United States
Environmental Protection
Agency
Environmental Monitoring and Support
Laboratory
Research Triangle Park NC 27711
EPA-600 4-79-014
February 1979
Research and Development
v>EPA
Summary of Audit
Performance
Measurement of
SO2, N02, CO,
Sulfate, Nitrate,
Lead, Hi-Vol Flow
Rate — 1977
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL MONITORING series.
This series describes research conducted to develop new or improved methods
and instrumentation for the identification and quantification of environmental
pollutants at the lowest conceivably significant concentrations. It also includes
studies to determine the ambient concentrations of pollutants in the environment
and/or the variance of pollutants as a function of time or meteorological factors.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
SUMMARY OF AUDIT PERFORMANCE: MEASUREMENT OF
S02, N02, CO, SULFATE, NITRATE, LEAD, HI-VOL FLOW RATE
1977
Steven M. Bromberg, Robert L. Lampe, and Berne I. Bennett
Quality Assurance Branch
Environmental Monitoring and Support Laboratory
Research Triangle Park, North Carolina 27711
ENVIRONMENTAL MONITORING AND SUPPORT LABORATORY
QUALITY ASSURANCE BRANCH
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Monitoring and
Support Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
-------
ABSTRACT
The Quality Assurance Branch of the Environmental Monitoring and
Support Laboratory, Research Triangle Park, North Carolina, conducts an
ambient air audit program. Measurement principles for sulfur dioxide,
nitrogen dioxide, carbon monoxide, sulfate-nitrate, and lead are audited
on a semiannual basis. Blind samples, the concentrations of which are
known only to the EPA, are sent to the participating laboratories. Hi-Vol
sampler flow rate audits are conducted annually using a modified orifice.
The analytical results are returned to the Quality Assurance Branch for
evaluation. After processing, an individual repdrt is returned to each
participant.
This report contains a summary of audit results obtained during
October-December 1976 and all of calendar year 1977.
-------
CONTENTS
Abstract iii
Figures v
Tables vi
Acknowledgment ix
1. Introduction 1
2. Program Coordination 3
3. Statistical Approach 5
4. Audit Materials '
S02 8
N02 9
CO 10
S0° - N0~ 11
Pb 12
Hi-Vol Reference Flow Device 13
5. Results
S02 16
N02 28
CO 39
50
61
Pb 71
Hi-Vol Flow Rate 78
References 85
Appendices 86
iv
-------
FIGURES
Number Page
1 Example of individual report 4
2 ReF mounted on Hi-Vol sampler 13
3 ReF with resistance plate 13
4 Plot of S02 results 19
5 Plot of S02 absolute percent differences 22
6 Plot of S02 slope and intercept 24
7 Plot of N02 results 31
8 Plot of N0« absolute percent differences 33
9 Plot of N02 slope and intercept 36
10 Plot of CO results 43
11 Plot of CO absolute percent differences 45
12 Plot of CO slope and intercept 48
13 Plot of SO^ results 53
14 Plot of S0| absolute percent differences 55
15 Plot of SOI slope and intercept 58
16 Plot of NO^ results 64
17 Plot of NO, absolute percent differences 66
18 Plot of NOg slope and intercept 69
19 Hi-Vol flow rate histogram - 1976 81
20 Hi-Vol flow rate histogram - 1977 82
-------
TABLES
Number Page
1 Mean and Standard Deviation of S02 Acceptance Analysis ... 9
2 Mean and Standard Deviation of N02 Acceptance Analysis ... 10
3 Mean and Standard Deviation of CO Acceptance Analysis ... 11
4 Sulfate and Nitrate Sample Values ............. 12
5 Lead Sample Values ..................... 13
6 S02 Agency Distribution ...... ............ 16
7 S02 Analytical Methods ... ................ 17
8 S0« Sample and Target Ranges ................ 18
9 SOp Absolute Percent Difference .............. 21
10 S02 Survey Statistics ................... 23
11 Summary of S02 Analytical Methods Used by Outlier Labs ... 25
12 Mean and Standard Deviation of S0« Results by Analytical . . 26
Method L
13 N02 Agency Distribution ............. ..... 29
14 N02 Analytical Methods ................... 29
15 N02 Sample and Target Ranges ................ 30
16 N02 Absolute Percent Difference .............. 32
17 N02 Survey Statistics ................... 35
18 Summary of N02 Analytical Methods Used by Outlier Labs ... 37
19 Mean and Standard Deviation of N0? Results by Analytical . . 38
Method
20 CO Agency Distribution ................... 40
vi
-------
Number Page
21 CO Analytical Methods ................... 40
22 CO Sample and Target Ranges ................ 41
23 CO Absolute Percent Difference .............. 44
24 CO Survey Statistics ................... 46
25 Summary of CO Analytical Methods Used by Outlier ..... 49
Instruments
26 Mean and Standard Deviation of CO Results by Analtyical . . 49
Method
27 S0| Agency Distribution .................. 51
28 SOJj Analytical Methods ............ ...... 51
29 SO^ Sample and Target Ranges ............... 52
30 SOj Absolute Percent Difference .............. 55
} i
31 SOj Summary Statistics .................. 57
32 Summary of S0| Analytical Methods Used by Outlier Labs . . 59
33 Mean and Standard Deviation of SOJ; Results by Analytical . 60
Method ^
34 NO Aenc Distribution .................. 62
NO Agency Distribution
35 NO^ Analytical Methods, .................. 62
36 NOg Sample and Target Ranges ............... 63
37 NO^ Absolute Percent Difference .............. 65
38 NOg Summary Statistics .................. 68
39 Summary of NO^ Analytical Methods Used by Outlier Labs . . 70
40 Mean and Standard Deviation of NO^ Results by Analytical. . 71
Method J
41 Pb Agency Distribution .................. 72
42 Pb Analytical Methods ................... 72
43 Pb Extraction Procedures ................. 72
vn
-------
Number Page
44 Pb Sample and Target Ranges 73
45 Pb Absolute Percent Difference 74
46 Pb Summary Statistics 76
47 Summary of Pb Analytical Methods Used by Outlier Labs ... 76
48 Mean and Standard Deviation of Pb Results by Analytical . . 77
Method
49 Hi-Vol Flow Rate Agency Distribution 78
50 Hi-Vol Flow Rate Measurement Methods 78
51 Hi-Vol Flow Rate Percent Difference 80
52 Summary of Hi-Vol Flow Rate Measurement Methods 83
vm
-------
ACKNOWLEDGMENTS
The authors would like to thank all the participants for their coopera-
tion during the past year. Also due a word of thanks are the diligent
programmers of the Statistical and Technical Analysis Branch, EMSL, for
developing the data management systems necessary to store and summarize
the audit data, and the technical reviewers for reviewing a very tedious
document.
-------
SECTION 1
INTRODUCTION
Since 1972 the Environmental Protection Agency (EPA) has been engaged
in a Performance Audit program of various monitoring groups throughout the
United States and in several foreign countries. The audit program is only
one part of an overall quality assurance program, so the results should not
be construed as an absolute indicator of data quality. Used along with infor-
mation obtained from an internal quality control program, however, the
conclusions can be quite meaningful.
The purposes of the audit program are twofold. The first, from a
participant standpoint, is most important. Agencies are furnished a means
of rapid self-evaluation of the specific operation under study. The
second objective of the program is to provide EPA with a continuing index
of the validity of data being reported to air quality data banks.
The program is being coordinated through the 10 EPA Regional Offices
(RO) by the Quality Assurance Branch (QAB) of the Environmental Monitoring
and Support Laboratory (EMSL), Environmental Research Center, Research
Triangle Park, North Carolina 27711. Comments, questions, or applications
to participate in the program should be sent to the above address.
This publication is the second of a continuing series of yearly
summary reports. The document covers the period from October 1976 through
December 1977.
Users of the information contained in this report should take note of
-------
some limitations imposed in gathering the data. With the exception of the
audit for carbon monoxide (CO), the surveys checked only a portion of the
entire system. The sulfur dioxide (SCL), nitrogen dioxide (N0?), sulfate-
nitrate (SOT-NOZ), and lead (Pb) surveys examined only the analytical portion
of the system. The Hi-Vol audit checked only the flow rate portion of the
method. No restrictions were placed on the methodology used by the parti-
cipants (other than that the method used had to be compatible with the
audit samples). However, to the extent possible, the various methods used
were documented.
The following sections include discussions of the program operation,
descriptions of the audit materials, the statistical approach used to
analyze the data, and the results. The Results Section includes data from
October 1976-December 1977. The appendix contains the raw data of the
audit sample analysis.
-------
SECTION 2
PROGRAM COORDINATION
Participants in the audits were selected by the Regional Quality
Control Coordinator in each of the 10 Regions. Once a potential participant
has received audit samples for a particular pollutant, he is automatically
notified of subsequent surveys for that pollutant. Additional participants
are obtained through individual requests and through notification in the
Journal of the Air Pollution Control Association. Participants are
assigned an identification number which remains with the agency throughout
all audits for all pollutants.
After the audit roster is completed for a particular survey, instruc-
tional materials and unknown samples are mailed. The participants are
allowed 5 to 6 weeks in which to return their results. After these
results are entered into a data bank, individual reports are returned
to the participants. This report indicates the acceptable ranges for
each sample as well as the value reported by the agency. Figure 1 is
an example of an individual report for S07.
In order to determine whether incorrect results are due to inherent
problems with a laboratory procedure or to just a "single event", recheck
samples are sent to laboratories whose results do not meet certain criteria.
Thus, by having a second chance to analyze a set of samples, real deficiencies
can be distinguished from one-time problems.
-------
123456
Mr. J. Doe
5432 Anywhere St.
Yourtown, USA
31.13
«U I I
5251
6116
7122
8112
t
I
'.
f
'•
""V
f
10*125?
!
?
-LAnORATORY STnOr RESULTS
I > T I ? 7 7 I
POLLUTANT - SOM
x
• X
" • ^ B 2 ? X
1
9. 9 DO
1 I
2.750
1 .760
5.30M 10.125
EPA RT5ULTS
18.80**
UNITc; -
C U n I C H F T F.
.000
I?. 8 25
?.850
_ .qop
I I »655
1^.175
17.010
.3. 150
Z«?05
9.M35
1 I .M75
_L3 * 7 7 °
2.550
1
1 ?.7-^,5
IS .5?5
f « . 6 1 0
3 . M S P
Figure 1. Example of individual report.
-------
SECTION 3
STATISTICAL APPROACH
Of primary concern at the onset of the program was the establishment
of some type of acceptance criteria to enable participants to judge their
performance. Initially, there was little guidance, and the choices of
acceptable limits were somewhat arbitrary. As more data have been gathered
through the audit program, more realistic acceptance limits have been set.
For convenience, two acceptable ranges have been defined. The Sample
Range contains variability attributable to the sample material. This
estimation was based on the repeated analysis of several samples from each
concentration range by one laboratory. Using the precision of the set of
samples under consideration plus past data from similar samples, the
Sample Ranges were chosen. The Sample Ranges should be goals of each
agency; it is not reasonable to expect all laboratories to fall with-
in these ranges on any given analysis. However, falling within these
ranges repeatedly indicates a laboratory with excellent precision and
accuracy. Falling outside the Sample Ranges does not necessarily indicate
problems.
The Target Range has been constructed to include sample variability
and inter!aboratory variability. A laboratory falling outside the Sample
Range but within the Target Range should feel comfortable with the results.
If, however, the results fall outside the Target Range, the accuracy of the
-------
analysis should ba considered suspect. The Target Range is based in part
on past performance of all participants. The percent difference between
the reported results and the true value (result determined by QAB) was
established for past surveys. Using this percent difference for all
samples over all surveys, an average difference was determined. This average
difference was used to calculate the Target Ranges.
When evaluating results based on the acceptable ranges, one must take
into consideration the uses of the data. For some purposes the Sample Range
may be too wide; for other uses the Target Range may be too narrow. Thus,
judgment should be excercised whenever comparing ranges with results.
A preferable method of evaluating performance "after the fact" is to
compare individual results with the respective tables entitled "Absolute
Percent Difference" found in the Results Section. These tables are
frequency distribution tables of the percent difference between EPA and
reported results. The relationship of individual performance to the over-
all study performance can be determined using these tables.
Two outlier tests were used as part of the overall analysis and to
screen the data for further analysis. The first test was used to screen
results that were grossly in error. To be eliminated from further analysis,
a laboratory had to report all samples outside the Target Ranges for the
respective pollutants. These data appear only in the Appendix and are
marked with an asterisk (*). No further statistical analyses in this report
contain these data. Data appearing in table columns labeled "All Data",
or containing no specific designation, are results subjected to the first
test.
A second outlier test, using a standard technique was also used.
-------
Depending upon the number of results for each concentration a factor,
ranging from 2.1 for 20 samples to 3.3 for 300 samples, was chosen. This
factor was multiplied by the standard deviation of the sample as determined
from the participant results. Results outside the range determined by the
expression [(factor) x (standard deviation)] ± study mean were identified as
outliers. These data are identified in the "Outliers/Removed" columns of
selected tables.
Several summaries have been used to condense the large amounts of data
into a more manageable form. The tables in the Appendix are sequential
listings of all data by sample concentration for each audit. Statistical
summaries of the sequentially listed data are presented.
-------
SECTION 4
AUDIT MATERIALS
AMBIENT SULFUR DIOXIDE SAMPLES
The commercially produced sample material consisted of freeze-dried
mixtures of sodium sulfite and potassium tetrachloromercurate (TCM) contained
in 5-ml glass vials with inert screw cap closures or in sealed glass ampoules.
Sets of samples were comprised of five vials/ampoules containing approxi-
mately 4 to 54 yg of S0« equivalent per container. Initially, vials were
chosen in place of ampoules to facilitate quantitative transfer of the
contents. Difficulty had been experienced in transferring powder from the
small ampoules due to the surface tension. However, because sample decay
appeared to be a problem, use of ampoules was reinstated. The samples
were placed in freezers immediately upon receipt with the expectation
that a low temperature would preserve the integrity of the material.
Initial EPA analyses were performed immediately after receipt. Subsequent
reanalyses of the samples after several months demonstrated that freezing
did not completely stabilize the sulfite content in that the S0? levels
continued to gradually decline. Subsequent analyses of the samples have
demonstrated that, while the decay is not completely eliminated, the rate
has been reduced substantially. The analyses were performed by the
reference method for the determination of S02 in the atmosphere (pararosani-
line method)(1). The sample, when dissolved in 0.04 N potassium tetrachloro-
mercurate (TCM) forms a dichlorosulfitomercurate complex. The complex is
8
-------
reacted with pararosaniline and formaldehyde to form intensely colored
pararosaniline sulfonic acid. The absorbance of the solution is measured
spectrophotometrically at 548 nm.
Ten samples from each concentration were analyzed. Table 1 lists the
results of these analyses. The values contained in Table 1 are those used
in determining the acceptable ranges reported to participants. The values
listed are based on the assumption that the sample was collected in 50 ml
of absorbing reagent and the total sample air volume was 300 £.
TABLE 1. MEAN AND STANDARD DEVIATION OF S02 ACCEPTANCE ANALYSIS (yg/m3)
S09 Audit
* 1176
S09 Audit
* 0477
S09 Audit
* 1077
Cone. 1
Mean
19.13
21.90
13.60
Std.
Dev.
0.87
1.33
0.50
Cone. 2
Mean
43.03
48.23
56.20
Std.
Dev.
1.07
1.33
1.31
Cone. 3
Mean
73.37
80.17
93.73
Std.
Dev.
2.40
1.30
2.23
Cone. 4
Mean
102.97
114.70
93.73
Std.
Dev.
3.22
3.03
2.23
Cone. 5
Mean
149.03
155.40
179.07
Std.
Dev.
2.04
2.27
1.00
AMBIENT NITROGEN DIOXIDE SAMPLES
The commercially produced samples consisted of 4 ml of aqueous sodium
nitrite in 5-ml glass vials with inert screw cap closures. A set of samples
consisted of five vials. The samples, when mixed with absorbing reagent,
simulated ambient samples ranging from 0.12 to 0.87 yg/ml.
EPA analysis of 10 samples from each concentration was performed using
a candidate method for the determination of N02 in ambient air (2). Measure-
ments were made on a Varian Model 635 UV-Vis spectrophotometer at 540 nm.
Aqueous calibration standards were prepared to encompass a linear range from
-------
0 to 1.6 yg/ml of N02. Table 2 lists the results of these analyses. The
values contained in Table 2 are those used in determining the acceptable
ranges reported to participants. The values listed are based on the assump-
tion that the sample was collected in 50 ml of absorbing reagent.
TABLE 2. MEAN AND STANDARD DEVIATION OF N02 ACCEPTANCE ANALYSIS (yg/ml)
N09 Audit
c 1276
N09 Audit
* 0677
N07 Audit
* 1277
Cone. 1
Mean
0.12
0.12
0.18
Std.
Dev.
0.01
0.02
0.04
Cone. 2
Mean
0.18
0.24
0.31
Std.
Dev.
0.01
0.02
0.03
Cone. 3
Mean
0.33
0.24
0,50
Std.
Dev.
0.01
0.02
0.02
Cone. 4
Mean
0.54
0.38
-
0.50
Std.
Dev.
0.05
0.02
0.02
Cone. 5
Mean
0.87
0.69
0.87
Std.
Dev.
0.02
0.04
0.03
AMBIENT CARBON MONOXIDE SAMPLES
The samples consisted of commercially produced mixtures of CO and arti-
ficial air. Mixtures also contained approximately 2 ppm methane arid 350 ppm
of C02- Specially treated aluminum cylinders were used to improve gas stabi-
lity. Sample concentrations ranged from approximately 3 to 40 ppm of CO.
Each participant received a set of three cylinders, one from each concentra-
tion level.
Analysis by EPA of cylinders from each concentration was performed
using a Bendix Model 8501 NDIR analyzer. National Bureau of Standards SRM
gases were used as the reference standards. Three SRM's at nominal concentra-
tions of 9, 46, and 95 ppm CO were used as the primary reference gases.
Ten samples from each concentration were analyzed. Table 3 lists the
results of these analyses. The values contained in Table 3 are those used
in determining the acceptable ranges reported to participants.
10
-------
TABLE 3. MEAN AND STANDARD DEVIATION OF CO ACCEPTANCE ANALYSIS (ppm)
CO Audit - 1076
CO Audit - 0377
CO Audit - 0977
Cone. 1
Mean
4.51
3.38
4.79
Std. Dev.
0.09
0.02
0.04
Cone. 2
Mean
20.89
15.87
16.59
Std. Dev.
0.09
0.04
0.02
Cone. 3
Mean
40.14
31.72
36.00
Std. Dev.
0.04
0.04
0.05
SULFATE-NITRATE SAMPLES
The commercially prepared samples consisted of 19 mm by 20 cm (0.75 x
8 in.) fiberglass filter strips with depositions of potassium sulfate (K
and lead nitrate (KNOo). Filter strip sample sets were comprised of
combinations of differing S07 and N03 concentrations. Each strip was
packaged in a plastic envelope. The concentration of sulfate ranged from
o
0 (blank) to 40 yg/m . Nitrate concentrations ranged from 0 (blank) to
3
15 yg/m . Concentrations were calculated assuming that the samples were
collected on a 20 by 25.4 cm (8 x 10 in.) filter with a total air volume
of 2000 m3.
It was felt that the gravimetric preparation of the solution deposited
onto the filter strips was more accurate than analysis using existing pro-
cedures. Thus, the values accepted as "true values" are the vendor certifi
cations. Verification analyses ensured that the accuracy and precision of
the samples were acceptable. Table 4 lists the concentrations of the
samples used during the audit.
11
-------
TABLE 4. SULFATE AND NITRATE SAMPLE VALUES (yg/nT)
Audit 3
1076
\ fl _. |\I fi
O v/ yi liwo
Audit
0277
•^ \J it ^l'|^y*\
Audit J
0877
Cone. 1
so4
0.00
4.47
0.00
N03
0.00
1.94
0.00
Cone. 2
so4
1.20
15.63
2.10
N03
0.54
5.17
0.72
Cone. 3 Cone. 4
so4
2.99
21.16
3.00
N03
3.45
7.38
1.32
so4
10.42
24.87
11.11
NO,
9.58
10.92
4.80
P.nnr. R Cone. D
so4
-
-
13.52
NO,
-
-
6.00
so4
-
-
16.20
NU3
-
-
7.81
LEAD SAMPLES
The commercially prepared samples consisted of 19 mm by 20 cm (0.75 by
8 in.) fiberglass filter strips with depositions of lead nitrate (Pb(NOo)o).
O £
Filter strip sample sets were comprised of combinations of differing lead
concentrations. Each strip was packaged in a plastic envelope. The concen-
o
trations of lead ranged from 0 (blank) to approximately 12 yg/m . Concentra-
tions were calculated assuming the samples were collected on a 20 by 25.4 cm
(8 x 10 in.) filter with total air volume of 2000 m3.
It was felt that the gravimetric preparation of the solution deposited
onto the filter strips was more accurate than analysis using existing
procedures. Thus, the values accepted as "true values" are the vendor
certifications. Verification analyses ensured that the accuracy and pre-
cision of the samples were acceptable. Table 5 lists the concentrations
of the samples used during the audit.
12
-------
TABLE 5. LEAD SAMPLE VALUES (yg/m3)
Pb Audit
0877
Cone. 1
x s*
0.00
Cone. 2
x s*
1.80 .01
Cone. 3
x s*
4.05 .01
Cone. 4
x s*
6.60 .01
Cone. 5
x s*
8.40 .01
Cone. 6
x s*
11.40 .01
*Based on weight of applied chemical
HI-VOL REFERENCE FLOW DEVICE (ReF)
One ReF was supplied to each participating agency. The organization
was instructed to check as many Hi-Vol sampling units as feasible within
the alotted time. The auditing unit received by each laboratory consisted
of a modified orifice (ReF), wind deflector, manometer, and resistance
plates (to change flow rates).
Fig. 2. ReF mounted
on Hf-Vol sampler.
Fig. 3. ReF with resistance plate
13
-------
When auditing the flow rate of a Hi-Vol sampler, the ReF was mounted on top
of the sampler, replacing the filter face plate (see Figure 2). The wind
deflector was necessary to prevent fluctuation in the readings due to wind
flowing across the orifice. The manometer was used to measure the pressure
drop across the orifice. The resistance plates, when inserted into the
ReF, simulated various filter loading conditions (see Figure 3).
By calibrating each ReF with a positive displacement meter (Roots meter),
and by measuring pressure drops and temperatures, an individual calibration
curve in the form of an orifice equation could be derived. The equation shown
below was used to determine "K" orifice constant for each unit.
where:
QI=AYC /AP Ti
pi
q
= volumetric flow at conditions of T-, and P-, (m /nrin)
2
A = area of orifice (in )
Y = expansion factor
C = orifice coefficient
AP = pressure drop across orifice (in. H
P-| = upstream pressure (barometric pressure -mm Hg)
T-j = upstream temperature (ambient temperature - K)
Because A is constant for a given orifice, and Y and C are essentially
constant over the flow range in question, a new orifice constant K was
defined as:
K = AYC
14
-------
Thus, the orifice equation becomes:
AP
During calibration of the ReF, Q-, , AP, T-, , and ?•• were measured. The constant
K is then determined by regressing a series of Q-, measurements onto the
square root of the values under the radical.
During an audit, field personnel measure AP, T-, , and P-j. By knowing K,
the "true flow" can be calculated. This flow is compared with the flow rate
as measured by the Hi-Vol sensor to determine the accuracy of the flow
rate measurements.
15
-------
SECTION 5
RESULTS
AMBIENT SULFUR DIOXIDE
Participant Characteristics
Sulfur dioxide study number 1176 began in November 1976. Out of 184
sample sets requested by participants, 151 sets of data were returned for
a response rate of 82%. Study number 0477 began in April 1977. Out of
173 sample sets requested, 140 sets of data were returned for a response
rate of 81%. Study 1077 began in October 1977. Out of 166 sample sets
requested, 126 sets of data were returned for a response rate of 76%.
Table 6 indicates the monitoring agency type distribution.
TABLE 6. S02 AGENCY DISTRIBUTION
Foreign
EPA
State
Local
Private
Total
S02 Audit - 1176
Agencies Requesting Samples
Agencies Returning Data
1
1
9
7
61
51
85
75
27
17
184
151
S02 Audit - 0477
Agencies Requesting Samples
Agencies Returning Data
1
1
8
8
57
49
78
62
29
20
173
140
S02 Audit - 1077
Agencies Requesting Samples
Agencies Returning Data
1
1
8
5
56
46
73
54
28
20
166
126
16
-------
Methods used to analyze the samples were grouped into three broad
categories: manual pararosaniline, automated pararosaniline, and other,
Table 7 lists the analytical methods used and the number of respondents
using the method.
TABLE 7. S02 ANALYTICAL METHODS
Method
Manual Para.
Automated Para.
Other
Number Agencies Using Method
Audit 1176
118
30
3
Audit 0477
107
31
2
Audit 1077
96
27
3
It should be noted that laboratories tend to define the method used in very
general terms. Thus, an agency reporting to have used the manual pararo-
saniline technique may not have used the method exactly as it appears in
40 CFR 50.11 (1).
Acceptable Ranges
As described in a previous section, two ranges were used as one
means of judging performance. The Sample Ranges and Target Ranges for the
studies are listed in Table 8. The ranges apply to sample values in
increasing concentration.
The Sample Ranges for Audit 1176 were determined using the method
described in the first document in this series (3). Sample Ranges for Audit
0477 were determined using the method described in the Statistical Approach
section of this report. The Sample Ranges, as used in Audit 1077, were not
ranges in the sense used above, but were indications of the precision with
17
-------
TABLE 8. S02 SAMPLE AND TARGET RANGES
Cone. 1
Cone. 2
Cone. 3
Cone. 4
Cone, t)
S02 Audit - 1176
Sample Range
Target Range
+ 9.8%
+ 20%
+ 5.4%
+ 20%
+ 7.0%
+ 20%
+_ 4.7%
+ 20%
+ 2.9%
+_ 20%
S02 Audit - 0477
Sample Range
Target Range
+ 10%
+ 20%
+ 10%
+ 20%
+ 10%
+ 20%
+ 10%
+_ 20%
+ 10%
+ 20%
S02 Audit - 1077
Sample Range
Target Range
+ 3.7%
+ 20%
+ 2.3%
+ 20%
+ 2.4%
+ 20%
+ 2.4%
+ 20%
+_ 0.6%
+ 20%
which QAB analyzed the samples. The percentages were the equivalent of
one standard deviation from the mean. As explained previously, it was
determined from earlier studies that the average percent difference between
reported results and EPA determined values was +20%. Thus, this value has
been used for all Target Ranges.
Using the above criteria, a tabulation was made of the number of
agencies reporting results within the ranges. For Audit 1176, 53 (35%) of
the agencies reported all five results within the Target Ranges, while
20 (13%) of the laboratories reported all results outside these ranges.
The corresponding figures for Audit 0477 showed 49 (35%) and 7 (5%),
respectively. Audit 1077 showed 36 (29%) and 11 (9%), respectively. These
results compare with previous audits in that 37% of the laboratories
submitting results in Audit 0576 reported all samples within the Target
18
-------
Ranges, while 10% reported all values outside the ranges. Figure 4 shows
a running tally of the above values.
CO
-------
ranges if the laboratory is performing similarly to most of the study
population. Thus, if the agency is not performing in a similar manner, its
data should not be included in the summaries.
The remainder of the discussion will be concerned with the reported
results minus the data sets that met the exclusion criterion. These data
points are marked in the Appendix by an asterisk (*).
Table 9 is a frequency distribution of the percent difference between
the reported and EPA values for each sample concentration. The differences
were calculated by the following equation:
Reported value - EPA value
Absolute Percent Difference =
x 100 .
EPA value
The frequency distribution was then constructed and appears below. For
example, for audit 1175, 50% of the reported results for sample Concentration
1 were less than or equal to an absolute percent difference of 17.6%. It
should be noted that the "All Samples" line is not an average of the
numbers appearing above it, but is the distribution resulting when all the
data is examined together regardless of its concentration.
The data in Table 9 are very useful for laboratories trying to
determine their performance relative to other participants. For example,
for Audit 0477, only 10% of the labs reporting results for Concentration
1 had a percent difference of 4.3% or less, while 50% of the laboratories
reported a percent difference of 21.0% or less for the same concentration.
The table also indicates the average percent difference for all laboratories
for all samples. These are presented graphically in Figure 5 along with
corresponding values from previous audits.
20
-------
TABLE 9. S02 ABSOLUTE PERCENT DIFFERENCE
No.
Min.
10%
30%
50%
70%
90%
Max.
Mean
S02 Audit - 1176
Concentration 1
Concentration 2
Concentration 3
Concentration 4
Concentration 5
All Samples
128
133
133
132
130
656
0.7
0.5
0.0
0.5
0.2
0.0
2.5
2.4
1.7
2.9
1.3
2.2
9.6
6.2
5.3
6.8
5.1
6.3
17.6
10.8
9.2
10.4
7.9
10.6
30.0
16.6
14.3
15.3
11.6
16.3
58.9
32.6
21.4
24.0
17.8
35.5
173.9
448.0
614.2
871.1
571.0
871.1
27.2
18.0
16.5
23.6
14.9
20.0
S02 Audit - 0477
Concentration 1
Concentration 2
Concentration 3
Concentration 4
Concentration 5
All Samples
128
132
132
131
133
656
0.5
0.0
0.3
0.0
0.0
0.0
4.3
2.2
1.5
0.6
1.2
1.5
15.3
5.6
3.9
3.1
2.9
4.1
21.0
10.3
6.4
4.8
6.1
8.6
31.5
18.2
11.5
9.3
9.7
16.3
55.5
31.6
23.0
17.2
23.1
32.3
851.1
75.8
61.6
43.9
67.2
.851.1
33.5
15.1
10.6
8.4
9.5
15.3
S02 Audit - 1077
Concentration 1
Concentration 2
Concentration 3
Concentration 4
Concentration 5
All Samples
103
in
112
111
no
547
0.0
0.2
0.2
0.2
0.0
0.0
2.9
2.1
1.5
0.6
0.8
1.4
13.0
4.6
3.2
2.7
3.3
3.7
22.9
8.5
5.6
5.3
4.9
6.8
46.1
13.4
8.8
8.4
7.5
13.3
69,1
34.7
18.4
17.4
16.4
39.8
134.4
715.5
51.3
89.1
97.2
715.5
44.7
20.8
8.4
8.4
9.3
17.9
All data received (with the exception of previously omitted data) were
grouped according to concentration. The results for each sample value are
listed in the Appendix in increasing concentration. The summary statistics
which appear in Table 10 and at the top of each listing in the Appendix do
not include the data sets identified as outliers.
21
-------
OJ
o
E
CD
60% i
50%
40% •
£ 30%
M-
•r—
Q
20% -
10%-
0
r-»
«>f
o
Study Number
Figure 5. Plot of SOp absolute percent differences.
Table 10 lists summary statistics based on the reported data. The "Out-
liers Removed" columns contain data on which the outlier criterion was applied;
all data which met the second criterion stated in Section 3 were removed. It
should be noted that the variation in the number of samples from concentration
to concentration is due to laboratory accidents and damaged samples.
An examination of Table 10 reveals that there are no apparent problems
with bias (as indicated by "Accuracy"). For only one sample is the "Accuracy"
figure greater than 10%. For all three audits the data are Normally
distributed (Skewness, see foot note of table).
Each laboratory data set was plotted against the corresponding EPA
data set, and the slope and intercept of the linear regression line were
determined. For audit 1176, the mean of 132 slopes was 0.912 with a standard
22
-------
TABLE 10. S02 SURVEY STATISTICS (jag/m3)
Concentration 1
All Outliers
Data Removed
Concentration 2
All Outliers
Data Removed
Concentration 3
All Outliers
Data Removed
Concentration 4
All Outliers
Data Removed
Concentration 5
All Outliers
Data Removed
S02 Audit - 1176
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var.
Skewnessi
Accuracy1"
128 125
19.1 19.1
17.5 16.8
17.6 17.4
52.2 36.5
7.6 6.1
43.1 36.0
1.1 -0.3
-8.1 -9.2
133 132
43.0 43.0
45.7 44.3
45.1 44.9
225.5 63.1
18.6 8.5
40.8 19.2
7.9 -0.5
4.9 4.4
133 132
73.4 73.4
80.9 77.5
77.8 77.8
484.1 93.5
40.4 11.7
50.0 15.1
9.9 0.5
6.0 6.0
132 130
103.0 103.0
123.1 112.7
111.8 111.8
950.2 191.3
89.8 18.4
73.0 16.3
8.2 2.5
8.6 8.6
130 129
149.0 149.0
163.0 156.6
158.4 158.3
999.1 250.0
77.6 23.4
47.6 14.9
9.6 -1.9
6.3 6.2
S02 Audit - 0477
Number
True value
Mean
Median
Jtenge
Std. Dev.
Coef. Var.
Skewness0
Accuracy1"
128 127
21.9 21.9
29.4 18.9
18.3 18.3
207.2 44.6
18.2 7.1
89.0 37.5
8.7 0.9
-16.4 -16.4
132 130
48.2 48.2
45.3 45.8
45.7 45.8
60.0 51.5
9.6 8.8
21.2 19.2
-0.2 0.3
-5.2 -5.1
132 129
80.2 80.2
77.8 78.2
78.2 78.3
83.5 62.0
12.1 10.5
15.5 13.4
-0.7 -0.4
-2.5 -2.3
131 128
114.7 114.7
115.2 114.1
113.6 113.0
86.1 75.0
13.9 12.1
12.1 10.6
0.9 0.4
-0.1 -1.5
133 130
155.4 155.4
158.6 159.6
158.4 158.6
177.7 101.0
22.2 17.8
14.0 11.2
-0.9 0.1
1.9 2.1
S02 Audit - 1077
Number
True value
Mean
Median
Ran ye
StdT. Dev.
Coef. Var.
Skewness0
Accuracy""
103 102
13.6 13.6
15.4 13.6
r~ 13.1 13.1
193.9 29.1
18.9 5.8
122.9 42.5
8.5 0.5
-3.7 -3.7
in no
56.2 56.2
57.4 53.8
54.6 54.6
437.3 91.2
40.3 12.4
70.2 23.0
8.9 1.0
-2.8 -2.0
112 110
53.7 93.7
91.9 92.6
91.8 92.3
76.0 56.8
11.1 9.7
12.0 10.5
-0.9 -0.3
-2.0 -1.4
111 108
93.7 93.7
94.8 94.8
94.3 94.3
123.4 51.1
13.0 8.9
13.6 9.4
1.9 0.3
0.6 0.6
110 106
179.1 179.1
167.8 173.4
172.9 172.9
206.6 115.1
32.6 15.3
19.4 8.8
-3.5 -1.4
-3.4 -3.4
ro
CO
1
Skewness - A statistic that indicates the lack of symmetry in a distribution.
this value is near zero.
For a normal distribution
"Accuracy -
Median - True value
True value
x 100
-------
deviation of 0.209. The mean intercept was 3.93 with a standard deviation
of 12.00. Audit 0477 had a mean slope of 0.944 with a standard deviation
of 0.148. The mean intercept was 6.44 with a standard deviation of 10.75.
Audit 1077 had a slope of 0.957 with a standard deviation of 0.241. The
intercept was 6.02 with a standard deviation of 21.36. Figure 6 shows a
graphical plot of these results.
r 7
O)
Q.
O
1.10
1.05
1.00
0.95
0.90 -
0.85
Intercept
. 6
. 5^
Slope
. 4
. 3
. 2
. 1
E
•^
o>
O.
O)
O
LO
o
o
Study Number
Figure 6. Plot of SOo slope and intercept.
Analytical Method Summary
A check was made to determine whether the data sets excluded from the
summary were related to the method used for analysis (i.e., was any one method
responsible for most of the data outliers). The following table (Table 11)
resulted.
24
-------
TABLE 11. SUMMARY OF S02 ANALYTICAL METHODS USED BY OUTLIER LABS
Method
Total Number
Using Method
Number Identified
as Outliers
Percent of Total
as Outliers
S02 Audit - 1176
Pararosani 1 i ne-Manual
Pararosani 1 i ne-Automated
Other
118
30
3
17
3
-
14%
10%
-
S02 Audit - 0477
Pararosani 1 i ne-Manual
Pararosani 1 i ne-Automated
Other
107
31
2
7
-
-
S02 Audit - 1077
Pararosani 1 i ne-Manual
Pararosani 1 i ne-Automated
Other
96
27
3
11
2
-
7%
-
-
11%
7%
-
As can be seen, no one method contributed an unusually large portion of the
outlier data.
To determine whether a particular analytical method produced biased
results, Table 12 was developed. This table contains the mean and standard
deviation of each sample concentration for each method used.
No pattern was established for the manual method in any of the audits,
with the exception of 1176. In this audit, the lower values were lower
than the "true values", while the higher concentrations were greater than
the "true values". The same pattern held for the automated method for both
1176 and 0477
25
-------
TABLE 12. MEAN AND STANDARD DEVIATION OF S02 RESULTS BY ANALYTICAL METHOD (yg/rn )
Cone. 1
Mean Std. Dev.
Cone. 2
Mean Std. Dev.
Cone. 3
Mean Std. Dev.
Cone. 4
Mean Std. Dev.
Cone. 5
Mean Std. Dev.
S02 Audit - 1176
Manual
Automated
Other
True value
17.5 7.6
15.7 5.2
16.7 3.2
19.1
43.9 9.2
45.3 5.0
46.9 7.5
43.0
77.2 12.5
78.4 9.2
78.8 7.3
73.4
121.0 91.0
113.9 11.3
120.1 10.3
103.0
154.2 25.3
164.3 13.0
168.6 12.7
149.0
S02 Audit - 0477
Manual
Automated
Other
True value
21.5 20.7
17.5 3.6
8.7 10.7
21.9
45.6 10.7
45.0 5.1
35.6 0.2
48.2
78.4 12.4
77.8 7.7
47.8 24.1
80.2
115.7 15.2
114.0 9.3
108.5 2.1
114.7
159.2 22.2
159.8 17.0
110.9 55.0
155.4
S02 Audit - 1077
Manual
Automated
Other
True value
15.9 12.8
13.2 4.7
11.3 2.4
13.6
51.9 13.4
72.6 81.9
49.6 17.2
56.2
89.0 14.8
96.7 8.9
103.7 8.2
93.7
93.1 16.1
95.8 14.6
105.0 9.0
93.7
162.4 39.6
179.6 17.1
168.4 26.6
179.1
ro
-------
As would be expected, the automated method was more precise than the
manual method. However, in many cases the "Other" category was more
precise, as indicated by smaller standard deviations, than either the
automated or manual method.
Recheck Program
Starting with audit 1176, a recheck program was begun. Any laboratory
reporting at least three results that were greater than +20% of the EPA values
was sent a second set of samples. This procedure would distinguish
between labs having chronic problems and those who just had a "bad day."
For audit 1176, 34 laboratories received a second set of samples. Of the
24 groups returning data, 13 had corrected their problems to the extent
that at least three samples fell within +20% of the EPA values.
Twenty-three laboratories received recheck samples for audit 0477.
Sixteen of these groups returned data, of which 14 had corrected their
problems to the extent that at least three samples fell within +20% of the
EPA values. Fourteen agencies received recheck samples for audit 1077.
Eight returned data, of which six had corrected their problem to the extent
that at least three samples fell within +20% of the EPA values.
Nine agencies receiving recheck samples did so in at least two of
the above audits. Of these, four did not participate in the remaining audit.
Thus, it appears that a small percentage of the agencies performing S02
analysis still have basic analytical problems. It also appears that many
of the problems uncovered during the initial analysis were corrected
before the recheck samples were received.
Summary
The S02 audit starting dates covered in this report ranged from
November 1976 to October 1977. The number of participants varied from
27
-------
126 to 184. Foreign, EPA, State, local, and private laboratories submitted
data.
Three analytical methods were used. The majority of the laboratories
in all three audits used the manual pararosaniline method. Approximately 25%
of the laboratories used the automated pararosaniline procedure.
The overall results showed no bias of any practical significance between
the reported and EPA values. The statistical distribution of each audit
appeared to be normal.
As would be expected, the automated method proved to be more precise
than the manual method. Neither method appeared to be more accurate than
the other.
The average slopes (Reported vs. EPA) for each audit ranged from 0.912
3
to 0.957. The intercepts varied from 3.93 to 6.44 yg/m .
AMBIENT NITROGEN DIOXIDE
Participant Characteristics
Nitrogen dioxide audit number 1276 began in December 1976. Out of
130 sample sets requested by participants, 110 sets of data were returned
for a response rate of 85%. Audit number 0677 began in June 1977. Out of
123 sample sets requested, 95 sets of data were returned for a response
rate of 77%. Audit number 1277 began in December 1977. Out of 124 sample
sets requested, 90 sets of data were returned for a response rate of 73%.
Table 13 indicates the monitoring agency type distribution.
28
-------
TABLE 13. N02 AGENCY DISTRIBUTION
Foreign
EPA
State
Local
Private
Total
N02 Audit - 1276
Agencies Requesting Samples
Agencies Returning Samples
2
1
11
8
47
41
57
50
13
10
130
no
N02 Audit - 0677
Agencies Requesting Samples
Agencies Returning Samples
1
1
9
6
41
37
60
44
12
7
123
95
N02 Audit - 1277
Agencies Requesting Samples
Agencies Returning Samples
1
1
9
5
41
37
58
40
15
7
124
90
Methods used to analyze the samples were grouped into five broad
categories. Table 14 lists the analytical methods used and the number of
respondents using the method.
TABLE 14. N02 ANALYTICAL METHODS
Method
Saltzman-man.
Sodium Arsenite-man.
Sodium Arsenite-auto.
TGS - ANSA-man.
Other
Number Agencies Using Method
Audit 1276
3
72
28
3
4
Audit 0677
4
62
23
2
4
Audit 1277
4
56
24
2
4
It should be noted that agencies tend to define the analytical method used
in very general terms. A laboratory reporting the use of the manual sodium
arsenite procedure as the method of choice may have used that procedure with
various modifications. Thus, Table 14 should be interpreted as 72 laborator-
ies used a procedure approximating the manual sodium arsenite method during
audit 1276.
29
-------
Acceptable Ranges
As described in a previous section, two performance ranges were used
as one means available for judging performance. The Sample Ranges and
Target Ranges for NCL are listed in Table 15. The ranges apply to sample
values in increasing concentration.
TABLE 15. N02 SAMPLE AND TARGET RANGES
Sample Range
Target Range
Cone. 1
±10%
±20%
Cone. 2
±10%
±20%
Cone. 3
±10%
±20%
Cone. 4
±10%
±20%
Cone. 5
±10%
±20%
Neither the Sample Ranges nor the Target Ranges were determined using
the methods described under Statistical Approach. Both ranges were arbi-
trarily set at the values listed in Table 15. As more audits are conducted
and more data becomes available, the ranges will be refined.
Using the above criteria, a tabulation was made of the number of
agencies reporting results within the ranges. For audit 1276, a total of
61 (55%) agencies reported all five results within the Target Ranges, while
2 (2%) laboratories reported all results outside these ranges. The corres-
ponding figures for audit 0677 showed 59 (62%) and 2 (2%), respectively.
Audit 1277 showed 69 (77%) and 3 (3%), respectively. These results compare with
previous audits in that 51% of the laboratories submitting results in
study 0676 reported all samples within the Target Ranges, while 8% reported
all values outside the ranges. Figure 7 shows a running plot of the
above .values.
30
-------
80% 1
70% •
60%
50%
CO
OJ
I 40%
0)
.
CM
r-.
vo
o
C\J
Study Number
Figure 7. Plot of NO- results (five samples
within and outside Target Ranges).
Data Summary
Using the Target Ranges as one means of eliminating nonrepresentative
data, it was decided that any laboratory not reporting at least one value
within the Target Ranges would be considered an outlier. The Target Ranges
are sufficiently broad that, unless the analysis is totally out of control,
at least one value would fall within the ranges if the laboratory is
performing similarly to most of the study population. Thus, if the agency
is not performing in a similar manner, its data should not be included in
the summaries. The remainder of this discussion will deal with the reported
results minus the values identified as outliers. The values identified as
outliers are noted in the Appendix by an asterisk (*).
31
-------
Table 16 is a frequency distribution of the percent difference between
the reported and EPA values for each sample concentration. The differences
were calculated by the following formula:
Absolute Percent Difference =
Reported value - EPA value
EPA value
x 100.
The frequency distribution was then constructed and appears below. It
TABLE 16. N02 ABSOLUTE PERCENT DIFFERENCE
No.
Min.
10%
30%
50%
70%
90%
Max.
Mean
N02 Audit - 1276
Cone. 1
Cone. 2
Cone. 3
Cone. 4
Cone. 5
All Samples
106
105
108
108
107
534
0.0
0.6
0.0
0.2
0.0
0.0
0.8
0.6
0.9
0.8
0.5
0.8
5.0
5.0
2.4
3.0
1.7
2.8
8.3
8.8
4.9
6.0
3.0
6.1
16.7
11.6
7.4
8.2
6.3
10.1
30.8
22.6
16.6
14.2
13.8
22.5
80.0
81.8
32.2
40.5
34.1
81.8
15.3
11.4
7.1
7.3
5.4
9.3
N02 Audit - 0677
Cone. 1
Cone. 2
Cone. 3
Cone. 4
Cone. 5
All Samples
88
91
86
92
90
448
0.0
0.0
0.4
0.0
0.0
0.0
0.8
0.8
1.2
0.5
0.4
0.8
5.9
3.4
2.5
2.1
2.0
2.9
9.2
5.5
4.9
3.4
3.8
5.0
14.3
9.2
7.8
6.9
5.9
9.2
24.4
16.4
17.2
15.1
12.2
17.6
77.3
82.4
76.2
74.9
75.7
82.4
12.7
8.5
7.8
7.8
5.8
8.5
N02 Audit - 1277
Cone. 1
Cone. 2
Cone. 3
Cone. 4
Cone. 5
All Samples
84
86
87
86
88
431
0.0
0.3
0.2
0.0
0.1
0.0
0.6
1.3
1.8
1.0
0.9
1.0
2.2
3.5
3.4
2.2
1.8
2.6
5.0
4.5
4.8
3.4
3.1
4.2
7.7
7.4
6.8
5.4
5.2
6.6
22.1
15.8
12.5
11.9
9.2
13.7
58.6
35.7
19.4
20.6
96.1
96.1
9.0
7.1
6.0
5.0
5.3
6.5
32
-------
should be noted that the "All Samples" line is not an average of the numbers
appearing above it, but is the distribution resulting when all the data is
examined together, regardless of the concentration.
Table 16 is very useful for laboratories trying to determine their
performance relative to the other participants. For example, in audit
1276, only 10% of the labs reporting results for Concentration 1 had a
percent difference of 0.8% or less, while 50% of the laboratories reported
a percent difference of 8.3% or less for the same concentration. The table
also indicates the average percent difference for all laboratories for
all samples. These values are shown in Figure 8 along with corresponding
numbers from previous studies.
30%,
CD
O
c
-------
All data received (with exception of the previously omitted data) were
grouped according to concentration. The results for each sample value are
listed in the Appendix in increasing concentration. The summary statistics
which appear in Table 17 and at the top of each listing in the Appendix do
not include the data sets identified as outliers.
Table 17 lists summary statistics based on the reported data. The
"Outliers Removed" columns contain data on which the outlier criterion was
applied. All data that met the second criterion stated in Section 3 were
removed. It should be noted that the variation in the number of samples
from concentration to concentration is due to laboratory accidents and
damaged samples.
An examination of Table 17 shows that, with the exception of one sample,
all accuracy figures were less than +5%. All concentrations were normally
distributed. Two sets of duplicate samples were analyzed (audits 0677 and
1277). The results showed that the reproducibility was excellent.
Each laboratory data set was plotted against the corresponding EPA
data set, and the slope and intercept of the linear regression were determined,
For audit 1276, the mean of 108 slopes was 1.024 with a standard deviation
of 0.081. The mean intercept was 0.003 with a standard deviation of 0.026.
Audit 0677 had a mean slope of 0.970 with a standard deviation of 0.095. The
mean intercept was 0.000 with a standard deviation of 0.025. Audit 1277 had
a slope of 0.969 with a standard deviation of 0.153. The intercept was
-0.001 with a standard deviation of 0.068. Figure 9 shows a tally of
these results.
34
-------
TABLE 17. N02 SURVEY STATISTICS (yg/ml)
Concentration 1
All Outliers
Data Removed
Concentration 2
All Outliers
Data Removed
Concentration 3
All Outliers
Data Removed
Concentration 4
All Outliers
Data Removed
Concentration 5
All Outliers
Data Removed
N02 Audit - 1276
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Vair.
Skewness0
Accurac/l_
106 104
0.12 0.12
0.12 0.12
0.12 0.12
0.17 0.14
0.03 0.02 1
22.52 20.40
0.17 -0.57
0.00 0.00
105 103
0.18 0.18
0.17 0.17
0.17 0.17
0.24 0.15
0.03 0.02
16.48 13.49
-0.73 0.04
-3.87 -3.87
108 106
0.33 0.33
0.32 0.32
0.32 0.32
0.20 0.18
0.03 0.03
10.37 9.47
0.07 -0.52
-0.61 -1.07
108 107
0.54 0.54
0.52 0.51
0.52 0.52
0.36 0.20
0.05 0.04
9.40 8.34
0.60 -0.56
-2.99 -2.99
107 105
0.87 0.87
0.85 0.85
0.86 0.86
0.54 0.31
0.07 0.06
7.94 6.66
0.48 -0.20
-1.61 -1.61
N02 Audit - 0677
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var.
Skewness0
Accuracy"
88 86
0.12 0.12
0.13 0.13
0.13 0.13
0.15 0.10
0.02 0.02
J5.86 12.43
-1.29 0.37
8.40 8.40
91 89
0.24 0.24
0.25 0.25
0.25 0.25
0.30 0.12
0.03 0.02
12.14 8.04
1.69 -0.71
4.62 4.62
86 85
0.24 0.24
0.25 0.25
0.25 0.25
0.26 0.13
0.03 0.02
11.95 9.22
2.05 -0.64
2.46 2.45
93 90
0.38 0.38
0.39 0.39
0.39 0.39
0.54 0.17
0.06 0.03
14.30 7.37
1.14 -0.17
2.38 2.38
90 88
0.69 0.69
0.72 0.71
0.71 0.71
0.71 0.24
0.07 0.04
9.64 5.38
3.72 -0.46
2.82 2.82
N02 Audit - 1277
Number
True value
Mean
Median
J^aruje
Std. Dev.
Coef. Var.
Skewness0
Accuracy"
84 81
0.18 0.18
0.19 0.19
0.1:9 O.lT
0.19 0.13
0.03 0.02
13.73 10.53
-0.31 0.28
2.49 2.21
86 83
0.31 0.31
0.33 0.33
0.32 0.32
0.18 0.13
0.03 0.02
8.05 6.52
0.15 0.21
4.18 4.18
87 85
0.50 0.50
0.52 0.52
0.52 0.52
0.19 0.14
0.03 0.02
5.42 4.47
-0.97 0.14
4.44 4.44
86 85
0.50 0.50
0.52 0.52
0.52 0.52
0.19 0.15
0.03 0.03
5.51 4.91
^0.70 0.08
3.17 3.17
88 87
0.87 0.87
0.88 0.89
0.89 0.89
0.96 0.29
0.10 0.04
11.51 4.99
-6.63 -0.85
2.30 2.30
1
'C*l^^.i.««-L*-k<-.<-. A r*4--\4--5r*4-T/^ 4" U> -\ 4- *iv^/4pir»^ Mr* v*m*\ T x4*?r»4-\st^l^i*4-4>iM
GO
en
this value is near zero.
"Accuracy -
Medi ajn j- True val ue
True value
x 100
-------
1.15n
1.10-
1.05-
£ LOO-
Q.
O
CO
0.
0.90-
0.85
Intercept
Slope
C\J
CM
r 0.04
• 0.03 ;_
E
0.02
o.
-------
TABLE 17. N02 SURVEY STATISTICS (yg/ml)
Concentration 1
All Outliers
Data Removed
Concentration 2
All Outliers
Data Removed
Concentration 3
All Outliers
Data Removed
Concentration 4
All Outliers
Data Removed
Concentration 5
All Outliers
Data Removed
N02 Audit - 1276
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Varr.
Skewness0
Accura£y^__
106 104
0.12 0.12
0.12 0.12
0.12 0.12
0.17 0.14
0.03 0.02
22.52 20.40
0.17 -0.57
0.00 0.00
105 103
0.18 0.18
0.17 0.17
0.17 0.17
0.24 0.15
0.03 0.02
16.48 13.49
-0.73 0.04
-3.87 -3.87
108 106
0.33 0.33
0.32 0.32
0.32 0.32
0.20 0.18
0.03 0.03
10.37 9.47
0.07 -0.52
-0.61 -1.07
108 107
0.54 0.54
0.52 0.51
0.52 0.52
0.36 0.20
0.05 0.04
9.40 8.34
0.60 -0.56
-2.99 -2.99
107 105
0.87 0.87
0.85 0.85
0.86 0.86
0.54 0.31
0.07 0.06
7.94 6.66
0.48 -0.20
-1.61 -1.61
N02 Audit - 0677
Number
True value
Mean
Median
Range_____
Std. Dev.
Coef. Var.
Skewnesso
Accuracy*"
88 86
0.12 0.12
0.13 0.13
0.13 0.13
0.15 0.10
0.02 0.02
15.86 12.43
-1.29 0.37
8.40 8.40
91 89
0.24 0.24
0.25 0.25
0.25 0.25
0.30 0.12
0.03 0.02
12.14 8.04
1.69 -0.71
4.62 4.62
86 85
0.24 0.24
0.25 0.25
0.25 0.25
0.26 0.13
0.03 0.02
11.95 9.22
2.05 -0.64
2.46 2.45
93 90
0.38 0.38
0.39 0.39
0.39 0.39
0.54 0.17
0.06 0.03
14.30 7.37
1.14 -0.17
2.38 2.38
90 88
0.69 0.69
0.72 0.71
0.71 0.71
0.71 0.24
0.07 0.04
9.64 5.38
3.72 -0.46
2.82 2.82
N02 Audit - 1277
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var.
SkewnessJ,
Accuracy1"
84 81
0.18 0.18
0.19 0.19
0.19 0.19
0.19 0.13
0.03 0.02
13.73 10.53
-0.31 0.28
2.49 2.21
86 83
0.31 0.31
0.33 0.33
0.32 0.32
0.18 0.13
0.03 0.02
8.05 6.52
0.15 0.21
4.18 4.18
87 85
0.50 0.50
0.52 0.52
0.52 0.52
0.19 0.14
0.03 0.02
r 5.42 4.47
-0.97 0.14
4.44 4.44
86 85
0.50 0.50
0.52 0.52
0.52 0.52
0.19 0.15
0.03 0.03
5.51 4.91
^0.70 0.08
3.17 3.17
88 87
0.87 0.87
0.88 0.89
0.89 0.89
0.96 0.29
0.10 0.04
11.51 4.99
-6.63 -0.85
2.30 2.30
CO
en
—: ;
Skewness - A statistic that indicates the lack of symmetry in a distribution. For a Normal distribution
this value is near zero.
"Accuracy- Med1 Rvalue*- x 1QO
-------
. 0.04
CO
CD
Q.
O
CO
1 . 10-
1.10-
1.05-
1.00-
0 Q*v
W • */ w
0.90-
0.85
\Intercept
\____________
<^ 10 r*** r^
*«^ i^^ r*^ f*w_
* \s • W~
• 0.03 _
E
en
- 0.02 *
CL
t^O CxJ VO CxJ
0 i—
O
Study Number
Figure 9. Plot of N02 slope and intercept.
Analytical Method Summary
A check was made to determine whether any relationship existed between
the laboratories whose data were omitted from the summaries and the analytical
method employed (i.e., was any one method responsible for most of the data
outliers). The following table (Table 18) resulted.
As can be seen, no particular analytical method was responsible for an
unusually large portion of the outlier data.
36
-------
TABLE 18. SUMMARY OF N02 ANALYTICAL METHODS USED BY OUTLIER LABS
Method
Sodium
Sodi urn
Sodi urn
Sodium
Sodium
Arsenite
Arsenite
Arsenite
Arsenite
Arsenite
Total Number
Using Method
Number Identified
as Outliers
N02 Audit -
- Man.
- Auto.
72
28
Percent of
as Outli
1276
1
1
N02 Audit -
- Man.
62
Total
ers
1%
4%
0677
2
N02 Audit -
- Man.
- Auto.
56
24
3%
1277
2
1
4%
4%
To determine whether a particular analytical method produced biased
results, Table 19 was developed. This table contains the mean and standard
deviation of each sample concentration for each method used. As can be seen,
the automated sodium arsenite method is the most precise of all the methods.
This procedure also appears to yield the most accurate results. However,
with the exception of the "Other" category, the remainder of the methods
also estimate the EPA value quite adequately.
Recheck Program
Starting with audit 1276, a recheck program was begun. Any laboratory
reporting at least three results that were greater than £20% of the EPA
values was sent a second set of samples. This procedure would distinguish
between labs having chronic problems and those that had a "bad day." For
audit 1276, six laboratories received a second set. Of the six groups
returning data, four had corrected their problem to the extent that at least
three samples fell within +20% of the EPA values.
Seven laboratories received recheck samples for audit 0677. Five of
these groups returned data, of which all had corrected their problems. Three
37
-------
TABLE 19. MEAN AND STANDARD DEVIATION OF N02 RESULTS BY ANALYTICAL METHOD (yg/ml)
Method
Concentration 1
Mean Std. Dev.
Concentration 2
Mean Std. Dev.
Concentration 3
Mean Std. Dev.
Concentration 4
Mean Std. Dev.
Concentration 5
Mean Std, Dev.
N02 Audit - 1276
Saltzman - Man.
Sodium Arsenite-Man.
Sodilum Arsenit6-Auto.
TGS - ANSA-Man.
Other
True value
Saltzman-Man.
Sodium Arsenite-Man.
Sodium Arsenite-Auto.
TGS - ANSA-Man.
Other
True value
0.11 0.02
0.12 0.03
0.12 0.02
0.14 0.06
0.12 0.01
0.12
0.16 0.03
0.17 0.04
0.16 0.05
0.21 0.04
0.17 0.01
0.18
0.34 0.01
0.32 0.04
0.32 0.02
0.35 0.05
0.31 0.01
0.33
0.49 0.04
0.51 0.06
0.52 0.03
0.54 0.04
0.54 0.01
0.54
0.81 0.05
0.85 0.08
0.86 0.04
0.89 0.04
L0.86 0.01
0.87
N02 Audit - 0677
0.13 0.00
0.12 0.02
0.13 0.01
0.15 0.02
0.12 0.02
0.12
0.25 0.00
0.25 0.04
0.24 0.02
0.26 0.02
0.26 0.02
0.24
0.25 0.00
0.25 0.04
0.24 0.02
0.25 0.02
0.26 0.01
0.24
0.38 0.00
0.39 0.06
0.38 0.03
0.52 0.19
0.40 0.04
0.38
0.68 0.02
0.73 0.08
0.70 0.06
0.69 0.01
0.72 0.02
0.69
N0£ Audit - 1277
Saltzman-Man.
Sodium Arsenite-Man.
Sodium Arsenite-Auto.
TGS - ANSA-Man.
Other
True value
0.19 0.02
0.19 0.03
0.18 0.01
h~0.20 0.00
0.22 0.03
0.18
0.33 0.02
0.32 0.03
0.32 0.01
0.32 0.02
0.34 0.02
0.31
0.52 0.04
0.52 0.03
0.52 0.02
0.52 0.01
0.53 0.03
0.52
0.52 0.02
0.52 0.03
0.52 0.02^
0.52 0.00
0.53 0.03
0.50
0.87 0.04
0.89 0.05
0.84 0.18
0.89 0.02
0.91 0.04
0.87
CO
00
-------
Most of the instruments classified as "Other" use an electrochemical
method of detection.
Agencies generally employ a liberal interpretation of the analytical
method used. However, in the case of CO analyses, the methods are clearly
defined. Unknowns that could effect results and are considered part of the
analytical system are purity of zero air and accuracy of calibration
standards.
Acceptable Ranges
r'. -*••
As described in a previous section, two performance ranges were used as
one means of judging performance. The Sample Ranges and Target Ranges for
the audits are listed in Table 22. The ranges apply to sample concentrations
in ascending order.
TABLE 22. CO SAMPLE AND TARGET RANGES
Concentration 1
Concentration 2
Concentration 3
CO Audit - 1076
Sample Range
Target Range
+11.1%
+11.1%
+ 2.4%
+ 10%
+_ 1 . 2%
+ 10%
CO Audit - 0377
Sample Range
Target Range
+ 12.9%
+ 12.9%
+ 2.7%
+ 10%
+ 1.4%
+ 10%
CO Audit - 0977
Sample Range
Target Range
+ 10.4%
+ 10.4%
+ 3.0%
+ 10%
+ 1.4%
+_ 10%
The Sample Ranges were not determined by the method described under
Statistical Approach. Because filling of the cylinders was done so pre-
cisely, the standard deviations of the verification analyses were small.
41
-------
Thus, using the method described under Statistical Approach resulted in
unreasonably small Sample Ranges for all concentrations. Rather than use
unrealistic values, the Sample Ranges were set at the QAB determined value
±0.5 ppm.
The Target Ranges were determined from previous CO audit results.
Earlier studies indicated that the average percent difference between re-
ported results and EPA determined values for all sample concentrations was
+10%. Thus, this value has been used for all Target Ranges with the
exception of Concentration 1. In that case, the Sample Range was used
because of its larger value.
Using the above criteria, a tabulation was made of the number of instru-
ments reporting results within the ranges. For audit 1076, a total of 103
instruments (42%) reported all values within the Target Ranges, while 16
instruments (6%) reported all results outside the Target Ranges. The
corresponding values for audit 0377 showed 153 (52%) and 16 (5%), respectively,
Audit 0977 showed 119 (46%) and 9 (4%), respectively. These results compare
with previous audits in that 46% of the instruments in audit 0476 reported
all values within the Target Ranges, while 9% reported all values outside
the Target Ranges. Figure 10 shows a running graphical plot of the above
values.
Data Summary
Using the Target Ranges as one means of eliminating nonrepresentative
data, it was decided that any instrument not reporting at least one value
within the Target Ranges would be considered an outlier. The Target Ranges
are broad enough that, unless an instrument is totally out of control, at
42
-------
60%
50%
43 40%
c:
o
Study Number
Figure 10. Plot of CO results (three samples
within and outside Target Ranges)
least one value would fall within the ranges. This should occur if the in-
strument is performing similarly to most of the study population. If the
instrument is not performing in a similar manner, its data should not be
included in the summaries. The remainder of the discussion will deal with
the reported results minus the values identified as outliers. These results
are eliminated from further data summaries. The values identified as out-
liers are indicated in the Appendix by an asterisk (*).
Table 23 is a frequency distribution of the percent difference between
the reported and EPA values for each sample concentration. The differences
were calculated by the following formula:
Absolute Percent Difference =
[Reported value - EPA value
EPA value
x 100
The frequency distribution was then constructed and appears below. It
43
-------
should be noted that the "All Samples" line is not an average of the
numbers appearing above it, but is the distribution resulting when all data
is examined together regardless of concentration.
Table 23 is very useful for laboratories trying to determine their per-
formance relative to the other participants. For example, in audit 1076
only 10% of the instruments reporting results for Concentration 1 had a
percent difference of 0.2% or less, while 50% of the instruments reported a
difference of 9.1% or less for the same concentration. The table also
indicates the average percent difference for all instruments for all samples,
TABLE 23. ABSOLUTE PERCENT DIFFERENCE
No.
Min.
10%
30%
50%
70%
90%
Max.
Mean
CO Audit - 1076
Concentration 1
Concentration 2
Concentration 3
All Samples
228
230
214
672
0.2
0.1
0.1
0.1
0.2
0.7
0.4
0.4
4.2
1.7
1.4
1.8
9.1
3.1
2.2
3.9
11.5
5.5
4.1
7.1
22.4
10.3
7.1
15.7
"842.1
89.4
32.5
842.1
"19.2
5.3
3.6
9.5
CO Audit - 0377
Concentration 1
Concentration 2
Concentration 3
All Samples
279
278
275
832
0.6
0.1
0.1
0.1
3.6
0.8
0.6
0.8
6.5
2.3
1.5
2.5
11.2
4.0
2.5
4.2
18.3
5.9
4.2
9.0
33.1
11.8
8.8
21.0
85.2
101.6
52.7
101.6
15.1
5.7
4.4
8.4
CO Audit - 0977
Concentration 1
Concentration 2
Concentration 3
All Samples
243
244
244
731
0.2
0.1
0.0
0.0
1.9
0.5
0.3
0.5
6.0
1.9
1.2
1.9
8.1
3.6
2.0
4.3
14.4
5.5
3.9
7.3
23.8
io.i
8.3
15.7
202.7
29.5
22.2
202.7
12.4
4.6
3.4
6.8
These values are tabulated in Figure 11 along with corresponding values from
previous audits.
44
-------
40%-
cu 30%
o
£=
I 20%
10%
0
i*** r--
•?*• O
O r—
CO
O
Study Number
Figure 11. Plot of CO absolute
percent differences
All data received (with the exception of the previously omitted data)
were grouped according to concentration. The results for each sample value
are listed in the Appendix, in increasing concentration. The summary statis-
tics which appear in Table 24 and at the top of each listing in the Appendix
do not include the data sets identified as outliers.
45
-------
TABLE 24. CO SURVEY STATISTICS (ppm)
Concentration 1
All Outliers
Data Removed
Concentration 2
All Outliers
Data Removed
Concentration 3
All Outliers
Data Removed
CO Audit - 1076
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var.
Skewness0
Accuracy1"
228 225
4.5 4.5
4.6 4.2
4.3 4.3
40.9 4.4
3.7 0.6
70.7 14.2
9.0 -0.7
-4.7 -5.5
230 227
20.9 20.9
21.2 21.1
21.0 21.0
23.7 10.8
2.0 1.3
9.3 6.3
4.2 -0.1
0.7 0.7
214 210
40.1 40.1
40.7 40.7
40.4 40.4
16.4 9.6
2.1 1.7
5.1 4.2
1.8 0.7
0.8 0.6
CO Audit - 0377
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var.
Skewness0
Accuracy1"
279 276
3.4 3.4
3.1 3.1
3.0 3.0
4.5 3.3
0.6 0.6
19.4 18.1
-0.2 -0.2
-10.1 -10.1
273 275
15.9 15.9
16.2 15.1
16.0 16.0
20.0 8.3
1.5 1.1
9.3 6.8
4.4 0.2
0.8 0.8
275 268
31.7 31.7
32.5 32.3
32.2 32.1
29.5 10.2
2.3 1.5
7.0 4.6
0.1 1.0
"1.5 1.2
CO Audit •- 0977
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var.
SkewnesSo
Accuracy"
243 242
4.8 4.8
4.6 4.6
4.5 4.5
12.0 4.8
0.9 0.7
20.2 15.2
4.9 0.4
-6.0 -6.0
244 242
16.6 16.6
16.7 16.7
16.7 16.7
9.3 6.0
1.0 1.0
6.2 5.8
-0.2 -0.0
0.6 0.6
244 239
36.0 36.0
36.7 36.6
36.4 36.4
13.2 9.4
1.7 1.4
4.7 3.9
1.1 0.4
1.1 1.1
1
Skewness - A statistic that indicates the lack of symmetry in a distribution
For a Normal distribution this value is near zero.
"Accuracy - Median - True value
True value
x 100
46
-------
Table 24 lists summary statistics based on the reported data. The
"Outliers Removed" column contains data on which the outlier criterion was
applied. All data that met the second criterion in Section 3 were removed.
^
It should be noted that the variation in the number of samples from concentra-
tion to concentration is due to laboratory accidents and damaged samples.
An examination of the data in Table 24 reveals no bias or skewness
problems in any of the audits. The greatest inaccuracy was -6.7%. The
near zero value of the skewness indicator shows that the distribution
was normal.
The EPA determined "True value" and the median of the study results agree
well. This is indicated by the "Accuracy" values. The study population is
normally distributed as indicated by the "Skewness".
Each instrument data set was plotted against the corresponding EPA data
set, and the slope and intercept from the linear regression were determined.
For audit 1076, the mean of 230 slopes was 0.963 with a standard deviation
of 0.125. The mean intercept was 0.642 with a standard deviation of 3.062.
Audit 0377 had a mean slope of 0.963 with a standard deviation of 0.067.
The mean intercept was 0.423 with a standard deviation of 0.926. A total
of 278 instrument results were used to determine these values. Audit 0977
had a slope of 0.973 with a standard deviation of 0.055. The intercept
was 0.324 with a standard deviation of 1.012. The number of instrument
results used to calculate these values was 248. Figure 12 shows a tally
of these results.
47
-------
CO
CU
Q.
O
oo
i.io-
1.05-
1.00-
0.95
0.90
0.85 L
Intercept
Slope
"3- O CO CT>
O •— O O
0.6
h 0.5
h 0.4 g;
0.3
0.2
^ 0.1
Study Number
Figure 12. Plot of CO slope and intercept.
Analytical Method Summary
A check was made to determine whether any relationship existed between the
instruments whose data were omitted from the summaries and the analytical
method employed (i.e., was any one method responsible for most of the data
outliers). The following table (Table 25) resulted. As can be seen, both
of the predominately used methods contributed approximately equally to the
outlier data.
48
-------
TABLE 25. SUMMARY OF CO ANALYTICAL METHODS USED BY OUTLIER INSTRUMENTS
Method
Total No.
Using Method
No. Identified
as Outliers
% of Total
as Outliers
CO Audit - 1076
NDIR
Flame lonization
Other
187
51
9
11
5
-
6%
10%
0%
CO Audit - 0377
NDIR
Flame lonization
Other
243
42
10
15
1
-
6%
2%
0%
CO Audit - 0977
NDIR
Flame lonization
Other
219
31
7
6
3
«,
3%
10%
0%
To determine whether a particular analytical method produced results that
contained a bias, Table 26 was developed. This table contains the mean and
standard deviation of each sample concentration for each method used.
TABLE 26. MEAN AND STANDARD DEVIATION OF CO RESULTS BY ANALYTICAL METHOD (ppm)
Method
Concentration 1
Mean Std. Dev.
Concentration 2
Mean Std. Dev.
Concentration 3
Mean Std. Dev.
CO Audit - 1076
NDIR
Flame lonization
Other
True value
4.8 4.2
4.0 0.9
4.4 0.5
4.5
21.3 1.9
20.8 2.2
20.9 0.7
20.9
40.6 1.6 !
41.2 3.3
39.4 0.8 1
40.1 1
CO Audit - 0377
NDIR
Flame lonization
Other
True value
NDIR
Flame lonization
Other
True value
3.1 0.6
2.9 0,3
3.5 0.8
3.4
16.2 1.0
16.5 3.2
15.6 1.0
15.9
32.4 1.5
32.7 4.6
31.6 1.5
31.7 !
CO Audit - 0977
4.6 1.0
4.0 0.5
5.0 0.8
4.8
16.5 2.5
15.9 1.4
17.2 0.6
16.6
36.5 3.1
36.5 1.7
36.6 1.9
36.0
49
-------
The data indicate that the NDIR method was the most accurate of the two
major methods. However, the "Other" category was the most accurate. The
method that showed the best precision varied from audit to audit. The "Other"
category was the most precise of the three methods.
Summary
The CO audit start dates covered in this report ranged from October 1976
to September 1977. Number of participants varied from 92 to 116. Number
of instruments checked varied from 247 to 295. Foreign, EPA, State, local,
and private laboratories submitted data.
Three analytical methods were used. The majority of the instruments
were NDIRs. Approximately 80% of the instruments were of this type.
The overall results showed no bias between the reported and EPA values.
The results appeared to be normally distributed .
Of the two most used procedures, the NDIR technique seemed to give the
most accurate results. However, the "Other" category was the most accurate
and precise.
The average slopes (Reported vs. EPA) for each audit varied from 0.963
to 0.973. The intercepts ranged from 0.324 to 0.642.
HI-VOL SULFATE
Participant Characteristics
Sulfate audit number 1076 began in October 1976. Out of 71 sample sets
requested by participants, 56 sets of data were returned for a response
rate of 79%. Audit number 0277 began in February 1977. Out of 71 sample
sets requested, 57 sets of data were returned for a response rate of 80%.
Audit number 0877 began in August 1977. Out of 85 sample sets requested,
57 were returned for a response rate of 67%.
50
-------
Table 27 indicates the monitoring agency type distribution.
TABLE 27. SO^ AGENCY DISTRIBUTION
Agencies Requesting Samples
Agencies Returning Data
Agencies Requesting Samples
Agencies Returning Data
Agencies Requesting Samples
Agencies Returning Data
Foreign
1
0
2
2
3
2
EPA
5
2
3
2
4 ,
3
State
SOj /
32
27
so*/
__JO_!
27
so;/
r 3T~n
23
Local
\ud1t - 1
23
19
\ud1t - 0
24
18
\udit - 0
30
21
Private
076
10
8
277
12
8
877
17
8
Total
71
56
71
57
85
57
Methods used to analyze the samples were grouped into nine broad
categories. Results were received from laboratories using eight methods and
are listed in Table 28.
TABLE 28. SOjj ANALYTICAL METHODS
Method
Methyl thymol blue - Man.
Methyl thymol blue - Auto.
Barium Chloride - Man.
Barium Chloride - Auto.
Sulfa-ver - Man.
Sulfa-ver - Auto.
Barium Chi orani late - Man.
Other
Number Agencies Using Method
Audit 1076
1
11
Audit 0277
_
13
20 i 18
4 i 2
17
-
-
3
14
1
Audit 0877
-
13
23
2
12
1
1 I
8 1 6 i
It should be noted that some agencies tend to define the analytical methods
used in very general terms. A laboratory reporting the use of the automated
methyl thymol blue procedure as the method of choice may have used that pro-
cedure with various modifications. Thus, Table 28 should be interpreted as
meaning that 11 laboratories in audit 1076 used procedures approximating the
automated methyl thymol blue procedure.
51
-------
Acceptable Ranges
As described in a previous section, two performance ranges were used
as one means of judging performance. The Sample Ranges and Target Ranges
are listed in Table 29. The ranges apply to sample concentrations in
ascending order.
TABLE 29. SO^ SAMPLE AND TARGET RANGES
Sample Range
Target Range
Cone. 1
+ 5%
+ 15%
Cone. 2
+ 5%
+ 15%
Cone. 3
+ 5%
+ 15%
Cone. 4
+ 5%
+ 15%
Cone. 5
+ 5%
+ 15%
Cone. 6
+ 5%
+ 15%
Neither the Sample Ranges nor the Target Ranges were determined using
the methods described under Statistical Approach. Both ranges were arbi-
trarily set at the values listed in Table 29. As more audits are conducted
and more data become available, the ranges will be refined.
Using the above criteria, a tabulation was made of the number of agencies
reporting results within the ranges. For audit 1076, five (9%) agencies re-
ported five or six results within the Target Ranges, while seven (12%)
laboratories reported all results outside these ranges. The corresponding
figures for audit 0277 showed 18 (32%) and 2 (4%), respectively. Audit 0877
showed 15 (26%) and 7 (12%), respectively. The improvement in performance
from the first to the second and third studies is probably due to the sample
concentrations being analyzed. During audit 1076, most of the samples were
near the minimum detectable limit of the methods in use.
The above results compare with previous audits in that 37% of the
laboratories reported five or six values within the ranges during audit 0476,
and 15% were not able to report any values within the ranges. Figure 13 shows
52
-------
a running graphical plot of the above values.
O)
(J
o>
60%-,
50%.
40%.
30%.
20%.
10%-
0
Five or Six Samples
In Target Ranges
All Samples Outside
Target Ranges
c\j
o
—r
r-«
r-.
oo
o
Study Number
Figure 13. Plot of SOJ; results (five or six
samples with'in Target Ranges and
six samples outside Target Ranges).
Data Summary
Using the Target Ranges as one means of eliminating nonrepresentative
data, it was decided that any laboratory not reporting at least one value
within the Target Ranges would be considered an outlier. The Target Ranges
are sufficiently broad that, unless a laboratory is totally out of control,
at least one value would fall within the ranges; this should occur if the
laboratory is performing similarly to most of the study population. If the
53
-------
lab is not performing in a similar manner, its data should not be included
in the summaries. Seven laboratories met this criterion for outlier re-
jection in 1076, two in 0277, and seven in 0877, and were eliminated from
further data summaries. The values identified as outliers are indicated in
the Appendix by an asterisk (*).
Table 30 is a frequency distribution of the percent difference between
the reported and EPA values for each sample concentration. The differences
were calculated by the following formula:
Absolute Percent Difference =
Reported value - EPA value
x 100 .
EPA value
The frequency distribution was then constructed and appears below. It should
be noted that the "All Samples" line is not an average of the numbers appear-
ing above it, but is the distribution resulting wh.en all data is examined
together regardless of concentration.
Table 30 is very useful for laboratories trying to determine their
performance relative to other participants. For example, only 10% of the
labs in Audit 0277 reporting results for Concentration 2 had a percent
difference of 1.7% or less, while 50% of the laboratories reported a percent
difference of 6.6% or less for the same concentration. The table also
indicates the average percent difference for all laboratories for
all samples (except Concentration 1). These values are tabulated in Figure
14 along with corresponding values from previous audits.
Concentration 1 was a blank for Audits 1076 and 0877 and was not
included in the "All Samples" distribution. Because small concentration
differences result in large percent differences, it was felt that the
numbers would unduly distort the study results.
54
-------
TABLE 30. S0| ABSOLUTE PERCENT DIFFERENCE
No.
Min.
10%
30%
50%
70%
90%
Max.
Mean
SO;J Audit - 1076
Concentration 1
Concentration 2
Concentration 3
Concentration 4
All Samples
Blank
48
48
99
241
0.2
0.4
0.1
0.1
6.0
0.4
1.2
1.9
26.4
5.6
3.9
6.6
47.2
14.5
7.5
14.5
92.6
31.1
11.3
38.0
205.8
66.0
23.3
97.0
546.0
133.1
82.4
546.0
83.9
28.8
10.3
42.7
S0= Audit - 0277
Concentration 1
Concentration 2
Concentration 3
Concentration 4
All Samples
53
107
111
56
327
0.7
0.1
0.0
0.5
0.0
2.2
1.7
0.8
0.9
1.3
6.0
3.8
3.5
2.3
3.6
12.8
6.6
6.3
4.0
6.3
25.7
10.6
8.8
6.7
11.0
77.2
24.3
20.2
22.5
28.7
838.7
234.4
241.2
48.4
838.7
53.7
14.5
10.4
8.8
18.5
SO^ Audit - 0877
Concentration 1
Concentration 2
Concentration 3
Concentration 4
Concentration 5
Concentration 6
All Samples
Blank
48
50
51
51
50
250
2.9
0.0
0.0
0.0
.0.0
0.0
4.8
1.3
0.8
0.9
1.6
1.4
10.0
7.0
2.8
3.0
4.8
5.4
19.0
10.7
7.0
6.7
6.7
9.0
38.1
33.3
10.8
11.1
8.8
14.3
98.1
73.7
18.9
18.4
16.7
50.5
205.2
120.0
75.7
1260.0
72.2
1260.0
39.2
26.7
10.4
34.8
10.2
24.1
60% -
50% -
40% -
30%-
cu
o
c
CD
S-
OJ
q-
g 20%
10% H
0
vo
o
C\J
o
CO
o
Study Number
Figure 14. Plot of SO^ absolute percent differences.
-------
All data received (with the exception of the previously omitted data)
were grouped according to concentration. Results for each sample value are
listed in the Appendix in increasing concentration. The summary statistics
which appear in Table 31 and at the top of each listing in the Appendix do
not include the data sets identified as outliers.
Table 31 lists summary statistics based on reported data. The "Outliers
Removed" column contains data on which an outlier test has been applied. All
data that met the second criterion in Section 3 were removed. It should be
noted that the variation in the number of samples from concentration to
concentration is due to laboratory accidents, damaged samples, and inclusion
of duplicate samples.
An examination of Table 31 shows that all the data are normally distri-
buted. Inaccuracy at some of the concentrations is larger than would be
desired. However, most of the inaccuracies occur at low concentrations.
Generally, there tends to be a positive bias in the lower concentrations
and a negative bias in the higher concentrations.
Each laboratory data set was plotted against the corresponding EPA data
set, and the slope and intercept of the linear regression line were determined,
For Audit 1076, the mean of 50 slopes was 1.050 with a standard deviation of
0.256. The mean intercept was -0.558 with a standard deviation of 2.262.
Audit 0277 had a mean slope of 0.934 with a standard deviation of 0.306. The
mean intercept was 1.411 with a standard deviation of 5.538. Audit 0877 had
a slope of 1.036 with a standard deviation of 0.196. The intercept was -0.263
with a standard deviation of 1.770. Figure 15 shows a plot of these results.
56
-------
TABLE 31. SO^ SUMMARY STATISTICS ( yg/m3)
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var,
Skewness0
Accuracy"
Cone. 1
a3 b4
Cone. 2
a b
39 38
0.00 0.00
1.24 1.12
0.90 0.85
5.76 3.64
1.26 1.02
101.2 91.1
1.51 0.89
- -
94 92
1.22 1.22
1.96 1.84
1.65 1.59
7.81 5.38
1.43 1.17
72.8 63.0
1.8 1.2
34.9 30.0
Cone. 3
a b
Cone. 4
a b
Cone. 5
a b
Cone. 6
a b
SO^ Audit - 1076
48 46
3.01 3.01
3.26 3.10
3.10 3.10
6.93 6.21
1.32 1.09
40.6 35.3
0.8 0.3
3.1 2.9
99 96
10.2 10.2
10.2 10.1
10.1 10.1
14.0 6.6
1.6 1.2
15.9 11.5
-0.5 0.5
-1.2 -1.2
SO^ Audit - 0277
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var.
Skewness
Accuracy
53 51
4.5 4.5
6.1 5.1
4.6 4.6
41.6 20.6
6.4 2.8
104.1 56.1
4.1 3.6
3.4 2.9
107 105
15.6 15.6
16.2 15.5
15.9 15.9
51.1 23.4
5.6 2.7
34.5 17.1
4.5 -1.8
1.7 1.7
111 110
21.2 21.2
21.3 20.9
21.1 21.0
58.2 13.0
5.4 2.4
25.4 11.5
7.3 -0.5
-0.3 -0.5
56 54
24.9 24.9
24.2 24.2
24.4 24.4
23.2 17.3
3.6 2.9
14.8 11.9
-0.4 -0.8
-1.8 -1.8
SOJ Audit - 0877
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var.
Skewness
Accuracy
34 33
0.0 0.0
1.0 0.9
0.6 0.6
5.0 3.0
1.1 0.8
107.5 93.9
1.8 1.2
-
48 46
2.1 2.1
2.5 2.4
2.4 2.3
6.4 5.0
1.2 1.0
47.9 40.5
1.1 0.5
11.9 9.5
50 49
3.0 3.0
3.2 3.1
3.0 3.0
6.1 5.5
1.2 1.1
37.9 35.7
0.5 0.2
1.5 1.0
51 49
11.1 11.1
10.9 10.8
10.8 10.8
13.7 6.8
1.8 1.2
16.8 11.0
1.7 0.5
-2.7 -2.7
51 50
13.5 13.5
16.5 13.2
13.3 13.3
181.0 18.7
24.0 2.3
145.4 17.8
6.6 -1.0
-1.3 -1.3
50 48
16.2 16.2
16.1 15.7
15.4 15.4
15.9 9.0
2.7 1.6
16.6 10.0
2.5 0.6
-5.0 -5.2
1
Skewness - A statistic that indicates the lack of symmetry in a distribution.
For a Normal distribution this value is near zero.
"Accuracy - Median - True value
" ' "-wT ^ ' ""*•• """ " ~ J
True value
5A11 Data
Outliers Removed
lnn
I \J\J
57
-------
(1-41)
Q.
O
1.15
1.10
1.05
1.00-
0.95
0.90
0.85
Intercept
\
0.50
CO
E
0.00 ?
D_
OJ
O
-------
TABLE 32. SUMMARY OF SO^ ANALYTICAL METHODS USED BY OUTLIER LABS
Method
Total No.
Using Method
; No. Identified
as Outliers
% of Total
as Outliers
SO^ Audit - 1076
Methyl thymol blue-Auto.
Barium Chloride-Man.
Sulfa-ver-Man.
Other
11
20
17
3
1
3
2
1
9%
15%
12%
33%
SO^ Audit - 0277
Sulfa-ver-Man.
14 i
2
14%
SO^ Audit - 0877
Barium Chloride-Man.
Sulfa-ver-Man.
Sulfa-ver-Auto.
Other
23
12
1
6
1
2
1
3
4%
17%
100%
50%
While none of the methods are responsible for a large percentage of the
outlier data, the manual barium chloride and manual Sulfa-ver methods are
responsible for outlier data in at least two of the three audits.
To determine whether a particular analytical method produced biased
results, Table 33 was developed. This table contains the means and standard
deviations of each sample concentration for each method used.
In general, the automated methythymol blue procedure was the most
accurate and precise. There was little difference in the accuracies of
both barium chloride methods and the manual Sulfa-ver method. The automated
barium chloride was more precise than the manual barium chloride. None of
the methods showed a large bias.
Summary
The S04 audit start dates covered in this report ranged from October
1976 to August 1977. The number of participants remained constant at 57.
59
-------
TABLE 33. MEAN AND STANDARD DEVIATION OF SOj RESULTS BY ANALYTICAL METHOD (yg/m3)
Method
Cone. 1
x s
MTB - Man.
MTB - Auto.
Bad - Man.
Bad - Auto.
Sulfa-ver -Man.
Other
True value
5.80
0.84 1.17
0.95 0.98
1.10 1.30
1.09 1.02
0.10 0.13
0.00
Cone. 2
x s
Cone. 3
x s
Cone. 4
x s
Cone. 5
x s
Cone. 6
x s
SOJj Audit - 1076
4.27 0.72
1.35 0.54
2.35 1.82
1.93 1.29
1.43 1.04
2.70 2.14
1.22
5.80
3.38 1.07
3.27 0.88
3.55 2.38
3.12 1.51
1.79 0.50
3.01
10.76 0.38
10.46 1.04
10.24 2.22
10.96 1.42
9.77 0.90
8.66 1.32
10.22
SOJj Audit - 0277
MTB - Auto.
Bad - Man.
Bad - Auto.
Sulfa-ver -Man.
Sulfa-ver -Auto.
Ba Chlor.-Man.
Other
True value
.4.47 0.78
5.77 5.40
3.05 0.07
6.45 5.04
4.50
10.08
9.13 13.27
4.47
16.28 1.04
15.14 2.88
14.94 1.13
14.64 3.67
17.10 1.27
18.22 1.02
20.75 12.16
15.63
20.92 1.94
22.15 8.74
19.37 3.66
20.02 2.26
21.00 8.48
24.12 0.51
22.22 1.92
21.16
24.97 1.14
23.18 3.80
24.64 1.93
24.24 4.97
21.90
32.10
25.10 2.4-9
24.87
SOJj Audit - 0877
MTB - Auto.
Bad - Man.
Bad - Auto.
Sulfa-ver -Man.
Other
True value
0.36 0.49
0.91 1.33
0.00 0.00
0.75 0.97
0.54 0.57
0.00
1.98 0.66
2.44 1.57
1.02 1.39
2.99 1.45
2.28 0.52
2.10
2.92 0.43
3.24 1.56
1.40 0.86
3.29 1.46
3.04 0.29
3.00
10.51 1.07
10.93 3.16
10.60 0.84
10.54 2.56
10.24 0.73
11.10
12.87 1.07
20.60 36.64
13.31 0.98
13.04 2.17
12.51 0.72
13.50
15.55 1.08
16.30 4.98
15.03 1.46
15.34 1.94
15.44 0.47
16.20
-------
Foreign, EPA, State, local, and private laboratories submitted data.
Eight analytical methods were used. The majority of the laboratories
used either the automated methylthymol blue, manual barium chloride, or
manual Sulfa-ver methods.
The overall results showed no bias between the reported and EPA values.
The statistical distributions appeared to be normal. Audit 1076, which had
most of the sample concentrations near the detectable limits of the methods,
showed the greatest inaccuracies.
The automated methythymol blue method exhibited the greatest accuracy
and precision. The barium chloride and Sulfa-ver methods were approximately
equal in accuracy.
The average slopes (Reported vs. EPA) for each audit ranged from 0.931
to 1.05. The intercepts varied from -0.558 to 1.411.
HI-VOL NITRATE
Participant Characteristics
Nitrate audit number 1076 began in October 1976. Out of 56 labora-
tories returning sulfate data, 41 submitted nitrate results. Audit 0277
began in February 1977. Out of 57 sulfate results received, 47 returned
nitrate data. Audit 0877 began in August 1977. Out of 57 laboratories
returning sulfate data, 39 of these laboratories also submitted nitrate
results.
Table 34 indicates the monitoring agency type distribution.
61
-------
TABLE 34. NO^ AGENCY DISTRIBUTION
Foreign
EPA
State
Local | Private
Total
NO^ Audit - 1076
Agencies Requesting Samples
Agencies Returning Data
1
1
1
1
20
20
12
12
7
7
41
41
N0~ Audit - 0277
Agencies Requesting Samples
Agencies Returning Data
2
2
3 21
3 21
13
13
8
8
47
47
NO^ Audit - 0877
Agencies Requesting Samples
Agencies Returning Data
3
3
2
2
16
16
10
10
8
8
39
39
Methods used to analyze the samples were grouped into five broad
categories. Table 35 lists the analytical methods used and the number of
respondents using the method.
TABLE 35. NO" ANALYTICAL METHODS
Method
Cadmium reduction - Man.
Cadmium reduction - Auto.
Hydrazine reduction - Man.
Hydrazine reduction - Auto.
Other
Number Agencies Using Method
Audit 1076
4
15
3
3
16
Audit 0277
6
16
2
5
18
Audit 0877
4
11
1
7
16
It should be noted that some agencies tend to define the analytical methods
used in very general terms. A laboratory reporting the use of the automated
cadmium reduction method as the procedure of choice may have used that
method with various modifications. Thus, Table 35 should be interpreted as
meaning that 15 laboratories in audit 1076 used procedures approximating the
automated cadmium reduction method.
Acceptable Ranges
As described in a previous section, two performance ranges were used as
62
-------
one means of judging performance. The Sample Ranges and Target Ranges for
N03 are listed in Table 36. The ranges apply to sample concentrations in
ascending order.
TABLE 36. NO^ SAMPLE AND TARGET RANGES
Sample Range
Target Range
Cone. 1
+ 5%
+_ 15%
Cone. 2
+ 5%
±15%
Cone. 3
+ 5%
+ 15%
Cone. 4
+ 5%
+ 15%
Cone. 5
+ 5%
+_ 15%
Cone. 6
+ 5%
+ 15%
Neither the Sample Ranges nor the Target Ranges were determined using
the methods described under Statistical Approach. Both ranges were arbitrarily
s,1
set at the values listed in Table 36. As more audits are conducted and more
data become available, the ranges will be refined.
Using the above criteria, a tabulation was made of the number of
agencies reporting results within the ranges. For audit 1076, a total of
10 (24%) agencies reported five or six results within the Target Ranges,
while four (10%) laboratories reported all results outside these ranges.
The corresponding figures for audit 0277 showed 22 (48%) and 7 (15%),
respectively. Audit 0877 showed nine (23%) and four (10%), respectively.
The above results compared with previous audits in that 59% of the
laboratories reported five or six values within the ranges during audit 0476,
and 17% were not able to report any values within the ranges. Figure 16
shows a graphical plot of the above values.
Data Summary
Using the Target Ranges as one means of eliminating nonrepresentative
data, it was decided that any laboratory not reporting at least one value
within the Target Ranges would be considered an outlier. The Target Ranges
are sufficiently broad that, unless a laboratory is totally out of control,
63
-------
at least one value would fall within the ranges; this should occur if the
laboratory is performing similarly to most of the study population. If the
lab is not performing in a similar manner, its data should not be included
in the summaries. Four laboratories met this criterion for outlier rejections
t
in audit 1076, seven in 0277, and four in 0877, and were eliminated from
further data summaries. The values identified as outliers are indicated in the
Appendix by an asterisk (*).
60% -t
50%-
en
-------
Table 37 is a frequency distribution of the percent difference between
the reported and EPA values for each sample concentration. The differences
were calculated by the following formula:
Absolute Percent Difference =
Reported value -,EPA value
EPA value
x 100
The frequency distribution was then constructed and appears below. It should
be noted that the "All Samples" line is not an average of the numbers appear-
ing above, but is the data examined together regardless of the concentration.
TABLE 37.
ABSOLUTE PERCENT DIFFERENCE
Concentration 1
Concentration 2
Concentration 3
Concentration 4
All Samples
Concentration 1
Concentration 2
Concentration 3
Concentration 4
All Samples
Concentration 1
Concentration 2
Concentration 3
Concentration 4
Concentration 5
Concentration 6
All Samples
No.
72
74
37
183
40
76
38
76
230
34
34
35
35
35
173
Min.
1.8
0.4
0.4
0.4
0.0
0.0
0.1
0.2
0.0
0.0
0.0
0.0
0.9
0.6
0.0
10%
7.3
1.8
1.3
1.8
2.1
0.8
0.3
1.1
1.3
2.8
0.8
1.9
0.3
0.9
1.3
30%
NO-
Bl
11.1
"1781
3.1
6.2
NO"
6.7
2.5
3.8
2.6
3.1 "1
NO-
B!
16.7
3.8
3.1
2.0
2.2
3.3
50%
Audit •
_ANK
27.8
7.1
12.1
18.4
Audit •
13.9
7.2
J.7
5.3
7.1
Audit •
.ANK
26.4
9.2
5.0
[ 4.2
4.1
7.7
70%
- 1076
53.7
24.0
29.9
44.4
• 0277
20.6
11.8
10.7
9.9
13.2
- 0877
47.1
17.7
8.5
9.8_j
8.2
22.2
1 90% ]
77.8
76.1
78.6
77.8
36.1
23.0
20.3
20.8
28.4
90.3
69.2
45.0
42.3
43.1
77.8
Max.
203.7
155.0
97.9
203.7
99.5
49.7
38.2
46 a
99.5
266.7
89.2
97.9
90.8
93.0
266.7
Mean
40.2
26.2
28.1
32.1
19.1
10.0
9.8
9.4"
11.4
52.3
21.9
18.2
16.5
16.5 i
25. OJ
Table 37 is very useful for laboratories trying to determine their per-
formance relative to other participants. For example, only 10% of the labs
in Audit 1076 reported results for Concentration 2 that had a percent
difference of 7.4% or less, while 50% of the laboratories reported a percent
65
-------
difference of 27.8% or less for the same concentration. The table also
indicates the average percent difference for all laboratories for all samples
These values are tabulated in Figure 17 along with corresponding values from
previous audits.
Concentration 1 was a blank for Audits 1076 and 0877 and was not in-
cluded in the "All Samples" distribution. Because small concentration
differences result in large percent differences, it was felt that the numbers
would unduly distort the study results.
60% -.
50% -
40% -
g 30%
s-
cu
M-
5 20% -
0
•vf o c\j
O r- O
CO
o
Study Number
Figure 17. Plot of NO^ absolute percent differences
66
-------
All data received (with the exception of previously mentioned data)
were grouped according to concentration. Results for each sample value are
listed in the Appendix in increasing concentration. The summary statistics
which appear in Table 38 and at the top of each listing in the Appendix do
not include the data sets identified as outliers.
Table 38 lists summary statistics based on reported data. The "Outliers
Removed" column contains data on which an outlier test has been applied. All
data that met the second criterion described in Section 3 were removed. It
should be noted that the variation in the number of samples from concentration
to concentration is due to laboratory accidents, damaged samples, and
inclusion of duplicate samples.
The data in Table 38 indicates that the results are normally distributed
(skewness near zero). Certain samples seemed to cause accuracy problems.
In general, however, the accuracy was within 5%.
Each laboratory data set was plotted against the corresponding EPA data
set, and the slope and intercept of the linear regression line were determined.
For Audit 1076 the mean of 37 slopes was 1.393 with a standard deviation
of 1.339. The mean intercept was -0.047 with a standard deviation of 1.098.
Audit 0277 had a mean slope of 1.076 with a standard deviation of 0.242.
The mean intercept was -0.375 with a standard deviation of 1.006. Audit
0877 had a slope of 1.832 with a standard deviation of 0.459. Figure 18
shows a plot of these results.
The mean slopes for audits 1076 and 0877 are much greater than is desired.
However, in each case, the slopes were influenced by a few very large
laboratory slopes. This can be seen in the standard deviation of the slopes.
67
-------
TABLE 38. NO" SUMMARY STATISTICS ( yg/m3)
Cone. 1
a3 b4
Cone. 2
a b
Cone. 3
a b
NO^ Audit
Number
True value
Mean
Median
Rancje
Std. Dev.
Coef. Var.
Skewness0
Accuracy1"
30 30
0.00 0.00
0.23 0.23
0.26 0.26
0.46 0.46
0.16 0.16
70.1 70.1
-0.02 -0.02
-
72 71
0.54 0.54
0.65 0.64
0.64 0.64
1.51 0.97
0.27 0.24
41.1 37.9
0.36 -0.31
18.5 18.5
74 71
3.4 3.4
3.3 3.0
3.4 3.4
8.1 4.6
1.5 1.1
46.0 35.3
1.0 -1.1
1.8 1.8
Cone. 4
a b
Cone. 5
a b
Cone. 6
a b
- 1076
37 37
9.4 9.4
7.2 7.2
8.4 8.4
11.5 11.5
3.3 3.3
45.9 45.9
-0.8 -0.8
-10.8 -10.8
N0~ Audit - 0277
Number
True value
Mean
Median
Rancje
Std. Dev.
Coef. Var.
Skewness
Accuracy
40 38
1.94 1.94
2.27 1.94
2.21 2.16
2.20 1.19
0.43 0.30
19.1 13.6
1.57 0.33
13.9 11.3
76 74
5.2 5.2
5.3 5.2
5.3 5.3
4.0 3.0
0.7 0.6
13.9 12.1
0.4 -0.4
2.5 2.2
38 37
7.4 7.4
7.2 7.3
7.4 7.4
4.3 3.8
1.0 0.8
13.2 11.7
-0.8 -0.5
0.2 0.3
76 74
10.9 10.9
10.7 10.7
11.0 11.0
9.8 6.6
1.6 1.4
14.6 12.7
-0.6 -1.2
0.7 0.7
NO" Audit - 0877
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var.
Skewness
Accuracy
29 28
0.00 0.00
0.43 0.27
0.30 0.20
4.94 1.28
0.91 0.28
212.0 106.0
4.2 1.9
-
34 32
0.72 0.72
0.96 0.86
0.86 0.86
2.57 2.13
', 0.55 0.39
57.7 45.3
1.4 0.7
20.1 19.4
34 32
1.30 1.30
1.34 1.41
1.32 1.33
2.08 1.94
0.45 0.35
33.8 24.8
-0.9 -0.2
1.9 2.3
35 33
4.80 4.80
4.14 4.38
4.65 4.68
5.67 5.30
1.49 1.14
35.9 2611
-1.7 -2.0
-3.1 -2.5
35 32
6.00 6.00
5.43 5.58
5.82 5.86
9.75 5,45
1.78 1.07
32.7 19.2
-1.0 -2.3
-3.0 -2.2
35 33
7.80 7.80
6.47 6.92
7.44 7.48
7.42 6.15
2.00 1.40
30.5 20.2
-1.8 -2.0
-4.6 -4.1
1
Skewness - A statistic that indicates the lack of symmetry in a distribution.
For a Normal distribution this value is near zero.
>
'Accuracy - Median - True value lnn
True value x IUU
*A11 Data
Outliers Removed
68
-------
CO
-------
TABLE 39. SUMMARY OF NO" ANALYTICAL METHODS USED BY OUTLIER LABS
Method
Total No.
Using Method
Iad. Reduction-Man.
ad,. Reduction-Auto.
ther
4
15
16
No. Identified
as Outliers
% of Total
as Outliers
NO" Audit - 1076
1
2
1
25%
13%
6%
NO^ Audit - 0277
ad. Reduction-Man.
ad. Reduction-Auto.
ther
6
16
18
2
2
3
33%
12%
17%
NO^ Audit - 0877
Cad. Reduction-Auto.
Hyd. Reduction-Man.
Hyd. Reduction-Auto.
Other
11
1
7
16
1
1
1
1
9%
100%
14%
6%
To determine whether a particular analytical method produced biased
results, Table 40 was developed. This table contains the means and standard
deviations of each sample concentration for each method used.
The manual hydrazine reduction appears to be the most precise. The
automated hydrazine reduction method was more precise than the automated
cadmium reduction method, but less precise than the manual hydrazine procedure,
The automated hydrazine reduction is also the most accurate, with the
automated cadmium reduction next. As pointed out earlier, however, the
important point is that most laboratories are not using any of the defined
methods.
Summary
The N03 audit start dates covered in this report ranged from October
1976 to August 1977. The number of participants submitting ranged from 39
to 47. Foreign, EPA, State, local, and private laboratories returned results.
70
-------
TABLE 40. MEAN AND STANDARD DEVIATION OF NO" RESULTS BY ANALYTICAL METHOD
Method
Cone. 1
X
s
Cone. 2
X
s
Cone. 3
X
s
Cone. 4
X
s
Cone. 5
x
s
NO" Audit - 1076
Cad. Red. -Man.
Cad. , Red. -Auto.
Hyd. Red. -Man.
Hyd. Red. -Auto.
Other
True value
0.23
0.20
0.30
0.16
0.21
0.18
0.17
0.20
0.10
0.19
0.00
0.81
0.56
0.60
0.60
0.71
U.46
0.28
U.lb
0.16
0.22
0.54
2.86
2.77
2.95
3.49
3.81
0.99
1.37
O.b9
0.74
1.78
3.39
7.85
6.00
7.47
9./B
7.44
2.85
3.52
2.04
i./i
3.46
L 9.42
Cone. 6
x
s
NO^ Audit - 0277
Cad. Red. -Man.
Cad. Red. -Auto.
Hyd. Red. -Man.
Hyd. Red. -Auto
Other
True value
2.54
2.32
1.90
2.02
2.27
0.89
0.26
0.02
0.25
0.45
5.79
5.64
3.97
5.11
5.03
0.92
0.54
0.15
0.35
0.65
1.94 1- 5.17
7.72
7.36
5.37
7.59
7.16
1.01
0.73
0.38
0.43
1.04
7 38
10.59
11.32
6.95
10.92
10.59
0.98
1.65
0.24
0.57
1.21
10.92
- - -
j
s
t
t
*
\
NO" Audit -0877
Cad. Red. -Man.
Cad. Red. -Auto.
Hyd. Red. -Auto.
Other
[True value
0.51
0.14
0.23
0.51
0.62
0.22
0.15
1.24
0.00
1.62
0.81
1.00
0.81
0.95
0.66
0.26
0.36
0.72
1.53
1.29
1.43
0.45
0.49
0.11
1.2810.52
1.30
3.93
4.32
4.89
3.78
1.55
1.11
0.26
1.90
4.80
5.13
5.49
6.73
4.96
1.15
1.50
1.77
1.96
6.00
6.76
6.57
7.69
6.06
7.
1.26
2.04 !
0.32 I
2.42
80
Four analytical methods were used (plus a category, "Other"). Most of
the samples were analyzed by methods in the "Other" category.
The overall results showed no bias between the reported and EPA values.
The statistical distribution of the results appeared to be normal.
The automated hydrazine reduction method showed the best accuracy,
while the manual hydrazine reduction procedure had the best precision.
The average slopes (Reported vs. EPA) for each audit ranged from 1.08
to 1.83. The intercept varied from -0.047 to -0.375.
HI-VOL LEAD
Participant Characteristics
Lead audit number 0877 began in August 1977. Out of 89 sample sets
71
-------
requested by participants, 67 sets of data were returned for a response
rate of 75%. Table 41 indicates the monitoring agency type distribution
TABLE 41. Pb AGENCY DISTRIBUTION
Agencies Requesting Samples
Agencies Returning Data
Foreign
0
0
EPA
P
6
5
State
b Audit
45
33
Local
- 0877
31
25
Private
7
4
i Total
89
67
Methods used to analyze the samples were grouped into four broad
categories. Results were received from laboratories using two methods; these
are listed in Table 42.
TABLE 42. Pb ANALYTICAL METHODS
Method
Atomic Absorption
X-Ray Floresence
Number Agencies Using Method
66
1
It should be noted that some agencies tend to define the analytical
methods used in very general terms. A laboratory reporting the use of the
atomic absorption method procedure as the method of choice may have used
that procedure with various modifications. Thus, Table 42 should be
interpreted as meaning that 66 laboratories used procedures approximating the
atomic absorption procedure.
A tabulation was made of the methods used to extract the material
from the filter. Table 43 shows the various methods in use.
TABLE 43. Pb EXTRACTION PROCEDURES
Method
Hot Acid Extraction
Cold Acid Extraction
Ultrasoni cation
Other
NumDer Agencies Using
56
3
1
7
Method
72
-------
Acceptable Ranges
As described in a previous section, two performance ranges were used
as one means of judging performance. The Sample Ranges and Target Ranges
are listed in Table 44. The ranges apply to sample concentrations in
ascending order.
TABLE 44. Pb SAMPLE AND TARGET RANGES
Sample Range
Target Range
Cone. 1
± 5%
± 10%
^Conc. 2
± 5%
± 10%
Cone. 3
± 5%
± 10%
Cone. 4
± 5%
± 10%
Cone. 5
± 5%
± 10%
Cone. 6
+ 5%
± 10%
Neither the Sample Ranges nor the Target Ranges were determined using
the methods described under Statistical Approach. Both ranges were arbitrarily
set at the values listed in Table 44. As more audits are conducted and more
data become Available, the ranges will be refined.
Using the above criteria, a tabulation was made of the number of
agencies reporting results within the ranges. Thirty-two agencies (51%)
reported five or six samples within the Target Ranges. Five laboratories (8%)
reported all results outside the Target Ranges.
Data Summary
Using the Target Ranges as one means of eliminating nonrepresentative
data, it was decided that any laboratory not reporting at least one value
within the Target Ranges would be considered an outlier. The Target Ranges
are sufficiently broad that, unless a laboratory is totally out of control,
at least one value would fall within the ranges; this should occur if the
laboratory is performing similarly to most of the study population. If the
lab is not performing in a similar manner, its data should not be included
in the summaries. Five laboratories met this criterion for outlier rejection
73
-------
and were eliminated from further data summaries. The values identifiec as
outliers are indicated in the Appendix by an asterisk (*).
Table 45 is a frequency distribution of the percent difference between
the reported and EPA values for each sample concentration. The differences
were calculated by the following formula:
Absolute Percent Difference =
Reported value - EPA value
EPA value
x 100
The frequency distribution was then constructed and appears below. It should
be noted that the "All Samples" line is not an average of the number appear-
ing above it, but is the distribution resulting when all data is examined
together regardless of concentration. ;'
TABLE 45. Pb ABSOLUTE PERCENT DIFFERENCE J
I No.
Min.
10%
30%
50%
-70% I 90%
Max.
Mean
Pb Audit - 0877 !
i
Concentration 1
Concentration 2
Concentration 3
Concentration 4
Concentration 5
Concentration 6
All Samples
51
51
51
62
62
58
284
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.6
0.6
0.7
0.8
0.2
0.5
0.7
3.9
3.9
2.5
2.0
2.0
3.0
2.8
5.6
5.6
5.4
4.6
4.9
4.7
5.4
10.0
10.0
11.1
9.1
7.9
7.0
9.1
45.0
45.0
23.5
19.6
13.1
14.2
23.5
327.8
327.8
250.6
95.0
91.7
94.6
327.8
20.0
20.0
16.0
10.2
9.5
9.2
12.6
Table 45 is very useful for laboratories trying to determine their per-
formance relative to other participants. For example, only 10% of the labs
reporting results for Concentration 2 had a percent difference of 0.6% or
less, while 50% of the laboratories reported a percent difference of 5.6% or
less for the same concentration. The Table also indicates that the average
percent difference for all laboratories for all samples (except Concentration
1) was 12.6%.
Concentration 1 was a blank and was not included in the "All Samples"
distribution. Because small concentration differences result in large
74
-------
percent differences, it was felt that the numbers would unduly distort the
study results.
All data received (with the exception of the previously omitted data)
were grouped according to concentration. Results for each sample value are
listed in the Appendix in increasing concentration. The summary statistics
which appear in Table 46 and at the top of each listing in the Appendix do
not include the five data sets identified as outliers.
Table 46 lists summary statistics based on reported data. The "Outliers
Removed" column contains data on which an outlier test has been applied. All
data that met the second criterion mentioned in Section 3 were removed. It
should be noted that the variation in the number of samples from concentration
to concentration is due to laboratory accidents, damaged samples, and inclusion
of duplicate samples.
With the exception of Concentration 1, the EPA determined "True value",
and the mean and median of the study results agree well. This is indicated
by the "Accuracy". The greatest difference is approximately 3.5%. The
relatively poor agreement between the EPA and reported values for concentra-
tion 1 is an indication of the minimum detectable limits of the various
analytical methods used during the audit.
Each laboratory data set was plotted against the corresponding EPA
data set, and the slope and intercept from the linear regression were
determined. The mean of 58 slopes was 1.138 with a standard deviation of
0.752. The mean intercept was 0.242 with a standard deviation of 0.898.
75
-------
TABLE 46. Pb SUMMARY STATISTICS (yg/nT)
Number
True value
Mean
Median
Range
Std. Dev.
Coef. Var.
Skewness0
•Accuracy"
Cone
a3
28
0.00
0.04
0.03
0.14
0.03
80.8
1.3
-
. 1
b4
27
0.00
0.04
0.03
0.09
0.03
71.2
0.8
— I
Cone
a
51
1.8
1.8
1.8
7.6
1.0
54.0
4.6
-2.8
. 2
b
50
1.8
1.6
1.8
2.0
0.4
25.9
-2.1
-2.8
Cone
a
51
4.0
4.0
4.0
14.0
1.7
41.9
4.1
-1.7
:. 3
b
Pb Au<
50
4.0
3.8
4.0
5.3
0.8
21.8
-2.6
-2.0
Con
a
Jit -
62
6.6
6.1
6.4
8.0
1.2
19.8
-2.9
-2.9
c. 4
b
0877
60
6.6
6.3
6.4
4.5
0.7
11.2
-1.0
-2.6
Cone
a
62
8.4
7.8
8.2
8.8
1.6
20.6
-3.3
-2.5
:. 5
b
59
8.4
8.2
8.3
3.6
0.6
7.3
-0.7
-1.6
Cone
a
58
11.4
10.6
11.0
12.2
2UO
18.4
-3.4
-3.7
. 6 ""
b "^
56
11.4 ~
11.0
11.0
5.5
1.0
8.8
-1.1
-3.5
1
Skewness - A statistic that indicates the lack of symmetry in a distribution.
For a Normal distribution this value is near zero.
"Accuracy - Median - True value
True value
x 100
°A11 Data
4
Outliers Removed
Analytical Method Summary
A check was made to determine whether any relationship existed between
the five laboratories that submitted data that were omitted from the summaries
and the analytical method employed (i.e., was any one method responsible for
most of the data outliers). Table 47 resulted from this check.
TABLE 47. SUMMARY OF Pb ANALYTICAL METHODS USED BY OUTLIER LABS
Method
Total Number
Using Method
Number Identified
as Outliers
% of Total
as Outliers
Pb Audit - 0877
Atomic Absorption
X-Ray Floresence
66
1
5
-
Qo/
O/o
rw
U/o
76
-------
As can be seen, all of the outlier labs used the atomic absorption method.
However, only one lab used another procedure.
To determine whether a particular analytical method produced biased
results, Table 48 was developed. This table contains the means and standard
deviations of each sample concentration for each method used.
TABLE 48. MEAN AND STANDARD DEVIATION OF Pb RESULTS BY ANALYTICAL METHOD (yg/m3)
Method
Cone. 1
x s
Cone. 2
x s
Cone. 3
x s
Cone. 4
x s
Cone. 5
x s
Cone. 6
x s
Pb Audit - 0877
AA
XRF _,
True value
0.02 0.03
0.00 -
0.00
1.77 0.95
0.85 -
1.80
3.93 1.76
3.10 -
4.05
6.16 1.18
3.80 -
6.60
7.81 1.62
8.40 -
8.40
10.62 1.96
12.60 -
11.40
The AA procedure produced the most accurate results at the lower concen-
trations. The XRF method had the least accurate results at the lower concen-
trations. The AA method, overall, produced the most accurate results.
Summary
The Pb Audit Survey 0877 began in August 1977. Eighty-nine laboratories
requested samples, of which 67 returned results for a response rate of 75%.
EPA, State, local, and private laboratories submitted data.
Two analytical methods were used. Ninety-nine percent of the laboratories
used the atomic absorption procedure. One laboratory used the X-ray florescence
method.
The overall agreement between the reported values and the EPA values
was good. No bias was evident, and .the results were normally.distributed.
The average percent difference between the EPA results and reported results
was 12.6%. This figure does not include the blank data.
77
-------
HI-VOL FLOW RATE
Participant Characteristics
Hi-Vol flow rate audit numbers 0576 and 0577 began in May 1976 and
May 1977, respectively. Out of 175 agencies requesting to participate in
audit number 0576, 146 responded with data, for a response rate of 83%. A
total of 1,021 Hi-Vol units were tested. Out of 198 requests to participate
in audit 0577, 167 agencies responded with data, for a response rate of 85%.
A total of 1,016 Hi-Vol samplers were checked.
Table 49 indicates the monitoring agency type distribution.
TABLE 49. HI-Vol F10W RATE AGENCY DISTRIBUTION
Foreign j EPA
State
Local
Private
Total
Hi-Vol Audit - 0576
Agencies Requesting ReF
Agencies Returning Data
1
1
11
7
64
53
98
84
1
1
175
146
Hi-Vol Audit - 0577
Agencies Requesting ReF
Agencies Returning Data
2
2
9
6
64
57
116
97
7
5
198
167
Methods used to measure the flow rate were grouped into three categories:
rotameters, pressure transducers, and other. Table 50 lists the measurement
methods and the number of units using the method.
TABLE 50. HI-VOL FLOW RATE MEASUREMENT METHODS
Rotameter
Pressure
Transducer
Other
Number Units
Audit 0576
593
344
84
Using Method
Audit 0577
445
395
176
Table 50 indicates that a trend may be developing to move away from
the rotameter measurement method to other more precise and accurate methods
78
-------
Acceptable Ranges
As described in a previous section, two ranges were used as one means of
judging performance. The Sample Ranges and Target Ranges were calculated as
+5% and +9%, respectively.
Using this criterion, a tabulation was made of the number of samplers
reporting results within the ranges. For Study 0576, a total of 526
samplers (52%) reported four or five readings within the target ranges, while
168 units (16%) reported four or five readings outside the Target Ranges.
Fifty-six agencies accounted for 108 units that reported all values outside
the Target Ranges. The corresponding results for Study 0577 showed that
680 units (60%) reported four or five readings within the Target Ranges,
while 101 units (10%) reported four or five readings outside the Target
Ranges. Thirty-two agencies accounted for 74 units that reported all values
outside the Target Ranges.
Data Summary
Table 51 is a frequency distribution of the percent difference between
the reported and EPA values for each measurement pair. The differences were
calculated using the following formula:
n 4. n-.c.e EPA value - Reported value .
Percent Deference = EPA value
An iterative routine was used to eliminate outliers while constructing the
distribution. During each pass of the data, a check was performed to determine
which reported values met the second criterion mentioned in Section 3. These
values were removed, and the procedure repeated until no outliers were
identified. This is the distribution that is titled "Outliers Removed" in
Table 51. A total of six passes were required in audit 0576 to remove the
outliers, while eight passes were required in audit 0577 to achieve the same
79
-------
results. As can be seen, 164 values were removed from audit 0576, while
204 values were removed from Study 0577.
TABLE 51. HI-VOL FLOW RATE PERCENT DIFFERENCES
No.
Min.
10%
30%
50%
70%
90%
Max.
[Mean
Hi-Vol Audit - 0576
All Values
Outliers Removed
5014
4850
-166
-27
-12
-11
-5
-5
-1
-1
2
2
9
8
68
24
-1
-1
Hi-Vol Audit - 0577
All Values
Outliers Removed
4934
4730
-504
-23
-11
-10
-4
-4
-0.9
-0.8
2
2
6
6
72
20
-2
-1
As can be seen, removal of the outliers had little effect on the main body
of the Table.
Table 51 is useful when evaluating the overall results of the audits.
Excluding outliers, of all the observations reported for Audit 0577, 80% were
within +10% of the EPA calculated value. The overall average difference was
-2%.
Histograms were constructed of all reported values for both audits and
appear in Figures 19 and 20. As can be seen, there are slight negative
bias (reported values higher than EPA). The distributions appear to be
normal, with slight skewing to the negative side. However, considering the
diversity of the measurement sources, the results appear well behaved and
show excellent precision.
All reported value pairs for each audit were summarized using a linear
regression equation of the form,
y = mx + b
where:
y = reported value
80
-------
oo
.
H5.0
38.1!
31.7
31 .2
27.8(
17.5!
10.6'
1 7.2"! "
3.8
.3
-3.0
-9.9
-13.3
-16.7
-23.6
[- 2 7 • 0
J-30.H
-3-J t^
-37.3
-HO . /
' - M 7 . 6
-51 .0
-57.9
-61. J
-6H.8
-71 ,6
-7 5 . r
-78.5
-BI .V
-85. H
-68. 8
OR
• r
• f
. ;
-.'3
- } 9 9 H
-f 1
•! 1
-f 1
-f 3
•f3
-•6
-f 1
- f **
- J 94
- J98
- !999
-j 99999996
-f 979999999999993
-! 9999999999999999999992
i
>
-J999999999999999 9 99999999999999999999999999999999999999999999999999999999999999999996
-•9999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999998
-f99999999999999999999999999999999999999999999999999999999999999999 99 99999999999999 99 9999999999999999999999
-.'999999999999999999999999999999999999999999999999999999999999999995
-f 999999999999999999999999999999999999991
- '99999999999996
-•779979995
-f 99992
V ' *
- ! y ¥3
-?993
-f 6
- ! 5
- ?6
-f2
-» J
.-!•*
F-ffI
-fZ
. i
-!1
i-f
-r
-."*
EACH X REPRESENTS 9 VALUES
Figure 19. Hi-Vol flow rate histogram - 1976.
-------
00-
ro
0)
o
£
CD
-!
53.3
50.3
47.4 I-'
44.4
41.4
3fc.5
35.5 I-*
32,5
29.6
26.6
23.6
20.7
17.7
14.7
11.7
8.8
5.8
2.8
-.0
-3.0
-6.0
-8.9
-11.9
-14.9
-17.9
-20.8
-23.8
-26.6
-29.7
-32.7
-35.7 -!4
-38.6 ~!3
-41.6 -!4
-44.6 -!4
-47.5 -!2
-50.5 -!4
-53.5 -!1
-56.5 -!1
-59.4 -!3
-62.4 -!1
-65.4 -!3
-68.3
-71.3 I-'
-74.3
-77.2
OR
EACH X REPRESENTS
!X5
!X8 _ .
!XXX4 v
IXXXX2
!XXXXX&
•XXXXXXXXXXXXXXX4 . . ._ _ __ . . .... _ . .. . _.
!XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX3
fxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx?
•xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
fxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxe
IXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX1
•XXXXXXXXXXXXXXXXXXXXXXXXXX8 . .. .._
fxxxxxxxxxxxxxxxxxxxs
!XXXXXXXXXXX5
!xxxxxx8
!xxxx6 ... . ... .. .
!XXX
!XX3
!X4
!X2
10VALUES
Figure 20. Hi-Vol flow rate histogram - 1977.
-------
x = EPA value
The resulting equation for 0576 and 0577, respectively were:
y = 0.920 x +3.669 (0576)
y = 0.936 x +3.204 (0577)
The equations appear to indicate, because of the large intercept, a larger
bias than actually exists. Since the equation is valid only over the range
for which the test was conducted (30-60 ft3/min), the influence of the inter-
cept is not as great as is indicated.
Measurement Method Summary
To determine whether a measurement method produced biased results and
to gain an indication of the accuracy and precision of the methods in use,
the numbers in Table 51 were separated by method. Table 52 resulted.
TABLE 52. SUMMARY OF HI-VOL FLOW RATE MEASUREMENT METHODS
Rotameter
P. Transducer
Rotameter
P. Transducer
f No.
2785
1665
2037
1870
Min.
-29
-21
-26
-21
10%
-13
-8
-12
-9
30%
Hi-Vo'
-6
-2
Hi-Vo'
-5
-2
50%
1 Audit
-2
1
I Audit
-2
0.1
70%
- 0576
2
4
- 0577
1
3
90%
8
8
5
7
Max.
25
22
21
20
Mean!
j
-2.1
0.5
-2.8
-0.2
The numbers generated in the above table are the result of several iterations
similar to the ones needed to derive the "Outliers Removed" lines of Table 51
It is obvious from Table 52 that the pressure transducer method of
measurement is more accurate than the rotameter method. The table also shows
that the apparent bias revealed in Table 51 is due almost entirely to the
negative bias of the rotameter readings.
83
-------
Forty-five percent of the units using the rotameter were able to report
four or five values within the Target Ranges for audit 0576, while 60% of the
units using the pressure transducer reported four or five values within the
Target Ranges. Corresponding figures for Study 0577 showed 51% and 63%,
respectively.
Linear regression equations for each method were derived as described
in the preceding sections The resulting equations appear below:
Year Equations Measurement Method
1976 y = 0.904 + 4.734 rotameter
1976 y = 0.924 + 2.568 pressure transducer
1977 y = 0.892 + 5.361 rotameter
1977 y = 0.951 + 2.161 pressure transducer
As can be seen, the pressure transducer measurement pairs result in a slope
closer to unity and an intercept closer to zero than do the rotameter pairs.
Summary
The Hi-Vol Flow Rate audits covered in this report started in May 1976
and May 1977. Number of participants ranged from 146 in 1976 to 167 in 1977.
Foreign, EPA, State, local, and private laboratories submitted data.
The results from both audits showed a slight bias between EPA and
reported results. The distributions were skewed slightly. Considering the
number of units checked and the number of participating personnel, the bias
and skewing are insignificant.
The slope of the equation representing all values for audit 0576 was
0.920 with an intercept of 3.669. The corresponding values for audit 0577
were 0.936 and 3.204.
84
-------
SECTION 6
REFERENCES
1. Code of Federal Regulations (40 CFR) 50.11 Appendix A, pp. 5-11.
2. Federal Register, Vol. 38, No. 110, June 8, 1973, pp. 15175-15176.
3. Bromberg, S., Bennett, B., and Lampe, R. "Summary of Audit Performance,
Measurement of S0?, N0?, Sulfate, Nitrate - 1976. EPA 600/4-78-004,
January 1978.
85
-------
INTER-LA MORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 1
1176
128
N
TRUE-VALUE
MEAN
MEDIAN
DATA IN ASCENDING ORDFR
IT. 14
17.
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE :
STD. OE1V.
CO€F. VAR
4S.48-
48.98
7.flG
40 .8 4
C.I.CUPPERI
C.I.1CLOUE
SKEWNESS
ACCURACY
18.35
15*93
.52
-8.85
07®
.19
1.17
1.19
2.62^
2.6241
2.65
6.00
6.08
7.00
7.50
7.36
7.85
7.85
7.85
8.34
8.34
9.20
9.32*
9.52
11*00
1 1;. oo
11-10
11-11
11:. 9B
12.17.
12.73
13.10.
13.33
13.39
13.53
14.60
14.50
14.B2
14.65
14.93
15.00
15.32
15.33
15.40
15,63
15.72
15.72
15.72
15.72
15.72
15.72
15.76
15.80
16.30
16.33
16.52
16.67
16.67
16.57
16.67
16.. 70
16.90
1T.OO
17.00
1T.OO
17*02
IT. 10
IT. 17
17.37
17*50
17*67
17*81
1 3*t}0
13*00
13*00
18*22
18.34
18.34
18.34
18.34
18.34
18*34
13*43
18.53
18*67
13.S7
18*70
13*75
Id* 80
18.90
18.92
18.92
19.00
' 19.50
19.60
19.61
19.80
20.00
20 .00
20.00
20 .34
20.60
20 .67
20.83
20.96
20 .96
20 .96
20.96
20.96
20.96
21.17
22.00
22.20
22.20
22.27
22.50
23.00*
23.00.
23.30
21.5©
23.64
24. 16
25.67
•2li*;04t..
2€. 17
26.20
27.00
27.07
27.30
27.80
28.48©
29.82
35.70®
36.67
40.75^
41.20
45.67
52.40
33.87^
206.670
275.10®
464.560
86
-------
INTER-LA BIORATORT STUDY
POLLUTANT - S02
SAMPLE NUM3ER - 2
UNITS - WICROCRAHS PEP CUBIC METER
N 133
TRUE-VALUE 13.03
MEAN 11* 01
MEDIAN 44.75
DATA IN ASCENDING
.060
.130
.160
2.620
3.300
4.110
5.240
10*25
16*67
?1*00
23.50
25.18
27.77
28.82
29*00
29*00
29*50
31.11
31.11
3 3'. 30
33.33
31.06
35*21
36*00
36.10
36.68
36,70
36*77
37.09
38-. 00
3 8'. 05
ORDER
38,33
38.33
38,38
38, '38
38,67
39.22
39,30
39,30
39,30
39,30
39.50
39.93
10.31
10.83
11.00
11.00
11.13
41.66
11.86
11,75
11,83
41.92
11.92
41.92
42.00
42,00
42.33
12.72
42,82
13,30
43,33
RANSE
VARIANCE :
STD. 0E1V,
COfF, V4R.
68.03
83 .17
9.12
20.72
13*10
13.68
13*70
13*99
11.00
11*00
11*33
11*51
11*51
11.51
11*51
11;. 61
11*75
15*13
15.27
15*33
15*33
15*38
15* SO
15*63
15*70
15*80
16*09
16*^7
16.S7
16*70
IE* 83
17*00
17*00
171*00
17*16
C.I.f UPPER! 45 ,56
C.I.ftOWE RS 42,46
SKE VINES S -.32
ACCU&8EY 1,00
17.16
47.16
17,16
47.30
47.31
47.80
47.90
48,33
48,97
49.00
49 .DO
49,60
49,78
49,78
49,83
50.00
SO. 00
50.00
50.00
50.00
50.08
50.17
50.27
50.57
50.71
50.83
51.00
51.00
51.33
51.B7
52.08
52.25
52.40
5?. 50
53.33
53.83
54,16
54.60
54.75
55.02
55.02
57.57
57.64
57,80
59.33
60.760
61. 000
66.820
88.60
73.33
73.480
100.600
119.300
193.330
235.80 •
473.330
696.84*
749.320
87
-------
INTER-L4BORATORY STUDY
1176
POLLUTANT - S02
SAWPLE NUMBER -
N
TRUE-V«LU£
MEAN
MEDIAN
133
73.37
76.97
77.73
DATA IN ASCENDING ORDER
UNITS - HICROGRAMS PER CUBIC METER
RANGE
VARIANCE
STD. 0E!V.
COEF. VAR.
127.53
175 .02
13 .'2 3
17.19
C.I.fUPPERJ 79.22
C.I.CLOWE Rl 7*.72
SKEWNESS -.76
ACCURACY 5.94
,14*
IS*
5.24*
5.50*
14:. 67*
20.96*
29.00®
39.30
41i. 10
12.31
54f.74*
5 5». 50
551.83
5 8f. 10
50'. 33
50.70
52.50
52.93
52.94
5 5 . 50
5 5 . 67
56.00
56.67
57.50
57.80
58.00
5S.12
58.12
58.12
58.1?
58.48
69.06
69.50
70.07
70.10.
70.30
70.7%
71.00
71.83
72.10:
72.33
73.00
73.36
73.36
73.36
73.^0
73.60
73.79
73.79
73.81
7*. 00
7«l,65
75.00
75.00
75.00
75.10
75.10:
75.29
75.35
75.98
76.00 .
76.10 .
76.33 '
76.45 «
7&.S7
76.60 .
76.70
77.00 ;
77.12 .
77^17
77.30
77.37
77.73
77.77 •
77.92 .
78.00
78.15 (
78.20
78.38
78.44 !
78.60 :
73.00
79.06
79.33
79.88
8C.OD
80.10
80.20
80.33
80.50
80.57
80.70
80.74
80.74
81.22
81.22
81.22
81.22
81 .E7
81.79
82.59
82.82
83 .30
83.33
83.40
83.67
83.84
83.84
83.54
84.00
84.00
84.00
84.00
84 .07
84.37
84.83
85.00
85.00
85.50
85.56
86 .46
86.46
8S.46
86.46
86.67
87.00
83.90
88.90
89.00
89.08
89.08
89.10
91.33
31.87
99.5S
99.89
103.33
104.00^
104.170
11S.149
117.21*
119.18*
121.50
132.77
155.00*
189.64*
351.67*
524.00
803.33*
1000.00*
-------
INTER-LA8CRATORY STUDY
POLLUTANT - SQ2
SAMPLE NUMBER - *
N
TRUE-VALUE
MEAN
MEDIAN
132
102.97
115.60
111.82
DAT* IN ASCENDING ORDER
117S
UNITS - MICRQGRAHS PER CUBIC METER
RANGE
VARIANCE:
STD» 0EV.
COCF. VftR,
992.13
6390.11
79 .9H
67*40
C.I.fUPPERI 132.24
C.I.fLOHE Rl 10% .96
SKEWNESS 10.15
ACCURACY 8.59
• 1S0
. 29 0
.360
7.860
9.900
48.000
49.78
571.640
72.60
78.20
78*600
79.00
79.16
83.00
91.60
91.70
91.87
92.50
94.3?
94.57
95.50
96.94
97.87
98.10
99.17
99.56
99.90
99.99
100.48
100.77
101.74
103.48
104.23
104.49
104.80
105.00
105.00:
105.00
105.17:
105.73
106.46
10S.70
106.82
106.88
107.00
107.14:
1O7.42
107.57
108.00
108.06
108.10
108.50
109.00
109.00
109.20
109.33
109.57
109.77
109.80
110.00
110.00
110.00
110.04
110.04
110.04
110.04
110.04
110.04
111.00
111.22
111.3B
111.48
111.80
111.84
111.84
112.51
112.65
112.65
113:. 00
113.. 00
113.08
113*54
113.67
113.90
113.97
114.00
114.00
114.58
114.93
115.02
115.28
115.28
115.23
115.33
116.07
116.70
116.70
116.83
116.87
117.33
117.86
118.33
118.33
118.65
118.73
118.80
119.00
120.00
120.00
120.00
120.52
120.83
121.50
121.83
122.00
122.46
122.60
122.93
123.14
173.14
123.69
124.00
124.00
124-40
124.67
125.40
125.76
126.70
127.75
123.38
129.25
13C.210
133.33
136.63
139.370
141.50
144.10
147.000
152.14
163.210
176.40
134.900
189.070
241.04
241.650
505.000
602.60
999.99
1000.000
1000.000
89
-------
INTER-LABORATORY STUDY
1176
POLL'JTflNT - S02
SAMPLE NUMBER -
N
TRUE-VALUE
MEAN
MEDIAN
129
:. 03
156.56
158.33
UNITS - WICROGRANS PER CUBIC METER
RANGE
VARIANCE
STO. 0E!V.
COEF. VAR.
215.96
516.21
23 .37
C.I.f UPPER! 160.59
C.I .if LO HE RJ 152.52
SKEWNESS -1.30
ACCURACY 6.21
DATA IN ASCENDING ORDER
.23*
.1?*
.91
11.30*
70.71 .
91.00*
97.11*
102.18*
106.51*
123.08
128.38
129.17
131.00
131.67
131.77
132.50
133.20
133.62
135.88
137.00
138.17
133.67
138.70
139.61
115.30
116.26
1161.66
IIS.72
116.72
116.72
118.00
118.61
119.31
119.10
119.83
150.00
150.00
150.00
150.50
150.57
150.83
151.00
151.00
151.17
151.96
152.17
152.50
152.51
152.56
153.00
157.29
153.33
151.00
151.00
151.00
151.00
151.10
151.58
155.10
15 5'. S3 -
1551.56
156.25 (
15&.60
157'. 20
15 7'. 20
157.20
157.20
157.20
158.00
158.33
158.55 t
158.58
153.92
159.30
159.37 .
153.53
159.78
159.82
159.82
159.83
159.89
l&G.OO
160.00
160.50
160.83 '
160.85 !
161.90
162.11
162.11
162.56
163.00
167.22
163.83
161.29
161.36
161.89
165.00
165.06
165.06
165.06
165.26
165.26
166.00
1S6.31
166.67
166.70
166 .TO
167.33
167.96
168.52
168.67
168 .'BO
168.90
170.00
170.30
170.97
171.10
172.00
17?.92
173.00
173.76
175.00
175.00
17*5,51
175.67
176.12
176.67
175.85
178.16
180.00
188.15*
190.77
190.*7*
202.00*
213.17
228.98
213.73*
211.50*
217.56*
250.90
331.80*
635.00*
1000.00
1000.00*
1000.00*
90
-------
INTER-LABORATORY STUDY
POLLUTANT - S02
SAMPLE NUMBER - 1
N
TRUE-VALUE
MEAN
MEDIAN
21*30
?€.18
18.30
DATA IN ASCENDING ORDER
477
UNITS - HICR*)GRA*S PER CUBIC METER
RANCE
STD. OEiV.
C0ۤ* VAR
207.20
310.1?
17 .84
C,I.f UPPER* 23.20
C.I.ICLOWE R> 17,16
SKEWNESS 8.77
ACCURACY -15.44
1.10
2.S2
3.78
4.70
4.93
7.830
8.34
8.75
9.76
101.31
10*48
10;. 48
11*10
11*80
11*89
11*90
12*57
13* 10
131.37
13*87
13*99
14*00
14*00
14*00
14!. 28
1 4* 33
14.44
14.55
14.82
14.83
14.94
15.00
15.00
15.72
15.72
15.82
16.00
16.04
16.10
16.131
16.27
16,33
16.43
16.55
16*67
16.S7
16.67
16.70
16.70
16.90
17.00
17.00
17.14
17.30
17.35
IT. 39
17*80
17*83
IT. 90
18*00
18*00
IS. CO
18*00
18.17
18*25
18.30
la. 30
13*30
18*30
ia*30
1 8* 30
18*31
18.34
18*34
18.34
18.54
18*81
id. 32
19.00
19.03
19*09
19*40
19.50
19.64
19.80
19.80
19.92
20.00
20.04
20.30
20.34
20 .SO
20.80
20.30
20.96
20.96
20.96
21.18
21.33
21.60
22.00
22.00
22.00
22.13
22.72
22.49
22.50
22,78
23.19
23 .20
23.56
23.
24.10
25.00
25.00
25.98
2S.OO
26.20
26.20
26.20
26.90
27.52
27.60
28.82*
30.63
31.44
34.06
34.06
34.06
38.53
41.20
41.67
45.70
188.
208.30
91
-------
INTER-LAB0RATOKY STUDY
477
POLLUTANT - S02
SAMPLE NUMBER -
N
TRUE-VALUE
MEAN
MEDIAN
138
18.23
45.08
45.22
DATA IN ASCENDING ORDER
UNITS - HICROGRAHS PER CUBIC METER
RANSE
VARIANCE
STD. QEV.
COEF. VAR.
60.00
83.03
9.38
20.81'
C.I.f UPPER) 15.65
C.I.if LOWE Rt 43.52
SKEWNESS -.22
ACCURACY -B.23
.290
11.67
13*10
17*17
2 a* 20
24.00
26.46
27.90
29.00
32.88
3 3-. 00
33.15
33.30
33.HO
33.50
34.06
35.04
35.50
3 5'. 52*
35.75
36.68
37 . 52
38.00
38.09
38.30
38.37
T8.83
39.17
39.30
39.30
39.30
39.44
40.00
40.00
40.21
40.30
40.50
40.70
40.76
41.00
41.20
41.20
41.39
41.70
41.83
41.92
41.34
42.33
42.67
43.00
43.00
43.01
43.17:
43,28
43.30
43.30
43.30
43.33
43.40
43.67
4 3. 80
44.00
44.00
44.17
44.33
44.34
4 4:. 44
4 4 . 54
44.60
44.95
45.00
4 5.. 00
45.45
45.6S
45.77
4 5. 80
45.83
16.00
46.15
46.20
46.25
46.30
46.40
46.67
46.63
46.70
46.70
46.70
46.77
46.90
47.00
47.00
47 .00
47.00
47.16
47.16
47.16
47.16
47.16
48.00
48.06
48.09
48.22
48.40
48 .80
48.83
49.03
49.29
49.60
49.78
49.78
50.00
50.00
50.95
51.19
51.50
51 .SO
51.96
52.11
52.40
52.40
52.40
52.40
53.50
53.75
54.00
54.00
54.98
55.02
59. 4-2
60.26
60.26
61.16
61.30
S2.88
65.20
66.00
67.00
68.30
68.30
71.67
104.80^
508.28*
92
-------
INTER-LABIORATORt STUDY
POLLUTANT - S02
SAMPLE NUMBER - 3
N
TRUE-VALUE
ME AH
MEDIAN
139
80.17
77.53
7S.2D
DATA IN ASCENDING ORDER
UNITS - HICROGRA1S PER CUBIC METER
RANSE
VARIANCE
STD, OtfV.
COEF. VAR.
7% .20
130.23
11 .HI
11 .72
C.I.I UPPERS 79.13
C.I.fLQHE R? 75.63
SKEWNESS -.32
ACCUSAEt -2.1C
.3?*
17!. 280
30.80
31.33
3 9-. 40
13,00
501.00
51.59
5SI.11
5 at. 10
581.73
51.73
52.07*
52.10
52.88
SI. 83
56.70
57.00
57.33
58.00
59.05
59.89
70.00
70.10
70.10
70.52
70.71
70.71
70.97
71.10
71.60
72.00
72.30
72.50
72.80
72.81
73.00
73.00
73.27
73.36
71.00
71.20
71.70
71.80
71.90
75.00
75.00
75.00
75.00
75.08
75.20
75.15
75.50
75,50
75.56
75.73
75.71
75,90
76,00
76.50 :
76.67 .
76.70 .
76.90
77.00 .
77.00 :
77.15 i
77.37 .
77.17
77.53 '
77.75 l
77.83
78.00
78. ?Q
78.30
78.30
78.50
78.50
78.60
78.60
78.60
73.60
78.62
78.62
79.00
79.20
79.20
80.00
80.10
80.12
80.18
80.50
80.52
81.00
81.22
81.22
81.22
81.22
81.10
81.67
82.00
82.00
82.00
82.05
82.17
83.30
83.33
83.81
83.81
83.81
83.81
83.81
81.18
81.70
81.89
85.00
85.31
85.60
86.16
86.16
86.61
87. 17
87.30
88.88
89*01
89..0B
90.00
93.00
93.23
93.90
91.30
91.32
91.3?
91.80
95.32
95.65
97.50
99.00
101.00
101.17
105.00
196.
251.
937.960
93
-------
INTER-LABORATORY STUDY
477
POLLUTANT - S02
SAMPLE NUMBER - 4
N
TRUE-VALUE
MEAN
MEDIAN
138
114.70
114.54
113.00
UNITS - MICROGRAMS PER CUBIC METER
RANSE
VARIANCE
STD. QE!V.
COEF. VAR.
85 .06
17ft .17
13.20
11.52
C.I.fUPPERi 11&.74
C.LiCLOtlE R» 112.
SKEWNESS .83
ACCURACY -1,48
DATA IN ASCENDING ORDER
.34%
49.8!
79.00
80.90
82.08
87.369
91.06
94.3?
96.94
96.94
96.94
98.64
98.95
99.20
99.54
99.56
100.10
100.17
100.50
102.18
103.23
103.46
104.00
104.60
104.80
105.00
los:.oc
105.20
105.47
105,80
106.70
106.32
107.00
107.00
107.00
107.00
107.10
108.00
108.00
108.17
108.51
108.70
109.20
109.30
109.50
109.80
109.90
110.00
110.00
110.00
110.00
110.00
110.00
110.04
110.17:
110.20
110.66
111.00
111.00
111.02
111.10
111.11
111.17
111.17
111.35
111.67
111:. 70
112.00
112.6B
112.66
112.69
113.00
113.00
113.00
113. 06
113.55
113.80
114.00
114.27
114.44
114.45
114.73
114.8B
115.00
115:. 01
115.24
115.77
115.23
115.28
115.28
115.45
115.68
116.00
116.60
116.67
116.70
116.88
116.90
117.50
117.90
117.90
117.90
117.90
118.33
118.T9
118.83
118.91
119.00
119.37
119*95
120.00
120.00
120.80
121.00
122.70
122.96
123.14
123.40
124.10
124.58
125.66
126.67
128.38
129.46
130.20
131.00
131.00
132.06
132.83
133.12
13 3. ,32
134«37
134.77
136.00
140.67
142.03
147.00
148.00
154.00
158.00
165.06
277.72f
427.06%
1000.00%
94
-------
INTER-LA BflRATORY STUDY
POLLUTANT - S02
SAMPLE NUHSER - 5
N 1*D
TRUE-VALUE 155.*Q
MEAN 157J.7I
MEDIAN
DATA IN ASCENDING ORDER
177
UNITS - KICR0GRA1S PER CUBIC HETER
RANGE
STE. OE!V.
COEF. VAR.
153.130
* 53.31
21.23
13.50
C.I.ffUPPERJ 161.2*
C.I.CLOWE Rl 15*.18
SKEWNESS -1.22
ACCUSA.CY 1.81
31.860
5H.OO
68.00*
72.00
105.76
108.00
103.60
112.66
122.*5*
125.85
128.38
123.70
131*. 00
131.00
171.77
135.30
137.50
138.00
1*1*36
1*1.*8
1*1.*8
1*1.62
1*2.50
1*2.83
;.oo
U10
1*5^00
1*5.0*
1*5.16
1*6.30
1*6.36
1*7.00
1*7.*2
1*8.00
1*9.00
1*9.00
1*9.83
1*9.90
150.22
150,30
151.00
151.83
151.86*
151.96
151.96
152.33
153.00
153.90
15*.23
15*.57
155.00
155.20
155.20
155.*0
155.5*
155.93
156.05
156.11
156.83
1571.00
157>.00
157.20
157.20
157.20
157.20
157.20
157.60
157.67
157.95
157.93
158.GO
158.00
1581.13
158.30
158.*0
15 8'. 80
159.00
159.30
153.30
159>.30
159.82
1591.95
160.00
150.00
160.00
160.00
IB0.00
161.00
161.00
161.10
161.11
161.50
161.62
1S1.67
161.70
162.00
162.30
162.**
162.58
162.95
16*.82
165.06
165.*0
166.10
166.20
166.67
167.00
167.09
167.68
167.82
168.50
169.20
169.67
169.80
170.00
170.0*
170.30
170.50
171.02
171.10
171.13
172.96
173.30
173.61
17*. 00
17*. 07
175.5*
176.66
177.50
180.00
183.93
186.56
188.09
190.00
191.26
192. *0
135.58
196.50
20O.OO
200.3O
20*. 00
205.80
209.00
662.86*
1000.00%
95
-------
INTER-LABORATORY STUDY
1077
POLLUTANT - S02
SAMPLE NUMBER -
N
TRUE-VALUE
MEAN
MEDIAN
UNITS - MICROGRAHS PER CUBIC
13.60
15.15
13.10
RANGE
VARIANCE
STD. OEY.
C0EF. VAR(
194.30
330.69
13 .IS
120.02
C.I.If UPPERI 18. 49
C.I.
-------
STUDY
1077
POLLUTANT - SQ2
SAMPLE NUMBER - 2
UNITS - WICR;0GRA*S PER CUBIC METFR
N
TRUE-VALUE
HE AW
MEDIAN
125
56.20
56.01
53.91?
RAHBE
VARIANCE
STt). 0EV.
COEF. VAR
IN ASCENDING ORDER
441.60
147S.B2
38.4 2
68 .60
C.I.f UPPER! 62.7*1
C.I.SLOWE Rl 49.27
SKEHNESS 9.18
ACCURACY -4 .04
.480
• 570
5. (190
16*70
20.95
20*960
21*60
24.33
26*270
27*50
30.17
31.44
35.000
36.00
36.65
36.68
39^.20
39.60
39.74
42*30
44*00
44*20
44*52
44.54
44.54
44.80
45.00
45.25
45.85
47.16:
47.16:
47.51
47.56
47.60
48.20
48.70
48.73
49.00
49.08
49.11
49.37
49.44
49.75
49.78
49.78
50.00
50.03
50.50
50.93
51.00
51.07
51.70
51.76
5Z.OO
52.10
52i.20
52W40
5 2 . 83
52.98
53.00
53.17
53f*56
53^.70
53.89
531.93
54*11
54W20
54W33
54.45
54'. 50
54.61
54.80
55.00
55.0D
55.00
55.02
53.02
5S.02
55.02
55.02
55.37
55.48
55.00
55.31
55.32
56.70
56.94
57.39
57.50
57.60
57.64
57 .64
57.70
*7 .00
58.30
58.60
58 .70
58 .SO
59.75
59.90
60.00
60.10
60.26
60.26
60.26
60 .26
60.26
60.30
60. 7€
61.00
51.00
51.15
51.62
62.00
52.96
63.17
63.47
56.60
56.60
S6.76
68.12
58.12
69.800
88.35
89.00
94.32
112.13
172. 960
419.200
458.30
1000.000
1000.000
1000.000
97
-------
INTER-LABJDRATORY STUDY
1077
POLLUTANT - S02
SAMPLE NUMBER -
N
TRUE- VALUE
MEDIAN
126
93.73
91.18
91.70
DATA IN ASCENDING ORDER
UNITS ~ WICROGRAHS PER CUBIC METER
RANGE
VARIANCE
STD. OOf.
COEF. VAR,
83 .
161.25
12.70
13.93
C.I. If L0 ME
SKEWNESS
ACCURA3CY
93.40
R) 88.97
-1.19
-2 .IT
9.36%
20.11%
39.30%
39.73
44*44%
IS* 64
16:. 70
SOU00%
581.30
5 4 . 36
68.1?
68.33
63.57
71.50
75.20
75.98
76.78
77-. 40
77*95 (
78.00
78.40
78.60
78.70
82.01
82.10
83.00
83.33
83.50
83.84
83.84
83.84
84.32
85.40
85.60
86.46
86.46
86.46
87.00
87.00
87.20
87.69
88.10.
88.50
88.96
89.33
89,37
89.65
90.00
90.00
90.30
90.33
90. (12
90.42
90.50
90.85
91.00
31.27
91.47
91.55
91.63
91.66
91.70
91.70
31.70
91.70
91.70
91.70
91.70
91.70
91.80
92-00
92.85
93.18
93.20
93.96
94.10
34.32
94.32
94.32
94.32
34.40
34.90
95.10
95.17
95.40
95.67
36.00
96 .25
96 .30
98.62
S6.TO
36.80
36.94
97 .00
97.13
97.30
97.40
98.00
38 .00
38.30
98 .^53
98.49
33.00
39.00
39.10
33.30
39 .42
93.56
33.67
93.83
100.00
100.91
101. 5G
101.67
102.00
102.00
102.25
102.50
103.30
104.80
105.70
106.00
106.00
110.04
111.00
111.70
111.94
113.03
121.68
123.14
145.10%
252. 52%
432^84%
759.800
1000.00*
1000.00%
98
-------
INTER-LABORATORY STUDY
1077
POLLUTANT - S02
SAMPLE NUMBER - 4
UNITS - HICROGRAHS PER CUBIC METER
N
TRUE-VALUE
MEAN
MEDIAN
125
93.73
33.63
9*1.3?
DATA IN ASCENDING ORDER
RANGE
VARIANCE
STO. 0E1V.
C0EF. VAR.
133. .08
14.53
15 .S3
C.I.fUPPERI 98.19
C.I.ILOUE RI 91.06
SKEWNESS .87
ACCURAEf .63
.82®
1.02®
9.41®
19.05$
3 9'. 300
16-. 13
53U30
531.80
571.26
5 7s. 69
5 3. 300
70. CO
72.52
73.36
75.20
77.50
78.00
78.96
80.80
81.22
81.22
83.31
83.84
8*. 10
84.50
84.90
85.50
85.55
85.82
36.00
86.09
86.40
86.57
87.00
87.31
88.00
88.10
88.30
88.30
88.83
89.00
89.03
89.08
89.08
89.08
89.33
89.83
90. OO
90.30
90.80
91.00
91.20
91.47
91.57
91.70
91.70
91.70
92.00
32.82
33.01
S3. OS
33.20
93.50
33.50
93.92 :
33.95
94.29
94.32 :
94.32 :
34.32 :
34.32
34.32
34.32
34.32
34.32
94.64 .
35.00 :
35.00
35.00 :
35.00
35.24 :
35.42 -
35.58 '
36.00
96.07
96.11
96.80
96.94
97.00
97.33
97.34
97*40
97.51
98.00
98 .20
98.40
98.49
98 .'70
99 .00
99.12
99.30
99.56
39.56
99.^6
39.67
S9 .83
100.00
100.00
100.00
101.83
102.18
1C3.00
104.00
104.10
105.00
106.00
106.58
106.70
107.00
108.09
110.04
110.71
111.00
111.94
113.00
114.21
115.10
120.59
121.10
128.36
161.20®
177.21
342.46®
864.60®
1000.00®
1000.00®
1000.00®
99
-------
IN1TR-HBDRATORY STUDY
1077
POLLUT*NT - S02
SAMPLE NUMBER - 5
N
TRUE-VALUE
MEAN
MEDIAN
1^9.07
171.21
172.92
UNITS - MICROGRAWS PER CUBIC HETER
RANGE
VARIANCE
STO. OEV.
COEF. VAR>.
995.03
6511.76
81.S2?
46.78
C.I.CUPPER* 188.59
C.I.CLONE R) 159.89
SKEWNE5S 8.11:
ACCURACY -3.13
DATA IN ASCENDING ORDER
1.6lft
1.90ft
1.97
1 5* <*0
1 71. 73ft
22.20
36.68
38. 51ft
70.71ft
82.30ft
83.33
96.52
125.00
132.01
133.65
133-.1C
110.30ft
1*1.05
1*3.30
1*9*76
1501.00
151!. 15
1551.30
157'.20
1591.67
150.00
151.00
152.11
162.70
163.70
161.01
161.30
161.60
161.62
161,91
165.00
165.00
165.06
165.15
165. S5
165.70
166,31
167.00
167.00
167,23
167.58
167.68
168.00
168.30
168.30
168.31
168.83
169.36
169.39
170,10
170.30
170*30
170.30
170.91
171.00
171.72
172.13
172.57
172.60
172.92
172.92
172.92
172.92
172*92
172.92
172.92
173.59
173.70
175.12
17S.50
175.51
175.51
175.51
175.68
176.00
176.12
176.70
177.68
177.80
178.16
178.25
178.50
178.51
178.61
178.80
178.96
179.00
179.19
179.65
179.90
180.00
180.91
180.91
181.00
181.00
181.50
182.61
182.92
183.30
183.33
183.35
183.10
181.33
185.00
186.00
186.57
186.70
187.00
187.05
187.35
187.50
188.17
188.30
188.61
188.61
191.26
192.00
193.00
196.50
197.17
200.00
209.60
210.00
210:* 30
211.60
258.00ft
180.83ft
1000.00ft
1000.00ft
1000.00ft
1000.00ft
1000.00ft
TOO
-------
INTER-HB0RATORY STUDY
POLLUTANT - N02
SAMPLE NUMBER - 1
1275
UNITS - MICRIOGRANS PER CUBIC METER
N
10B
TRUE-V
HE AN
MEDIAN
ALUE *12
.1?
.1?
DATA IN ASCENDING
.04
.05
.05
.06
.06
.07
.09
.09
.09
.09
.09
.09
.09
.09
.09
.09
.10
.ID
.10
.10
.10
.10
ORDER
.10
.10
.10:
.10:
.10:
.11
.11:
.11:
.11
.11:
.11:
.11:
.11
.11
.11:
.11.
.11:
.11:
.11:
.12
.12
.12:
RANGE.
VARIANCE
STD. OEV.
COEF. VAR.
.17
.00.
.03
22 .5 2'
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
C.I.C UPPER) .12
C.I.ICLOWE R» .11
SKEWNESS .IT
ACCURACY .no
.13 .14
.13 .14
.13 .14
.13 .14
.13 .14
.13 .14
.1* .15
.13 .15
.13 .15
.13 .15
.13 .16
.13 .16
.13 .16
.13 .16
.13 *17
.13 .18
.13 .19*
.13: .20 :
.14 .22
.14 .24*
.14
101
-------
INTER-LABORATORY
1275
POLLUTANT - N02
SAMPLE NUMBER -
N
TRUE-VALUE
MEAN
MEDIAN
DAT* IN ASCENDING ORDER
UNITS - HICROGRAMS PER CUBIC METER
105
.18
RANGE
VARIANCE
STD. OEV.
COEF. VAR
.00
.03
C.I.fUPPERJ .18
C.I.CLOWE R> .17
SKEWNESS -.73
ACCU34SY -3.87
.0?
.11
.11
.12
.13
.14
.1*
.1^
.1*
.1*
.1*
.15
.15
.15
.15
.15
.15
.15
.16
.16
.IE
.16
.16
.16:
.16:
.16
.16:
.16
.16:
.16:
.16:
.16
.16:
.16
.16
.16:
.17:
.17:
.17:
.17
.17;
.17:
.17
.17:
17 i
17 i
17 .
17 i
17 .
17 i
17 .
17 ;
17 ;
17
17 ,
18 '
18
18
18
18
IB
18
18
18
18
18
.13
.18
.18
.18
.18
.18
.19
.19
.19
.19
.19
.19
.19
.19
.19
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.,20
.20
.21 :
.21 :
.21
.21
.22
.21
.25
. 27
.34*
.1164
102
-------
INTER-LABORATORY STUDY
1275
POLLUTANT - N02
SAHPLF NUMBER - T
N
TRUE-VALUT
MEAN
MEDIAN
108
.32
.3?
DATA IN ASCENDING ORDER
UNITS - .MICROGRAMS PER CUSTC METfR
RAN6E
VARIANCE i
STO. OEV.
COEF. VAR.
.20
.00
.03
10.37
C.I.fUPPERJ .53
C.I.it LOWE R> .32
SKEWNESS .07
ACCURACY -.61
23
i
23
24
24
25
25
27
27
28
28
28
29
29
29
30
30
30
30
30
30
30
30
30
.30
.30
.30
.31
.31
.31
.31
.31
.31
.31
.31
.31
.31
.32
.32
.32
.32
,32
.32
.32
.32
.32
.32
.3Z
.32
.32
.32
.3Z
.3Z
.32
.32
.32
.3Z
.33
.33
.33
.33
.33
.33
.33
.33
.33
.33
.33
.33
.33
.33
,33
.33
.33
.33
.33
.34
.34
.34
.34
.34
.34
.34
< .34
.34
,34
.34
.34
.34
.34
.34
,34
.35
.35
.36
.35
.35
.35
.35
. 36
.36
.36
.37
.37
.36
.40
.41
.43
.43
.47
.57
103
-------
INTER-LABORATORY STUDY
1276
POLLUTANT - N02
SAMPLE NUMBER - 4
N
TRUE-VALUE
MEAN
MEDIAN
108
.54
.52
.52
UNITS - MICROGR41S PEP CUBIC METER
RANGE
VARIANCE
STO. OEW.
COEF. VAR.
.36
.00
.05
9.<10
C.I.CUPPER! .52
C.I.ICLOWE Rl .51
SKEHNESS .BO
ACCURACY -2.99
DATA IN ASCENDING ORDER
39
39
40
47
45
45
45
45
45
45
46
45
45
47
47
47
43
43
43
43
48
.48
.43
.49
.49
.49
.49
.49
.49
.49
.50
.50
.50
.50
,50
.50
.50
.50:
.50
.50
.50
,50
.50
.51
51
51
51
52
52
52
52
52
,52
,52
52
52
52
52
52
52
52
52
52
,53 '
,53 *
53 '
.53
,53
,53
.54
.54
.54
.54
.54
.54
.54
.55
,55
.55
.55
.55
.55
.55
.55
,55
.55
.55
.56
.56
.55
.58
.55
.56
.57
.57
.57
.57
.56
.53
.58
.59
.59
. sef
.75
104
-------
INTER-LA BIORATORY
127S
POLLUTANT - N02
SAMPLE NUMBER - 5
UNITS - MICRQGRAMS PER CUBIC METFR
N
TRUE-VALUE
MEAN
MEDIAN
107
.87
.36
DATA IN ASCENDING ORDER
RANGE
VARIANCE
STD. DEV.
C0EF. VAR
.54
.00
.07
7.34
C.I.CUPPER* .86
C.I.CLOWE RJ .84
SKEWNESS .48
ACCURACY -1.S1
.63
.70
.72
.72
.73
.73
.74
.75
.75
.75
.75
.77
.78
.73
.79
.79
.SO
.80
.80
.80
.80
.81
.81
.31
.31
.32
.82
.82
.82
.33
.83
.83
.83
,54
.84
.34
.34
.84
.34
.34
.85
.85
.85
.85
85 i
85 I
35
85
85
85
85
85
88
8&
35
85
85
85
35
85
35
85
37
87
87
.87
.87
.87
.87
.37
.87
.87
.37
.88
.88
.38
.38
.38
.88
.38
.88
.88
.88
.38
.39
.39
.89
.89
.89
.89
.90
.90
.90
.90
.9?
.9?
.94
.9G
.98
1.00
1.01
1.150
1.17
1.670
105
-------
INTER-LABORATORY STUDY
POLLUTANT - NC2
677
SAMPLE NUMBER -
N
TRUE-VALUE
MEAN
MEDIAN .13
DATA IN ASCENDING ORDER
94
.12
UNITS - MICROGRAMS PER CUBIC METFR
RANGE
VARIANCE
STD. OEV.
COEF. VAR.
.15
.00
.02
17.51
C.I.CUPPER1 .13
C.I.CLOWE R> .12
SKEWNESS -1.39
ACCURACY 8.40
• Olw
.019
.03
.04
.07
.08
.09
.09
.10
.10
.10
.11
.11
.11
.11
.11
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
.12
,12
.12
.12
.12
.12
.12
.12
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.14
.14
.14
,14
.14
.14
.14
.14
.14
.14
.14
.14
.14
.14
.14
.14
.14
.15
.15
.15
.17
.17
.17
.18
.18
106
-------
INTER-1.ABORATORY STUDY
677
POLLUTANT - N02
SAMPLE NUMB
N
TRUE-VALUE
MEAN
MEDIAN
- 2
9T
.2*
.25
DATA IN ASCENDING ORDER
UNITS - MICROS RAMS PER CUBIC METER
RANGE
VARIANCE
S7D. 0EV.
COEF. VAR.
.30
.00
.03
11.85
C.I.CUPPERJ .?5
C.I.CLOWE R> -24
SKEWNESS 1.73
ACCURACY 4.20
.02*
.02 •
.13
.17
.19
.20
.21
.21
.21
.22
.22
.23
.23
.23
.23
.23
.23
.23
.23
.23
.23
.2*
.24
.24
,24
.2*
.24
.24
.24
.24
.2*
.24
.2*
.24
,24
.24
.24
.2*
.24
.24
.24
.25
.?5
.25
.25
.75
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
,25
.25
.26
.26
.26
.26
.26
.26
.26
.26
.26
.26
.26
.27
.27
.27
.27
.27
.27
.28
.28
.28
.29
.29
.29
.29
.43
.26
107
-------
INTTR-LABORATORY STUDY
677
POLLUTANT - N02
SAMPLE NUMBER - 3
N 9?
TRUE-VALUE .24
MEAN .2*
MEDIAN .25
DATA IN ASCENDING ORDER
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
STD. OEV.
COEF. VAR
.26
.00
.03
11.59
C.I.CUPPER* .26
C.I.CLOWE RJ .25
SKEWNESS 2.11
ACCURACY 2.46
.02*
.02*
.17
.18
.19
.20
.21
.21
.21
.23
.23
.23
.23
.23
.23
.23
.23
.23
.23
.23
.24
.24
.24
.24
.24
.24
.24
.24
.24
.24
.24
.24
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
,?5
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.75
.26
.26
.26
.26
.26
.26
*?6
.26
.26
.26
.26
.27
.27
.27
.27
.27
.27
.27
.27
.28
.28
.23
.28
.29
.29
.29
.30
.30
.43
108
-------
INTER-LABORATORY STUDY
677
POLLUTANT - N02
E NUMBER - 4
N
TRUE-VALUE
MEAN
MEDIAN
99
.38
.39
•39
DATA IN ASCENDING ORDER
UNITS - MICR08RAWS PER CUBIC METER
RANGE
VARIANCE
S7D. 0€V.
COEF. VAR.
.54
.00
.05
11.12
C.I.CUPPER* .40
C.I.fLOHE RJ .33
SKEWNESS 1.16
ACCURACY 2.38
.03*
.04%
.13
.30
.31
.31
.32
.32
.32
.34
.34
.35
.35
.35
.36
.36
.36
.37
.37
.37
.37
. 37
.37
.37
.37
.38
.38
.38
.38
.38
.38
.38
.38
.38
.38
.38
.38
.38
.38
.38
.38
.38
.33
.38
.38
.38
.38
.38
.38
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.39
.79
.79
.39
.40
.40
.40
.40
.40
.40
.40
.40
.40
.40
.41
.41
.41
.41
.41
.41
.4?
.42
.4?
.42
.43
.43
.43
.43
.45
.46
.48
.65
.66
109
-------
INTER-LABORATORY STUDY
677
POLLUTaNT - N02
SAMPLE" NUMBER - rj
N 96
TRUE-VALUE .59
MEAN .71
MEDIAN .71
DATA IN ASCENDING ORDER
UNITS - MICROGRAHS PFR CUBIC METER
RANGE
VARIANCE
STD. OEV.
COEF. VAR
.71
.00
.07
C.I.I UPPER I .73
C.I.CLOWE R) .70
SKEWNESS 3.82
ACCURACY 2.75
07
07
50
57
59
,65
,55
,55
,6G
,56
,67
,67
,68
,58
,69
,69
,69
,69
,69
.69
.'69
.69
.69
.69
.69
.70
.70
.70
.70
.70
.70
.70
.70
.70
.70
.70
.70
.70
.70
.70
,70
,70
,70
,71
,71
,71
,71
,71
,71
,71
,"'1
,71
,71
,71
,71
,71
,71
,71
,71
.72
,72
.72
.72
.72
.72
.72
.72
.72
.72
.73
.73
.73
.73
.73
.73
.73
.74
.7*1
.75
.75
,75
.75
.75
.75
.7S
.7B
.77
.77
.78
.79
.8P
.80
.81
1.21
no
-------
INTER-LA 8 ORATORY STUDY
POLLUTANT - N02
1277
UNITS - HICROGRAMS PER CUBIC HETER
SAMPLE NUMB
N
TRUE-VALUE
MEAN
MEDIAN
- 1
96
.18
.18
RANGE
VARIANCE
STB. DEV.
COEF. VAR.
.36
.00
.04
21.58
DATA IN ASCENDING ORDER
C.I.CUPPERI .19
C.I.CLOWE R» .13
SKEWNESS -.11
ACCURACY 1.9*
.01
.09
.09
.11
.11
.12
.13
.13
.14
.15
.15
.15
.15
.15
.15
.17
.17
.17
.17
.1T
.17
.17
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.13
.18
.13
.18
.18
.18
.18
.16
.18
.18
.18
.18
.18
.18
.18
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
.19
,19
.19
.19
.19
.19
.19
.20
.20
.20
.20
.70
.20
.20
.20
.20
.21
.21
.2?
.22
.22
.2?
.23
.24
.24
.25
.2B
.27
.37
.40'
9.00'
in
-------
INTER-LABORATORY STUDY
1277
POLLUTANT - N02
SAMPLE NUMBER - 2
UNITS - MICROGRAMS PER CUBIC METER
N
TRUE-VALUE
HEAN
HEDIAN
98
.31
.32
.3?
RANGE
VARIANCE
STO. DEV.
COEF. VAR.
DATA IN ASCENDTNG ORDER
.20
.00
11.09
.10*
.2?
.23
.24
.24
.25
.25
.25
.25
.25
.26
.2S
.27
.29
.29
.29
.30
.30
.30
.30
.30
.30
.31
.31
.31
.31
.31
.31
.31
.31
.32
.32
.32
.32
.32
.32
.32
,32
.32
.32
.32
,32
.32
.32
.32
.32
.32
.32
.32
.72
.32
.32
.32
.32
.72
.72
.32
.32
.32
.33
.33
.33
.33
C.I. CUPPER*
C.I. .CLOWE R>
SKEHNESS
ACCURACY
33
33
33
73
33
33
33
73
33
73
33
33
34
34
34
34
74 1*;
.33
.31
-.46
4.13
.34
.35
.35
.35
.35
.36
.36
.36
.37
.37
.37
.38
.39
.,42
*4?
.66*
;.oo*.
,3*
112
-------
INTER -LABORATORY STUDY
POLLUTANT - N02
SAMPLE NUMBER - 3
1277
UNITS - MICROGRAMS PER CUBIC HETER
N
TRUE-VALUE
MEAN
MEDIAN
DATA IN ASCENDING ORDER
99
.50
.51
RANGE
VARIANCE
STD. 0EV.
COEF. VAR
.57
.00
.06
11.84
C.I.fUPPERI
C.I. CLOWE R|
SKEWNESS
ACCURACY
.52
.50
.91
.29
.3S
.37
.33
.39
.39
.41
.43
.45
.47
.48
.48
.43
.49
.49
.49
.50
.50
.50
,50
.50
.51
.51
.51
.51
.51
.51
.51
.51
.51
.51
.51
.51
.51
.51
.51
.51
.51
.52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
,52
.52
.52
.52
.52
,53
.53
.53
.53
.53
.53
.53
.53
.53
.53
.53
.53
.53
.54
.54
.54
.54
.54
.54
.55
.55
.55
.55
.56
.56
.56
.57
.57
.57
.57
.59
.87
1.06©
.00*
113
-------
INTER-LABORATORY STUDY
1277
POLLUTANT - N 02
SAMPLE NUMB
- 4
UNITS - MICROGRAMS PER CUBIC METER
N
TRUE-VALUE
MEAN
MEDIAN
98
.5P
.52
.52
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
.69
• OG
.07
13.45
C.I.fUPPER) .53
C.I.CLOWE R> .50
SKEMNESS 3.33
ACCURACY 2.98
DATA IN ASC BIDING ORDER
,33
,37
,38
,39
,40
,40
,40
,41
,42
,45
,48
,48
,48
,49
,50
,50
,50
,50
,50
50
50
50
50
50
51
51
51
51
51
51
51
51
51
51
51
51
51
52
52
52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
.52
,52
,52
,52
,53
,53
,53
,53
,53
,53
,53
,53
,53
,53
,53
,53
54
,54
,54
,54
,54
,54
.54
.55
.55
.55
.56
.56
.58
.57
.57
.57
.57
.58
.58
.59
1.02
25. 00*
114
-------
INTER-LABORATORY STUDY
POLLUTANT - N02
SAMPLE NUMBER - 5
1277
UNITS - MICROGRAMS PER CUBIC METER
N
TRUE-VALUE
MEAN
MEDIAN
10D
.87
.87
.89
RANGE
VARIANCE
STO. DEV.
COEF, VAR
1*03
.01
.11
12.37
DATA IN ASCBMDTNG ORDER
C.I.IUFPERI .89
C.I.CLOWE RJ «85
SKEHNESS -*ii.75
ACCURACY 2.2*
.03
.369
.70
.70
.70
.72
.72
.73
.71
.7*
.75
.77
.80
.81
.81
.82
.83
.8*
.84
.85
.85
.85
,85
.85
.86
.86
. 38
. 86
.86
.87
.87
,87
.88
,88
.88
.88
.83
.88
.88
. 88
.88
. 88
.83
.38
.88
.88
.88
.88
.83
.83
.83
.89
.83
.89
.89
.89
.83
.83
.83
.89
.90
.90
.90
.90
.90
.90
.90
,90
,90
,90
,90
,91
.91
.91
.91
.91
.91
.91
.92
.92
.92
.92
,92
.92
.93
.93
.93
.9*1
.9*
.95
«<**>
.95
.95
.96
.96
.97
.97
.97
.98
.99
1.1?
1.83
%3.00
115
-------
INTER-LABORATORY STUDY
POLLUTANT - CO
SAMPLE NUMBER - 1
1076
N
228
TRUE-VALUE 4.51
MEAN
MEDIAN
4.63
4.30
DATA IN ASCENDING
1.00® 3.70
1.50* 3.7C
1.60 3.7C
2.CC® 3.75
2.00 3.75
2.00 3.80
2.80 3.80
2.80
2.96
3. CD
3.00
3.20®
3.20
3. 20
3.20
3. 3D
3.30
3.39
3.409
3.40
3. 48
3.49*
3.50*
3.50
3. 50
3.50
3.5C
3.5C
3.5P
3.5C
3.50
3.5G
3.50
3.55
3.57
3.58
3.60
3.60
3.60
3.60
3.67
3.69
3.70
ORDER
3.80
3.80
3.80
3.80
3.81
3.85
3.90
3.90
3.90
3.90
3.90
3,91
3.99
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.03
4.08
4.10
UNITS - MICROGRAHS PER CUBIC METER
C.I. (UPPER) 5.11
C,I. (LOWER) 4.15
SKEWNESS 9.00
ACCURACY -4.66
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
4.10
4. 10
4.10
4.10
4.10
4.10
4.14
4.20
4.20
4.20
4.20
4.20
4.20
4.20
4.20
4.20
4.20
4.20
4.22
4.23
4.25
4.25
4.25
4.25
4.25
4.25
4.25
4.26
4.30
4.30
4.30
4.30
4.30
4.30
4.30
4.31
4.33
4.35
4.35
4.35
4.40
4.40
4.40
4.40
4.41
4.42
4.44
4.45
4.45
4.47
40.89
13,65
3.69
79.76
4.47
4.50
4.50
4.50
4.50
4.50
4 .50
4.50 4.65 .73 H2.49
4.50 4.69 .75 43.31*
4.50 4.70 .75
4.50 4.70 .85
4.50 4.70 .90
4.50 4.70 4.90
4.50 4.72 5.00
4.50 5.00
4.50 5.00
4.50 5.00
4.50 5.00
.50 5.GO
.50 5.00
,50 5.00
.50 5.CO
.50 5.CO
.50 5.00
.50 5.00
.50 5.00
.50 5.00
.50 5.00
.50 5.09©
.50 5.10
.50 5.15
.50 5.20
,50 5.20
,50 5.22©
.50 5.30
.50 5.33©
.50 5.36
4.53 5.40
4.55 5.50
4.60 5.50
4.60 5.60®
4.60 5.70
4.60 5.76®
4.60 6.CO®
.60 6.00
.60 6.01
,60 6.25®
.El 21.85
.62 23.480
•65 40.00
116
-------
INTER-14BORATORY STUDY
POLLUTANT - CO
SAMPLE NUMBER - 2
TRUE-VALUE
MEAN
MEDIAN
230
20.86
2.1.23
21.00
OAT 4 IN ASCENDING ORDER
1076
UNITS - MICROGRft«S PER CUBIC METER
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
23.74
3.92
1.98
9.32
C.I. (UPPER) 21.48
C.I. (LOWER) 20.97
SK^WNESS 4.16
ACCURACY .67
10. 80* 20.00
15.76 20.00
1S.19* 20.00
16. 5C 20. CO
17.000 20. OG
17.10 20.10
17.70*
18 .00*
18.12
18. 3D
18.50
18. 5C*
18.60
18. 6C
18. 6C
18.66
18.75
18.75
19. CC
19. CC
19. CC
19.10
19.30
19.42
19. 5C
19. 5C
19.50
19. S7
19.57
19.64
19.70
19.70
19.72
19.75
19.85
19.90
19 .9 C
19.90
20.00
2G .CO
20.00
20.00
20.00
20,11
20,20
20.20
20. 25
20,25
20, 30
20,30
20.30
20.30
20,30
20,30
20.34
20, 39
20.40
20. 40
20.40
20.40
20.47
20,50
20.50
20. 50
2C. 50
20,50
20.60
20.60
20.60
20.60
20.61
20.65
20,70
20.70
20,70
20.70
20.74
20.75
20.75
20.75
20.75
20.75
20,80
20.80
20.80
2C. 80
20.80
20.80
20.80
20,87
20,90
20.90
20.30 21.30 21.40 22.00 22.17 27.79
20.90 21.30 21.40 22.00 22.20 3Q.OO®
20.9S 21.30 21.42 22.00 22.22 32.00
21.00 21.31 21.48 22.00 22.42 39.50
21.00 21.31 21.49 22.10 22.50 <*
21.00 21.40 21.50 22.11 22.50
21.00 21.50 22.50
21.00 21.50 22.50
21.00 21.50 22.50
21.00 21.50 22.50
21.00 21.50 22.50
21.00 21.50 ?2.50
21.00 21.50 22.50
21.00 21.50 22«50
21.00 21,50 22.52
21.00 21.50 22.55
21.00 21.50 22.SO
21.00 21.50 22.65
21.00 21.50 22.70
21.00 21.50 22.80
21.00 21,50 22.30
21.00 21,50 23.CO
21.00 21.50 23.00
21.00 21.50 23.00
21.00 21.50 23.00
21.00 21.53 23.00
21.00 21. SO 23.00
21.00 21.65 23.16
21.02 21.65 23.20®
21.04 21.72 23.48®
21.10 21.75 23.48
21.10 21.80 23.48
21.10 21.86 23.50
21.20 21.90 24.CO*
21.20 21.90 24.10
21.20 21.95 24.25*
21.20 22.00 24.38
21.20 22.00 24.50
21.20 22.00 24.50 »
21.20 22.00 24.56*
21.20 22.00 24.71*
21.25 22,00 26.60
21.30 22.00 27.
117
-------
INTER-LABORATORY STUDY
1076
POLLUTANT - CO
SAMPLE NUMBER - 3
N 214
TRUE-VALUE 40.14
MEAN 40.73
MEDIAN 40.45
DATA TN ASCENDING ORDER
UNITS - MICR06RAHS PTR CUBIC
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
16.39
4.30
2.07
5.09
C.I. (UPPER) 41.01
C.I. (LOWER) 40.46
SKEWNESS" 1-76
ACCURACY .78
21.00$ 79.10
31 .00* 33.10
35.300 39.10
35.500
3S.CC*
36.00*
36 .C 3*
36.30
37.20
37.3*
37.50
37.5C
37.5C
37. 5C
37.50
37.8G
37.90
38 .CO
38.00
38 .CO
38.02
38.10
38.20
38.22
38.30
38.30
38.50
38.50
38.5*
38.7!:
38.80
38.80
38.90
38.90
38.96
39.00
39.CC
39.00
39.00
39.00
39.00
39.OF
39.1C
39. 20
39.2^
39.26
39.30
39.40
39.50
39.50
39.50
39.50
39.50
39.50
39.50
39.50
39.50
39.50
39.50
39.50
39,50
39*50
39.60
39.60
39.69
39,70
3^,70
39.70
35?, 75
39.75
39,80
39.80
39.90
39.91
39.98
40.00
40.00
40.00
40. 00
40.00
40.00
40.00
40,00
40.00
40.00
40,00
40.00 40.00 40*80
40,00 40.00 40.87
40.00 40.00 40.90
40.05
40.07
4C.10
40.10
40.20
40.20
40.20
40.20
40.20
40.25
40.25
40.30
40.30
40.30
40.30
40.40
40.40
40.40
40.44
40.47
40.43
40. SO
40,50
40.50
40.50
40. SO
40.50
40.50
40. SO
40. SO
4C.50
40.50
40.50
4C.53
40. £0
40.70
40.75
40.80
40*80
4C*80
40.98
40.99
41,00
41.00
41.00
41. 00
41*00
41.00
41.00
41,00
41*00
41.00
41.00
41.00
41.00
41.00
41.13
41. 2Q
41*25
41,30
41.30
41,31
41.40
41. £0
41,50
41.50
41.50
41.50
41.50
41.57
41, CO
41, SO
41.SO
41.65
41.70
41.75
41.80
41, 8€
41.80
41.84
41.90
42. (TO
42.00
42.00 ,,2.09
42.00 )2.12
42.00 n^g
42.38
42*50
42.52
42.61
42.62
42.66
42.80
42.81
42.96
43.00
43.00
43.00
43.00
43.00
43.00
43.20
43.22
43.46
43.50
43.50
43.80
43.95
44*00
44*35*
44.50*
44.50
45.00
45.00
45.29
45.70
45.80*
46.10
46.39
46.80*
47 .63*
47.70
47.75
48.10*
53.19
80.00*
118
-------
INTER-t*8CR*TORY STUDY
POLLUTANT - CO
377
UNITS - MICR00R4KS PFR CUBIC
SAMPLF NUMB
TRUE-VALUE
MEAN
MEDIAN
- 1
279
3.38
3.08
3.04
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
4.50
.36
.60
19.37
C.I. (UPPER)
C.I. (LOWER)
SKEWNESS
ACCURACY
DATA IN ASCENDING ORDER
3.15
3.01
-.23
-TO.06
.50
1.000
1.50
1.50
1.52
1.650
1.7*
1.75
1.80
1.80
1.8*
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.10
2.10
2.10
2.11
2.20
2.2*
2.25
2.25
2.29
2.30
2.30
2.30
2.32
2.37
2.400
2.40
2.40
2.50
2.50
2.50
2.50
2.50
2.50
2.50
2.50
2.50
2.50
2fff, ^^
.50
2.50
2.50
2.50
2.50
2.52
2.56
2.60
2.60
2.60
2.60
2.6X10
2.65
2.65
2.70
2.70
2.70
2*70
2.70
2.70
2.75
2.75
2.75
2.77
2.80
2.80
2.80
2,80
7.84
2.86
2.88
2.89
2.90
2.90
?. 90
?. 90
2.90
2.90
2,90
2.90
2.90
2.90
2.92
2.95
2,96
2.96
2.98
2.98
2.39
3.00
3.00
3.00
3.00
7.00
7.00
3.00
3,00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3,00
7.00
3.00
3.00
3.00
3,00
7.00
7.00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.CD
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.00
3.04
3.04 .
3.04
3.05
3.08
3.C8
3.08
3.08
3.10
3.10
3.10
3.10
3.10
3.10
3.10
3.10
3.10
3.10
3.10
3.10
3.10
3.15
3.16
3.17
3.18
3.20
3.20
3.20
3.20
3.20
3.20
3.20
3.20
3.20
3.20
3.20
3.25
3.25"
3.25
3,25
3.25
3.25
3.25
3.25
3.25
3.25
3.25
3.25
3.25
3.70
3.30
3.30
3.30
3.70
3.30
3.30
3,30
3.30
3.30
7.70
3,40
3.40
3.40
3.40
3*40
3.40
3.40
3.40
3.45
3.46
3.47
3,48
3,50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3.50
3,50
3,50
3,50
3,50
3.50 4
7.50 4
3.50 4
3.50 4
3.50 *
7.50 4
7.50 *
3.52 4
3.56 4
7.58 5
7.60 5
7.60 5
7.65 8
7.7H 27
7.70 27
7.70
7.75
3.75
3.80
3.81
7.90
7.91
7.91
3.91
7.91
4.0H0
4.000
4.000
4.000
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.100
4.10
4.10
.12
.17
.30
.3P
.35
.50®
.50«
.50
.80
• GO*
.00
.00
.800
.390
.830
119
-------
TNTrR-L480R*TORY STUDY
377
POLLUTANT - CC
SAMPLT NUHBER -
N
TRUE-VALUE
MEAN
MEDIAN
DAT 4 IN
•y
7
8
12
12
13
13
13
13
13
13
13
14
1ft
1ft
1ft
1ft
1ft
1ft
1ft
1ft
1*
1ft
1ft
1ft
1ft
14
14
1ft
15
15
15
15
15
15
15
15
15
16
16
278
.87
.20
.00
tSCENDTNG
.040
.480
.850
.00
.29
.000
.100
.15
.60
.75
.30
.910
.00
.00
.00
.00
.07
.34
.38
.40
.50
.50
.60
.60
.72
.72
.72
.36
.90
.00
.00
.00
.00
.00
.00
.CO
.00
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
.09
.10
.13
.15
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.20
.25
15. CO
15.00
15
.CD
15.00
15,
.00
15.00
RANGE
VARIANCE
ORDfTR
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15,
15.
15.
15.
15.
15.
15.
15,
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
15.
25
26
28
30
34
36
40
45
46
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
60
60
60
60
70
70
70
70
70
70
70
75
75
75
80
80
STD.
COEF.
15.
15,
15.
15.
15.
15.
15.
15.
15,
15,
15,
15,
16,
16.
16,
16,
DEV.
VAR.
80
80
80
80
as
90
90
90
90
90
90
90
00
00
00
00
20.00
2.28
1.51
9.31
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
18
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
.20
.25
.25
.25
.25
.25
.29
.30
.3C
.30
.30
.^0
.30
.36
.36
.40
16.00
16
16
16
16
16
16
16
16
16
16
16
16
16
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
16.00
16
16
16
16
16
16
16
16
16
16
16
16
.04
.09
.10
.10
.10
.10
.15
.16
.17
.20
.20
.20
UNITS - MICROGR4MS PtR CUBIC METTR
C.I. (UPPER) 16.38
c.i. (LOWER) 16.03
SKEWNESS 4.41
ACCURACY .82
16.40
16.40
16.40
16.40
16.42
16.45
16.50
16.50
16.50
16.50
16.50
16.50
16,50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.52
16.52
16.52
16.58
16.60
16.60
16,60
16,60
16.60
16.60
16.60
16.60
16.66
16.56
16.70
16.70
16.70
16,70
16.80
16.80
16.82
16.82
16.86
16.87
16.93
16.96
17.00
17.00
17.00
17.00
17.00
17.00
17.OO
17.00
17.00
17.CO
17.00
17.00
17.00
17.00
17.07
17.25
17.25
17.25
17.28
17.30
17.30
17.30
17.30
17.30
17.30
17.39
17.39
17.40
17.40
17.50
17.50
17.50
17.50
17.50
17.70
17.75
17.75
17.83
17.83
17.90
18.00ft
18.00*
18.000
18.100
18.25 '
18.300
18.500
18.50
18.50
18.60
18.70
19.ftO
19.501
19.81
20.00
20.26
21.00
21.501
21.81
21.901
29.501
32. CO
3ft.C0<
120
-------
INTER -LABOR*! DRY STUDY 377
POLLUTANT - CO UNITS - HICROGRMS PER CUBIC METER
SAMPLE NUMBER - 3
N 275 RANGE 29.50 C.I. (UPPER) 32.73
TRUE-VALUE 31.72 VARIANCE 5.17 C.I. (LOWER) 32.19
MEAN 32.46 STD. DEV, 2.27 SKEWNESS .09
MEDIAN 32.20. COEF. VAR. 7.00 ACCURACY 1.51
OAT ft IN *SCE
3.550
13.480
17.480
14.00
15.00 •
18*050
27*500
27.610
28*500
28*50
29.00
29.00
29.50
29.60
29.90
30.00
30.00
30.00
30.00
30.10
30.10
30.20
30.20
30.20
70.25
. 30.25
30.40
30.41
30.41
30.41
30.50
30.50
30.50
30.50
30.54
30.54
30.60
30.60
30.60
30.64
30.67
3C.80
30.92
NDTNG ORDER
31,
31
31
31
31
31
31
31
31
31
31
31
31
31
31
31
.00
.00
.00
.00
.00
.00
.00
.GO
.00
.00
.00
.00
.00
.00
.00
.00
3 1. 00
31.18
31,20
31,20
31,20
31,20
31.25
31,27
31,30
31,30
31.40
31.40
31.40
31,48
31,50
31*50
31*50
31*50
31.50
31.50
31.5O
31*50
31.50
31.50
31,50
31.50
31. 50
31.56
31.60
31.60
31.60
31.63
31.70
31.70
31.70
31.70
31*70
31.75
31.75
31.75
31.75
31.78
31*80
31.
31.
31.
31.
31*
31.
31.
31*
31.
31*
31.
31.
31.
31.
31.
31.
80
80
80
80
80
80
81
85
90
90
90
90
90
90
90
90
31.97
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.00
32.10
32.10
32.20
32.20
32.20
32.20
32. ?0
32.25
32.25
32.25
32.75
32.25
32.70
32.30
32.30
32.30
32.75
32.40
32.40
32.40
32.40
32.40
32.40
32.49
32.50
32.50
32.50
32.50
32.50
32.50
32.50
32.50
32.50
32. SO
32.50
32,50
32.50
32*50
32.50
32.50
32. SO
32,50
32.59
32.60
32.60
32,60
32.61
32,61
32.61
32.67
32.70
32.70
32.70
32.75
32.78
32.80
32.80
32.80
32.86
32.90
32.90
32.94
32.97
73.00
33.00
33.00
33.00
33.00
33.00
33.00
33.00
33.00
33.00
33.00
33.04
73.05
33.09
33,10
33.20
73,20
33.20
33.40
33.40
33.40
33.41
33.50
33,50
33*50
33.50
33.50
33,50
33,50
73,50
33,50
37.54 37
33.70 37
33.75 78
77.76 38
3^.80 79
37.80- 39
37.88 40
37.90 41
73.91 4?
34.00 47
34.00 4U
74.00 69
34.00
34.00
34.00
34.18
34.20
34.26
34.31
74.37
34.50
34.50
34.50
34.60
74.60
34.78
34.78
75.000
35.000
35.00
3^.00
35.22
35.25
35.500
35.60
36.00
36.300
36.30
36.40
36.500
36.61
37.00
37.00
.33
.50
.200
.70
.63
.79
.50
.00
.000
.49
.50
.000
121
-------
INTER-tABORATORY STUDY
977
POLLUTANT - CO
SAMPLE NUMBER -
UNITS - HICROGRAMS PER CUBIC WETFR
N
TRUE-VALUE
MEAN
MEDIAN
DATA
243
4.79
4.60
4.50
IN ASCENDING
?.5C
2.70
2.ac
2.89
2.90
2.95
3.00
1.04
3.20
3.20
3.20
3.30*
3.30
3.40
3.48
3.50^
3.50
3.50
3.50
3.65
3.66
3.709
3.70
3.73
3.75
3.759
3.80
3.80
3.80
3.80
3.80
3.80
3.80
3.80
3.88
3.90
3.90
3.90
3.96
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.01
4.01
4.01
4.02
4. 07
4.09
ORDER
4.10
4.10
4,10
4.10
4.10
4.10
4.10
4,10'
4.10
4.10
4.11
4.12
4.14
4.15
4.15
4.20
4.20
4.20
4.20
4.20
4.20
4.20
4.20
4,25
4.27
4. 30
4.30
4.30
4. 30
4.30
4,30
4.30
4.30
4.3C
4.30
4. 30
4. 30
4.35
4,35
4.35
4.35
4. 39
4.40
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
4.40
4.40
4«40
4.40
4.40
4.40
4.40
4.40
4.40
4.43
4,44
4.44
4.44
4.44
4.44
4.50
4.50
4.50
4.50
4.50
4.50
4.50
4.50
4.50
4.50
4.SO
4.50
4.50
4.50
4.50
4.50
4.50
4.58
4.50
4.50
4.50
4.50
4.50
4.50
4.50
4.50
4.^0
4.50
4.50
4.50
4.50
4.*0
4.50
4.50
4.50
4.50
4.50
4.53
4.55
4.55
4.55
4.56
4.59
122
12.00
.88
.94
20.42
4.59
4.SO
4.60
4.60
4.60
4.60
4.60
4.60
4.69
4.70
4.70
4.70
4.70
4.70
4.70
C.I. (UPPER)
C.I. (LOWER)
SKEWNESS
ACCURACY
4.70
4.70
4.70
4.70
4.70
4.71
4.71
4.74
4.75
4.75
4.78
4.80
4.80
4.80
4.80
4.80
4.80
4.80
4.80
4.80
4.80
4.80
4.80
4.81
4.81
4.82
4.83
4.84
4.87
4.90
4.90
4.90
4.92
4.96
5.00
5.00
5.00
5.00
5:. 00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
5.00
4.72
4.49
4.87
-6.05
5.00
5.09
5.10
5.10
5.10
5.10
5.10
5.15
5.15
5.15
5.20
5.20
6.0
5.0
8,0
6.0
6.3
5.3
6,5
6.7
6.7
7.3
8.3
9.3
5.20 13.5
5.20 14.5
5.20
5.20
5.22
5.22
5.25
5.25
5.25
5.27
5.3P
5.43
5.489
5.50*
5.50
5.50
5.50
5.50
5.50
5.50
5.50
5.60
5.60
5.7H
5.70
5.70
5.80
5.90
5.90
6.00
6.00
-------
INTrR-LABORATORY STUDY
POLLUTANT - CO
SAMPLE NUMBER - 2
N
TRUE-VALUE
"EAN
MEDIAN
244,
16.59
16.71
16.68
DATA IN ASCENDING
977
UNITS - MICROGRAMS PFR CUBIC METER
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
9.30
1.07
1.04
6.20
11.70
13.50
13.70
13.900
14.000
14.20
14.500
14.75
14.78
14.30
14.90
14.90
14.90
14.90
14.90
15.00
15.00
15.05
15 .OS
15.0S
15.10
15.20
15.20
15.40
15^49
15.50
15.50
15.50
15.50
15.50
15.50
15.50
15. 5S
15.60
15. SO
15.64
15.70
15.70
15.70
15.75
15.80
15.30
15.80
15.80
15.80
15. 8S
15.89
15.99
15.99
16.00
IS. 00
16.00
16.00
IS. 00
16.00
1S.OC
IS. 00
16.00
IS. CO
15.00
IS, 00
16.00
IS, DC
IS, 00
16,00
IS, 00
IS. 00
IS. 09
IS. 10
IS. 10
IS. 10
IS. 10
16. 13
IS. 16
IS. 17
IS, 20
IS. 20
IS, 20
IS, 20
IS, 20
16.20
16.20
IS. 20
IS. 20
IS. 25
16,27
IS. 30
IS. 30
IS. 30
IS. 30
IS. 30
16.30
IS. 30
IS. 30
IS. 30
16.30
IS, 30
IS. 30
IS, 30
16.38
IS. 38
IS. 38
IS, 40
IS. 40
16.40
IS. 40
IS. 41
16.47
IS, 47
IS, 50
IS. 50
16.50
16*50
IS. 50
IS. 50
IS. 50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.50
16.55
16.55
16. SO
16. SO
16.SO
16. SO
16. SO
16. SO
16. SO
16.62
16.65
16.67
16. S3
16. S3
16.70
16.70
16.70
16.70
16.70
16.70
16.73
16.73
16.74
16.75
16.75
16.76
16.77
16.80
16.80
16.30
16.80
16.80
16.30
18.30
16.82
16.82
16.S5
16.90
16.90
16.90
16.90
16.90
16.90
16.99
17.00
17.00
17.00
17.00
17.00
17.00
17.00
17.00
17.00
17,00
17.00
17.00
17,00
17.00
17.00
17.00
17.00
17,00
17.04
17.08
17.10
17.10
17.10
17.11
17.20
17.20
17.20
17.20
17.30
17.30
17.30
17.30
17.30
17.30
17.30
17.30
17.30
17.30
17.30
17.30
17.39
17.40
17.44
17.44
17.45
17.50
17.50
17.50
17.50
17,50
17.50
17.50
17.50
17.50
17.50
17.50
17.50
123
-------
INTER-LABORATORY STUDY
POLLUTANT - CO
977
UNITS - HICROGRAMS: PFR CUBTC METER
SAMPLE NUMBER -
N
TRUE-VALUE
MEAN
MEDIAN
244
36.00
36.68
36.40
RANGE
VARIANCE
STD. DEV.
COEF. VAR
1
•
3.20
3.00
1.73
4.72
C.I. (UPPER)
C.I. (LOWER)
SKEWNESS
ACCURACY
»._* * " '
36.90
36.46
1.07
1.12
DATA IN ASCENDING ORDER
29.000
30.80
31.400
32.100
32.17
32.30
33.10
33. 3C
33.50
33.90
34.00
34.20
34.30
34.50
34.50
34.50
34.50
34.50
34.60
34.60
34.62
34.65
34.70
34.90
34.90
34.95
35.00
35.00
35.00
35.00
35.00
35.00
35.00
35.01
35.10
35. ID
35.20
35.20
35. 2D
35.20
35.25
35.29
35.30
35.30
35.32
35.40
35.40
35.41
35.45
35.50
35.50
35.50
35.50
35.50
35.50
35.50
35.50
35.50
35. 50
35.50
35.57
35.58
35. 60
35. 60
35. 60
35. 60
35.68
35.68
35.70
35.80
35. 80
35.80
35. 80
35,80
35. 80
35.82
35. 82
35.83
35. 85
35.90
35.90
35.90
35.90
35.90
35.90
35,90
35.91
35.91
35.92
35.95
35.99
35.99
36. 00
36.00
36.00
36.00
36.00
36.00
36. GO
36. CO
36. 00
36,00
36.00
36.00
36.00
36.00
36.00
36.00
36.09
36.10
36.10
36.10
36.10
36.10
36.10
36.12
36. 17
36.20
36.20
36.20
36.20
36.20
36.20
36.22
36.24
36.25
36.25
36.25
36.25 »
36.25
36.28
36.30
36.30
36.30
36.30
36.30
36.31
36.31
36.36
36.38
36.40
36.40
36.40
36.41
36.45
36.45
36.50
36.50
36.50
36.50
38.50
36. SO
36.50
36.50
36.50
36.50
36.50
36.50
36. SO
36.50
36.50
124
36.50
36.51
36.51
36.60
36.60
36.60
36.60
36.60
36.62
36.66
36.70
36.70
36.70
36.70
36.70
36.70
36.713
36.70
36.80
36.90
36.90
36.90
36.95
36.96
37*00
37.00
37.00
37.00
37.00
37.00
37.00
37.00
37.00
37.00
37.00
37.PO
37.00
37.00
37.00
37.00
37.00
37.00
37.10
37.10
37.20
37.20
37.20
37.20
37.2TJ
37.39
37.40
37.40
37.40
37.50
37.50
37.50
37.50
37.50
37.50
37.50
37.50
37.50
37.50
37.50
37.50
37.60
37.60
37.70
37.70
37.80
37.80
38.00
38.00
38.00
38.00
38.00
38.00
38.00
38.00
38.00
38.02
38.10
38.20
38.20
38.21
38.40
38.40
38.40
38.50
38.50
38.50
38.60
38.70
38.80
38.80
39.00
39.00
39.00
39.00
39.00
39.10
39.10
39.13
39.30
39.30
39.44
39.44
39.50
39.50
39.800
40.00
40.00
40.00
40.00
40.50
40.900
41.00
41.55
42.001
43.001
43.20|
43.25
43.48
43.50
44.00
44.00
44.09
45.89
51.22
-------
INTER-LABORATORY STUDY
POLLUTANT - SCI
SAMPLE NUMB
- 1
1076
UNITS - MICROGRAKS PER CU8TC METTR
N
91
TRUE-VALUE 1.22
MEAN
MEDIAN
DATA IN
1.9S
1.65
ASCENDING
.09
.09
.339
.339
.39
.10
.10
.10
.53
.60
.60
.60
.60
.70
.80
.80
.30
.83
.90
.96
.96
.99
ORDER
1.00
1.00
1.01
1.01
1.05
1.10
1,15
1.17
1.19
1,20
1.22
1.23
1,21
1,28
1.29
1.30
1.30
1,35
1.15
1,50
1.50
1.50
RANGE
VARIANCE
STD. OEV.
COEF. VAR
7.81
2.05
1.13
72.85
1.51
1.51
1.56
1.56
1.62 .
1.61*
1.68
1.70
1.70
1.75
1.77
1.77
1.80
1.80
1.80
1.80
1.80
1.820
1.90
1.92
1.92
1.95
C.I.CUPPERt
C.I.tLOWE R>
NEWNESS
ACCURACY
2.25
1.68
1.96
2.02
2.22
2.23
2.29
2.10
2.10
2.10
2.10
2.10
2.10
2.15
2.50
2.50
2.70
2.70(
2.71
2.90
3.00
3,00
3.20
3.30
3.30
3.71
3.75
3.76
1.35
1.11
1.69
1.78
5.10
5.17
6.960
7.61
7.90
7.950
9.360*
11.00^
11.109
125
-------
INTER-LABORATORY STUDY
1076
POLLUTANT - S04
SAMPLE NUMBER - 3
UNITS - MICROGRAWS PER CUBIC METER
N 39
TRUE-VALUE .00
MEAN 1.2*
MEDIAN .90
DATA IN ASCENDING ORDER
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
5.76
1.58
1.26
101.17
C.I.CUPPER) 1.6*
C.I.CLOWE RI .85
SKEWNESS 1.51
ACCURACY .00
.00
.05
.06
.11
.12
.16
.19
.2*
.25
.27
. 30
.30
.12
.60
.60
.80
.80
.90
.90
.96
.98
1.20
1.30
1.19
1.50
1.620
1.68
1.73
1.75
2.50
2.85
3.20
3.21
3.630
3.68
4.800
5.80
6.19*
11.100
126
-------
INTFR-LABORATORY STUDY
POLLUTANT - S04
SAMPLE: NUHBER - n
N 48
TRUE-VALUE 3.01
MEAN 3.26
MEDIAN 3.10
DATA IN ASCENDING ORDER
107G
UNITS - HICROGRAMS PER CUBIC METER
RAN€E
VARIANCE
STD. 0EV.
COEF. VAR.
6.93
1.75
1.32
40.62
C.I.tUPPERI 3.64
C.I.CLOWE Rl 2.89
SKEMNCSS ,8*
ACCURACY 3.09
.09
• G0
1.30
1.8941
2.1*1
2.2*
2.40
2.45
2. 5S
2,61
2. B7
2.70
2.86
2.88
2.90
2.91
3.00
3.00
3,00
3.00
3.10
3.10
3.11
3.12
3.18
3.18
3.21
3.25
3.30
3.30
3.30
3.36
3.45
3.48
3.50
3.SSQ
3.95
4.05
4.08
4.10
4.330
4.44
4.50
5.00
5.80
6.30
6.90
7.02
8.690
8.380
9.00ft
127
-------
INTER-LABORATORY STUDY
1D76
POLLUTaNT - SC4
SAMPLE NUMBER - 8
N 99
TRUE-V*LUE 10.22
MEAN 10.15
MED UN 10.10
DATA IN ASCENDING ORDER
UNITS - MICROGRAHS PEP CUBIC METER
RANGE
VARIANCE
STD. OEY.
COEF. VAR.
14.01
2.61
1.62
15.93
C.I.CUPPER) 10.47
C.I.CLOWE Rl 9.83
SKEWNESS -.49
ACCURACY -1.16
1.80
3.85*
5.540
7.50
7.52
7.52
7.68
8.58
8.70
8.70
8.70
8.39
9.00
9.00
9.00
9.00
9.C6
9.10
9.12
9.12
9.15
9.15
9.20
9.24
9.30
9.30
9.30
9.33
9.38
9.40
9.40
9.45
9.50
9.59
9.60
9.64
9.66
9.69
9.70
9.75
9.75
9,79
9.80
9.80
9.81
9.83
9.84
10.00
10.08
10.10
10.10
10.10
10.19
10.20
10.20
10.20
10.21
10.33
10.37
10.38
10.38
10*39
10.40
10.40
10.42
10.49
10.50
10.50
10.50
10.53
10.60
10.60
10.60
10.62
10.63
10. 65
10.65
10.70
10.70
10.93
11.00
11.03
11.23
11.25
11.25
11.49
11.50
11.52
11.52
11.52
11.55
11.64
11.70
11*82
12.
12.1?
12.60
12.60 •
12.729
13.00*
13.33
14.10
14.400
15.120
15.30
15.81
17.640
18.630
20.000
20.900
128
-------
INTER-LA 8 OR AT DRY STUDY
277
POLLUTANT - SOI
SAMPLE NUHBER - C
H
TRUE-VALUE
MEAN
MEDIAN
110
21.15
21.311
21.10
DATA IN ASCENDING ORDER
UNITS - NICROGRAHS PER CUBIC METER
RANGE
VARIANCE
STD. 0EV.
COEF. VAR
58.21
29.69
5.45
25.53
C.I.fUPPERI 22.36
C.I.fLOWE RI 20.33
SKEHNESS 7.30
ACCURACY -.28
IT. 96
11.16
15. OC
15.15
15.504
15.66
15.77
15.81
16.32
16.95
17.20
17.20
17.13
17.90
18.70
18.90
19.04
19.26
19.50
19.50
19.50
19.50
19.62
19,65
19.75
19.76
19,76
19.80
19.80
19. 80
19.88
19.88
19.92
19.95
19.95
20.00
20.10
20.15
2D. 18
20,10
20, 10
20. 10
20.10
20.18
20,50
20.51
20.60
20.70
20.71
20.76
20.79
20.81
21.00
21.00
21.00
21.10
21.10
21.11
21.16
21.20
21 .?Q
21.2D
21.21
21.36
21.10
21.13
21.18
21.70
21*72
21.75
21.80
21.90
21.90
21.90
21.90
21.90
22.00
22.00
22.00
22.00
22.05
22.05
22.15
22.20
22.50
22,50
22.50
22.50
22,50
22.56
22.56
22.75
22.86
22.90
?3.0?
23.02
23.20
23.30
23.10
23.76
23.90
21.00
21.31
21.18
21.18
25.30
25.11
25.56
25.90
27.00
31.80<
38.10<
72.20
129
-------
INTtR-tABORATORY STUDY
277
POLLUTANT - SC4
UNITS - HICROGRAMS PER CUBIC METER
SAMPLE NU*3ER
N
TRUE-VALUE
MEAN
MEDIAN
v>
107
15.63
IS .20
15.90
RANGE
VARIANCE
STD. DEV.
COEF. VAR
51.12
31.19
5.58
34.43
C.I.fUPPERI 17.26
C.I.fLOWE Rl 15.
SKEHNESS 4.54
ACCURACY 1.73
DATA IN ASCENDING ORDER
1.14
4.95
9.00
10.80(
11.40
11.40
13.05
13.08
13.24
13.35
13.35
13.44
13.50
13.50
13.60
13.80
14.00
14.00
14.10
14.30
14.32
14,34
14.40
14.40
14.40
14. 40
14. 46
14.70
14.79
14.85
14.90
15.00
15.00
15.00
15.06
15.07
15.10
15.12
15.27
15,30
15.36
15.45
15.50
15.55
15.56
15.59
15.64
15.75
15.80
15.83
15.89
15.90
15.90
15.90
15.90
16.00
16.02
16*04
16.07
16.10
16.15
16.17
16.20
16.20
16.20
16.23
16.25
16.30
16.30
16.30
16.32
16.35
16.35
16.40
16.50
16.50
16.50
16.50
16.50
16.66
16.72
16.80
16.80
16.98
17.10
17.20
17.21
17.2&
17.28
17.28
17.28
17.30
17.30
17.49
17.52
17.59
17.85
18.00
18.00
19.34
19.34
19.43
19.44
20.99
24.57
28.50*
29.40*
51.17
52.26
130
-------
INTER-H8ORATORY STUDY
POLLUTANT - S04
SAMPLT NUMB
- 7
277
UNITS - MICROGRAMS PER CUBIC HETFR
N
TRUE-VALUE
MEAN
MEDIAN
56
24.87
24.25
24.41
RANGE
VARIANCE
STO. DEV.
COEF. VAR.
23.18
12.88
3.59
14.80
C.I.fUPPERI 25.19
C.I.CLOWE Rl 27.31
SKEMNESS -.37
ACCURACY -1.83
DATA IN ASCENDING ORDER
12.82
1U.8D
15.69
17.10
20.10
20.82
21.70
21.90
21.90
22.20i
22.70
23.02
23.21
23.27
23. <*0
23. «T8
23.52
23.60
. 30
2«l. 00
24, GC
24.00
21.20
. 25
24.30
24.32
24.38
24.45
24.50
24.51
24.60
24.75
25*01
25.10
25.10
25.10
25.20
25.20
25.30
25.43
25.44
25.51
25.80
25.86
25.92
25.95
26.00
26.40
26.40
26.70
27.60
28.70
30.46
32.10
36.00
46.20<
131
-------
INTFR-LABORATORY STUDY
277
POLLUTANT - SC4
SAMPLT NUMBER - 9
N 53
TRUE- VALUE 4.47
MEAN £.14
MEDIAN 4.62
DATA TN ASCENDING ORDER
UNITS - MICROGRAWS PFR CUBIC METER
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
41.64
40.87
6.39
104.11
C.I.IUPPER! 7.86
C.I.CLONE R) 4.42
SKEHNESS 4.OS
ACCURACY 3.36
.32
1.02
2.06
3.00
3.05
3.10
3.301
3.35
3.75
3.90
3.98
4.06
4.10
4.11
4.20
4.20
4.20
4.30
4.40
4.50
4,50
4.50
4.57
4.60
4.60
4.50
4.60
4.62
4.63
4.80
4.80
4.86
4.95
4.98
5.04
5.18
5.20
5.29
5.30
5.37
5.40
5,43
5.55
5.62
5.P8
6.00
6.24
6.40
6.47
10.08
11.40
20.97
25.64
41.96
132
-------
INTER-LABORATORY STUDY
877
POLLUTANT - SO*
SAMPLE NUW9ER - 3
N 40
TRUE-VALUE .00
MEAN .99
MEDIAN .60
DATA IN ASCENDING ORDER
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
ST!. OEV.
COEF. VAR.
4.96
1.12
1.06
105.75
C.I.(UPPER) 1.72
C.I.CLOWE Rl .66
SKEWNESS 1.81
ACCURACY .00
,09
18
18
18
20
25
25
28
30
30
30
30
,55
,60
,60
,60
50
,73
.80
.90
.99
1.08
1.20
1.20
1,40
1.44*
1.50
1.72
1.82
1.88
2.16
2.34
2.40
3.00
3.10
3.12*
5.05
133
-------
INTER-LABOR AT DRY STUDY
877
POLLUTANT - S04
SAMPLE NUMBER - 4
N 59
TPUE-VALUE 11.10
MEAN 10 .76
MEDIAN 10.83
DATA IN ATCENDTNG ORDD?
UNITS - MICROGRAMS PER CUBIC METER
RANSE
VARIANCE
S^T). DEV.
COEF. VAR.
11.20
3.66
1.91
17.78
C.I.CUPPER)
C.I.fLOME R) 10.27
SKEWNESS .98
ACCURACY -2.43
2.400
5.30
5.76
6.900
7.55
7.80
8.580
9.00
9.00
9.55
9.60
9.60
9.60
9.67
9.80
9.82
9.84
9.30
9.30
9.99
10.00
10.02
10.03
10.09
10.10
10. 1H
10.35
10.50
10.70
10.70
10.73
10.73
10.30
10.83
10.87
11.00
11.00
11.03
11 .OH
11.06
11.10
11.10
11.18
11.19
11.25
11.28
11.40
11.40
11.60
11.69
11.71
11.80
11.80
11.88
11«88
12.00
12.23
12.36
12.90
12.360
13.51
14.55
19.50
134
-------
INTER-LABORATORY STUDY
877
POLLUTANT - S04
SAMPLE NUMBER - 5
UNITS - MICROGRAWS PER CUBIC METER
N
TRUE-VALUE
MEAN
MEDIAN
60
13.50
15.70
13.26
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
181.02
192.95
22.20
141. 45
C.I.CUPPERJ 21.31
C.I.CLOWE Rl 10.08
SKEWNESS 7.70
ACCURACY -1
DATA IN ASCENDING ORDER
2.58
2.70
2.32®
6.90*
7.50
9.06 '
10.20
10.20
ai.oc
11.16
11.84
11.90
12.00
12.00
12.06
12.19
12.30
12.30
12.38
12.15
12.60
12.60
12.60
12.62
12.71
12.78
12.82
12.84
12.32
12.92
13.10
13.15
13.20
13.33
13.33
13.34
13.38
13.48
13.49
13.50
13.59
13.68
13.70
13,80
13.80
13.80
13.80
13.95
14.00
14. 00
14.11
14.41
14.46
14.55
14.90
15.00
15.00
15.06
15.43
15.
15.91
15.98
21.30
183.60
135
-------
INTER-i-ABORATORY STUDY
877
POLLUTANT - SOI
SAMPLE NUMB
UNITS - MICROGRAMS PER CUBIC METER
N 58
TRUE-VALUE 1^.20
MEAN 15.91
MEDIAN 15.45
DATA IN ASCENDING ORDER
RANGE
VARIANCE
STD. DEV.
COEF. V4R.
17.90
S.93
2.64
15.61
C.I.fUPPER)
C.I.CLOWE
NEWNESS
ACCURACY
16.59
> 15.23
2.18
-4.60
.780
3.220
9. 000
10.00
12.00
12.30 •
12.48*
12.77
13.50
13.30
14.00
14.09
14.15
14. 20
14.55
14.82
14.90
14.97
15.00
15.01
15.02
15.06
15.10
15.12
15.12
15.14
15.15
15.18
15.24
15.30
15.30
15.34
15.36
15.42
15.49
15.57
15.72
15.74
15.77
15.80
15.90
15.94
16.00
16.06
16.12
16.20
16.20
16.23
16. 35
16.43
16.50
16.68
16.80
17.59
17.60
17.62
17.73
18.00
18.24
18.54
20.97
24.480
25.30
27.90
136
-------
INT FR-LABORATORY STUDY
877
POLLUTANT - SO*
SAHPLF NUMBER - 7
N 58
TRUE-VALUE 3.00
MEAN 3.02
MEDIAN 3.01
DATA IN ASCENDING ORDER
UNITS - MTCROGRAWS PER CUBIC METER
RANGE
VARIANCE
S-H?. DEV.
COEF* VAR.
1*60
1.27
41.86
C.I.fUPPER) 3.
C.I.CLOWE Rl 2.70
SKEWNESS .13
ACCURACY 1.50
.20
.30
.51
.79
.30
1.Q5
1.68
1.80
1.98
2.00
2.02
2.25
2,31
2. GO
2.68
2.70
2.70
2.73
2.73
2.75
2.81
2.90
2.93
2.95
2.96
3.00
3.00
3.00
3.02
3.03
3.06
3.12
3.12
3.11
3.15
3.16
3,20
3.21
3.21
3.27
3,30
3.30
3.36
3.39
3, SO
3.70
3. TO
3,76
3.81
1.00
1.10
1.20
1.370
1.11
5.05
5.18
5.32
6.00
6.60
7.200
137
-------
I NT FR-LABOR AT DRY STUDY
877
DOLLUTftNT - SOU
SAMFLF NUMBER - 8
N
TRUE-VALUE
WEAN
MEDIAN
2.10
2.50
2.30
DATA IN ASCENDING OROET?
UNITS - MICROGRAMS PFR CUBIC METTR
RANGE
VARIANCE
ST. OEV.
COTF. VAR.
6.37
1.33
1.13
47.03
C.I.CUPPER! 2.81
C.I.CLOWE R> 2.19
NEWNESS 1.07
ACCURACY 9.52
.04
.43
.50
.60
1.30
1.40
1.58
1.71
1.78
1.80
1.80
1. 89
1.92
1.97
1.98
2.00
2.00
2.00
2.04
2.16
2.18
2.22
2,23
2.26
2.28
2.28
2.30
2.30
2.40
2.40
2.40
2.40
2.40
2.40
2.50
2.55
2.55
2.55
2. SO
2.60
2.70
2.70
2.88
2.94
3.00
3.10
3.16
3.21
3.590
3.84
4.320
4.68
4.70
5.05
5.86
6.41
3.50
138
-------
INTER-LABORATORY STUDY
1075
POLLUTANT - N03
NUMBER - 1
N 72
TRUE-VALUE .51
HEAN .6*
MEDIAN .61
DATA IN ASCENDING ORDER
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
DEV.
. VAR.
1.51
.07
.27
11.06
C.I.fUPPERI
C.I. (LOWE R>
SKEW NESS
ACCURACY 18.5?
53
.13
.13
.15
.17
.18
.30
50
50
50
50
51
55
55
56
57
57
,58
,58
,59
,53
, EC
60
60
60 .
60
61
61
61
61
61
61
66
,66
,66
,63
,70
,71
.72
,75
.76
.78
.80
.80
.80
.81
.81
.83
.83
.87
.87
.90
.90
.95
.96
.98
.99
1.00
1.02
1.01
1.05
1.10
1.61
4.32*
6.11*
7.01*
139
-------
INTTR-LA30R ATO^Y
1076
POLLUTANT - N03
SAMPLr NUMBER - 3
N 30
TRUE-VftLUE .00
MEAN .23
MEDIAN .26
DATA IN ASCENDING ORDER
UNITS - MICROGRAMS PER CUBIC METH?
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
• 03
• 16
70*11
C.I*CUPPER> .29
C.I.CLOWE Rl ,17
SKEWNESS -.02
ACCURACY .00
.01
.01
.01
.02
.02
.OS
.0?
07
10
10
1C
11
18
19
.25
.27
.28
.30
.33
.33
,78
3.67*
140
-------
INTER-IABORATORY TTUOY
POLLUTANT - N01?
SAMPLT NUMBER - $
N 37
TRUE-VALUE 9.42
MEAN 7.1R
MEDIAN 8.40
DATA IN ASCENDING ORDER
1075
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
STO. DEV.
COEF. VAR.
11.50
10.79
3.28
^5.83
C.I.fUPPERJ 8.22
C.I.fLOWE R> <5.
SKEWNtSS -.80
ACCURACY -1D.S2
.20
.22(
.28
1.8C
1.95
2.02
2.15
2.22
3.12
4.56
5.07
e. oo
E. 25
£. EC
7.35
7. 90
8.10
8.20
3.28
3.40
S.40
8.90
9.13
9.25
9.27
3.30
9.38
9.47
9.56
9.57
9. GO
9.65
9.70
9.80
10.C?0
10.30
11.37
11.70
13.901
27.23<
141
-------
INTER-LABORATORY STUDY
1075
POLLUTANT - N03
UNITS - MICROGRAMS PER CUBIC METER
SAMPLE NUMBER - 8
N 74
TRUE-VALUE 3.39
MEAN 3.27
MEDIAN 3.45
DAT ft TN ASCENDING ORDER
RANGE
VARIANCE
STfl. DEV.
COEF. VAR.
8.07
2.26
1.50
46.01
C.I. CUPPER*
C.I.CLOWE R>
SKEWNESS
ACCURACY
3.51
2.93
1.05
.57
.57
.75
.77
.80
.31
.81
.82
1.04
1.05
1.40
1.56
1.S2
7.19
2.22
2.22
2.SO
2.70
2. 70
3.18
3. 19
3.20
3.20
3.20
3.20
3.20
3. 20
3. 2*»
3.26
3,30
3. 33
3. 3*
3.40
3.40
3.42
3.45
3.45
3.45
3.46
3.48
3.48
3.48
3.50
3.50
3.50
3.51
3.54
3.55
3.55
3.56
3.58
3.59
3.60
3*61
3.63
3.63
3.65
3.72
3.72
3.74
3.80
3.90
3.97
4.01
4.08
4.13«
4.20
4.2fl
4.27
4.32
4.34
5.19
7.280
7.280
8.30
8.40
8.64
11.350
13.420
142
-------
INTER-LABORATORY STUDY
POLLUTANT - N03
SAMPLE NUMBER - 0
277
UNITS - MICROGRAWS PEP CUBIC METTR
N
TRUE-VALUE
HEAN
MEDIAN
76
10.32
10 .6?
11.00
DATA IN ASCENDING ORDER
RANGE
VARIANCE
STD. DEV.
CCEF. VAR.
9.75
2.41
1.55
14.55
C.I.(UPPER! 11.02
C.I.CLOWE R* 10.32
SKEWNESS -.60
ACCURACY .73
.740
.83®
2.13*
2.250
2.52®
2.52®
2.52®
2.550
5.20
6.SO
7.02
7.05
7.1«>
7.44
7.74
8.65
9.04
9.12
9.26
9.57
9.60
9.60
9.72
9.82
10.02
10.22
10.22
10. 30
10.31
10.40
10,40
10.50
10.58
10.60
10.61
10.59
10.70
10.74
10.78
10.30
10.80
10. 80
10.84
10.94
10.98
10.98
11.02
11.09
11.10
11.10
11.10
11.13
11.15
11.15
11.17
11.17
11.20
11.20
11.
11. ?2
11. T5
11. TO
11.31
11.35
11.42
11.50
11.51
11.59
11. SO
11.61
11.70
11.78
11.89
11.98
12.00
12
12
12
12
12
12
13
,00
,07
,07
,36
,75
8?
18
95
17
19
99,
122
133,
173,
,30®
00®
79®
,00®
,03®
143
-------
INTER-LABORATORY STUDY
277
POLLUTANT - N03
SAMPLE" NUMBER - 5
N 75
TPUE- VALUE 5.17
MEAN 5.27
MEDIAN 5.30
DATA TN ASCENDING ORDER
UNITS - MICROGRAHS PER CUBIC METER
RANGE
VARIANCE
STO. DEV.
COEF. VAR.
4.C4
.54
.73
13.83
C.I. CUPPER*
C.I. CLOWE R>
SKEWNESS
ACCURACY
5.44
5.11
.44
2.51
.54 %
.71*
1.29
1.32%
1.34%
T.70
3.78
3.78
3.93
4.05
4.11
4.20
4.20
4.25
4.25
4.54
4. BO
4.63
4.70
4.30
4.80
4.80
4.95
5.02
5.05
5.07
5.10
5.10
5.10
5.10
5.10
5.12
5.13
5.15
5.17
5.18
5.18
5.20
5.20
5.24
5.25
5.27
5.309
5.30
5.30
5.30
5.30
5.37
5.38
5.39
5.40%
5.40
5.42
5.45
5.45
5.47
5.50
5.53
5.53
5*55
5.55
5.55
5.56
5.60
5.63
5.S4
5.GS
5.
5. 35
5.85
5.58
6.10
5.10
5.11
6.20
6.36
6.60
6.67
7.46
7.74
9.78
11.34
40.939
56.Q00
57.079
132.00%
144
-------
INTTR-LABORATORY STUDY
POLLUTANT - N03
SAMPLE NUMBER - 7
N 40
TRUE-VALUE 1.94
MEAN 2.27
MEDIAN 2.21
DATA IN ASCENDING ORDER
277
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
2.20
• 19
.43
19.03
C.I.(UPPER! 2.40
C.I.CLOWE R> 2.13
SKEWNESS 1.57
ACCURACY 13.92
.15
.57
.60
1.S7
1.72
1.73
1.39
1.30
1.90
1.90
1.91
1.92
1.91
2.00
2.00
2.00
2.05
2,07
2.07
2.10
2.10
2.12
2.20
2.22
2.25
2.25
2.30
2.30
2.31
2.34
2.34
2.39
2.42
2.50
2,50
2.56
2.50
2.61
2.64
2.80
2.86
3.36
3.3^ -
6.07*
18.93*
22.00*
145
-------
NTER-LA50RATQRY STUDY
277
POLLUTANT - N03
NUMBER -
N
TRUE-VALUE
MEAN
MEDIAN
38
7.33
7.21
7.39
DATA TN ASCENDING ORDER
UNITS - MICROGRAHS PFR CUBIC METER
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
1.32
.91
.95
13.17
C.I.fUPPERI 7.55
C.I.CLOWE Rl 6.
SKEWNESS -.77
ACCURACY .20
.55*
1.680
1.711*
1.770
1.56
5.10
5.78
6.15
6. 38
6.33
6.13
E. 70
f. 75
6. 80
6.87
7. DO
7.02
7.20
7.20
7.21
7.36
7.39
7.10
7.15
7.59
7.61
7.65
7.6S
7.70
7.81
7.59
7,90
7.95
7.96
8.00
8.10
8.17
8.33
8.51
8. 66
8.38
11.100
65.190
81. 00*
146
-------
INTER-LABORATORY STUDY
POLLUTANT - N03
SAMPLE NUHB
N
TRUE-VALUE
MEAN
MEDIAN
38
.00
.37
.19
DATA IN ASCENDING ORDER
877
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
S7t>« DEV.
COEF. VAR.
.85
.81
213.02
C.I.eUPPERJ .S2
C.I.CLOWE Rl .11
NEWNESS *».88
ACCURACY .00
,02
,02
,02
,0?
,07
,05
,06
.07
.08
.08
.08
.09
.09
.10
.10
.32
,13
,19
,20
,20
.40
.72
1.30
1.9G
21.300
25
30
.50
147
-------
INTER-LABORATORY STUDY
877
POLLUTANT - N03
SAMPLE NUMBER -
N
TRUE-VALUE 6.00
MEAN 5.3?
MEDIAN 5.78
DATA IN ASCENDING ORDER
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
STD. DEV.
COEF. VAR
9.80
3.58
1.89
35.57
C.I.CUPPER) 5.88
C.I.CLOWE Rl «U76
SKEWNESS -1.07
ACCURACY -3.58
.50
.55
.60
1.31
1.32
1.359
1.1 00
3.5*
«U 67<
4.90
5.10
5.28
5.15
5. 1G
5. 43
5.58
5, 7C
5.73
5.75
5.77
5.77
5.78
5.72
5.82
5.91
5.95
6.00
6.01
6.02
6.03
6.08
6.10
6.10
6.12
6.19
6.19
6.2%
6.59
S.SO
S.76
7.00
7.12
10.30
20.38$
148
-------
INTER-LABORATORY STUDY
POLLUTANT - N03
877
SAMPLE NUMB
N
TRUE- VALUE
MEAN
MEDIAN
DATA IN ASCENDING ORDER
44
4.80
4.00
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
STf). DEV.
COEF. VAR.
5.67
2.50
1.58
39.55
C.I.tUPPER* 4.47
C.I.fLOWE RI 7.53
SKEWNESS -1.40
ACCURACY -3
.10
.25
.38
.47
1.03
1.19
1.52
2.64
2. 66
3.00
3.
3.75
4.10
4,38
4. 33
4.48
4.51
4.53
4. EC
4.63
4.65
4.65
4.68
4.68
4.70
4.72
4.78
4.80
4.89
4.90
4.93
4.94
4.94
4.93
5.00
5.04
5.04
5.OS
5.09
5.10
5.15
5.61
5.77
24.000
149
-------
INTER-LABORATORY
877
POLLUTANT - N03
SAMPLt NUMBER - 6
N 11
TRUE-VALUE 7.80
MEAN 6.17
MEDIAN 7.31
DATA IN ASCENDING ORDER
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
STD. DEV.
COEF. VAR
8.65
4.85
2.2C
. 01
C.I.CUPPER! 7.12
C.I.fLOWE R) 5.82
SKEWNESS -1.65
ACCURACY -5.90
.55
.78
1.S8
1.82
1.359
3.83
*.<*«»
5.05
5.37
5. 70*
5. 8C
S. 1C
S. 42
6. 51
6.60
6.66
7.16
7.20
7.23
7.23
7.21
7.30
7.71
7.18
7.18
7.S1
7.51
7.55
7.59
7.K8
7.73
7.73
7.73
7.85
7.87
7.90
7.90
7.95
7.96
7.97
7.97
8.30
9.19
18.18*
150
-------
INTER-LABOR ATOPY STUDY
POLLUTANT - N03
SAMPLE: NUMBER - 7
N
TRUE-VALUE
HEAN
MEDIAN
.72
1.08
.86
877
UNITS ^ MICROGRAMS PER CUBIC METER
R AN GE
VARIANCE
STD. DEV.
Cf€F. VAR.
8*45
1.63
1.27
118.51
C.I.CUPPER* 1.4S
C.I.tLOWE Rl .£9
SKEMNESS 4.65
ACCURACY 19.44
DATA IN ASCENDING ORDER
.07
.08
.16
.17
.19
.200
.
.50
.SO
-GO
.60
.66
*70
.71
.72
.75
.75
,8C
.81
,81
.84
,85
.86
,86
,87
,88
,89
,90
,9D
.91
.91
.94
.96
.98
1.00
1.06
1*10
1.10
1.15
1.28
1.50
2.20
2.48
2.61
8.52
151
-------
INTERH.AB0RATORY STUDY
877
POLLUTANT - PB
SAMPLE NUMBER - 0
N 54
TRUE-VALUE 1.80
MEAN 1.75
MEDIAN 1.75
DATA IN ASCENDING ORDER
UNITS - MICROGRAMS PER CUBIC METER
RANEE
VARIANCE
STD. DEV.
COEF. VAR
7.59
-85
.92
52.53
C.I-.tUPPERl 2.00
C.I.fLOHE R! 1.51
SKEWNESS 4.71
ACCURACY -2.78
.11
.11
.85
.88
.94
.99
1.28*
1.3?
1.37
1.40
1.5E
1.5B
1.61
1. 65
1.S5
1, EG
1.67
1.E8
1, 68
1.68
1.68
1.70
1.70
1.70
1.71
1.75
1.75
1.75
1.75
1.75
1.77
1.77
1.78
1.79
1.80
1.80
1.80
1.81
1.82
1.87
1.88
1.88
1.88
1.88
1.90
1.90
1.91
1.94
1.98
1.98
2.01
2.14
2.16
7.70
18.24*
23.40^
152
-------
INTER-LABOR AT ORY STUDY
POLLUTANT - PB
SAMPLE: NUMBER - i
N 29
TRUE-VALUE .00
MEAN .04
MEDIAN .07
DATA IN ASCENDTNC ORDER
877
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
STD. DEV.
COEF. VAR
.00
.03
78.54
C.I.fUPPERI .05
C.I.CLOWE R> .03
SKEWWESS 1.28
ACCURACY .00
.01
.01
.01
.01
.01
.01
.02
02
02
02
02
02
03
03
.03
.03
.05
.05
.05
.OB
.06
,06
,06
,06
.10
.130
39.00*
,08
,03
,10
153
-------
INTO?-LABORATOPY STUDY
877
POLLUTANT - PB
SAMPLr NUMBER - 2
N 55
TRUE- VALUE 8.40
MEAN 7.83
MEDIAN 8.18
DATA IN ASCENDING ORDER
UNITS - MICROGRAWS PER CUBIC METER
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
8.80
2.18
1.57
2D.12
C.I.CUPPER! 8.21
C.I.fLOWE Rl 7.44
SKEMNESS -3.41
ACCURACY -2.62
.70
1.15
1.79
5.031
5.93
7.05<
7.1*
7.20
7.21
7.35
7.47
7.56
7.5S
7.60
7.68
7, 70
7.71
7.71
7.71
7.7*1
7.80
7.80
T. 80
7. 83
7.88
7.83
7.92
7.95
7.97
8.01
8.10
8.11
8.12
8.16
3.18
8.22
8.23
8.27
8.28
8.30
8.30
8.30
8.31
3.38
8.38
8. 10
8.40
8.40
8.45
8.46
8.47
8.43
8. 50
8.57
8.64
8.70
8.70
8.79
8.79
8.80
8.81
8.86
8.89
9.on
9.36
9.50
10.08®
10.20*
61.20*
82.80*
154
-------
INTER-LABORATORY STUDY
POLLUTANT - PB
SAMPLE NUMBER - 5
N
TRUE-VALUE:
MEAN
MEDIAN
61
11.40
10.66
10.96
877
UNITS - HICROGRAKS PER CUBIC METER
RANGE
VARIANCE
STD. DEV.
COEF. VAR.
12.17
3.87
1.31
17.36
C.l.fUPPER* 11.14
C.I.CLOWE Rl 10.18
SKEMNESS -3.52
ACCURACY -3.86
DATA IN ASCENDING ORDER
.62
2.76
7.27
9.36
9.604
9.78
9.85
9.9S
10.10
10.17
10. 2S
10.32
10.32
10.60
10.60
10.61
10.62
10.71
10.71
10.78
10. 80
10.80
10.80
10.87
10,90
10.90
10.92
10.93
10.95
10.96
10.99
11.00
11.00
11.03
11.05
11.07
11.16
11*22
11*25
11.29
11.36
11.40
11.43
11.45
11.46
11.50
11.52
11.52
11.52
11.62
11.63
11.77
11.84
12.00
12.10
12.39
12.48
12.60
12.79
13.50*
104.0441
106.20*
155
-------
INTTR-LABORATORY STUDY
877
POLLUTANT - PB
SAMPLT NUMBER - 7
N 54
TRUE-VALUE 4.05
MEAN 3.99
MEDIAN 3.98
DATA TN ASCENDING fRDER
WITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
STD. DEV.
CQEF. V4R
14.03
2.64
1*£2
C.I.CUPPER1
C.I. CLOWE R>
SKEWNESS
ACCURACY
3.56
.17
2.79
2.96
2.98
3.10
3.12
3.2*
3.21
3.4G
3.54
3.6C
3.60
3.60
3.66
3.76
3.78
3. 81*
3. 8«l
3.86
3.89
3.90
3.90
3.90
3.90
3.96
3.98
3.98
4.02
«».03
«l.05
1.07
4.07
4.09
4,10
4.10
4.10
4.14
4.15
4.16
4.74
4.?4
4.27
4.30
4.31
4.38
4.39
4.44
4.50
4.52
5.45
14.20*
41.04*
156
-------
INTER-LABORATORY STUDY
877
POLLUTANT - PB
SAMPLE NUMBER - B
N 65
TRUE-VALUE 6.60
MEAN S.I**
MEDIAN 6.42
DATA IN ASCENDING ORDER
UNITS - MICROGRAMS PER CUBIC METER
RANGE
VARIANCE
STD. OEV.
COEF. VAR.
7.35
1.41
1.19
13.33
C.I.fUPPER) 6,13
C.I.CLOWE RI 5.85
SKEMNESS -3.00
ACCURACY -2.73
.33
1.1G
3.80
4.05
4.504
4.74
5.23
5.31
5.52
5.75
5.79
5.73
5.81
5.83
5.83
5.31
6.00
6.00
6.00
6.00
6.00
6.00
6,01
6.06
6.12
6.14
6.21
6.30
6.30
6.30
6.37
6.40
6.40
S.42
6.43
6.50
6.5*
8.5tfr
6.57
S.53
6.53
6.53
6.60
6.60
6.60
6.65
6.65
6.66
6.67
6.63
6.70
6.TO
6.72
6.73
6.73
6.75
6.88
6.83
6.3C
6.95
6.38
7.02
7.20
7.50
7.7**
8.280
48.00*
66.Cn*
157
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO 2
EPA 600/4-79-014
4. TITLE A\D SUBTITLE
SUMMARY OF AUDIT PERFORMANCE: MEASUREMENT
N02, SULFATE, NITRATE, LEAD, HI VOL FLOW F
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
' OF S02, February 1979
IATE - T977. 6. PERFORMING ORGANIZATION CODE
7 AoTHORiS) 8. PERFORMING ORGANIZATION REPORT NO.
Steven M. Bromberg, Robert L. Lampe, Berne I. Bennett
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Environmental Monitoring and Support Labor
Quality Assurance Branch
Research Triangle Park, NC 27711
10. PROGRAM ELEMENT NO.
1HD 621
"atory 11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS 13. TYPE OF REPORT AND PERIOD COVERED
Environmental Monitoring and Support Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
14. SPONSORING AGENCY CODE
EPA 600/08
»
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The report summarizes the results of the quality assurance audits for
the period from October 1976 through December 1977. Pollutants for which
audits were conducted and results reported are SO^s NOp, sulfate, nitrate,
lead, and hi vol flow rate. The operation of the EPA audit program is also
described.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
air pollution
quality control
13. DISTRIBUTION STATEMENT
Release to Public
b. IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
ambient air 68 A
criteria pollutants 43 F
19. SECURITY CL.ASS (This Report) 21. NO. OF PAGES
Unclassified 160
20. SECURITY CLASS (This page) 22. PRICE
Unclassified
EPA Form 2220-1 (9-73)
------- |