally, person i's position over time is given by [x(t), y(t), z(t)]t for any spatial and temporal
resolution of interest. However, since (1) it may not be possible to determine either human activity
patterns or pollutant concentrations c(x,yj,f) with fine spatial or temporal resolution, and (2) pollutant
concentrations may be homogeneous in certain locations, a "discrete model" for exposure to air pollution
was proposed,19"21 using the microenvironment concept in which a person is exposed to a uniform air
pollutant concentration c in a certain location and block of time. Ott21 defines a microenvironment as a
discrete event that "denotes a clearly defined air pollution problem and a unique class of activities".
Similarly, Duan20 defines a microenvironment as "a location of homogeneous pollutant concentration that
a person occupies for some finite period of time". Only Behar et al.n describe a microenvironment as a
combination of a location and an activity.
A-l
-------
[ What? 1 [where?
Figure A-1. The most basic questions posed in human activity pattern (HAP) analysis: The What? and
Where? questions specify the microenvironmental factors, Who? and When? determine the
background (gender, age, census region) and time (time-of-day, day-of-week, season) factors,
respectively, and How Long? and How Often? determine the duration D and frequency of
occurrence O of the microenvironments, respectively.
Typically, human activity patterns such as NHAPS are analyzed and incorporated into exposure
studies through the microenvironment concept.5-6'11"13119"21'27"29 In broadening the original definition, we
propose the generalized n-factor microenvironment, which: (1) in addition to locations includes specific
pollutant-generating activities in the definition of a microenvironment to make full use of detailed HAP
surveys like NHAPS; (2) expands and generalizes the microenvironment concept for estimation of
exposures from any pollutant source and through any pathway; (3) is not limited to air pollutants that are
homogeneously distributed; and (4) mathematically formalizes the microenvironment concept for use in
scientific estimates of total human exposure. As the foundation of our definition, we establish the
following three ideas:
• Whereas the term "environment" encompasses all factors that influence an individual life, a
microenvironment is some fraction of all possible discrete events (or episodes) that occur
during a defined portion of time in an individual life.
• A microenvironmental factor is a discrete event related to human behavior such as a location, an
activity, some other simultaneous activity, or a background activity.
• Microenvironments are fully characterized by a single microenvironmental factor or
combination of any number n of microenvironmental factors and the corresponding time in
which each microenvironmental factor or combination of factors occurs.
Examples of microenvironmental factors are: (1) bedroom, office, automobile, etc. for locations: (2)
cleaning the floor, watching TV, mowing the lawn, etc. for activities: (3) using air freshener while
watching TV, smoking a cigarette while cleaning the floor, applying pesticides while mowing the lawn,
etc. for simultaneous activities that the respondent is doing directly; and (4) presence of a smoker,
A-2
-------
washing machine running, kerosene heater on, etc. for possible background activities that the respondent
is not directly controlling. Any one of these microenvironmental factors or any conceivable combination
of them may comprise the n-factor portion of our generalized microenvironment definition. In this report
we present results for all location, activity, and smoker-present categories and location x activity and
location x smoker-present microenvironments. Section 5 of this report presents the location x activity
and location x smoker-present microenvironments that were used.
Once the microenvironment factor(s) of interest has(have) been specified, the characterization of the
microenvironment is completed by assigning the appropriate starting and ending times for which the
microenvironmental factor(s) occurs(occur) over the desired time period. A specific occurrence of a
microenvironment — some combination of microenvironmental factors, a beginning time, and an ending
time — is called an episode. In this report human microenvironment patterns are analyzed by their
beginning and ending times (by episodes) and in increments of one minute over one 24-hour diary day
(by time-of-day).
A microenvironment is an abstract notion whose exact definition depends on the relevant micro-
environmental factors and the time frames of interest. The microenvironment is a "window of exposure"
during which human exposure to any pollutant source (exhaust, paints/solvents, aerosol sprays, etc.), via
any pathway (air, water, soil, food), and in any locale (home, automobile, playground, etc.), can occur.
The detailed mechanisms of exposure for each microenvironment, which follow from the relevant
microenvironmental factors, are experimentally and/or theoretically characterized by scientists. For
example, once a person is determined to be inside a large room with multiple cigarette smokers, a
physical mass flow model can be applied to calculate the typical exposure that a person will receive.35
A2. A Mathematical Formalism for Microenvironments
Our generalized n-factor microenvironment M is a relation between the set of n combined
microenvironmental factors F = Fj* F2\ ... x. Fn and the set T of their corresponding beginning times (B,
...B^) and ending times (E,... £„):
M:T-F
F = F1xF2xF3x...xFn
r={B1,..Bn,E1,..,En}
where
F = the set of n combined microenvironmental factors
T = the set of corresponding beginning and ending times
The microenvironmental factor portion is the same for any microenvironment formulation, but the
formulation may change depending on the time period or time scale of interest. For example, if
researchers wish to study occurrences of exposure events on an absolute scale of small fixed-length
intervals over a 24-hour period — as in a time-of-day study - then a different formulation is required (the
time-of-day formulation) than if researchers simply desire a tabulation of starting times, durations, or
number of occurrences of exposure events over a day, week, month, or year (using the episodic
A-3
-------
formulation). An episode is an unspecified occurrence of an exposure event — with beginning and
ending times — that is classified according to the microenvironment concept.
Microenvironmental Factors
Each combined microenvironmental factor can correspond to multiple beginning and ending times.
Each microenvironmental factor Ft has ak elements, and there are a, x a2 x,..., x an possible combinations
of microenvironmental factors denoted by
F • e F
W""V fc r
where
(2)
For one-factor microenvironments the F portion of M consists of a single microenvironmental factor
(location, activity, background activity, etc.). For two-factor microenvironments, F consists of the set of
all possible combinations of the two microenvironmental factors F, and F2 (location and activity, location
and background activity, etc.)
F F F
M.I M.2 — M
F F F
r2,i rz,2 ••• ^^
'•• •• F, t ;
F F F
a,.2 a,4 '" a
= F
where
Fj = microenvironmental factor #1
F2 = microenvironmental factor #2
GI = number of elements in Fj
02 = number of elements in F2
(3)
A-4
-------
Likewise, the F portion of M for three-factor microenvironments consists of all combinations of F,, F2,
and F3 and can be represented as follows (each cell contains the appropriate F subscript only):
111
211
311
a, 11
1113 jr 12fl3 jS s^ 11
X
X
2fl22
X
3fl22
X
X
X
X
X
X
202*,
X
X
0,421
The Episodic Formulation
The most condensed format for human microenvironment data — which we call the episodic
formulation — contains the combined microenvironmental factors linked with the times that they begin
and end. It is most useful for storing the data and for calculations of exposure-event duration and
frequency of occurrence over a 24-hour period (or week, month, year, etc.). The NHAPS data is stored in
this format by a simple list of each beginning and ending time and multiple variables that specify each
microenvironmental factor (location, activity, and presence of a smoker).
Mathematically, we write the episodic formulation of the set of all generalized n-factor
microenvironments M, (the e subscript stands for episodic) for person / and episode j as follows:
Me (i,j;kj,...,kj
where
N = number of individuals
(4)
/, = number of episodes for person i each with beginning time Bj and ending time E}
ki,...,kn = subscripts of the combined microenvironmental factors Fkl ^
*, e {I,2,...a,},fc2e {I,2,...a2},....^n e {l,2,...an}
al,...,an = number of elements in each microenvironmental factor
A-5
-------
The Time-of-Day Formulation
In exposure modeling, we need a microenvironment formulation that allows us to conduct time-of-
day comparisons of human exposure across arbitrary time periods. Instead of studying exposure events
by episodic "chunks", we create an absolute time scale that provides a homogeneous base for
comparisons. For example, by assigning exposure events — combinations of microenvironmental factors
- to every minute of the day, we may calculate a time series of exposure for each person and
subsequently derive hourly, 8-hour, 12-hour, or 24-hour exposures which can be compared across the
population.
This time-of-day formulation of the generalized n-factor microenvironment consists of sequential
time intervals of fixed length over a given time period. While the episodic formulation contains only
beginning and ending times mapped to each combined microenvironmental factor, this formulation is a
mapping of each combined microenvironmental factor f ^^...jc^to each time interval j. Different time
intervals j can point to the same fkljkr..^kf . For each person i, the set of all generalized n-factor
microenvironments M, (the subscript t stands for time-of-day) in the time-of-day formulation is:
M, (ij;k,,...,kj
where
N = number of individuals
;e{l,2,3,...,7} (5)
/ = number of time intervals in time period, e.g., 1440 minutes in 24-hours
fcj,... Jcn = subscripts of the combined microenvkonmental factors Fu fa
fc, € {U,...a,},^e { U,...^},-.-^ e {l,2,...aj
£,...,£ = number of elements in each microenvironmental factor
M, is an expansion of Mt from a relation that links each existing microenvironmental factor with its
starting and ending times to an onto mapping from the time segments to the microenvironmental factors.
As long as the length of the fixed time segments in M, are at the same time resolution as the starting and
ending times specified in Mf then the set M , has 2+n dimensions which fully describes the human
microenvironmental patterns of a population of N people by retaining all the information stored in M,,
and the time-of-day formulation can be converted back to the episodic formulation since every beginning
and ending time corresponds to some time interval. However, if the time intervals j (typically one
minute) are of greater resolution than the durations of each microenvironment, then duration information
is destroyed within each time interval for each episode whose ending time exceeds a time interval's right
boundary.
This formalism is helpful in efforts that apply human microenvkonmental patterns to exposure
modeling, especially when n z 2. It can be used to develop computer algorithms to calculate total human
exposure for populations.
A-6
-------
A3. The Microenvironment Calculations in This Report
In this report we calculate for all 9,386 NHAPS respondents: (1) durations D and frequencies of
occurrence O of microenvironments for the 24-hour diary day using the episodic formulation of
microenvironments, and (2) frequencies of occurrences O of microenvironments for each minute and
each 3-hour segment of the 24-hour diary day using the time-of-day formulation of microenvironments.
For episodic calculations, the episodes for each person are sorted according to microenvironmental factor
combinations ^fc,,^,...,*,,. Since each episode is sorted together with its beginning and ending times, we
-------
where
/ (1 10}
j {1....../J (7)
Ji {5,5,3,1,2,5,2,3,1,3}
Bj {l2:OOam,...,ll:59am},Ej {12:01om,...,12:00pm}
k {1,2}
We can represent the series of episodes including beginning times Bj, ending times £,, and location codes
(k = 1 for indoors and Jfe = 2 for outdoors) in a table with a variable number of records per person:
Table A-1. Episodic Formulation of Ten Persons in Indoor (k=1) and Outdoor (k=2) Microenvironments Over a
12-hr Period
Person f1
Bi
12,-OOa
8:01 a
8:31 a
10:31 a
11:01 a
Ey
8:00 a
830 a
10:30 a
11:00a
12.-OOP
k
1
2
1
2
1
Person #2
s/
12:00 a
5:31 a
9:31 a
10:01 a
11:01 a
Ey
5:30 a
9:30 a
10:00 a
11:00a
12:00 p
k
1
2
1
2
1
Person #3
B,
12:00 a
8:31 a
10:46 a
Ey
830 a
10:45 a
12,-OOp
Person #4
k B, E, k
1 12:00 a 12:00 p 1
2
1
Person f 5
Bj E, k
12:00 a 11:00 a 1
11:01 a 12:00 p 2
Person f6
BJ Ey
1230
3:31
4:01
6:31
1131
a
a
a
a
a
330 a
4:00 a
630 a
11:30a
12*»p
fr
2
1
2
1
2
Person #7 Person #8
B, $ k Bj E,
12:00 a 450 a 2 12£0a
4:31 a 12:00 p 1 631 a
11:46 a
630 a
11:45 a
12:00 p
Person #9
k 8, E, k
1 12:00 a 12:00 p 1
2
1
Person f 10
12:00 a
10:46 a
11:01 a
10:45 a
11:00a
12:00p
k
1
2
1
The cumulative duration (in minutes) and frequency of occurrence for outdoors and indoors for the
ten people over the 12-hour period - calculated after the people are sorted by location — are D (i,l)
{660,420,585,720,660,330,450,405,720,705} for time indoors, D (i,2) {60, 300, 135,0,60,390,
270,315,0,15} for time outdoors, O(i,l) {3,3,2,1,1,2,1,2,1,2} for occurrences indoors, and O(i,2)
{2,2,1,0,1,3.1,1,0,1} for occurrences outdoors. The average durations are 566 minutes for indoors and
155 minutes for outdoors, and average frequencies of occurrence are 1.8 for indoors and 1.2 for outdoors.
By adding up all the time spent in the indoor (5,655 minutes) vs. outdoor (1,545 minutes)
microenvironments (10 people x 720 minutes = 7,200 total minutes), we determine that for the
population of ten people, 79% of the time was spent indoors.
To calculate the proportion of microenvironmental factor occurrences across the population for each
half-hour, the microenvironments for the ten people are converted into the time-of-day formulation at
A-8
-------
one-minute resolution (Table A-2). Notice that whereas the episodic formulation requires only 30 set
elements (10 people x 3 variables), the time-of-day formulation requires 10 people x 24 time segments =
240 set elements. Also notice that we lose exact duration information by choosing half-hour time units,
which are less than the one-minute time resolution of the beginning and ending times. The frequencies of
indoor and outdoor microenvironments in half-hour increments over all ten people (Table A-2) can be
converted into percentages and plotted to present the movement of the population in and out of doors
over the 12-hour study period (Figure A-2). For most of the morning, 80% of the people are indoors,
with between 60 and 80% indoors from 8:00 am to noon.
A4. Modeling Human Exposure
Since exposure is usually reported in pollutant concentrations c,, before total human exposure can
be estimated for a population, we require knowledge of the pollutant concentration in each
microenvironment;. Concentrations are either measured directly using personal monitoring surveys, or
predicted with validated microenvironmental exposure models that incorporate mass balance descriptions
of pollutant flow.33'36
Duan20 has proposed a model that expresses the integrated exposure Ei for a population ofN
people's exposure to air pollutants over a specified time period T:
where
Et = the integrated pollutant exposure of person i
i = 1,2,3,.. .,N people
c = the pollutant concentration encountered in microenvironment j
tfj = the time spent by person i in microenvironment j
J - the number of microenvironments visited
Person f s integrated exposure over all microenvironments; is computed by multiplying the pollutant
concentration c, by the microenvironment' s duration tif. The average exposure of person i over the time
period T is E/T. This model - based on the episodic formulation of microenvironments - does not
explicitly incorporate a time sequence of microenvironments or a generalized microenvironmental
definition that accounts for all kinds of simultaneous exposure.
Klepeis, Ott, and Switzer13 have presented a framework for using microenvironmental data in
modeling human exposure to particles on personal computers - based on the time-of-day formulation of
microenvironments - that constructs 1440-minute arrays containing location, activity, and smoker-
present codes for each person over a 24-hour period: the Total Human Exposure Model (THEM). The
exposure E(ij) of an individual i at any minute; is a function of L, A, and S arrays, which contain
location, activity, and smoker-present codes, respectively:
ij), Sd\ffl
A-9
-------
Table A-2. Tlme-of-Day Formulation for Ten People (M..10) With Frequencies of Indoor (k=1) vs. Outdoor (k=2) Mlcroenvironments Over 12-Hours in Half-Hour
Increments __^____
Tlme/« 12:00 12:30 1:00 1:30 2:00 2:30 3:00 3:30 4:00 4:30 5:00 5:30 6:00 6:30 7:00 7:30 8:00 8:30 9:00 9:30 10:00 10:30 11:00 11:30
am am
/=1
2
3
4
5
6
7
8
9
10
0(/,D=
0(/,2)=
fc=1
1
1
1
1
2
2
1
1
1
8
2
1
1
1
1
1
2
2
1
1
1
8
2
1
1
1
1
1
2
2
1
1
1
8
2
1
1
1
1
1
2
2
1
1
1
8
2
1
1
1
1
1
2
2
1
1
1
8
2
1
1
1
1
1
2
2
1
1
1
8
2
1
1
1
1
1
2
2
1
1
1
8
2
1
1
1
1
1
2
2
1
1
1
8
2
1
1
1
1
1
1
2
1
1
1
9
1
1
1
1
1
1
2
2
1
1
1
8
2
1
1
1
1
1
2
1
1
1
1
9
1
1
1
1
1
1
2
1
1
1
1
9
1
1
2
1
1
1
2
1
1
1
1
8
2
1
2
1
1
1
2
1
1
1
1
8
2
1
2
1
1
1
1
1
2
1
1
8
2
1
2
1
1
1
1
1
2
1
1
8
2
1
2
1
1
1
1
1
2
1
1
8
2
2
2
1
1
1
1
1
2
1
1
7
3
1
2
2
1
1
1
1
2
1
1
7
3
1
2
2
1
1
1
1
2
1
1
7
3
1
1
2
1
1
1
1
2
1
1
8
2
1
1
2
1
1
1
1
2
1
1
8
2
2
2
1
1
1
1
1
2
1
2
6
4
1
1
1
1
2
1
1
2
1
1
8
2
Note: 0(/, 1) = Frequency of Being Indoors (/r = 1) for half-hour Increment /. 0(/, 1) = Frequency of Being Outdoors (k = 2) for half-hour Increment /.
D Indoors I Outdoors
»ij •—« «—l
«
Time of Day
Figure A-2. Percentage of people In indoor vs. outdoor microenvironments by time-of-day in half-
hour Increments over a 12-hour period.
-------
Each different kind of exposure to particles (tobacco smoke, wood smoke, cooking, cleaning, etc.) is
calculated by a subprogram (TS(ij), WS(iJ), CK(ij), CL(ij), etc.) that depends on the microenvironment that
each person i visits during a series of minutes;. The total exposure E(ij) is the sum of all of these different
kinds of exposure according to the "superposition hypothesis".2738 Such a framework is very versatile for
calculations of average total exposure over any time period (hourly, 8-hour, 24-hour) since it contains minute-
by-minute exposure information. It can be improved by generalizing the kinds of exposure to any kind of
simultaneous microenvironments (including exposure to ambient pollutant sources) as we have done above
with the generalized microenvironment M.
Building on the time-of-day microenvironmental formulation we have developed above, total human
exposure E(i,f) is a function of the generalized set of microenvironments M for person i and time intervals;
(minutes):
E(iJ)=f[M(iJ;k,,...,kn)]
M can be decomposed into microenvironmental factors F for person i and time interval;:
U M(ij;F,,...^) =^(1^ ..... JJM(ijf^...\)M(ij-fat^) (11)
=,...,£„=!
Example
Consider a two-factor location x smoker-present microenvironments each with two elements (fc, = 1 for
indoors, k: = 2 for outdoors, ^ = 1 for no-smoker-present, and fcj = 2 for smoker-present):
U M(ij;F ^MdjfUMd j;F1)2)UM(ij;F2 ..jUMft/^ (12)
M is based on the union of each combination of microenvironmental factors (Table A-3 illustrates M(iJ; F)
for person #1 from the example in Table A-2). The set M(l j;Fu) corresponding to "no smokers indoors" for
person #1 (Table A-3) contains the most elements (14) with M(\j;F^ for "smoker present indoors" having
the next highest number of elements (6).
To determine exposures for each combination of location and smoker-present categories for person i at
minute; the corresponding exposure array element E(iJ) is filled in using appropriate exposure subprograms
that operate on each element of M(ij; F) (fn for indoors/no-smoker-present,/12for indoors/smoker-present, /21
for outdoors/no-smoker-present, and/^ for outdoors/smoker-present) - assigning null values if no exposure
has taken place. The total human exposure E(ij) for person i over the time intervals; is the union of the
resulting exposures e(ij\ Ftljk2) (Table A-3):
u) U e(iJ;F^ U e(iJ;Fu) U
A-ll
-------
Table A-3. Example (n=2) Breakdown of the Time-of-Day Formulation of the Generalized n-Factor
MicroenvJronment Mjj; F = FMJ(2) for Person /= 1, and Location (Indoor/Outdoor) and Smoker-
Present Microenvironmental Factors (fr, = 1 for indoors, /c, = 2 for outdoors, fc, = 1 for smoker-present,
1% = 2 for no-smoker-present)
.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
w
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
^
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
1
1
1
2
2
1
\ *
*f
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
1
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
«*U
j
1
2
3
4
5
6
7
8
9
10
11
12
19
24
W
1
1
1
1
1
1
1
1
1
1
1
1
1
1
k, i j k, k, i j k, k, i j k, k,
1
1
1
1
1
1
1
11
1
1
1
1
1 13 1 2
1 14 1 2
1 15 1 2
1 16 1 2
1 17 2 2
1 18 2 2
1
1 20 1 2
1 21 1 2
1 22 2 1
1 23 2 1
1
In Table A-4 the elements of the exposures e(lj; Fu) and e(\ J; F^) for person #1 all contain exposures
of 0 mg/m3 since there is no smoker present, while the elements of e(\ j; F1-2) and e{\ J; F2>2) contain indoor
and outdoor smoking exposures calculated from the/12 and/^ subprograms, respectively - note that these are
hypothetical values.
In general, the exposure subprograms can either be physical submodels that predict exposures based on
microenvironment characteristics (air flow, pollutant source strength, etc.)33"36 or probabilistic submodels that
randomly sample typical exposures from monitoring data (using the Monte-Carlo method). The same
subprogram can be used for different elements of M(ij; F), and some subprograms (e.g.,/n,/21) may simply
function to fill in every element in the exposure array with values of zero. For microenvironments that
incorporate simultaneous exposures (e.g., exposure to particles from both a smoker and someone vacuuming
carpet), exposure values are summed to give the total exposure (by the superposition principle).
A-12
-------
Table A-4. Breakdown of the 12-Hour (24 half-hour intervals y) Total Exposure Array E(/,/) for Person /= 1 into
the Exposure Arrays e(ij, FM ^) Containing Hypothetical Exposure Concentrations c, (mg/m3)
Calculated from Exposure Subprograms f11f /12, f21, fa for Each Different Combination of
Microenvironmental Factors (ft, = 1 for indoors, fr, = 2 for outdoors, ^=1 for no-smoker-present, and
fr,=2, for smoker-present)
.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
™
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
?4
c,
0
0
0
0
0
0
0
0
0
0
0
0
20
20
140
0
40
50
0
130
140
0
0
0
e(
1
1
1
1
1
1
1
1
1
1
1
1
1
1
j
1
2
3
4
5
6
7
8
9
10
11
12
19
24
c, / j c, / j c, / j c,
0
0
0
0
0
0
0
0
0
0
0
0
1 13 20
1 14 20
1 15 140
1 16 120
1 17 40
1 18 50
0
1 20 130
1 21 140
1 22 0
1 23 0
0
Note: Instead of containing location or smoker-present codes the exposure arrays e(/,/;FMJB) contain exposure concentrations
in mg/m3. Microenvironments that contain multiple-simultaneous exposures (e.g., smoker-present, dust, car exhaust,
etc.) are treated by each exposure subprogram (f^,fn,fa,f^, by branching to cigarette, dust, or other exposure
submodels - that account for pollutant cross-effects or assume simple superposition of exposures - or by sampling from
distributions of typical exposures. Each possible combination of exposure factors can theoretically give a different kind
of exposure.
A-13
-------
Appendix B
Detailed Examples of the
Methodology Used for Calculations in this Report
(With SAS31 Computer Code)
Bl. Episodic: Duration and Frequency of Occurrence of Microenvironments
The episodic formulation of microenvironments (beginning and ending times with matched
microenvironment codes for each record) is useful for calculations of the percentage of time spent, mean
24-hour durations, and mean 24-hour frequencies of occurrence for microenvironments occurring
throughout the diary day. An example of the episodic format, which is the native NHAPS format, is
found in Table B-l. Each record contains the beginning and ending times of each episode with the
corresponding original NHAPS microenvironment variables (FACT, RACT, WHR, SMK) and the
receded variables (NEWACT, REGACT, and NEWLOC). The episode duration is also included
(ETOA). There is a new record (= microenvironment) every time the respondent entered a new location,
activity, or smoker-present category. Counts of the number of records containing each location or each
activity can greatly overestimate their 24-hour frequency of occurrence because: (1) the same location
can occur consecutively when the respondent experiences different activities and vice versa; and (2) after
the original microenvironment codes were receded, it was possible for the identical microenvironment to
occur in two consecutive records, although this phenomena seems to occur mosdy when REGACT = 0.
The two columns in Table B-2 compare the 24-hour frequency computed by simply counting the
consecutive occurrences only versus counting records with the same consecutive locations, activities,
etc., only once, i.e., merging diese records. The 24-hour duration of each microenvironment and the
percentage of time spent in each microenvironment is calculated for Person ED#1.1 (Table B-2). The
mean 24-hour duration and 24-hour frequency of occurrence of each microenvironment over the NHAPS
respondents were obtained by averaging each individual's 24-hour duration (SAS code in Figure B-l).
The percentage time spent in each microenvironment over all respondents was obtained by dividing the
amount of time everyone spent in each microenvironment (calculated with the SAS code in Figure B-l)
by the total amount of time spent over die entire diary day by all respondents (9,386 x 1440 min =
13,515,840 min, unweighted) or a subgroup of the respondents (by age, gender, etc.).
B2. Time-of-Day I: Fraction of Respondents in Each Microenvironment by Minute
The time-of-day formulation of microenvironments (sequential fixed-length time segments matched
with microenvironment codes) is used for calculations of the fraction of respondents experiencing each
microenvironment during every minute of the diary day. The original episodic data storage format was
converted into a one-minute fixed time length format so that each record of the data base specified the
microenvironments diat a person entered in each hour of the day and for which of the 60 minutes he/she
was there (see SAS code in Figure B-2 and example output in Table B-3).
B-l
-------
Format Conversion
In the conversion process from episodic to time-of-day formats, the input file containing the original
episode and demographic information for all respondents was Nl .TIMRES2. The output data files
containing the one-minute fixed time length format for each hour of the diary day were Nl .SUMACTO,
N1.SUMACT1 N1.SUMACT23 with Nl.SUMACT containing all 24 hours concatenated into one
large data set (see SAS code in Figure B-3). In the episodic data format, the variables STRMINU and
ENDMINU store the beginning and ending times of each episode over the diary day with values ranging
from 0 to 1440 minutes. Note that the value of STRMINU has been considered exclusively and the value
of the ENDMINU has been considered inclusively. For example, an episode with STRMINU = 0 and
ENDMINU = 25 occurred at a particular location and activity between 00:01:00 AM and 01:25:00 AM
on the diary day.
Table B-1. Example 24-Hour Diary File in the Episodic Format for Person ID# 1.1 Containing Beginning &
Ending Times, Original Codes, Regrouped Codes, and Durations of 22 Microenvironments Visited on
the Diary Day (see Section 5 for code explanations).
No. ID START END
FACT RACT NEWACT REGACT WHR NEWLOC SMK ETOA
1
2
3
*4
*5
„*«
7
8
*9
*io
*«
12
13
14
15
*1€
17
18
19
20
21
22
1.1
1.1
1.1
1.1
1.1
<14
°iTi
1.1
1,1
1.1
.I-,1
1.1
1.1
1.1
1.1
"i.f
1.1
1.1
1.1
1.1
1.1
1.1
0:00
1:45
2:00
11:00
11:05
•11^15'
lV:25
11:30
13337^
13:37"
13:44
13:54
13:57
15:30
15:33
" 16:30
17:00
19:00
19:10
19:25
19:35
21:00
1:45
2:00
11:00
11:05
tins
, f135
li:30
11:37
|3:37
13:44
13S4
13:57
15:30
15:33
16:30
moo
19:00
19:10
19:25
19:35
21:00
24:00
at night clubl
traveled home after night club
Sleeping or Napping
1 Wished teeth
Preparing Meals or Snacks
, v' , , ; Eating Meats or Snacks
Dressing or Personal Grooming
traveled to play football
i . ' • 1 pSayfrig flag football
travel to a bar
." Preparir^Meafe or Snacks
travel from bar
at bar
travel from bar
Watching TV
• . BaitogorSnowering
WatchingTV
travel to shopping
shopping for food
travel related to shopping for food
Watching TV
studied
77
79
45
' 44
10
' 43
47"
89
< 80
79
"10
79
77
79
91
; 40'
91
39
30
39
91
54
0
0
31 -
18
is-- .;
6
0
sfib -,°
0
,'< "' 18 " /*
0
0
0
0
~_30 ~~? "
0
0
0
0
0
0
0
0
0
40
10
70\
0
0
80
b"
10
0
0
0
0
" ' 40™'".
0
0
0
0
0
0
405
301
105
104
.101
!Q2
102
306
SJ7
" "306
201
301
405
301
102
, 104-
102
301
414
301
102
102
90
30
10
, 10
, 10
-,-tO1
10
40
&b x.
' '""40"
10,,
30 "'
90
30
10
-: ~ {Q".
10
30
90
30
10
10
1
0
0
0.
0
' 0
o"
0
', b
0
'"0
6
1
0
0
• v
0
0
0
0
0
0
105
15
540
< 5
' 10
10
5
7
_ " 120 '
7
/10
3
93
3
57
~< 30
120
10
15
10
85
180
Note 1: The shaded records with an asterisk (*) beside their number designate location x activity microenvironments that were
classified as exposure-related - based on the exposure activity, i.e., microenvironments with REGACT = 0 are not
shaded. Exposure could be occurring in the other microenvironments, but it must be determined based on locations
alone, e.g., exposure to automobile exhaust in vehicles, since the exposure activity classifications lack detailed
information.
Note 2: FACT = verbatim activity description; RACT = original NHAPS activity code; NEWACT = receded activity #1 (16
categories); REGACT = receded activity #2 (8 categories); WHR = original NHAPS location code; NEWLOC = receded
location (10 categories); SMK = original smoker-present code; ETOA = duration of microenvironment (in minutes). The
respondent with Person ID# 1.1 was an Hispanic male from Connecticut between 18 and 24 who was interviewed on a
weekend in the fall.
B-2
-------
At the heart of the algorithm is the assignment of a "1" (designated by | in Table B-3) for every
minute of each one-hour time block that the respondent is in the corresponding microenvironment and a
missing value (designated by 1 in Table B-3) whenever the respondent is not in the corresponding
microenvironment. The respondent must be in at least one microenvironment over the 60 minutes of
every hour, but can conceivably be in up to 60 different microenvironments in that hour so that 60
different time-of-day records would be necessary to describe their microenvironments by time-of-day -
one record for each different microenvironment. Each record would have one "1" for the minute they
were in the corresponding microenvironment and the remaining minutes would be filled with missing
values. However, more typically, each respondent is in much fewer than 60 different microenvironments
in any given hour. For example, in Table B-3 there are two records for hour #1 corresponding to the
Bar/Restaurant-No-Exposure-Activity-Smoker-Present (90-0-1 = NEWLOC-REGACT-SMK) and In-
Vehicle-No-Exposure-Activity-No-Smoker-Present (30-0-1) microenvironments. The first
microenvironment — in the second record - occurs from minute 1 through minute 45 (designated by |'s
with i 's inserted afterward) with the second microenvironment - in the third record - occurring from
minute 46 to minute 60 (designated by |'s with || 's inserted previously).
Table B-2. Example Microenvironment Calculations of Percentage of Time Spent, 24-Hour Duration, and 24-Hour
Frequency of Occurrence Using Person ID# 1.1
Locations
Activities
Smoker-Present
Location x Activity
Location x Smoker-Present
10. Res.lnd.
20. Res. Out
30. In Vehicle
40. NearVeh.
50. Other Out
60. Office/Fact
70. Mall/Store
80. Public BMg.
90. Bar/Rest.
100. Other Ind.
00. Non-Exposure
10. FoodPrepar.
20. Dishes/Laundry
30. Housekeeping
40. Bathing/Hygiene
50. YardTMainten.
60. Sports/Exercise
70. Eating/Drinking
Smoker-Present
No-Smoker Present
10-40
10-10
10-70
50-60
90-01
Percentage of
Time Spent
73.06
-
2.85
0.97
8.33
-
-
-
14.79
-
87.15
1.39
-
-
2.43
-
8.33
0.69
13.75
86.25
2.43
1.39
0.69
8.33
13.75
24-Hour Duration
(minutes)
1052
-
41
14
120
-
-
-
213
.
1,255
20
-
-
35
.
120
10
198
1,242
35
20
10
120
198
24-Hour
Frequency
11
.
5
2
1
-
.
-
3
.
16
2
-
-
2
.
1
1
2
20
2
2
1
1
2
Merged 24-Hour
Frequency
4
.
5
2
1
.
.
-
3
-
5
2
-
-
2
.
1
1
2
2
2
2
1
1
2
Note: The respondent with Person ID# 1.1 was an Hispanic male from Connecticut between 18 and 24 who was interviewed on a
weekend in the fall. Percentages of time spent over multiple respondents were done by taking the total amount of time
spent in a microenvironment and dividing by the total time = 1440 x 9,386 respondents. Mean 24-hour durations were
obtained by averaging over multiple respondents. There is a 'consecutive error' that arises when consecutive
microenvironments have the same location, activity, or smoker-present category. The merged 24-hour frequencies of
occurrence (column 5) do not count locations, activities, or smoker-present categories more than once if they occur
consecutively in different location x activity x smoker-present microenvironments. In this report 24-hour frequencies of
occurrence were only presented for location x activity and location x smoker-present microenvironments.
B-3
-------
The time-of-day format consists of 24 one-hour time blocks (0,...,23) specified by the variable H
with each of the 60 minutes in each one-hour time block corresponding to a microenvironment. The
variable a is the starting time of this one hour time block in the 1440-minute scale (i.e., 1, 61,121,...) and
b is the ending time of the one-hour time block in the 1440-minute scale (i.e., 60, 120,...). The code in
Figure B-2 must be repeated for all 24 one-hour time blocks. The "1" flags that show whether the
respondent is present in a microenvironment for a particular minute (tl, t2,...) are stored in the array tij.
The elements of the array are assigned by determining the starting point strtpt and ending point endpt of
the microenvironment within the limits of each one-hour time block beginning at a and ending at b, and
filling in "l'"s whenever the microenvironment occurs. The microenvironment starting time STRMINU
may occur before a, at a, or after a, and the microenvironment ending time ENDM1NU may occur before
b, at b, or after b.
libname n2'e:\s2df ;run;
options ps=66 ls=120;
proc sort data=n2.timres2 out=n2.timloc; by fid newloc;
run;
data n2.timloc; set n2.timloc; by fid newloc ;
if firstnewioc then do;
ctimloc=0;
cnumloc=0;
end;
ctimloc+etoa;
cnumloc+1;
if last.newkx then output;
run;
proc tabulate data=n2.timloc;
class newloc;
var ctimloc;
table all*ctimloc*(nf=5.0 sum*f=8.0 pctsum*f=6.2 mean*f=6.2
std*f=6.2) .newloc all;run;
proc tabulate data=n2.timloc;
dass newloc;
varcnumloc;
table all*cnumloc*(n*f=5.0 sum*f=8.0 pctsum*f=6.2 mean*f=6.2
std*f=6.2) .newloc all;run;
proc tabulate data=n2.timloc;
dass newloc;
var ctimloc; weight weight4;
table aB*ctimloc*(sumwgt*f=6.1 sum*f=8.0 pctsum*f=6.2 mean*f=6.2
std*f=6.2) .newloc all;run;
proc tabulate data=n2.timloc;
dass newloc;
varcnumloc; weightweight4;
table aB*cnumloc*(sumwgr*f=6.1 sum*f=8.0 pctsum*f=6.2 mean*f=6.2
stdf=6.2) .newtoc alljrun;
Figure B-1. The SAS code for calculating total time spent, 24-hour durations, and 24-hour frequencies of occurrence
of mtcroenvironments using the episodic formulation.
B-4
-------
The Fraction of Respondents Experiencing Each Microenvironment During Each Minute
From the concatenated file Nl.SUMACT containing the 24-hour time-of-day format for all 9,386
respondents, the number of times a microenvironment occurred during each minute of the day was
tabulated (see SAS code in Figure B-4). These frequencies were then divided by the total number of
microenvironments occurring in each minute to convert them into fractions (see example in Table B-4).
Since the episodic format had a resolution of one minute, our 1440-minute time-of-day format contains
all the information in the episodic format, and each microenvironment's fraction of the total
microenvironments occurring in each minute is equal to the fraction of respondents experiencing that
microenvironment during that minute.
libnamenl 'E:\S2DT;
PROC SORT DATA=N1.TIMRES2; BY FID ACTNO;RUN;
"time from 1 to 60;
data nl.sumactO ; set N1.TIMRES2;
KEEP H FID NEWLOC REGACT SMK ENDMINU STRTMINU REGIONC-smkexp TI1--TI60;
H=0;
a=(H*60)-t-1;
b=a+60;
if a gt strtminu then strtpt=a;
else if a=strtminu then strtpt=strtminu+1;
else if a It strtminu then strtpt=strtminu+1;
if b gt endminu then endpt=endminu;
else if b = endminu then endpt=endminu;
else if b It endminu then endpt=b;
array time{61} til -ti61;
do i= strtpt-(a-l) to endpt-(a-l);
time{i}=1;
end;
IF SUM(OF TI1-TI60) THEN OUTPUT;
run;
Repeat for every hour, H = 1, 2,3,4 23
Figure B-2. The SAS code for converting the 24-hour diary data from its episodic format into the time-of-day format.
Only the code for the first hour is shown (H = 0, output file = sumactO). For subsequent hours the values
of H range from 1 to 23 and the output files are sumactl,..., sumact23.
Time and Hardware Requirements
The complete conversion from the episodic to the time-of-day format (for all 24 hours and all 9,386
respondents) was accomplished with less than 60 minutes of CPU time on a 486 IBM-compatible
personal computer. Each hour of the time-of-day formulation required nearly 10 MB of hard disk space
- with approximately 10 demographic variables for each respondent - for a total of almost 240 MB over
all 24 hours (concatenated file). Thus, a total of almost 0.5 GB is required for the calculation since the
concatenated file must coexist on the hard disk with the hourly files. Overall calculations of the number
of respondents in each microenvironment for each of the 1440 minutes on the diary day each took
approximately 15-20 minutes on a 486 IBM-compatible personal computer. Slightly more time was
required for subgroup calculations (divisions by age, gender, region, etc.).
B-5
-------
B3. Time-of-Day EL: Duration of Microenvironments by Eight 3-Hour Segments
Since duration information was destroyed within each time unit of the time-of-day format by minute,
the time unit was increased to 3-hours, giving eight segments over the 24-hour diary day: (1) midnight-3
AM; (2) 3 AM-6 AM; (3) 6 AM-9 AM; (4) 9 AM-noon; (5) noon-3 AM; (6) 3 PM-6 PM; (7) 6 PM-9
PM; and (8) 9 PM-midnight. The durations of microenvironments originating in each of the eight 3-hour
segments that comprise the 24-hour diary day were calculated in the same way as 24-hour durations were
calculated (see SAS code in Figure B-5) except that the entire duration of each microenvironment was
used even if its ending time extended beyond the 3-hour segment. For example, the ninth
microenvironment for the NHAPS respondent with Person ID# 1.1 (Sports/Exercise in the Other Outdoor
location) began at 11:37 AM (in the fourth 3-hour segment) and continued until 1:37 PM (in the fifth 3-
hour segment). This microenvironment would be attached to the fourth 3-hour segment — its time
segment of origin — and assigned a duration of two hours even though it lasted 1 hour and 37 minutes
into the fifth 3-hour segment.
B-6
-------
Table B-3. Example Time-of-Day Format for Person ID# 1.1. | = Person in Corresponding Location x Activity x Smoker-Present Microenvironment at the
Specified Hour and Minute, 1 = Person Not in Corresponding Microenvironment at Specified Hour and Minute
W
F
I
D
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
N
E
W
L
O
C
90
90
30
10
10
10
10
10
10
10
10
10
10
10
10
10
R
E
G
A
C
T
40
10
70
S
M
K
S
T
R
T
M
I
N
U
105
120
120
120
120
120
120
120
120
120
660
665
675
685
E
N
D
M
I
N
U
105
105
120
660
660
660
660
660
660
660
660
660
665
675
685
690
H
O
U
R
MINUTE
10
11
12
13
14
15
16
17
18
19
20
21
22
23
nnni inrinnnririnnnnnnnnnnnn
i ii H H H H H H H H ii ii H H ii H H H H H ii H H i
H ii ii ii ii ii n ii ii ii ii ii H H H H H H H ii H H i
ii n n ii n n n n n ii n n n n n n n n n n n n i
n n n n n n n n n ii ii n n n n n n ii n ii n n i
n n n n n n n n n n ii n n n n n n n n n n n i
n n ii n n n n n n ii ii ii n n n n n n n n n n i
n n n n n n n n n n ii ii n n n n n n n n n n i
n n n n n n n n n ii ii n n n n n n ii n n n n i
n n n n n n n n n n n n n n n n n n n n n n i
n n n ii n n n n n ii ii n n n n n n n n n n n i
H
O
U
R
11
11
^••m
11
40
690
697
11
11
50
50
50
40
60
60
60
697
817
11
697
817
12
697
817
13
II II II II II II II II II II II II II II II II II II II II II II
II II II II II II II II II II II II II II II II II II II II II II
817
824
13
11
12
13
13
10
10
824
834
13
13
1.1
30
834
837
13
13
1.1
1.1
1.1
1.1
90
90
90
30
837
930
13
837
930
14
837
930
15
II II II II II II II II II II II II II II II II II II II II II II I
I II II II II II II II II II II II II II II II II II II II II II II I
930
933
15
JE; *?; IE;
13
15
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
10
10
10
10
10
30
90
30
40
933
990
15
933
990
16
II II II II II II II II II II II II II II II II II II II II II II I
990
1020
16
1020
1140
17
1020
1140
I II II II II II II II II II II II II II II II II II II II II II II I
I II II II II II II II II II II II II II II II II II II II II II II I
1140
1150
1150
1165
1165
1175
19
15
19
19
19
1.1
1.1
1.1
1.1
1.1
10
10
10
10
10
0
1175
1260
19
1175
1260
20
1260
1440
21
1260
1440
22
1260
1440
23
II II II II II II II II II II II II II II II II II II II II II II i
ii n ii ii ii ii ii ii ii ii ii ii ii ii ii ii ii n ii n ii ii i
n n n n n n n n n ii ii ii n n n n n n n n n n \
n n n n n n ii n n n n ii ii n n ii n n n n n n •
19
Note: See Section 5 for an explanation of the microenvironmental variable codes (NEWLOC, RAGACT, SMK). STRMINU and ENDMINU are the starting and ending minutes of
each hour (HOUR).
-------
Table B-3. Continued
H
O
U
MINUTE
H
O
U
15
19
19
19
24
25
I II II II II H H II II H II II II II II II II II II II II II I
56
57
58
59
60
H ii H ii ....... , H i, I. , ..... „ „ „ „ Minn
H ii H H H H H ...... H ...... H ........... 1 1!
i i i i i i i H H H ii ii ...... n i ..... H H H H H H H H U U H U i i i
!! !! !! !! !! !! !! !! !! I1 " " " " " " ..... ' " " " " "
! !! !! ! !! ! ! !! !! ! ! ! ! !! " " " " " '• » » » " » " n " " " » n n " n
! !! !! !! !! ! !! !! ! ! " " " " » " » » " » " « " » » » " » » " » H n n n
' " " " " ' ....... » " " " » » » » i ....... "
ii n n ii n n n ii n n n n n n n n n n n n n
II II II II II II
i .....
!! !! ! !! ! !! !! ! ! ! ! ! !! !! " " " " " » » »
Li » » " n n n " « » n H H H n n
n i ...... i
» » » » n
» "
i .......
n n n n H n n n ii n n n n n ii i
I II II II II II II II II II I
sa;
II II
II II
II II II II II II II II II II II II II II II
•• ii ii ii ii ii •• n n n n n n n n
.: i: :: :: :: :: i: :: " ::
II II II II II
:i :: :: :: :: :: i: » ::
" " !! » « " « « "
n ;: :::::: » i
» " « !! « " i
18
19
-------
LIBNAME NE'F:\S2DT';RUN;
DATA NE.SUMACT;
SET N1.SUMACTO N1.SUMACT1 N1.SUMACT2 N1.SUMACT3 N1.SUMACT4 N1.SUMACT5
N1 .SUMACT6 N1 .SUMACT7 N1 .SUMACT8 N1 .SUMACT9
N1.SUMACT10N1.SUMACT11 N1.SUMACT12 N1.SUMACT13N1.SUMACT14
N1.SUMACT15 N1.SUMACT16 N1.SUMACT17 N1.SUMACT18 N1.SUMACT19
N1.SUMACT20 N1.SUMACT21 N1.SUMACT22 N1.SUMACT23;
RUN;
Figure B-3. The SAS code used to concatenate the 24 files containing the time-of-day formulation for each hour of the
diary day. The resulting output file containing all 24 hours for all 9,386 respondents is SUMACT which
took up approximately 240 MB of hard disk space.
B-9
-------
Sbname n21-.\s2dt1;njn;
lenameoul 'c:\ttocaIIJst1;
filename ou2 'c:Mlocsex.lsf ;
options te=120 ps=72 nocenten
OveralL
Unweighted:
proc printto print=ou1 ;
itte'Aclivity Study by every minute unweighted toe1;
proc tabulate data=fl2.sumact;
class Hnewloc;
var61-660;
table H*(ti1 ti2ti3ti4ti5«6ti7ti8ti9titO
611 612 613 614 615616 617 618 619620
621 ti22ti23ti24ti25ti26ti27ti28ti29ti30
1J31 632 633 634 635 636 637 638 639 ti40
ti41 642643 644 ti45 646647648649650
ti51 652653654655856657658659660
)*(n*f=6.0),newtocai;
run;
Weighted:
ttte'Acfivity Study by every minute weighted toe1;
proctabutetedata=n2.sumact;
class Hnewloc;
varti1-tJ60; weight weighH;
taoleH*(ti1 62 63 6465 666768 69 610
till ti12ti13ti14ti1Sti16ti17ti18ti19ti20
ti21 1i22ti23ti24ti25ti261i27ti28ti29ti30
ti31 632tt33ti34ti35ti361i37ti38ti39ti40
b41 M2ti43ti44t)45b46ti47t)48t)49ti50
t»51 «52«53ti54ti55«56«57ti58tiS9ti60
)*(8umwgTM.1 ),newtoc al;
run;
By Gender.
Unweighted:
proc printto pnnt=ou2;
trbe'AcSvity Study by every minute unweighted toe';
proc tabulate data=n£sumact;
dass rsex H newtoc;
varti1-ti60;
table rsex*H*(ti1 «2 fi3 M K tt t7 ti8 «9 ti10
i11 1i12ti13til4ti15ti16ti17ti18ti19ti20
S21 S22ti23ti24ti25ti26tiZ7ti28ti29ti30
631 632ti33ti34ti35ti36ti37ti38li39ti40
641 tt42tt43644ti45ti461i47ti48ti49ti50
«S1 652ti53ti54ti55ti56ti57tJ58tJ59ti60
run;
WeigMed:
titte'Activity Study by every minute weighted toe';
proc tabulate data=n£sumact;
dass rsex H newtoc;
var tit-ISO; weight weighM;
table rsex*H*(ti1 «2 «3 1§4 85 86 17 68 ti9 1110
fill ti12ti13ti14ti15ti16ti17ti18ti19ti20
ti21 «22tt23ti24ti25t)26ti27ti28ti29ti30
«31 «2tC31J34tJ35ti36ti37ti38ti39ti40
«41 «42ti43tt44ti45ti46ti47ti48ti49ti50
651 652 653 6546556566571158 659 660
run;
The unweighted and weighted analyses were repeated
for the other subgroups: age, region, weekday vs.
Bkendandi
Mson.
Figure B-4. The SAS code used to calculate the unweighted and weighted (using WEIGHT4) numbers of respondents
(= fraction of microenvironments at the one-minute time resolution of the study) in each location (NEWLOC
category) for every minute of the 24-hour diary day, overall and for mates and females (RSEX). The same
code was changed slightly to do analyses by exposure activities, smoker-present categones, and by age,
region, weekday vs. weekend, and season.
B-10
-------
Table B-4. An Example Output Data File Containing the Overall Weighted Fraction of Respondents in Each
Location (NEWLOC category) for Every Minute of the Diary Day Calculated Using the Time-of-Day
Formulation of Microenvironments (see Section 5 for code explanations)
Minute
1
2
3
4
5
6
7
8
9
10
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
Time
12:01 AM
12:02 AM
12:03 AM
12:04 AM
12:05 AM
12:06 AM
12:07 AM
12:08 AM
12:09 AM
12:1 0AM
11:50 AM
11:51 AM
11:52 AM
11:53 AM
11:54 AM
11:55 AM
11:56 AM
11:57 AM
11 -.58 AM
11:59 AM
12:00 PM
12:01 PM
12:02 PM
12:03 PM
12:04 PM
12:05 PM
12:06 PM
12:07 PM
12:08 PM
12:09 PM
12:10 PM
11:50PM
11:51 PM
11:52PM
11:53 PM
11:54 PM
11:55 PM
11:56 PM
11:57PM
11:58 PM
11:59PM
12:00 AM
10
0.937
0.937
0.937
0.937
0.937
0.937
0.937
0.937
0.937
0.937
0.367
0.369
0.369
0.369
0.369
0.369
0.370
0.370
0.370
0.370
0.370
0.365
0.365
0.367
0.368
0.367
0.374
0.374
0.375
0.375
0.375
0.934
0.935
0.935
0.935
0.935
0.935
0.936
0.936
0.936
0.936
0.936
20
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.069
0.069
0.069
0.069
0.069
0.069
0.069
0.069
0.069
0.069
0.069
0.055
0.055
0.055
0.055
0.056
0.057
0.057
0.057
0.057
0.057
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
30
0.012
0.012
0.012
0.012
0.012
0.012
0.012
0.012
0.012
0.012
0.067
0.067
0.067
0.067
0.067
0.067
0.064
0.064
0.064
0.064
0.064
0.116
0.116
0.115
0.113
0.113
0.102
0.102
0.102
0.101
0.101
0.012
0.012
0.012
0.012
0.012
0.012
0.010
0.010
0.010
0.010
0.011
40
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.032
0.031
0.031
0.031
0.030
0.030
0.031
0.031
0.031
0.031
0.031
0.040
0.040
0.039
0.038
0.038
0.033
0.033
0.033
0.033
0.033
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
NEWLOC Code
50 60
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.044
0.043
0.043
0.043
0.043
0.043
0.044
0.044
0.044
0.044
0.044
0.038
0.038
0.039
0.039
0.039
0.039
0.039
0.039
0.039
0.040
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.110
0.110
0.110
0.110
0.110
0.110
0.110
0.110
0.110
0.110
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.007
70
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.053
0.053
0.053
0.053
0.053
0.053
0.053
0.053
0.053
0.053
0.053
0.046
0.046
0.047
0.047
0.047
0.047
0.047
0.047
0.047
0.047
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
80
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.009
0.179
0.179
0.179
0.179
0.180
0.180
0.181
0.181
0.181
0.181
0.181
0.168
0.168
0.168
0.168
0.168
0.169
0.169
0.169
0.169
0.169
0.007
0.007
0.007
0.007
0.007
0.007
0.007
0.008
0.008
0.008
0.008
90
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.031
0.032
0.032
0.032
0.032
0.032
0.031
0.031
0.031
0.031
0.031
0.032
0.032
0.032
0.033
0.033
0.039
0.039
0.039
0.039
0.039
0.018
0.017
0.017
0.017
0.017
0.017
0.018
0.018
0.018
0.018
0.018
100
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.011
0.031
0.031
0.031
0.031
0.031
0.031
0.032
0.032
0.032
0.032
0.032
0.029
0.029
0.029
0.029
0.029
0.029
0.029
0.029
0.029
0.029
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
0.008
B-ll
-------
libname n2"e:\s2df ;run;
options ps=66 ls=120;
data timres2;set n2.tjmres2;
Overall.
gsthMtoor(sthr/3);
proc sort data=timres2 out=n2.timtloc; by gsthr fid newloc ;
run;
data n2.tifntk)c; set n2.timHoc; by gsthr fid newloc ;
if firstnewtoc then do;
ctimk)c=0;
cnumtooO;
end;
ctimtoc+etoa;
cnumtoc+1;
if lastnewloc then output;
run;
Unweighted:
proc tabulate daJa=n2.timtloc;
class newloc;
varctenlcc;
table aTctimtoc*(nf =5.0 sumf =6.0 pctsumf =62 meanf =6.2
sttf=6.2) .newtoc att;run;
Weighted!
proc tabulate (Jata=n2.timtkx;
class newloc;
varcfenkx; weight weigrrM;
table aJTctimloc<(sumwgrfs6.l sum-f»=8.0 pctsumf =6.2 mean'f=6^
s*Tf=6.2) .newloc aD;run;
By gender.
Urranighted:
options te=146 ps=56 nocenter
proc tabulate data=n2.timtk>c;
dass rsex newloc gsthr;
label gsthr=t
rsex^sex1;
varcttrOoc;
table rsex*gsthrctimloc*(nf =5.0 sumf =8.0 pctsumf =6^ mean*f=62
s«^=6i) aircSmloc*(n<1=5.0 sumf=8.0 pctsumf =6^ meanf=6i
sttf=6.2),
newloc aH/rtssZ;
keylabel pctsum=<%sm1
sum='sm';
run;
Weighted:
proc tabulate data=n2.tirntkx:;
dass rsex newloc gsthr;
label gsttv»r
rsexs'sex';
varcbmloc; weight weighM;
table rsex-gsthr-ctwrtoc*(sumwgtf=6.1 sumf=8.0 pctsumf =€J2 meanf =6^
stdf-6^) aTcti»nloc*(sumwgtf-6.1 sumf =8.0 pctsumf =6.2 meanf =6^
newloc aR/rts=2;
keylabel sumwgt='wsni'
pctsum=>%sm';
run;
Figure B-5. The SAS code used to calculate weighted and unweighted overall mean durations in each of the eight
3-hour segments (GSTHR) in the 24-hour diary day, overall and by gender. Similar code was used in
analyses for the other subgroups.
B-12
------- |