United States
             Environmental Protection
             Agency
              Mobile Source
              Enforcement Division
              Technical Support Branch
              Washington DC 20460
November, 1978
&EPA
Motor Vehicle
Tampering Survey (1978)

-------
                                                 021835
               1978

          MOTOR VEHICLE

         TAMPERING SURVEY
U.S.  Environmental Protection Agency
 Mobile Source  Enforcement Division
      Technical Support Branch
          November,  1978

-------
First Printing	November,  1978



Second Printing	December,  1978

-------
             TABLE   OF   CONTENTS
          EXECUTIVE SUMMARY

I.        Introduction
II.       Design of the Survey
III.      Tampering Inspection and Tampering Rationale
IV.       Site Data
V.        Major Results
VI.       Further Results and Analysis
VII.      Tampering and Inspection-Maintenance Programs
VIII.     Emission Effects of Tampering
IX.       Future Plans
X.        Conclusion

          APPENDIXES

-------
                     Executive Summary

     The Technical Support Branch of EPA's Mobile Source
Enforcement Division has completed a survey which 1) assesses
the rate of vehicle emission tampering on a nationwide basis,
2) assesses the types of tampering most prevalent, and 3) quantifies
the relationship between tampering and idle test emission
scores.  During the period May to August, 1978; 1,953 vehicles
comprising model years 1973 through 1978 were inspected at six
sites.  These sites, Delaware, Maine, Virginia, Washington,
Tennessee, and Texas, were chosen to represent various national
geographic categories.  Vehicles were observed while participating
in a non-voluntary program such as a mandatory state safety
inspection.  The sites were selected such that the sample
reasonably represents vehicles not currently subject to an
emission oriented inspection-maintenance program.

     Employees of the Technical Support Branch working with  an
expert consultant and NEIC personnel conducted the  inspections.
The same consultant was used at eacn site to assure a consistent
application of the  inspection criteria.

     At each site 200 to 400 vehicles were  inspected for  tampering
of 15  emission related components.   In general,  each inspection was
no longer  than five Minutes.  In  addition to component  inspection,


-------
idle hydrocarbon and carbon monoxide emissions were measured and
recorded.  Fuel  samples were taken from vehicles which required
unleaded gasoline.

     The results of the survey classified each car into one
of four mutually exclusive groups: tampered  (at least one
case of obvious tampering), arguably tampered (potential, but not
clear cut tampering), malfunctioning (some component not properly
functioning), or "O.K." (all components properly functioning).
The results show an overall tampering rate of 19% of the 1973
through 1978 vehicles in the nation.  An additional 48% showed at least
one item in the arguably tampered category (e.g., limiter cap removal)^
JTh,
     These results are shown in Table 1, which displays  the degree
of tampered, arguably tampered, malfunctioning, and  "O.K." cars
by model year.
                Table 1 - Tampering Summary

Age  Model     Tampered  Arguably  Malfunctioning    O.K.

0-1
1-2
2-3
3-4
4-5
5-6
Year
78
77
76
75
74
73

7%
10%
18%
22%
33%
32%
Tampered
31%
44%
53%
53%
57%
56%

4%
1%
1%
2%
1%
3%

58%
45%
28%
22%
9%
9%
Total         19%        48%            2%           31%
                          -11-

-------
     There is a direct relationship between the tampering rate
and the age of the vehicle as shown in Figure 1.  The projected rate
of tampering increases to approximately 50% as the fleet approaches
100,000 miles.  The rates of tampering at different sites were
all within 5% of the 19% national average.


    I The rate of tampering correlates well with increased idle
emissions scores.  For each model year under consideration, idle
emissions were significantly higher for tampered cars than for
"O.K." cars./  EGR tampering and vacuum spark retard tampering
were the most prevalent forms of tampering.  High idle carbon monoxide
levels were found to be indicative of NO  related EGR tampering.
                                        X
This is a secondary effect, however, indicating that a person
who tampers with the NO  control system also tampers with the
                       A
idle settings.  A relationship was also found between limiter
cap removal and higher idle carbon monoxide readings.  Significantly,
72.5% of cars with limiter caps in place were below a 1% idle
CO level, while among vehicles with missing caps only 39.1% were below 1%
CO.  The positive relationship between idle scores and tampering
demonstrates the ability of an idle test to detect tampered vehicles,
and a corresponding benefit of an inspection-maintenance program.
                              -111-

-------
i
!-••
<
    T
         i caca.ia
        9ta.ca
        BB.CB
        •70.13
         Bta.ca
         SEB.BI
         HB.BI
         2H.CJ
          ita.ta
                                       Figure  1
                                  Plot of Tampering vs. Age
                                         (N=1953)
                             Y - 3.627 + 5.574(X)
                             R2= .9550
                             F = 84.95
5         s         a         ?         a
                               fin
                                                                                                       a         a

-------
     From  the subset of  vehicles  inspected  which  required
unleaded gas, an overall  fuel  switching  rate  of  approximately 5%
can be estimated.  This  is  the  percentage of  vehicles  with leaded
fuel in the tanks  (as determined  by  sample  analysis) and/or tampered
filler neck restrictors.  Somewhat surprisingly,  75% of  the cars
with evidence of leaded  gas  in  their tanks  did not  have  tampered
fillernecks, indicating  either  alternate methods  of filling a car
with leaded gas (e.g. funnels,  slow  fill),  or nozzle violations  at the
gas pump.  Conversely, 62% of  the vehicles  with  tampered  filler  neck
restrictors did not have  leaded gas.   Presumably,  this indicated only
sporadic use of leaded gas,  or  some  other rationale such  as second
ownership  of the vehicle.   In either case,  whatever motivated
the removal of the filler neck  did not apparently motivate
continued  switching.  The 6% fuel switching rate differs  from the
estimate of 10% found in an  MSED  gas station  observation  survey.
Differences in the vehicle mix  observed  in  each  survey may explain
some of the discrepency  .  In order  to assess the reasons  for these
differences further sampling will be instituted  concentrating
on 1977 and 1978 model year  vehicles.  It takes a minimum  of three
unleaded fills on top of a new  empty tank (1  gallon left)  to lower
the tank lead level to less  than  .05g/gal lead after one  fill of
leaded @ 2.0g/gal. of lead  .

     The emission effect of  an  overall tampering rate  of  19% is
significant.  Approximate calculations indicate  that by the
                             -v-

-------
time a vehicle reaches 50,000 miles, the average hydrocarbon and



carbon monoxide emissions will be nearly four times that of a



vehicle with no tampering or malmaintenance.
                               -vi-

-------
                     I.   Introduction

     The f'.ooile Source Enforcement Division of tne L'.'J. El'A
is responsible for enforcement of the mobile source provisions
of the Clean Air Act, as amended August, 1977.  Sections
203(a)(3)(A) and 203(a)(3)(B) of the Act prohibit the removal
or rendering inoperative of emission controls by dealers and
the service industry, and are generally referred to as  the
"taripering" provisions of the Act.

     Prior to 1378,  the Division had collected evidence
indicating that tampering was occurring.  However, these
data were very difficult to quantitatively assess Because of
the variability anong inspection procedures and inspectors.
In conjunction with  this variaoility was the ausence of a
specific method by which a non-voluntary vehicle sample
could oe inspected.  The qualitative data that was available
originated nostly from I/A areas and a voluntary sample, and
was assuned to reflect less tampering than woulcl be expected
in non-I/M areas and with a non-voluntary sanple.  A sunr.iary
of these data is presented in Table 1.1.

     In early 1973,  the Technical Support Branch undertook a
systematic national  tampering survey of non-I/Li areas in
                            -1-

-------
Table
III. Mobile Source In-Use
Year Site/Study
1977-78 New Jersey1
1977-78 Fuel Switching
(Nationwide)
1977-78 VW Field Fix
Survey
1977-78 Ford Recall Data
1977 Portland, Oregon
1976-77 CARB
1976-77 MSAPC Restor.
Maint.
1976 Chrysler Recall
Data
1975 MSAPC Emission
Factors Program
1975 New Jersey
1974 New Jersey
1974 D. C.
1973 EPA Rpt. Auto.
Exh. Emission
Surv. (1973)
1 Preliminary results
2_ Rabbits and Sciroccos, 19
3_ 3 of 15 randomly selected
4_ Checked only for catalyst
restriction
c \ w
•H £ -0
11 a £ AJ
J"± US U  M ^'^
W Qt -H O -H dt/l
o e e e oj e H
it m -H to 3 o e
Vehicle MY1 s ° * J * r* u *
1975-78
1975-1978
1975-762
1977
1975-77
1975-1976

1975-76
1975
1972-76
1970-1976
1970-1974
1970-74
1968-71
14%
—
100%
20%3
8%4
15%

-30%*
24%5
—
6%
11%
15%
—
51%
—
__
—
—
—

45%
— —
—
57%
53%
33%
—
—
10%
__
—
—
—

—
—
—
—
—
—
—
—
—
__
—
—
47%

42%
—
23%6
—
—
—
37%7
of 19 randomly selected vehicles
vehicles
, unleaded labels, fuel inlet
5_ 7 of 29 randomly selected vehicles
6 By city: Denver-7% Los Angeles-33% Che. Houston,
St. Louis, Phoenix, D.C.-31%
By model year: 1976-47%, 1975-28%, 1974-13%, 1973-13%, 1972-17%
7 HC and CO only; By city:
Denver 6%, By model year:
1969-70%, 1970-81%, 1971-62%.
* Unconfirmed data
-2-

-------
order to obtain data sufficient to  (1) quantitatively project
the incidence of tampering on a national level with high
confidence, (2) to determine the most prevelant types of
tampering, (3) and to determine the relationship between
tampering and idle emissions.

     The original strategy was to "tag-on"  to mandatory
state/city safety inspection programs where a representative,
non-voluntary vehicle sample would  be expected. This  strategy
was subsequently modified to include  "tagging along"  on
state police license/safety roadside  checks for the  same
reason.  The strategy called for one  expert to supervise  the
inspection teams throughout the survey,  in  order to  minimize
variability.
                  II.   Design  of  the Survey

      The  major  elements  of  design were as follows:

      1.    A minimum of six  (6)  sites in areas with  mandatory
           safety inspections  would be required for  pro-
           jection to a national  level.
                             -3-

-------
2.    300 to 500 vehicles per site would yield the
     1800 to 3000 vehicles necessary for a confident
     national projection.  Survey dates were to be
     chosen for peak flow such as at the end of the
     month when inspection stickers expire.

3.    1973 and newer light duty vehicles and light duty
     trucks would be the target population.  1973 was
     the first model year that these vehicle classes
     possessed significant emission controls.

4.    The sample would be completely random.  No attempt
     would be made to approximate the  national vehicle
     mix according to manufacturer, model, or model
     year.  The main goal was  to survey as many vehicles
     as possible, and not skip any because they were
     rare  or  too many of one  type had  already been
     inspected.

5.   Representative geographic diversity  would  be
     essential  and sites would be chosen  from major
     geographic  areas of  the  country.

6.   The  inspection must be  thorough  and  check  all
     visible  emission control systems, record  basic
                        -4-

-------
     data, obtain  a  fuel  sample,  and  require  less  than
     four  to  five  minutes.   It  was  felt  that  this  speed
     would be necessary  if  the  cooperation  of the
     safety agencies was  to be  maintained.  A detailed
     description of  the  inspection  is given later
     in  this  report.

 7.   At  least two  inspectors would  be required,  and
     in  the busiest  cases,  three or four.

 8.   Idle  HC  and CO  emissions would be measured  with
     portable analyzers.

 9.   A single expert should perform the inspection
     of  the major  emission control devices throughout
      the survey in order to maintain consistency.   This
     person must have a thorough knowledge of the
      emission control systems used by the major domestic
      and foreign manufacturers.

10.    The inspection would objectively record the con-
      dition of the emission controls.  A method would
      be developed for determining which conditions
      constituted tampering.
                -5-

-------
    11.   The survey should be completed by September, 1978
     An "expert" in emission controls was recruited from the
private sector to conduct the summary.  Jack Gockel, a
recognized emission expert, was under contract to MSED and
agreed to perform the survey.

     In arranging survey sites, the agencies responsible for
the vehicle safety inspections in the following areas were
contacted:

               New York State
               Georgia
               Delaware
               Nebraska
               Chattanooga, Tenn.
               Memphis, Tenn.
               Washington, D.C.
               Virginia
               Florida
               Texas

     Not  all  non-I/M  areas with mandatory  safety  inspections
were contacted.  The  above were judged  to  be  the  most desirable
from which  to select  a geographically diverse  sample.  Contact
                        -6-

-------
was initially made by phone.  Letters requesting permission
to conduct the survey were sent to each along with a list of
the items to be inspected in each car.  It was emphasized
that this was an informational survey and that no enforcement
action would be taken.

     It should be noted that the term "tampering" was  avoided
in oral and written communication with  the agencies.   The
survey was described as a "check of  the emission control
equipment."

     All agreed to participate except Memphis and Florida.
Memphis has a high volume 5 minute safety check where  a
vehicle stops for one minute at each of 4 positions  along
the safety lane.  Memphis was concerned that  the tampering
inspection would be too disruptive to the flow.  Florida
declined because of current controversy in  the  state
legislature over the  state  safety  inspection.   The  Florida
state  police did not  want  the safety inspection to  become
involved  in any activity  that could  result  in adverse
publicity.

      Delaware, Tennessee  (Chattanooga), Texas,  and  Virginia
were  chosen  as survey sites because  they  represented a
                             -7-

-------
reasonable geographic diversity.  However, it was judged



that at least two additional sites in other areas of the



country were necessary.  Thus, additional states were contacted



concerning the possibility of "tagging along" on state



police roadside license and safety checks.  Using this



approach, roadside surveys in Maine and Seattle, Washington



were arranged.







     In principle, participation in the tampering survey was



voluntary.  That is, the driver was given a choice before



his car was inspected.  However, by virtue of the safety



inspection or roadside check by police, the audience was



"captive" or non-voluntary and  thus most people were encouraged



and inclined to participate.
     III. The Tampering Inspection and Tampering Rationale







     The circumstances under which the sample was  obtained



in each area will be described  in a  later  section.  The



following is a description  of  the inspection procedure,  and



the  rationale used  to determine tampering.  The  forms  used



to record data  (figures 3.1 and 3.2)  were  essentially  identical



to forms used by  the New  Jersey Department of Environmental
                            -8-

-------
    FIGURE  3.1
HPA Vehicle Tampering Study
                                                     Date:

-------
-10-
     FIGURE 3.2

EPA Vehicle Tampering  Study
                                                                                                Codes:

                                                                                               Jisconnectcd Unit
                                                                                               ~Tl octrical  - 1
                                                                                                Vacuum  -  2
                                                                                                Mechanical  - 3

                                                                                               loses
                                                                                                Incorrect Rt -  4
                                                                                                Rev.Operation -5
                                                                                                Bypass - 6

                                                                                               For Is
                                                                                                Missing - 7
                                                                                                Hi sadjus ted - 8
                                                                                                Malfunction - 9
                                                                                               Functioning
                                                                                                  Properly - 0

                                                                                               lion-Stock Equip.-
                                                                                                          99

-------
Protection in performance of a prior tampering survey for
MSED.  These forms covered all the items to be checked in
the tampering inspection, and did not need modification
except for the addition of space to record odometer readings
The functional check of emission components was performed
mostly by Jack Gockel.

     Arrangements were made with the Region III lab in
Wheeling/ West Virginia to analyze the  fuel samples for
lead content.
A.   Equipment

The following equipment was used:

     1.   Leaded nozzle to check the  fuel  filler  inlet  for
          penetration.

     2.   Hand fuel pump with  a 3 ounce  sample  bottle
          and approximately three feet of  hose.

     3.   Flashlight.

     4.   Hand held mirror.
                             -11-

-------
5.   Hand vacuum pump and gauge for use in the EGR
     functional check, including hose and adapting
     nipples.

6.   Fender covers.

7.   Screwdriver and pliers for disconnecting and
     reconnecting vacuum hoses.

8.   Two Horiba Mexa 300 A emission analyzers

9.   Power converters to run  the analyzers off the
     car batteries when out in  the field.

10.  Gasoline  can  for replacement  fuel  and fuel  to
     flush the fuel  line after  each sample.

11.  Spare fuel line.

12.  Extension cords.

13.  200-300 sample  bottles per site.   (1973  and 1974
     vehicles  did  not  require fuel samples.)

14.  Boxes and tape  for  packing fuel  samples.
                       -12-

-------
     15.   Labels  for fuel  sample bottles.
     16.   Shipping  labels to Wheeling,  W.Va. fuel analysis lab,
     17.   Flammable liquid labels for fuel shipment,
     18.   Clip boards.
     19.   Spare data forms,
     20.   Spare parts for the fuel pump.
     21.   Spare fuel pump,
     22.   Spare leaded nozzle,
     23.  Pens for recording data (black ink, not blue)
     24.  Cloth gloves.
E.   Data for Vehicle Identification







     The following data were recorded on  the  form  presented




in figure 3.1.






                           -13-

-------
 1.    Date
 2.    Vehicle  identifying  survey number -  vehicles were
      numbered sequentially as they were inspected,
      and  this number was  followed by a site identifying
      letter.   For example, "S" was used for the Seattle
      area.
 3.    Make - determined by observation.
 4.    Model -  determined by observation.
 5.    Model year - obtained from driver and verified by
      underhood emission label.
 6.    Odometer mileage
 7.    Engine  family/CID as recorded from the underhood
      vehicle  emission control information label.
 8.    HC in  ppm and CO in % for idle and high rev
      (approximately 2500 rpm).
 9.    The presence of smoke.
10.    Carburetor - if the carburetor was original
      equipment a "P" was used to  indicate that  it was
      a production unit.  If fuel  injection was  used then
      "FI" was recorded.  If the carburetor had  been
      replaced with a non-stock unit,  then a "99"  was
      used.   Jack Gockel's experience  was relied  upon
      to determine if a carburetor appeared to be  a
      non-stock unit.  The number  of barrels was  easy
      to determine by the  size and shape of the  carburetor
                          -14-

-------
          body.  A one barrel carburetor had a one "jug"
          body and one idle mixture screw.  Two and four
          barrel carburetors had two idle mixture screws
          and two and four "jugs" respectively.  The
          number of barrels was recorded under "carb
          model".

C.   Emission Control Checks, Data Recording, and Tampering

     The form presented in Figure 3.2 was used to record  the
inspection of the emission control devices.  The following
codes, as shown on the form, were used.

Code
 0   - item is functioning properly
 1   - electrical disconnect
 2   - vacuum disconnect
 3   - mechanical disconnect
 4   - incorrectly routed hose
 5   - hoses connected so that  an item  operates  in a  reverse
       manner  (special case of  incorrectly  routed hoses,
       mostly applicable to vacuum delay  valves).
 6   - hoses routed  such that an  item  is  bypassed  (another
       special case  of incorrectly routed hoses).
 7   - missing  item
                            -15-

-------
 8   - misadjusted item
 9   - malfunctioning item
99   - non-stock equipment

     The codes are designed so that the  inspector can ob-
jectively record the condition of a device, and he does
not have to make an "on the spot" judgement with respect  to
tampering.

     The observations were broken down  into five categories:
tampered, arguably (or questionably)  tampered, malmaintenance,
malfunctioning, and "O.K."  Most of the  items considered
tampering are obvious.  However, some are  not so obvious  and
the rationale for placing them in the tampering category  is
presented.  Figure 3.3 shows  how the  codes were categorized
for each  item.  Note that not every code applies to  each
item.

     No malmaintenance, defined by  figure  3.3 as a misadjusted
air injection system or vacuum spark  retard, was found.   Thus,
the results discussed  in Section V, Major  Results, will
consist of only  four categories,  "O.K.", tampered, arguably
tampered,  and malfunctioning.

     The  tampering  rationale  is  based on the  following
sections  of  the  Clean  Air Act:   Section 203(a).  The following

                              -16-

-------
-17-
FIGURE -3.3
                                      EPA Vehicle Tampering Study
                                                                                              Codes:

                                                                                             D\sconncctod Unit
                                                                                             Tlf.'ctrica! -  1
                                                                                              Vacuum - 2
                                                                                              Mechanical -  3

                                                                                             Hoses
                                                                                               ncorrect Rt  - 4
                                                                                              Rev.Operateon -5
                                                                                              Bypass  -  6

                                                                                             Farts
                                                                                              Missing - 7
                                                                                              Mi sadjus ted - 3
                                                                                              Malfunction - 9
                                                                                         ^ Q Functioning
                                                                                         gigf    Properly - 0
                                                                                         3' S   //OA' ^
                                                                                                      Equip.
                                                                                                       99
                                                                                                SPECIAL
                                                                                                CASE
                                                                                                AIR CLEANER
                                                                                                TAMPERING

-------
acts and the causing thereof are prohibited —

     (3)(A)  for any person to remove or render inop-
     erative any device or element of design installed
     on or in a motor vehicle or motor vehicle engine
     in compliance with regulations under this title
     prior to its sale and delivery to the ultimate
     purchaser, or for any manufacturer or dealer
     knowingly to remove or render inoperative any
     such device or element of design after such^sale
     and delivery to the ultimate purchaser; or
        (B)  for any person engaged in the business of
     repairing, servicing, selling, leasing, or trading
     motor vehicles or motor vehicle engines, or who
     operates a fleet of motor vehicles, knowingly to
     remove or render inoperative any device or element
     of design installed on or in a motor vehicle or
     motor vehicle engine  in compliance with regulations
     under this title following  its sale and delivery  to
     the ultimate purchaser.

     The "knowingly" element does not  imply  intent to  tamper,
but  rather knowledge that  an act occurred.   For example,
some people have  inverted  the  top of  the carburetor air
cleaner under  the assumption that  it  will  improve engine
                             -18-

-------
performance.  This renders inoperative the positive  crank-
case ventilation system, the evaporative control  system  in
most cases, and the heated air intake system, and while  it
may not be done to tamper per se,  it is none  the  less  a
prohibited act.

     The following items were inspected.  (See figures 3.2
and 3.3).  The applicable codes and inspection criteria  are
noted, and tampering is discussed.

     1.   Idle stop solenoid - This solenoid  provides  an idle
          stop for maintaining idle speeds to the higher speeds
          needed to minimize CO emissions.  On some  vehicles,
          it is used to close the  throttle and thus  prevent
          run-on when the engine ignition is  turned  off.  On
          vehicles with air conditioning, it  is used for in-
          creasing engine idle speed to compensate for a
          decrease in idle speed when the air conditioner is
          engaged.

          (a)  The electrical connection was  checked.
               If it was disconnected or broken - code 1.
               A disconnected connection is considered
               obvious tampering.  A broken connection,
                              -19-

-------
          could have been deliberate or accidental.
          Both cases were considered tampering because
          it is unlikely they could have occurred except
          by a deliberate act.

     (b)   With the air conditioner on, (or in non-air
          conditioned vehicles) the solenoid should
          activate and contact the throttle linkage.
          With the air conditioning turned off there
          would be a small gap between the solenoid
          stop and the throttle linkage.  If the
          appropriate condition was not observed, the
          solenoid was recorded as malfunctioning -
          code 9.  Time did not permit a check to see
          if the solenoid was defective or just mis-
          adjusted.

     (c)   Only codes 0, 1, 8, and 9 were observed.
          (Only one code 8 was observed, and it was
          grouped with code 9.)

2.   Heated air intake - provides warm air to carburetor
during cold engine operation.
                          -20-

-------
(a)   If the vacuum line to the vacuum override
     motor (top of air cleaner horn) was dis-
     connected or Hissing, then code 2.  It
     normally requires a deliberate act by a
     mechanic or owner to disconnect this line,
     and thus, it is considered tampering even
     though failure to reconnect might be due
     to negligence.

(b)   If the "stovepipe" (the flimsy olack paper/
     foil connection between the exhaust manifold
     shroud and the air cleaner horn) was
     missing - code 7.  If it was present but
     not properly connected, (hanging  from one
     connection, for exanple) it was considered
     disconnected - code  3.  If it  was  torn  or
     deteriorated,  it  was also considered dis-
     connected - code  3.  u>out half of  the  ve-
     hicles had metal  stovepipes that  were not  as
     susceptible to these problems.

 (c)  If  any problems v/ere evident  with the butterfly
     plate  in the  air  cleaner horn which appeared
     to  restrict  its movement,  it  was  considered  to
     be  malfunctioning -  code 9.
                   -21-

-------
(d)   If the air cleaner top had been inverted, or
     holes were punched in the air cleaner, the
     heated air intake was considered disconnected
     - code 3.  This is considered tampering because
     the heated intake is deliberately defeated
     and it also causes the PCV system and some
     evaporative control systems to malfunction.
     Malfunctions are registered through the code
     9's on the PCV and on some storage canisters
     which purged with a line connected to the
     air cleaner.

(e)  The question of tampering for a mechanically
     disconnected or missing stovepipe (codes 3 and 7)
     is controversial.  Certainly an improperly
     connected, missing, or torn stovepipe renders
     the heated intake inoperative.  However, no
     one who  understands engines would expect any
     improvement in performance or fuel economy due
     to its removal or disconnection.  These problems
     are the  result of fragile design and  careless
     replacement of the air cleaner, not of any attempt
     to tamper.  Many  times the foil just  fits over
     the sharp edge of the exhaust shroud, and  if  it
     is not pushed on  carefully,  it  easily tears.  The
                  -22-

-------
          incidence  of  codes  3  and  7  is  so  high  that it was



          determined that  it  would  be misleading to place



          them  in  a  tampering category.   Thus, codes 3  and



          7  are labeled as arguably tampered.  If code  3



          is coupled with  code 9  on the  PCV,  this indicates



          the altered air  cleaner,  and is tampering.








     (f)   Problems with the butterfly plate in the air



          cleaner  horn are considered defects,  and are



          labeled  as malfunctions.







     (g)   If a  non-stock (code 99)  air cleaner filter was



          installed  and this element was wider than the



          production design,  a gap was created in the air



          cleaner  chamber.  In this case, a 9 was recorded



          for the  PCV system, a 9 for some storage canisters,



          and a code 3, mechanical disconnect, for the



          heated intake.  This is considered tampering for



          the same reason as the inverted air cleaner top.







3.   Limiter caps -  plastic caps on  idle mixture screws



designed to limit carburetor adjustments.







     (a)  The  inspector checked for  the presence of the



          caps.  In most cases they  were missing - code  7.
                           -23-

-------
(b)   If caps appeared in good shape - code 0.

(c)   If the caps appeared damaged, they were
     checked to see if they could be rotated
     past the stops.  Caps with broken or bent
     tabs (allowing rotation past the stops) were
     given a 3.  In some cases, the data recorder
     designated a 9.  In this case, the code 3 and
     code 9 designate the same condition, a mech-
     anically disconnected and ineffective
     limiter cap.

(d)   Limiter cap removal is prevalent.  To place it
     in the tampering category would obscure the
     rest of the data.  However,  it has defeated
     an element of  emission control design and is
     done knowingly.  Thus, missing or disconnected
     limiter caps are considered  arguably tampered.
     This misadjustment usually causes a significant
     increase  in CO emissions.  Mechanics remove
     limiter caps to enrichen  the idle mixture for
     smoother  engine operation.   Enrichments pro-
     ducing greater than about 1% CO  in  the  exhaust,
     do  not provide enough  oxygen for  the correct
     oxidizing function  of  the catalyst.  As a
                          -24-

-------
          result, the vehicle usually exceeds EPA standards,
          These misadjustments are actions which render
          catalysts inoperative and represent the greatest
          single cause of maintenance related air pollution
          from post-1974 vehicles.

4.   Positive crankcase ventilation system - A typical
configuration for a V-8 engine consists of the PCV valve
connected to a valve cover and then connected to the
carburetor by a vacuum line.  The other part of the system
has a "fresh air tube" running from the air cleaner to the
other valve cover.

     (a)  If the line between the PCV valve and the
          carburetor was disconnected then code 2 des-
          ignates a vacuum disconnect.  If the "fresh
          air tube" was disconnected, code 3 was recorded.
          Some code 4's, incorrect routing, were also
          recorded for this  condition.

     (b)  Missing valve and  hoses - code  7.

     (c)  A code 9, malfunction, was  recorded  if the
          air cleaner top was  inverted,  if holes had
          been punched  in the  air cleaner, a  too thick
                            -25-

-------
     air filter was used,  or if there were any
     holes or tears in the PCV lines.

(d)   For add-on devices a non-stock equipment
     "99" was recorded.

(e)   The PCV system is difficult to categorize
     for tampering.  A good mechanic would not
     defeat it because it is common knowledge
     that it improves engine life by purging the
     crankcase of blow-by gases.  In many cases,
     it becomes disconnected when the air cleaner
     is removed and it is inadvertently left that
     way when the air cleaner is replaced.  How-
     ever, it is also the target for many "gasoline
     savings" devices that install in the vacuum
     hose leading to the carburetor base.  The PCV
     system is an important control that a) prevents
     crankcase emissions, b) is clearly visible,
     and c) is easily maintained.  Thus, codes
     2,3,4,7, and 99 are considered to be tampering.
     Code 9 is considered tampering only when it
     results from a tampered air cleaner.
                    -26-

-------
5.   Evaporative control system - Controls vapors from
the fuel tank and carburetor.  Some systems have two lines,
one from the fuel tank to the- canister, and one from the
canister to the carburetor or air cleaner to air purge  the
canister.  Other systems have a third line which is usually
connected to the carburetor.
     (a)  The lines were checked  for proper  routing  and
          connections.  If any were missing  or  disconnected,
          code 2 was  recorded for vacuum  lines  and code  3
          was recorded  for others.  Both  cases  are clearly
          tampering.

     (b)  If the canister was missing,  code  7 was recorded.
          This  is  clearly tampering.

     (c)  As discussed  under the  PCV system, if there
          were  problems with the  air cleaner,  then
          code  9 was  recorded  if  one of the  canister
          lines was connected  to  the air cleaner.
 6.    Tank cap
      (a)   The cap was checked for a good gasket that
           sealed properly.   If it did not seal properly,
                         -27-

-------
          code 9  was recorded.   If the cap was missing,
          code 7  was recorded.   (One code 3 was found, and
          it was  grouped with code 9.)

     (b)   The gas cap is also a controversial tampering
          item.  Its absence defeats the evaporative control
          system.  Yet, it is highly unlikely that it would
          be left off on purpose.  It was decided that
          there is no excuse for the cap to be missing, and
          that it should be considered as arguably tampered.
          A bad gasket is considered a malfunction.

7.   Air injection system - Consists of an air pump driven
by a crankshaft pulley which pumps air through a control
valve and lines connected to the exhaust manifold.

     (a)  The pump, belt, lines, and control valve were
          checked visually.  The most common problem was
          a missing drive belt which was usually recorded
          as code 7 for "missing".  Occasionally, the data
          recorder improperly designated a code 3 for
          mechanical disconnect.  Both codes designate the
          same condition.  Removal of the drive belt
          renders the pump and the control valve inoper-
                       -28-

-------
     ative.   Therefore,  a code 9  for "malfunctioning"
     pumps and  control  valves was recorded in most
     cases.

     The  second most common air system problem was
     the  removal of the air pump.  This was denoted
     by a code  7 for both the pump and the valve.   In
     many cases, the control valve is an integral  part
     of the pump -  thus, a 7 for the control valve.
     In a few cases, the recorder used 3 for mechanical
     disconnection  for  designating this same condition.

     Other more minor problems were the disconnection
     of hoses and lines.  If the vacuum line was
     disconnected to the control valve a code 2 was
     used.  If  the  inlet or outlet hoses to the pump
     were disconnected  a code 3 was used.

(b)   Codes 2,3, and 7 were considered blatant
     tampering.  While  there were inconsistencies
     in recording the codes for the individual
     components, an accurate percentage of total
     system tampering may be determined if the
     system is  analyzed as a whole.
                     -29-

-------
8.   Exhaust gas recirculation system - The standard con-
figuration consists of a vacuum line from the carburetor
to a sensor (used to detect temperature to activate the
EGR valve), and another vacuum line from the sensor to the
EGR valve.  Some systems have multiple sensors and thus
additional vacuum lines.  The system directs a portion
of the exhaust gases back into the cylinders for  the con-
trol of oxides of nitrogen.  This is one system where a
functional check was performed.  EGR valves are found in
sealed and non-sealed units, and require different function-
al checks.

     Non-sealed EGR valve functional check

     (a)  The  system was visually inspected to  see if
           the  valve, sensor(s) and hoses were  in  place.
           If any hoses  were disconnected, code  2  was
           recorded  for  the associated  valve or  sensor.
           Missing parts received  code  7.

      (b)   The  engine was revved  and  EGR valve  stem
           movement  was  checked visually or  by  touch.
           If stem movement was detected,  code  0 was
           entered  for  both  the valve and  the  sensor,
           and  the  system passed  the  check.
                          -30-

-------
(c)   If no stem movement was detected, the vacuum
     line to the valve was pulled off and checked
     for blockage.  If it was blocked, code 2 was
     recorded for the valve.  If it was not blocked,
     a hand vacuum pump was then connected to the
     sensor outlet and the engine revved.  If a
     vacuum was obtained, the sensor was good and
     the valve was given a 9 for a malfunction.  If
     no vacuum was obtained, then the line to the
     sensor from the carburetor was checked for
     vacuum while revving the engine.  If this line
     was blocked, then code 2 was recorded for the
     sensor and code 9 for  the valve, if the valve
     did not already have a code.  If this line
     to  the sensor had vacuum, then the sensor was
     bad and code 9 was  recorded for  both the sensor
     and the valve.

 (d)  Some  lines had a vacuum delay valve  in  them
     that  was  positioned backwards.   A reverse
     operation, code  5,  was recorded.

 (e)  On  some systems  with multiple  sensors  and
     multiple  nipples,  some code 4, mis-routings,
     were  found.
                     -31-

-------
(f)   A crushed nipple on the EGR valve or sensor
     resulted in a code 3 for that component.

(g)   For systems with multiple sensors, each sensor
     and hose was checked as above if no stem
     movement was detected.

Sealed EGR valve functional check

(a)   The system was visually inspected.

(b)   The vacuum hose to the EGR valve was discon-
     nected.  The hand vacuum pump was connected
     to the valve and vacuum applied with the
     engine running.  If idle speed dropped with
     the application of vacuum, the valve was
     good.  The vacuum pump was then inserted
     into the line leading to the valve's vacuum
     source.  The engine was revved to determine
     if vacuum was available.  If vacuum was
     available, the sensors were checked using
     the same procedure described for  the non-
     sealed unit.

(c)  Obviously, all codes  except 9 are  tampering
      (no code 6's were observed).  Note, if  a  sensor
                   -32-

-------
          was  bad,  for what ever reason,  the EGR valve
          was  coded  9  automatically and not usually checked,
          Thus,  the  number of malfunctioning EGR valves
          is the total of the EGR valves that failed the
          functional check and the EGR valves assumed to
          have malfunctioned due to a bad or tampered
          sensor.  The main reason for not checking each
          EGR valve  after finding a bad sensor was time.
          However,  EGR valves tend to freeze with dis-use
          and it would be impossible to determine which
          were defective and which froze from dis-use.

9.   Catalytic converter

     (a)  The converter was visually inspected  for  its
          presence and high temperature discoloration.
          The only applicable code was "missing" - code
          7.  This  is obvious tampering.   If  there was
          any uncertainty as  to whether or not  a vehicle
          was supposed to have  a  catalyst,  the  underhood
          vehicle emission  control  information  label  was
          checked.  On most vehicles,  the  emission  control
          systems are  listed  on this  label.   Chrysler is
          a notable exception.
                               -33-

-------
     (b)   It  is  possible  to remove the catalyst material
          from all General Motors'  catalysts through a plug
          in  the body of  the canister.  Our experience
          indicates that  some catalyst material removal can
          be  expected.  However, an inspection for this
          was not performed in this survey.

10.   Dash labels and tank labels

     (a)   The only applicable code is 7.  Removal of the
          labels, which are an element of emission control
          design, is considered arguably tampered.

11.   Filler Neck - Unleaded vehicles  only.

     (a)   The filler  neck  inlet restrictor was  checked
          with  a  leaded nozzle.   If  it  fit  (whether or
          not alteration was  visible) a code 3  was
          recorded.   If the  restrictor  was missing -
          code  7.

     (b)  It  was  difficult to determine which  cars
          require unleaded fuel.   Catalyst cars definitely
          required  unleaded  fuel.  However,  now several
          manufacturers produce non-catalyst cars that
                         -34-

-------
          require unleaded fuel.   The easiest method to



          clarify any doubts was  to check the dashboard



          since that label is rarely missing.







     (c)   Both acts are considered tampering.  (Code 3



          and 7.)







     (d)   One unleaded pick-up truck had an auxiliary fuel



          tank installed with a leaded inlet.  This was



          recorded as a non-stock 99, and was considered



          tampering.







12.  Vacuum Spark retard system.







     (a)   Visually checked for proper connections.







     (b)   Codes 2 or 3 - disconnected - tampering.







     (c)   Missing - 7 - tampering.







     (d)   Only codes 0, 2, 3, and 7 were observed.







13.  Tampering Source - with  time permitting, some



drivers with  tampered cars were asked if they knew  the



origin of the  tampering.  Little value  is given  to



these observations.
                            -35-

-------
                    0 - Owner
                    D - Dealer
                    OD - Other than dealer
                    DK - Don1t know

D.   Inspection Scenario

     A typical inspection required a four-man crew and proceeded
as follows:

     The driver was approached and greeted with:

          "Good morning, we  are with the U.S. Environmental
     Protection Agency and we are performing a  survey  to
     check the condition of  emission control equipment.
     If you have a few minutes, we would like to  look
     under your hood and see if everything is in  good
     shape, and also take a  3 ounce fuel sample to check
     for lead content."

     The general answer was  "yes" or "How  long  will  it
     take?"

     Those who questioned the time were  told  "three  or four
minutes."  Most who were apprehensive  at first  were  placated
by the short  time required.  If the driver asked  directly if

                                -36-

-------
he was required to participate, he was  informed  that  he
wasn't, but that it was an important survey and  it would  be
greatly appreciated if he would.  Some  people pleaded a case
of being late and left.  It  is estimated  that less than 5%
skipped the tampering inspection.  One  reason for not
participating used several times was that the driver  was  not
the owner of the car and that he didn1t want anything done
to the car without the owner's permission.

     Once the driver agreed, the car was  brought to the
inspection area and the hood was raised.   On unleaded
cars, one team member  would check the  dash unleaded
fuel label, the unleaded label at the filler inlet, the
size of the inlet with the leaded fuel  nozzle, to see
if it would fit and push open the flap, and visually
inspect the inlet, tank cap, and tank cap gasket.  He
would then draw out a fuel sample.

     The sample bottle was then tightly capped,  marked with
the survey identifying number, and placed in a container  for
shipment to the analysis laboratory.  When requested  by the
owner, approximately 3 ounces of unleaded fuel was poured
into his tank to replace the sample.  A small amount  of new
fuel was then poured through the hand pump line  to flush  any
leaded fuel residue that might have remained.
                              -37-

-------
     Some problems were encountered with locking gas caps.
If the driver had to shut off his engine to get the key to
open the cap and several vehicles were waiting, the locking
gas caps on leaded cars were not checked.

     When a damaged or missing gas cap gasket, a tampered
filler inlet, or missing labels were found, it was recorded
on the forms.

     As soon as a car arrived at the inspection point, one
team member would place the analyzer probe  in the tailpipe
and begin to fill out the vehicle data form.  When all the
vehicle data and inspection information except for the
emission values had been obtained, he would read and record
the analyzer at idle, then at a high rev.   He would then
remove the probe.

     Meanwhile, the consultant, using a flashlight and
mirror, would call out his underhood observations to another
team member who recorded them on the appropriate form.  When
the observations were complete and Jack had revved the
engine to complete the emission measurements  the car was
sent on  its way.  Time permitting, we would discuss any
tampering we had found with the driver.  Most drivers  were
very nice, cooperative, and interested.  The  few that  were
very adamant were sent on  their way without an  inspection.
                                  -38-

-------
                       IV.  Site Data
A.   Maine - Roadside Check.
     Dates -
May 16 through May 19, 1978
     Participants -
Maine State Police



Jack Gockel (Expert Consultant)



Mel Petroccia - EPA Region I



Randy Rice - EPA Region I



Robert Knowles - EPA Region I
     Samples -
338 vehicles
      Fuel  samples -
Maine refused to allow fuel sampling



for reasons of safety.
      Sites -
Tuesday - highway  in Bangor suburbs.



Wednesday - under  an overpass  in



     downtown Bangor.



Thursday - weigh station on highway



     between Gay and Springfield.



Friday - same as Thursday.
                                  -39-

-------
     The State Police conducted roadside registration and
safety checks and passed the vehicles along to the survey
team.  The drivers were made aware of the tampering survey
by the police and encouraged to participate.  The police
were very helpful and interested in the tampering survey.
The tampering survey was rained out on Tuesday afternoon and
a sheltered site was arranged for Wednesday, and it rained
most of the day.  Thursday was cloudy and Friday was sunny.
Temperatures were in the 60's° F.

B.   Virginia I - Private garage safety inspection.
     Dates -
 May 30  through June 2,  1978.
 June 19 through June 21,  1978
     Participants -
 Jack Gockel
 Paul Gesalman - EPA/MSED
 Jim Caldwell - EPA/MSED
 Steve Albrink - EPA/MSED
 Truman Wilson - NEIC
      Samples -
 98 vehicles (May 30-June 2)
 13_ vehicles (June 19-June 21)
111 Total
      Fuel  Samples  -
 39 analyzed
                               -40-

-------
                          9 destroyed in shipment via UPS.







     Sites -             Fairfax Exxon, 10480 Lee Highway,



                              Fairfax, Va. (May 30-June 2).



                         Exxon at Seven Corners, Arlington,



                              Va. (June 19-21).







     Weather -           Hot, sunny, 80's° F.
     Virginia has a semi-annual private garage safety in-



spection.  Flow is very low compared to a roadside check.



However, the State Police initially refused to allow us to



tag-on to a roadside check and that left only the garage



sites.  The peak inspection flow period is the final week of



the month since the inspection stickers run out the first of



the next month.  Jack Gockel and Paul Gesalman worked the



May 30-June 2 period and averaged about 25 vehicles a day,



with the station inspecting around 40 to 50 vehicles per



day, including pre-73 models.  The low flow rate in this



type of inspection requires only two inspectors.  Because of



extremely low flow rates, the inspection was discontinued after



the third day of the second week and the State Police were



contacted regarding a roadside check.  A roadside check was



arranged.
                                 -41-

-------
C.   Virginia II - roadside check.
     Dates -
Monday, July 24-July 28, 1978.
     Participants -
Virginia State Police
Jim Caldwell - EAP/MSED
Paul Gesalman - EPA/MSED
Larry Walz - NEIC
Ron Snyder - NEIC
Steve Sinkez - EPA/MSED
     Samples -
313 vehicles
     Fuel samples -
189 obtained and analyzed.
Paul Gesalman drove the samples to
Wheeling.
     Sites -
Monday - Parking lot of a dance hall
     on rural but busy State Route 17
     south of Warrenton.  Afternoon
     sampling was rained out.
Tuesday - weigh station on U.S. Route  1
     in Woodbridge - very busy road.
Wednesday -  State route 234, five
     miles south of Manassas.
                                    -42-

-------
                         Thursday - Route 1, five miles south
                              of Alexandria, in high school
                              parking lot.
                         Friday - same as Thursday.

     Weather -           Hot, mostly sunny, 80°F.
     One trooper was assigned to perform the check each day,
He was asked to send us only 1973 and later vehicles with
Virginia plates.  A few out of state cars were inspected.
The trooper's inspection was very quick and there was gen-
erally one car waiting for the tampering inspection at all
times.

D.   Wilmington, Delaware - State run safety lanes.

     Dates -             June 12-16, 1978.

     Participants -      Jack Gockel
                         John Fahrenback - EPA/MSED
                         John Davis - EPA/MSED
                         Janet Littlejohn - EPA/MSED
                         Charles Case - EPA/MSED
                         Paul Gesalman - EPA/MSED
                         Jim Caldwell - EPA/MSED
                                 -43-

-------
     Samples -
 360 vehicles.
     Fuel samples -
 56 analyzed
 83 are questionable because they
 were returned by UPS in poor condition
 and upon analysis, gave very atypical
 results.
 37  destroyed by UPS.
176  Total
     Site -
 Wilmington suburb, near airport.
     Weather -
 Hot, sunny, 80°F, moderate humidity,
     Delaware has an annual safety inspection performed at
four state operated sites.  The site chosen for an inspection
was the largest and had four lanes, only two of which were
being used.  The team used a vacant lane outside the inspection
building and cars were chosen by the team randomly from
those waiting in the other lanes.  The safety inspection
personnel were very accomodating.  In most cases, the vehicles
we took out of the safety inspection waiting line were let
back in behind the car originally in front of them.  The
flow was very heavy at times and overall just as good as  the
roadside check.
                                  -44-

-------
E.   Chattanooga, Tennessee - City run safety lanes,
     Dates -
June 26-30, 1978.
     Participants -
Jack Gockel



Paul Gesalman - EPA/MSED



Tom Newman - NEIC



John Dion - NEIC
     Samples -
325 vehicles.
     Fuel  samples -
23 analyzed



98 lost by UPS.
     Weather -
Hot,  sunny,  humid, 90°F, high  air



pollutant  levels.
      Chattanooga has a city-wide annual  safety inspection



 at one four-lane site.  Approximately 500  vehicles are



 checked per day in a four minute inspection.   The tampering



 inspection was performed in front of an  active lane and



 appropriate vehicles were directed to that lane by safety



 inspection personnel.  The tampering survey received local



 TV and press coverage.
                               -45-

-------
p.   Seattle, Washington - Roadside check.
     Dates -
Monday, July 10-July 14, 1978
     Participants -
Washington State Police
Jack Gockel
Bob Bernstein - EPA/MSED
Larry Walz - NEIC
Ron Snyder - NEIC
Jim Caldwell - EPA/MSED
     Samples -
323 vehicles.
      Fuel  samples -       161 were  taken and analyzed.
      Sites -
Monday - Rural highway  in Seattle
      suburbs.
Tuesday -  busy roadway  in industrial
      area.
Wednesday  -  intersection of  two  inter-
      state highways  near Seattle.
Thursday - exit  ramp of an  interstate
      highway in  Seattle suburbs.
Friday - local park  in  Seattle suburbs
      Weather -
 Hot,  sunny,  80°F,  low humidity,
                                 -46-

-------
     A roadside check was conducted by the Washington State



Police.  The police performed a license/safety check and



sent the vehicles to the tampering inspection team.  There



were generally two or three troopers present.  The assistance



from the troopers was excellent.  Region X was very interested



in the survey, sent several observers including a local



politician, and arranged local TV and press coverage.







G.   Houston, Texas - Private garage safety inspection.
     Dates -
Monday, August 28 - Saturday,



     September 2, 1978
     Participants -
Jack Gockel



Paul Gesalman - EPA/MSED



Herbert Vaughan - EPA/MSED



Ron Snyder - NEIC



Larry Walz - NEIC
      Samples -
218 vehicles.
      Fuel  samples  -
89 obtained and analyzed.
      Sites  -  Team  1  -
Monday - Mobile Station  in a depressed



     neighborhood of downtown Houston.



     There were many old vehicles.
                                   -47-

-------
                         Tuesday -  Exxon station on highway in



                              a Houston suburb.



                         Wednesday  - Texaco station in affluent



                              suburbs.



                         Thursday - Service station in downtown



                              Houston.



                         Friday, Saturday - same as Thursday.
          Team 2 -
Monday - same service station team 1



     surveyed on Thursday.



Tuesday - Saturday - Shell station in



     suburbs.
     Weather -
Hot, sunny, humid, 90°F.
     An attempt was made to conduct a roadside survey in



Texas.  However, the Texas Department of Public Safety did



not feel it would be appropriate to delay traffic beyond the



roadside license/safety check.  In order to increase the



sample size obtainable at private garages, two tampering



inspection teams were formed.  The State police took the



teams to the stations and stayed until  the tampering in-



spections got under way.
                             -48-

-------
     Jack Gockel lead team 2.  Paul Gesalman lead team 1 and
was replaced by Larry Walz for Thursday - Saturday.
                                   -49-

-------
                      V.  Major Results
     This section presents the basic aggregate results of
the survey.  Total tampering by site, model year, and type
of tampering are discussed.  As the data for each vehicle
were processed, the vehicle was classified into one of four
categories: "O.K.", tampered, arguably tampered, and mal-
functioning.  Since each vehicle inspected has various
components, each of which could be tampered, the vehicle  itself
is classified by the worst state of any component in the  vehicle.
The hierarchy is in the order: tampered, arguably tampered,
malfunctioning, not equipped, properly functioning.  Thus,  if  any
one component is tampered, the entire vehicle  is considered
tampered.   If one  component  is "arguably tampered"  and all  the
others are  functioning  properly, the entire vehicle is considered
"arguably  tampered".  Thus an "O.K." vehicle must have all
components  functioning  properly.

     An  "O.K."  vehicle  is  one with  no observable gross tampering,
arguable tampering, malfunctioning,  or missing equipment.  The
term,  "O.K.", does not  mean  that  the vehicle  is necessarily operating
properly.   For  example, a  spark plug or  coil  may  not be  performing
satisfactorily  resulting  in  a poorly operating and  excessively
polluting  vehicle. This  situation  would  still be  classified "O.K."
for  purposes  of this  tampering  report.
                                -50-

-------
£.        Aggregate Results

     The results for all vehicles tested are presented  in  Appendix
V-I and summarized in Table 5.1
               Table 5.1 - Aggregate Results
                         Percent                  Number
O.K. cars                 30.7%                   599
Tampered Cars             18.9%                   370
Arguably Tampered         48.4%                   945
Malfunctioning cars        2.0%                    39
                         100.0%                 1,953

     These results span several model years and the totals here
represent a snapshot of vehicles at a specific point in time
(Summer 1978).  As such, the numbers must be viewed with caution.
For instance, it would be incorrect to conclude that a mere 2.0% of
the vehicles had malfunctioning components.  This number appears
low because of the hierarchy explained above.  Many cars had com-
ponents malfunctioning but if at least one other component were
tampered or arguably tampered, the vehicle, as a whole, would
have been classified in the tampered or arguably tampered  state.
Thus as soon as an item like a missing limiter cap (arguably tampered)
is noted, the car cannot be classified as malfunctioning,  no
matter how many other components malfunction.

     A major concern is the degree of confidence that exists
in the 18.9% tampering estimate.  Because of the large size of our
sample, we are 95% confident that the true proportion of tampered
cars is between 17.1% and 20.7%.
                              -51-

-------
B.  Tampering by Vehicle Age
     Appendix V-l also presents a breakout of tamperi.no categories
by model year of the vehicle.  It is notable that the rate of
tampering increases with the age of the vehicle, and correspondingly,
the percentage of "O.K." cars decreases with the age of the
vehicle as shown in Table 5.2.
Age
—•^k«B
0-1
1-2
2-3
3-4
4-5
5-6
Tampered
7.4%
10.1%
17.7%
22.3%
32.6%
32.0%
O.K.
58.1%
44.6%
28.4%
22.3%
9.4%
9.1%
Figure 5.1 demonstrates  the  virtually  linear  nature  of  the  increase
in tampering  rate  over the life  of  the vehicle.

      This  tampering  increase with  age  is  significant.   The  19%
tampering  rate  represents  an average  tampering  rate  in  calendar
year  1978.  The data trend shows that  by  the  end of  six years
in the  life of  a model,  37%  of  the cars will  be tampered with.
After eight years  of driving (approximately 100,000  miles)
about 48%  of  the fleet will  be  in the  tampered  classification.
This  means that almost half  the  cars  will be  in the  tampered
group,  not just malmaintained or arguably tampered.
                               -52-

-------
tn
        i cscj.ta
       sca.ca
       BEJ.EJ
       •7G3.GJ
        Bca.ca
        srca.ta
        HG3.GJ
   EH
        20.13
         i ta.ca
   §    aca.ia
           Figure 5.1

Plot of  Tampering vs. Age
       (N=1953)
                        Y = 3.627  + 5.574(X)
                        R2= .9550
                        F = 84.95
	 1 	 	 — 1 	 1 	 	
•
1 ID K| ™
a — " m ri
	 , 	 1 	 H- 	 1 	
?aa S| B!
lA in r-
i 	 1 	 •
n ni —
                                                       Age (in years)

-------
Total cars
306
416
216
324
335
356
Tampered
15.7*
14.2%
22.2%
20.4%
20.9%
22.8%
Tampered
41.8%
49.8%
59.7%
44.4%
46.3%
51.1%
Malfunctioning
2.0%
2.6%
1.43
.9%
2.11
2.5%
O.K.
40.5%
33.4%
15.7%
34.3%
30.7%
24.2%
C.   Tanpering by Site

     Appendix V-2 breaks out tampering by site.  It can

be summarized in Table 5.3



               Table 5.3 - Tampering by Site

                                        Arguably
Site
Washington
Virginia
Texas
Tennessee
Maine
Delaware

Total         1953        18.9%         48.4%     2.0%            30.7%

     It can be seen that there is some variation from the  18.9%

nationwide average in the  different sites.  In an effort  to

ascertain if these departures from 18.9% are significant or if they

are the result of random fluctuations, a Chi-Square test at the 5%

level of significance was run.  The results proved just significant,

meaning randon chance could have resulted in these fluctuations

only 5% of the time.



     Interestingly, when the same test was run on the percentage

of "O.K." cars; the results proved very significantly different.

There are less than 5 chances in 1000 tnat such a wide discrepancy

occurs due to random chance.  The factor that appears responsible

for this is the low number of "O.K." cars in Texas.
                                    -54-

-------
D.   Tampering by Type of Vehicle



     In order to ascertain if any vehicle make exhibited more

tampering than others, a breakout by make was run and is shown

as Appendix V-3.  This has been summarized by manufacturer  in

Table 5.4:



           Table 5.4 - Tampering by Manufacturer

Manufacturer              Total cars         Tampering

GM                            882            20.1%
Ford                          491            20.2%
Chrysler                      242            20.1%
AMC                            64            31.3%
Foreign                       272             8.5%

     A Chi-Square Test was also run on  these numbers.  The

results were significant at a 5% level.  This indicates a

distinct difference in tampering level  amongst manufacturers.

This difference is probably due to the  low level of  tampering

emong foreign vehicles.


E.   Summary


          To summarize,  the national rate of tampering  in  1978  is

     18.9%.  Tampering increases with age, and as many  as  50% of  the

     vehicles which are  8 years old can be expected  to  have been

     tampered with.  The tampering rate appears  higher  for domestic

     makes  than for foreign makes, and  tends to  be  affected by

     geographic location.
                               -55-

-------
             VI.  Further Results and Analysis



     The previous section dealt with the major aggregate

tampering results by age, site, and manufacturer.  In this

section some further insight is provided by examining specific

individual forms of tampering,  as well as certain "groups"

of tampering such as fuel related tampering,  (fillerneck

restrictor tampering, gas cap tampering, and/or excess lead

levels) or EGR system tampering (EGR control valve or EGR

sensor tampering).  Further, the relationship between idle

emission levels and tampered vehicles is discussed.



A.   Individual Types of Tampering



     The rates of tampering by type are given in Table 6.1

Categories narked with an asterisk  (*) were not considered

as tampering.


          Table 6.1 -    Individual  Rates of Tampering

     Idle Stop Solenoid             0.74%
     Heated  Intake                  0.84%
     Limiter Cap Removal             *
     PCV Valve                      3.27%
     ECS Storage                    2.63%
     ECS Tank Cap                    *
     Airpump                       3.17%
     Air Control Valve              2.90%
     Air Pump Belt                  5.74%
     EGR Control Valve            11.87%
     EGR Sensor                     5.32%
     Catalytic Converter           1.21%
     Dashboard Label                 *
     Fuel tank label                 *
                             -56-

-------
     Filler Neck Removal           3.37%
     Vacuum Spark Retard          10.50%
     Air Cleaner                   0.87%


From these figures, it is apparent that EGR control and

vacuum spark retard tampering are the two major sources of

tampering.  It should also be noted that while limiter cap

removal was not included as definite tampering, this removal

occurred in 65% of the vehicles observed.  Assuming that a

removed limiter cap allowed the idle to be set rich, the

effect on emissions of such an action could be huge.



     Given the high rate of limiter cap removal, an

investigation was made on the effect of this on idle emission

scores.  Appendixes VI-1 and VI-2 demonstrate  the mean HC

and CO idle scores by model year  for cars with and without

limiter caps.  The summary is shown in Table 6.2.



          Table 6.2 - Limiter Cap vs. Idle Scores

Model Year
73
74
75
76
77
78
Mean
O.K.
182
180
122
145
128
54
Idle HC (ppm)
Missing
297
297
188
259
188
101
Mean Idle
O.K.
2.36
2.21
.99
1.35
.81
.52
CO (%)
Missing
3.41
3.35
2.19
2.52
2.07
1.40
                             -57-

-------
     It is quite clear that idle scores for both HC and CO

with limiter caps removed are considerably above those with

limiter caps in place for all model years.  Further, of the

1844 cars that had limiter caps required as equipment, a

test was made to see if their idle CO would have exceeded

1%.  The results were that 72.5% of the vehicles with

limiter caps in place would have passed a CO test with a 1%

CO idle cutpoint.  Conversely, 60.9% of those who removed

the limiter cap would have failed such a test.



     Since EGR tampering was the highest single  item of

tampering, it was desired to observe the idle test's ability

to identify this form of tampering.  Since EGR tampering

affects NO , one might expect little relationship between
          A
idle HC and CO scores and EGR tampering.  The results showed

differently however.  As demonstrated by Appendixes VI-3 and

VI-4 average HC and CO idle scores  increased as  the EGR

system was tampered with.  A summary is provided  in Table

6.3


          Table 6.3  EGR Tampering  vs. idle scores

Model Year
73
74
75
76
77
78
Idle
O.K.
247
295
164
235
152
67
HC (ppm)
Tampered
348
321
206
236
205
51
Idle CO
O.K.
3.04
3.07
1.73
2.19
1.40
.71
(%)
Tampered
4.28
3.68
2.17
3.00
1.75
1.21
                             -58-

-------
     Since, for physical reasons one would not expect an HC


or CO effect from NO  tampering, it appears that vehicles
                    x
with NO  system tampering also exhibit tampering with other
       x
systems or with limiter caps and thus degrade their HC and



CO controls as well as their NO  controls.
                               A
     To assess the degree of this collateral effect, the



data were examined for the amount of NO  tampering  that
                                       A


coincides with limiter cap removal, and the amount  of N0x



tampering that occurs with at least one other  forn  of tampering



or arguable tampering.






     The results showed  that in at least 80% of  the cases of



NO  tampering, limiter cap removal also occurred; and in
  A


86% of the 223 cases of  NOX tampering, at  least  one other



form of tampering or arguable tampering was present.  This



result tends  to support  the hypothesis that a  vehicle exhibiting



NO  tampering also exhibits other tampering.   In fact,  the
  A


EPA Restorative maintenance study has demonstrated  that even



if EGR disablement is the only form of tampering, resultant



HC and CO emissions will increase as well  as the anticipated



increase in NOx.






     B.   Relationship of Tampering and Idle Emissions







     Idle tests are  the  predominant test made  associated





                            -59-

-------
with inspection and maintenance programs.  To the extent



that idle test scores are capable of discriminating between



tampered and non-tampered cars, the more useful  the scores



become in an inspection maintenance application.  A few



examples have been brought to EPA's attention in which



extreme amounts of tampering including catalyst



 removal,  have  not  affected  a vehicle's ability to pass an



 idle test.   These  samples have shown the ability to circumvent



 an idle  test if  other settings were modified accordingly.



 This survey provided a unique opportunity to assess the



 ability  of  the idle  test to detect tampering on a large in-



 use sample  in  non  l/M states.   Thus the results would not



 be based on a  handful of specially prepared cars, but rather



 on a typical cross-section of in-use vehicles.







      In  order  to assess this ability of the idle test, low



 and high CO and  HC idle scores were measured on each vehicle.



 The following  analyses have considered only the low idle in



 assessing idle scores and tampering.







      Mean idle scores have been calculated for each of the



 four categories  of cars (O.K., tampered, arguably tampered,



 and malfunctioning)  for each model year in the survey.



 Appendixes VI-5  and VI-6 show the results.  "O.K." cars have
                             -60-

-------
Generally lower idle scores on both HC and CO than either

the tampered or arguably tampered cars for every model year,

For malfunctioning equipment, the idle scores are comparable

for some post-1975 models.



     The most important results are reproduced below  in Table

6.4

          Table 6.4 -  Relationship of tampering and  mean  idle  scores

Model Year     Mean Idle HC(ppm)             Mean Idle C0(%)

               O.K.      Tampered            O.K.     Tampered


73             124       328                 1.95     3.95
74             230       313                 2.63     3.43
75             115       211                 1.07     2.49
76             151       254                 1.43     2.66
77             109       221                 0.687    1.77
78              56        51                 0.483    0.591

     In an analysis of variance test,  idle scores proved  to be

very highly significantly correlated  to  tampered vehicles.



C.   Fuel Related Tampering



     Previous information indicated that a proportion of  the

population of vehicles requiring unleaded gas was being

fueled with leaded gas.  To evaluate  the extent of  this

phenomenon, fuel samples were taken from vehicles requiring

unleaded gas.  Due to some difficulties  in transporting  the

gas samples to laboratories, only 481 valid  samples could  be
                               -61-

-------
considered for the study.  From this smaller sample an



attempt was made to relate filler neck inlet tampering to



the evidence of leaded fuel (in excess of 0.05 grams per



gallon) in the tank.  The results appe'ar in Appendix VI-7.



Only 20 vehicles requiring unleaded gas (4.2%) had leaded



gas in the tank.  Among these 20, only 5 had  tampered



filler neck restrictors.  This appears to indicate the use



of some alternate means of filling the vehicle with leaded



gas such as funnels or a slow fill procedure.







     The percentage of the 481 vehicles  with leaded fuel



samples and/or tampered filler neck retrictors was 5.8%.



The 95% confidence  interval for this overall  fuel switching



potential  rate, based upon a sample of size 481  is from 3.6%



to 8.0%.







     Since this estimate differs  from another MSED study



showing approximately 10% fuel switching, further analysis



is being undertaken to examine any differences  in vehicle



populations observed that might account for this discrepancy,



Further, more  data  will  be obtained  in order  to  attempt



reconciliation of  the observed differences.   The additional



sampling will  concentrate on 1977 and later model year



vehicles.
                             -62-

-------
     VII.  Tampering and Inspection Maintenance Programs







     In previous sections, the relationship between idle scores



and the extent of tampering has been discussed.  The basic results



were that tampered cars had appreciably higher mean idle CO levels



and but for one model year had higher idle HC levels than



O.K. cars.  Further, idle HC and CO scores were also found higher



for tampered NO  related control devices such as the EGR system.  The
               A


probable reason for this is the concurrent tampering of the EGR



system with at least one other component affecting CO or HC, such



as limiter cap removal.  Additionally, a very high correlation was



found between limiter cap removal  and the inability to pass an



idle test with a cutpoint of 1% CO.







     The next analysis was to simulate these vehicles undergoing



a New Jersey inspection - maintenance program.  New Jersey



modified phase II cutpoints of SOOppm HC and 5.0% CO for model years



73 and 74; and SOOppm HC and 3.0% CO for model years 75 and later



were used.  The results are summarized in table 7.1







Table 7.1 - Relationship between Tampering and New Jersey Program



                    Pass          Fail





O.K.                86%           14%

Tampered            65%           35%

Arguably Tampered   68%           32%

Malfunctioning      80%           20%
                             -63 -

-------
As can be seen, 35% of the tampered cars failed the New Jersey
test, while only 14% of the "O.K." cars failed.  Thus there appears
to be a higher probablity of failing such a test with a
tampered vehicle.  It should be noted that the 65% of tampered
vehicles which pass New Jersey standards do not necessarily mean
that the idle test is incapable of identifying tampering, but
rather that the general New Jersey standards can be considered
somewhat lenient/  in that most cars, tampered or non-tampered,
appear to pass.

     To summarize, of the O.K. vehicles 86% pass New Jersey
standards, while vehicles in one  of the other three groups have
only a 67% chance  of passing.  The difference between the 86%  and
67%, based on  the  sample size of  this survey,  is statistically very
significant, demonstrating a marked inverse relationship between
the  rate of tampering and the ability to pass  the New Jersey  program.

     It  is important to remember  that these relationships are derived
from studying  a  set of vehicles not currently  subject to an  emissions
inspection maintenance program.   Any  secondary  effect such
as  the mere presence of such an I/M  program dissuading  tampering
actions  would  not  be reflected  in these  analyses.
                            -64-

-------
           VIII.   Emission Effects of Tampering

     The 19% overall tampering rate, along with projections
for increased tampering by age for each vehicle model year
have a considerable effect on resultant emissions.  To try
to quantify these effects, an emissions model which considers
each component with its tampered condition effect on emissions,
and the time phasing of different types of tampering is
required.

     While such a complete model does not now exist, an
approximation may be made using an emission model as described
in "Emissions from Catalyst Cars Beyond 50,000 Miles and the
Implications for the Federal Motor Vehicle Control Program",
(SAE paper 780027).  This model considers vehicles to be
classified in one of three states:  properly functioning,
poorly maintained, or disabled catalyst.  For purposes of
comparison, we have considered limiter cap removal as an
indicator of poor maintenance.  Further, not all  the tampering
observed in this study  is of the extreme nature as catalyst
disablement.  Thus, the limiter cap removal rate, which
increased by 10% of the vehicles per year was used as the
rate of malmaintenance.  The tampering rate which increases
at 4% per year  in a model's life was reduced to 3% to account
£or tampering incidents less egregious than catalyst dis-
ablement.
                              -65-

-------
     If these estimates are used in the approximation model,




the results show a four fold increase in the average carbon



monoxide and hydrocarbon emissions in the fourth year of the



model's life.  This means that by the time a fleet reaches



50,000 miles, the average emissions from this fleet will be four



times the average emissions of the same fleet having no malmainte-



nance and no tampering.  This estimate is a fairly gross



approximation and is, of course, dependent on the assumptions



made in the model.  However, it does provide a good "ball park"



assessment of the effects of tampering on vehicle emissions and



should serve to underscore the severity of the problem.







     The effect of EGR tampering on emissions has not been studied



extensively, but some data is available.  In a Restorative Maintenance



Program conducted by EPA thirty-seven 1975 and 1976 automobiles were



tested with their EGR systems deliberately plugged.  Emissions  increased



from the non-plugged condition by an average of 21%, 71%, and  123%  for



HC, CO, and NO  respectively.
              X
                          -66-

-------
IX   Future Plans








     The rates of tampering discussed  in  this paper were



found in the period May - August/ 1978.   The results represent



a rather large sample and reflect the  condition of vehicles  in



this time period.  While specific future  plans for tampering



assessment are still being designed, it is felt that the current



survey has adequately assessed the tampering situation at the



present time.








     Thus, the thrust of future plans will be to observe the change



in tampering rates observed as programs and resources are implemented



to curb this problem.  The specific recommendations are:







     1.    A survey in New Jersey of the same type



          performed in the other six sites, utilizing the



          same personnel, equipment, and definitions.  In this



          way a consistent comparison can be made between tampering



          rates and the implementation of an inspection maintenance



          program.







     2.    Future controlled surveys, probably using



          contractor assistance,  of sites in two EPA regions



          per calendar quarter.  This will serve as a monitor



          of the trend of future  tampering.  For regions earmarked



          for concentrated tampering enforcement,  or which have






                               -67-

-------
     on-going inspection maintenance programs/ this will



     enable a measurement of the effects of these programs.



     To provide for a consistent check of tampering



     levels, it is suggested that the EPA consultant



     currently supervising the surveys be utilized to train



     MSED,  contractor, and regional employees in the methods



     of tampering and the categorization of vehicles.







3.    Additional investigation in order to understand



     the differences between the fuel switching rate



     observed in this study and other studies.
                       -68-

-------
X    Conclusions



     A tampering study of 1,953 vehicles estimates the tampering

rate at 19% gross tampering in areas not currently subject  to  emission

inspection maintenance programs.  Since the vast majority of the

nation is comprised of non - I/M areas*, this  rate is considered  to

be a reliable nationwide estimate.  The tampering rate increases

with the age of the vehicle from about 5%  for  new vehicles  to  close  to

50% for 100,000 mile old vehicles.  The most prevalent forms of gross

tampering were the EGR system  (13%) and the vacuum spark  retard  (11%).

In addition, very high rates  (65%)  of  limiter  cap removal were observed

Vehicles with limiter caps removed  had  idle CO in excess  of 1%

most of the time  (61%).  The  tampering observed  has  a very  pronounced

emission effect.  Fuel switching as assessed by  this survey is

occuring in about 6% of  the vehicles  requiring unleaded  gas.



     Idle  test scores detect  tampered  cars.  In  fact,  idle  test

scores even catch EGR tampering  which  is N0x related, due

to  the high level of concurrent  presence of  EGR tampering and  at  least

one other  form of tampering.   This  tends  to  strengthen  the

usefulness  of  inspection maintenance  programs  due  to their

ability  to  discourage vehicle tampering.
           Current mandatory I/M areas are: New Jersey, Cincinnati,
           Portland (Oregon), Arizona (two counties), Nevada, and
           Rhode Island.  Approximately 6% of the nation's vehicles
           are included in such a program.
                                    -69-

-------
     The surveillance should continue with emphasis on  comparing



these results with those in areas having an on-going  inspection



maintenance program, and in investigating future  trends  in



tampering rates.  The reasons for the difference  between fuel  switching



observed in this study and other studies should be established.
                               -70-

-------
Appendix V-I

-------
TESTING OF DATA-AND PROGRAM



FILE   NONAME   (CREATION DATE 6 09/25/78)
                                                                     09/25/78
PACE   2S»
MYR

COUNT 1
R04 PCT ]
COL PCT
TOT PCT
73.



74".


J
J *
75.



76.



77.



78.




COLUMN
TOTAL

STATUS

OK

0.
23
9.1
3.8
1.2
26
9.4
4.3
1.3
61
22.3
10.2
3.1
112
28.4
18.7
5.7
204
44.6
34.1
10, a
173
58.1
28.9
8.9

599
30.7



TAMPERED

30.
81
32.0
21.9
4.1
90
32.6
24.3
4.6
61
22.3
16.5
3.1
70
17.7
18.9
3.6
46
10.1
12.4
2.4
22
7.4
5.9
1.1

370
16.9
C R 0 3 Z


ARGUABLE

31.1
141
55.7
14.9
7.2
156
56.5
16.5
8.0
146
53.3
15.4
7.5
209
52.9
22.1
10.7
201
44.0
21.3
10.3
92
30.9
9.7
4.7

945
48.4
i T A B U


MALFNCTr
0
33.
8
3.2
20.5
0.4
4
1.4
10.3
0.2
6
2.2
15.4
0.3
4
1.0
10.3
0.2
6
1.3
15.4
0.3
11
3.7
28.2
0.6

39
2.0
BY STATUS VEHICLE TAMPER STATUS


ROW
TOTAL

253
13.0


276
14.1
'

274
14.0


395
20.2


457
23. H

'
'. 298
[ 15.3
[
[
j
1953
100.0
CHI SUUARE
319.75342 WITH  IS DEGREES OF FREEDOM   SIGNIFICANCE &  0.0000

-------
Appendix V-2
     -72-

-------
TESTING OF DATA ,AND PRQGHAM




FILE   NONAME   (CREATION DATE = 09/25/78)
09/25/78
PAGE   46*
STATt

COUNT 1
ROW PCT ]
COL PCT
TOT PCT

wA



VA


1
•-J •
^J TX
1


TN



ME



DE



COLUMN
TOTAL

STATUS

OK

0.

124
40.5
20.7
6.3
139
33. 1
23.2
7.1
36
16.7
b.O
1.8
111
3U.3
lb.5
5.7
103
30.7
17.2
5.3
86
2Q.2
14.4
4.4
599
30.7



TAMPERED

30.1

46 1
15.7 ]
13.0
2.5
59
14.2
15.9
3.0
48
22.2
13.0
2.5
66
20.4
17.8
3.4
70
20.9
18.9
3.6
79
22.2
21.4
4.0
370
18.9
CROSS


ARGUABLE

31.1

128
[ 41.8
13.5
6.6
207
49. B
21.9
10.6
129
59.7
13.7
6.6
144
44.4
15.2
7.4
155
46.3
16.4
7.9
182
51.1
19.3
9.3
9«5
48.4
1 T A B U


MALFNCTN
D
33.

6
2.0
15. tt
0.3
11
2.6
28.2
0.6
3
1.4
7.7
0.2
3
0.9
7.7
0.2
7
2.1
17.9
0.4
9
2.5
23.1
0.5
39
2.0
BY STATUS VEHICLE TAMPER STATUS


ROM
TOTAL


306
15.7


416
21.3


216
11.1


324
16.6


335
17.2


356
18.2


1953
100,0
CHI SQUARE B    55.89305 WITH  15 DEGREES OF FREEDOM   SIGNIFICANCE =  0.0000

-------
Appendix V-3
     -74-

-------
  TESTING OF  DATA  AMI)  PROGRAM


  FILE   NONAME    (CREATION DATE
                                                    09/25/78
P^GE  162
09/25/78)
ui
I
MAKE

COUNT ]
RO-1 PCT ]
COL PCT
TOT PCT

VW



VOLV



TRIU



TOYO



SUBA



SAAB



RENA


m
COLUMN
TOTAL

STATUS
:
[OK

0.

13
35.1
2.2
0.7-
10
90.9
1.7
0.5
0
0.0
0.0
0.0
9
10.6
1.5
0.5
0
0.0
0.0
0.0
1
100.0
0.2
0.1
0
0.0
0.0
0.0 ]
599
30.7



TAMPERED

30.1

3 ]
8.1 1
0.8 1
0.2 I
1 J
9.1 ]
0.3 3
0.1 1
0 1
0.0 ]
0.0 ]
0.0
4
1.7
1.1
0.2
0
0.0
0.0
0.0
0
0.0
0.0
0.0
0
o.o
0.0
[ 0.0
370
18.9
C R 0 S 5


ARGUABLE

31.

18
48.6
1.9
0.9
0
[ 0.0
[ 0.0
0.0
3
[ 100.0
[ 0.3
0.2
72
84,7
7.6
3.7
8
100.0
0.8
0.4
0
0.0
0.0
0.0
1
100.0
0.1
0.1
945
48.4
J T A B U


MALFNCTN
0
33.

3
8.1
7.7
0.2
0
0.0
0.0
0.0
0
0.0
0.0
0.0
0
0.0
0.0
0.0
0
0.0
o.o •
0.0
0
o.o
0.0
0.0
0
0.0
0.0
0.0
39
2.0
BY STATUS VEHICLE TAMPER STATUS


ROM
TOTAL


37
1.9


11
0.6


3
0.2


85
4.4


8
0.4


1
0,1


1
0.1


1953
100.0
  (CONTINUED)

-------
TESTING OF DATA AND PROGRAM




FILE   NONAME   (CREATION DATE
                                                                                    09/25/78
                                                                                                     PAGE   163
                                09/25/78)
MAKE

COUNT 1
ROW PCT 1
COL PCT
TOT PCT
PONS



PONT


1
f PLYM
I


PEUG



OPEL



OLDS



MG



COLUMN
TOTAL

STATUS

OK

0.
1
50.0
0.2
0.1 '
47
35.3
7.8
2.4
26
24.5
4.3
1.3
0
0.0
0.0
0.0
I 1
I 100.0
0.2
0.1
52
36.9
8.7
2.7
0
0.0
0.0
0.0
599
30.7



TAMPERED

30.1
1 I
50.0 1
0.3 1
0.1 1
31
23.3
8.4
1.6
27
25.5
7.3
1.4
0
0.0
0.0
0.0
0
o.o
0.0
0.0
21
14.9
5.7
1.1
2
33.3
0.5
0.1
370
18.9
C H 0 S S


ARGUABLE

31.
0
0.0
0.0
0.0
S3
39.8
5.6
2.7
51
46.1
5.4

3
100.0
0.3
0.?.
0
0.0
0.0
0.0
66
46.8
7.0
3.4
4
66,7
0.4
0.2
945
48.4
TABU


MALFNCTN
D
33,
0
0.0
0.0
0.0
2
1.5
5.1
0.1
2
1.9
5.1
0.1
0
0.0
0.0
0.0
0
0.0
0.0
I 0.0
[ 2
1.4
5.1
0.1
0
0.0
0.0
0.0
39
2.0
BY STATUS VEHICLE TAMPER STATUS


ROM
TOTAL

2
0.1


133
6.8


106
5.4


3
0.2


1
0.1


141
7.2


6
0.3
[
[
1
1953
100,0
(CONTINUED)

-------
TESTING OF DATA -AND PROGRAM
FILE   NONAME   (CREATION DATE s 09/25/78)
09/25/78
PAGE  164'
MAKE

COUNT I
RQrt PCT 1
COL PCT
TOT PCT

MERS



MERC



MCOS
1
-J
-J
MAZD



MAZA



LINC



LANC



COLUMN
TOTAL

STATUS

OK

o.

1
100.0
0.2
0.1
20
26,0
5.3
1.0
I
100,0
0,2
0,1
3
27,3
0,5
0,2
1
100,0
0,2
0.1
8
38.1
1,3
o.«
1
100,0
0.2
0.1
599
30.7



TAMPERED

30.

0
0.0
0.0
0.0
21
27.3
5.7
1.1
0
0.0
0.0
0.0
0
0.0
0.0
0,0
0
0.0
0,0
0.0
3
1^.3
0.6
0.2
0
0.0
0.0
0.0
370
18,9
CROSS


ARGUABLE

31.

0
0.0
0.0
0.0
31
04.2
3.6
1.7
0
0.0
0.0
0.0
6
72.7 ]
O.A
0.4
0
0.0
0.0
0.0
9
12.9
1.0
0.5
• 0
0.0
I 0.0
I 0.0
945
08.4
TABU


MALFNCTN
D
33.1

0 ]
0.0
0.0
0.0
2
2.6
5.1
0.1
0
0.0
0.0
0.0
0
[ 0.0
I 0.0
[ 0.0
0
0.0
0.0
0.0
1
4.8
2.6
0.1
0
0.0
0.0
0.0
39
2.0
BY STATUS VEHICLE TAMPER STATUS


ROW
TOTAL-


1
0.1


77
3.9


1
0.1


11
0.6


1
0.1


21
1.1


1
0.1


1953
100,0
 (CONTINUED)

-------
TESTING OF DATA AND PROGRAM




FILE   NONAME   (CREATION DATE • 09/25/76)
09/25/78
PAGE  US
MAKE

COUNT 1
RO* PCT ]
COL PCT
TOT PCT
JEEP



INTE



MONO
]
i 1
£
I_
•
CMC



FORD



FIAT



DOOG




COLUMN
TOTAL

STATUS

OK

0.
2
50.0
0.3
o.i •
0
0.0
0.0
0.0
11
t 39.3
[ 1.8
[ 0.6
2
18.2
0.3
0.1
90
22.9
15.0
4.6
0
0.0
0.0
0.0
25
28.0
4.2
1.3

599
30.7



TAMPERED

30.
0
0.0
0.0
0.0
1
50.0
0.3
0.1
2
7.1
0.5
0.1
3
27.3
0.8
0.2
75
19.1
20.3
3.8
3
75.0
0.8
0.2
14
15.9
3.8
0.7

370
18.9
C R 0 3 {


ARGUABLE

31.
2
50.0
0.2
O.I
1
50.0
Oil
0.1
15
53.6
1.6
0.8
6
54. b
0.6
0.3
222
56. 5
2J.5
ll.o
1
25.0
0.1
0.1
48
54.5
5.1
2.5 ]

945
48.4
.TABU


MALFNCTN
D
33.
0
0.0
0.0
0.0
0
0.0
0.0
0.0
0
0.0
0.0
0.0
0
0.0
0.0
0.0
6
1.5
15.4
0.3
0
0.0
0.0
0.0
1
[ 1.1
[ 2.6
C 0.1

39
2.0
BY STATUS VEHICLE TAMPER STATUS


ROM
TOTAL

4
0.2


2
0.1


28
1.4


11
0.6


393
20.1


4
0.2


88
4.5



1953
100.0
(CONTINUED)

-------
TESTING OF DATA•AMD PROGRAM




PILE   NONAME   (CREATION DATE a 09/25/78)
09/25/78
PACE  1664
MAKE
COUNT ]
ROW PCT
COL PCT
TOT PCT
OATS
CHRY
CHEV
1
-J
X>
1 °
CADI
BUIC
BMW
AUDI
COLUMN
TOTAL

STATUS
[
[OK
0.
35
61.4
5,6
1.8 *
Ib
31.3
2.S
0.8
144
32.1
24.0
7.4
27
55.1
41
41.0
6.8
2.1
3
100.0
o.s
0.2
1
14.3
0.2
0.1
599
30.7

TAMPERED
30.
6
10.5
1.6
0.3
9
18.8
2.4
0.5
88
19.6
23.8
4.5
5
10.2
1.4
0,3
29
29.0
7.8
1.5
0
0.0
0.0
0.0
1
14.3
0.3
0.1
370
16.9
C R 0 S 5
ARGUABLE
31.1
14 J
24.6
1.5
0.7
23
47.9
2.4
1.2
203
«5.3
21.5
10.4
16
32.7
1.7
o.a
29
29.0
3.1
1.5
0
0.0
0.0
0.0
3
42.9 ;
0.3
0.2
945
46.4
5 T A B U
MALFNCTN
D
I 33.
[ 2
3.5
5.1
0.1
1
2.1
2.6
0.1
13
2.9
33.3
0.7
1
2.0
2.6
0.1
[ 1
[ 1.0
[ 2.6
E 0.1
t 0
[ 0.0
[ 0.0
[ 0.0
[ 2
[ 28.6
t 5.1
[ 0.1
39
2.0
BY STATUS VEHICLE TAMPER STATUS
ROM
TOTAL
57
2.9
48
2.5
446
22.9
49
2.5
100
5.1
3
0.2.
7
0.4
1953
100.0
(CONTINUED)

-------
 TESTING OF  DATA  AND  PROGRAM


 FILE    NONAME    (CREATION  DATE  B  09/35/78)
09/25/78
PAGE  167
MAKE


STATUS
COUNT I
ROW PCT IOK
COL PCT I
TOT PCT 0.
AMC
COLUMN
TOTAL.
13
1
0
8
.3
.3

TAMPERED
30.
33
S
1
20
.3
.4
.0
0 S
ARGUABLE
31.
3
1
32
!«
.6
599 370 945
30.7 18.9 48,4
S
I
I
I
I
I

TABU
MALFNCTN
D
33.
0
0.0
0.0
0.0
39
2.0

I
I
I
I
I
I
I
BY STATUS VEHICLE TAMPER STATUS
ROM
TOTAL
60
3.1
1953
100.0
 CHI  SQUARE  a    267.76416  WITH  105  DEGREES OF FREEDOM    SIGNIFICANCE  s  0.0000
oo
o
I

-------
Appendix VI-1
    -81-

-------
  TESTING 0^ DATA AND PNUUHAM

  FILE.   NONAME   (CHEAIIUN DATE = 09/26/70)
                                                                                      09/26/78
                                                                                PAGt
                                                                               22
CRITtKlUlN VARIABLE IHC
bKUKCN D04N BY LIMCAP
dY MYR
VAHIAbLE
FOR tNTiHt POPULATION
IDLE HC
LlMlTtR CAPS
CODE VALUE LABEL
OF S U b P
SUM
295109,0000

MEAN STO OtV VARIANCE
196,0857 259.4655 67322.3218

N
( 1505)
  LIMCAK
     MYK
     MfH
     MYN
     MYK
     MYK
     MYK

  LIMCAK
,     MYK
30    MYK
••^    MYK
1     MYK
     MYN
     MYK

  LIMCAH
     MYH
     MYK
     MYK
     MYK
     MYK
     MYK

     TOTAL CASES *
  MISSING CASES a
1953
 448 OH
           0.
          73.
          7«.
          75.
          76,
          77.
          12.
          73.
          7«.
          VS.
          76,
          77.
          78.

          98.
          73.
          71.
          75.
          76.
          77.
          78.
22.9 PCT.
         FUNCT  PROPERLY
         ARGUABLE  TAMPERING
         NOT EQUIPPED
 56391,
  5645.
  4511.
  5840.
 12485.
 20805.
  7105.
 0000
 0000
 0000
 0000
 0000
 0000
 0000
227108
 51695
 54038
 20166
 55959
 31280
  5970

 11610
  1190
  3745
  1850
  2600
  1615
   610
.0000
.0000
.0000
.0000
,0000
,0000
.ouuo

.0000
.0000
.0000
.0000
.0000
.0000
.0000
116.5103
162.0966
180.4400
121.6667
145.1744
127.638U
 54,2366

239.8164
29*.0977
296.9121
167.7733
259.0694
168.4337
101.1864

156.8919
236.0000
312.0633
142.3077
216.6667
 76.9046
 55.4545
163.7625
24J.26S2
211.9735
195.1931
160.2600
161.6020
90.2464
290.3218
342,6844
360.5927
225.2486
275.4513
212.9524
150.095t>
221.2878
171.0848
380.1403
207. 2098
209.2990
99.1019
77.0861
26818.1510
59 17 /.ViiVO
44932.7567
36100.3546
25689.6751
26115. 22UO
8144.41<:8
842B6.74.i9
117432. M973
130027. 11V3
50/36.9261
75U/3.4324
45346.7441
22528. 7405
48968. 
-------
TESTING OF DAT* ANO PMOGRAM


CRITtrflUN VARIABLE IHC
VARIABLE


LIMCAH

LIMCAH

LIMCAP
CODE


  0.
 12.
                                                        09/26/70
                                                                P*GE   23
....   ANALYSIS


 VALUE LAbEL


 FUNCT PROPERLY
 ARGUABLE TAMPERING
 NOT  fcUUIPPED
                         WITHIN GROUPS TOTAL
or v A K i
SUM
56491,0000
22/108,0000
11610,0000
MEAN
116.5103
239.6184
156,8919
STO DEV
163.7625
290.3218
221.2878
SUM OP SO
12953166.9403 (
79735258. /603 (
35/4685.1351 (
H
484)
94/)
74)
                                 295109.0000
                                           196.0857
253.1599  96263110.8438    (
1505)
 I
oo
OJ
 I

SOURCE
BETWEEN GROUPS
WITHIN GROUPS
ANALYSIS 0
SUM OF SUUARES
4989661,099
96263110.844
ETA s 0.2220
F V A
R I A N C £
D.F. MEAN SQUARE
2 **********
1502 64089.954
ETA SUUAREO s 0.0493
*
*
*
*
F SIG. *
*
38*927 0.0000 *
*
*
*
*
*

-------
Appendix VI-2
   -34-

-------
TESTING OF DATA'ANU PWBGRAM
                                                                                    09/26/78
                                                                                                    PA&E
FILE   NUNAME
                          OATE = 09/26/70)
CNlrtKlUN VARIABLE 1CU
tfKUKtN DQrtN dY LIMCAP
8Y MYK
VARIABLE
FUR ENTIRE POPULATION
LIMCAK
MYK
MYK
MYK
MYK
MYK
MYK
LIMCAH
M1K
, MYK
00 MYK
U1 MYK
1 MYK
MYH
LIMCAP
MYK
MYK
MYK
MYK
MYK
MYK
IDLE CU
LIMITER CAPS
CODE VALUE LABEL

0. FUNCT PROPERLY
73.

75!

77!
78.
12. ARGUABLE TAMPERING
'3.
74.
75.
76.
77.
78.
98. NOT EQUIPPED
73.
74.
75.
7<».
77.
78.
OF 5 u a P 0
SUM
3945.3997
633.U999
87.2000
70.6000
57.3000
147.2000
173.4000
98.0000
3136.6998
712.0000
764,8999
420.0000
665.0999
444.1000
130.6000
174.8000
25.0000
36.7000
25.6000
46.3000
28.4000
12.8000
PULAIIUNd
MEAN
2.0243
0.9920
2.3568
2.2125
0.98/9
1.3505
0.8065
0.5213
2.6117
3*4067
3.3548
2.18/5
2.5193
2.0656
1.4043
1.6037
4.1667
2.4467
1.0667
2.2048
1.0519
0.8000

STO DEV
2.4761
1.7758
2.3781
2.5494
1.5871
1.9M50
1.5411
1.3679
2.6355
2.7361
2.7872
2.5997
2.5294
2.3386
2.0227
2.1110
3.4679
2.4b9<4
1.6972
2.4622
1.1630
1.3317

VARIANCE
6.1310
3.1536
5.0553
6,4992
2.5190
3 . 9 <1 U 1
2.3750
1.8711
6.9461
7.4861
7. 7685
6./5B3
6.3979
ii ii /i y i
n U O 1 X
n fi ^ fa "I
12iu2o7
6.19/0
2. 8806
6.1615
1.3526
1./733

N
( 19«9)
( 639)
( 37)
( 32)
( 5d)
( 109)
( 215)
( 188)
( 1201)
( 209)
( 228)
( 192)
( 2b4)
( 215)
( 93)
( 109)
( 6)
( IS)
( 24)
( 21)
( 27)
( 16)
  TOTAL CASES =
MISSING CAStS a
                   1953
                      4 OK   0.2 PCT.

-------
TESTING UK DATA AND PHQGHAM




CKlTtKjON VARIABLE ICU
09/26/78
PAGE   25
VARIABLE
L1MCAI*
LIMCAP
LIMCAP


oo
CODE VALUt LAbEL
0. FUNCT PKUPEHLY
12. ARGUABLE TAMPERING
96. NOT EUUIPPED
WITHIN GROUPS TOTAL
ANALYSIS 0
SOURCE SUM OF SUUARES
BETWEEN GROUPS 1114.655
WITHIN GROUPS 10620.571
ETA c 0.3055
a i a u r VAKI
SUM
633.8999
3136.6998
171,6000
3945.3997
F VARIANCE
D.F. MEAN SQUARE
2 557.327
1946 5.565
ETA SQUARED = 0.0933
MEAN STD DEV
0.9920 1.775H
2.6117 2.6355
1.6037 2.1110
2.0243 2.3589
*
*
*
*
F SIG. *
*
100,157 0.0000 *
*
*
SUM Of- SO N
2011. 96V1 ( 639)
8335.3239 ( 1201)
481.<»7tJ5 ( '109)
10628.5714 ( 1949)



-------
Appendix VI-3
          -87-

-------
FILt
       Ut UAfA AMU


      NUMAME   (CKtAllUN UATfc = 09/26/78)
                                                                                  09/26/78
                                                                                                   PAGE
IB
CKlTtKlUN VAHlAHLt IMC
UKUKtN DO-IN dY tGKSl
6Y MYK
VAHlAbLt
FOK tMiHt POPULATION
EGHSUT
MYK
MYK
Mf K

MYK
MYK
EGHSTftF
MVr<
MYK
MYK
MYK
MYK
MYK
EtrtSTAT
MTK
MYH
MYK
MYK
MYK
MYK
EbHSUf
MYK
MYK
MYK
MYH
MYK
MYK
EUKSfAT
MYK
--•- OE3CMIPTION
IOLE HC
AT EliK SYSltM STATUS
CODE VALUE LABEL

0. OK
73.

yij
76J
77.
78.
30. TAMHtKEO
73.
74.'

Tb\
77.
78.
33. MALhNCTNO
73!

75.
/6,
77.
78.
98, NUT tQUIPPtO
73.

7«jt
/6.
77.
78,
9V, NON STOCK EQUIH
73,
OF 3 U 6 H
SUM
295109.0000
191546.0000
21517.0000
33382. oOC'O
21220 . OOUU
56962 ,0000
43615.0000
11850. 0000
498UO.OOOO
14951.0000
14//2.000U
6591.0000
8276.0000
49 40. 0000
360.000U
19256.0000
9960.0000
4735.0000
1255.0000
2811.0000
460. UOOO
15.0000
32051.0000
9706.0000
9405.0000
3/90.0000
2995.0000
4695.0000
1460.0000
23/6.0000
.2376,0000
OPULATIUN
MEAN
196. 085/
181.56U2
247.3218
295. U 159
163.6UU6
235,3802
151.9686
66.5730
266.7380
347.6977
321.1304
205.96B8
236, u5 /I
205.1 167
51,4286
200.5833
219.5000
109.4000
1 S9.4444
200. 7b5/
9^,0000
5.DOOO
194.24H5"
255.421 1
2hU, / 143
1/2.2/^7
130.2174
13a.08H2
112.30/7
1188.0000
1188.0000

STO DEV
259.4655
242.5808
296.5500
351.6840
212.4025
264.0120
183,9353
114.1713
329.77B7
390.6443
421.2063
226.1573
231.0092
249.4686
37.2731
195.1446
232.0665
153.55/3
161.13/4
iao.4551
59.3296
8.6603
258.2238
292.7691
3i'4.2603
2/2.0457
148,2U17
185.0963
115.8234
1149.7556
1149,7556

VARIANCE
67322.3218 (
58845.4269 (
87941. B952 (
123f.bl.63HO (
4511'i.tt2
-------
TESTING Uh  DATA «NO PKUUHAM



CHIUNIUN  VARIABLE IHC
                                                                                  OV/46/7B
19 ,
VARIABLE CUDt VALUE LABEL
EGH3IAT 0. UK
tliKSlAT 30. TAMHtHEO
tChSUT 33. MALFNCTND
fcGHSUT 98. NOT EulJIPPED
EkKSTAT 99. NON STUCK fcUUIP
WITHIN GHOUP3 TUTAL
ANALYSIS U
SOUHCE SUM OF SUUANtS
HETrttEN (jKUUPS 3126337.709
WITHIN GHOUP3 Vttl(26iJ34,2J4
tTA a 0.17*7
SUM MEAN STU DEV SUM UF SU N
1V1546.0000 1BI.S602 2')2.bHOB 6202307^.9200 ( lObb)
U9080.0UOO 266.73MO 329.77tt7 2022h240, 1604 ( 187;
320'3l!ouOO 194^248^ 2^8.2238 10935442.8121 ( 16SJ
23/6.0000 11H8.0000 1149.7556 132193d. 0000 C 2)
2V5109.0000 196.0057 21i5.7687 98126434.2338 { IbOS)
F VARIANCE
D.F, MEAN SQUARE F -SIC,
4 781b84,427 11,948 0.0000
liOO 6S017.623
tTA SUUArfED s 0.0309

                                                           -89-

-------
Appendix VI-4
       -90-

-------
TESTING U)- UATA AMU PHUteHAM

FlLt   NUMAME   (CHtAIIUN DATE a 09/26/76)
                                                                                    09/26/78
                                                                              20  „
CHlltMiUN VAHlAhLt ICU
riKUftLN DUNN dY tbHSlAT
dY MYH
VAHlAOLt
FUH tNMKt POPULATION
EGRSTAT
MYH
MYH
MYK
MYH
MYK
EGKS1AT
MYK
MfH
M f k
n i n
MYH
M f K
MYH
tUKSIAT
M YH
ni«i
MYH
MYK
MYK
MYH
MYK
fcGKSUT
MYK
MYK
MYK
MYH
MYK
MYK
MYH
- - - UESCK1PTIUN
IDLfc CU
LGH SYSlfcM STATUS
COOt VALUE LABEL

0, OK
73,
74,
7b.
76,
77
' ' ,
30, TAMPtKfcU
74!'
7«j,
76,
77,

33, MALf-NCTNU
74,
7b.

77,
78,
9B, NOT E&UIPPED
73.
• -* f
7 il
7b,
76,
77,
78,
99. NUN STOCK LUUIH
OF 3 U B P
SUM
394S.399/
2«b2.;-J9H
334.7000
441,bUOU
314,'luOO
529!bOOO
ib/.ouoo
bUb.OOOO
21d.'4UOll
202.2UOO
b2,-4UOO
123.2000
4 tt , *y o o o
10.9000
3//.6000
Ibb. luOO
lOb.bOUO
bl ,2oOO
34.9UOO
li.bUOO
lb. 200(1
406.4UOU
1U3,4UUO
122.0000
b4.9uOO
44.9UOO
53.9000
27,3000
12,6000
12,6000
UPULAT IONS
MEAN
2,0243
l,78Hb
3.0427
3,0667
1./2/5
2.1926
1.39/1
0,7110
3,0901
3|h/b4
2,1 bt34
3,0049
1 , 74<>4
1.2111
2.81/9
3.3717
2.9bll
2,bbOO
1 ,H3b8
1,9429
2.7000
1,8991
2,404/
3.0bUU
1.6147
1.2829
1 .2b35
1.43b8
6,3000
6,3000

STO UEV
2,47fal
2i684H
2.d23/
2.430b
2'.0704
I,b238
2,7842
3,0067
2./913
2 . 2248
2.76bb
2.2006
1.33b5
2.6/29
2.bilO
2.5314
3.3343
2.2234
1 ,9dUb
3.1b43
2,1416
2.193b
2.74i9
1 . 7245
1.5240
1.5956
2.2993
0.4243
0.4243

VAklANLt
6.1J10 (
5.7047 (
7)9 1 i4 (
5.9077 (
6.0557 (
4.2bb7 (
2.3220 (
/./Sltt (
7!791b (
4 , 94 Vb (
7,b4t>0 (
4 , H4f*6 (
l./bi6 (
7.14
-------
TESTING Uh  DATA  AND PROGRAM




CNITtHlUN VARIABLE ICU
                                                                                  09/26/78
PAGE   21
VAH1AULE CUDt VALUt LAUEL
tbMSIAT 0, UK
EGH6TAT 30. TAMPERED
EGH8TAT 33. MALFNC1NU
EGHSUT 98. NOT tUUlHPEO
EGHSUT 99. NON STOCK EQUIP
WITHIN GKUUPS TUTAL
ANALYSIS 0
SUUHCE SUM OF SQUANtS
BETWEEN liRUUKS 4b3.033
MIThIN GKUUPS 11490.193
t[A a 0.1V48
sis a ^ v A H i
SUM
6H6.0000
3//.6000
t06.4000
12.6000
3945.3997
f VARIANCE
O.F. MEAN SlJUAHt
4 11>.2S8
1944 b.911
ETA SQUARED * 0.0379
MEAN STU OEV SUM Uf SU N
1,/bttS 2.3«85 7tt49./HJO C 1377)
4.0901 2./042 1713. 13ol ( 222)
2.B1/9 2.6729 950.2369 ( 134)
1.8991 2.1416 976.91V7 ( 214)
6,3000 0.4243 0.1800 ( 2)
2.0243 2.4312 11490.1928 ( l"»49)

F SIG.
19.162 0.0000
                                                       -92-

-------
Appendix VI-5
      -93-

-------
TESTING OF DATA AND PKUtJRAM

ULE   NUNAMt   (CREATION DATt = 09/26/78)
                                                                                     09/26/78
PAGE
CKITtKlUN VAKIAULE 1HC IDLt HC
uKUKtN ooviH at &TATUS VEHICLE TAMPER STATUS
Q V U V U
O T " T «
VAKIAbLb
FOR tNTIHE POPULATION
STA!US>
MYK
MYK
MYK
MYK
MYK
MYK
STATUS
MYK
MYK
MYK
MYK
MYK
MYK
STATUb
MYK
MYK
MYK
MYK
MYK
MYK
STATUS
MYK
MYK
MYK
MYK
MYK
MYK
CODE VALUE LAUEL

0. OK
73.
74.
75.
76.
77.
78.
30. TAMPERED
73.
74.
75.
76.
77.
78.
31. ARGUABLE
73.
74.
75.
76.
77,
78.
33. MALFNCTND
73.
74.
75.
/6.
77.
78.
SUM
295109.0000
47266.0000
2350,0000
4606.0000
5060.0000
12385.0000
16200.0000
6645.0000
79569.0000
22637.0000
22532.0000
101 16,0000
14744.0000
8620.0000
920.0000
163359.0000
32723.0000
33126.0000
20520.0000
42865. OOOu
28530.0000
5595.0000
4915,0000
820.0000
2030.0000
140.0000
1050.0000
350.0000
525.0000
MEAN
196,0857
109.1594
123.6842
230,3000
115.4545
151.0366
108.7248
55.8403
261.7401
32*. 0725
312.9444
210.7500
254,2069
221,0256
51.1111
222.2571
282.0948
269.31 71
180.0000
252.1471
162.8846
99,9107
148,9394
136.6667
507.5000
28.0000
262.50UO
58.3333
65.6250
STD DEV
259.4655
155.9874
123,1738
295.1506
180.2770
166.1078
145.8551
93.0129
311.3740
368.4803
382.4806
204.6237
221.9911
252.2691
51.5796
272.9193
312.6801
333.9081
236.6011
287.5100
198.3798
153.5415
218.9027
68.0152
417.2429
40.8656
212.1910
49,1596
102,7264
VARIANCE N
67322.3218 ( 1505
24332.0602 ( 433
15171. /83t> ( 19
87113.9053 ( 20
32499.7866 ( 44
27591,6135 ( 62
21273.7008 ( 149
6651.4065 ( 119
96953.7771 ( 304
150916. 9211 ( oV
146291,^772 ( 72
41952.7447 ( 43
49260, 0617 ( 58
63639,7099 ( 39
2660.4575 ( 18
74484.9188 ( 735
97/66.6170 ( lit*
111494. t.2dl ( 123
56074.7768 ( 114
82662.0197 ( 170
39354.5285 ( 156
23574.9919 ( 56
47918.3712 ( 33
7746.6067 ( 6
174091,6667 ( 4
1670.0000 ( 5
45025. 0000 ( 4
2416.0667 ( 6
10553.1250 ( 8
  TOTAL CASES =
MISSING CASES 8
                    448 UK  22.9 PCT,
                                                             -94-

-------
TESTING OP DATA AND PROGRAM




CKIUK1UN VARIABLE IHC
09/26/78
PAGE
VARIAbLE CUD£ VALUfc LABEL
STATUS 0. OK
STATUS 30. TAMPtRED
STATUS 31. ARGUABLE.
STATUS 33. MALFNCTMD
WITHIN GROUPS TOTAL
ANALYSIS 0
SOURCE SUM OF SQUARtS
BETWEEN URUUPS 5159009.189
WITHIN GROUPS 9609 3762. 754
ETA c 0.2257
a I a u r VAKI
SUM
47266.0000
7*569,0000
163359,0000
4915.0000
295109.0000
F VARIANCE
O.F. MEAN SUUARE
3 **********
1501 64019,829
ETA SQUARED = 0.0510
MEAN STD OEV SUM OK SO N
109.1594 155.9874 10511450.0046 ( 433)
261.7401 311.3740 2937699(1. 4704 ( 304)
221.2571 2/2.9193 546M93G . 40UO ( 735)
148.9394 218.9027 1533387. b7H8 ( 33)
196.0857 253.0214 96093762.7538 ( 1505)

F S1G,
2 6'. 862 0.0000
                                                          -95-

-------
Appendix VI-6
      -96-

-------
TESTING OF DATA AMU PWOGHAM

FILE   NUNftME   (CHfeAFIUN DATE = 09/26/78)
                                                                                    09/26/70
                                                                              PAGE
CKITtKlUN VARIABLE KU
6KOKE.N DOrtN 6Y STATUS
8Y MYR
VAHIAbLE
FOR ENTIRE POPULATION
STATUS
MYK
MYK
MYK
MYK
MYK
MYK
STATUS
MYK
MYK
MYK
MYK
MYK
MYK
STATUS
MYK
MYK
MYK
MYK
MYK
MYK
STATUS
MYK
MYK
MYK
MYK
MYK
MYK
IDLE CO
VEHICLE TAMPER STATUS
CODE VALUE LABEL

0. OK
73,
74,
75.
76.
77.
78,
30, TAMPERED
73,
74.
75.
76.
77.
78,
31. ARGUABLE
73.
74.
75.
76,
77.
78,
33. MALFNCTND
73,
74,
75.
76,
77,
78.
OF S U B P
SUM
3945.3997
562,3999
44.0000
68.4000
65,bOOO
160.5000
140.2000
63.0000
1057.5999
315,9000
308.90UO
151.8000
186. SOUU
81.5000
13.0000
2259,8996
443.7000
4B2.0000
284.5000
503.5000
421.2000
125.0000
65.5000
19,8000
13.1000
1.1000
8.1000
3.0000
20.4000
OPULATIONS
MEAN
2.0243
0,9405
1.9478
2.6308
1,0738
1.4330
0.6873
0,4tt26
2.U661
J.9487
3.4322
2.4885
2.6643
1.7717
0,5909
2.3965
3.1468
3.1097
1.9406
2.4207
2,0955
1,3587
1.6795
2.47bO
3.2750
0.1833
2,0250
0.5000
1.8545

STO DEV
2.4761
1.6939
2.2078
2.7727
1.5467
2.0U39
1.3565
1,2278
2.7055
2,9465
2.6931
2.6204
2.50/9
2.0555
1,0479
2.5726
2.6111
2.7975
2.5519
2.5435
2,3550
2.0363
2.3095
2.0197
3.6234
0.4021
1.7251
0.7899
2.8991

VARIANCE
6.1310 (
2.8694 (
4,6744 (
7.6B78 (
2.3923 (
4.1777 (
1.6401 (
l.bU/5 (
7,3197 (
8.64J1B (
7,25cib (
6.6667 (
6.2696 (
4.22b2 (
1,09150 (
6.6185 (
6 . B 1 7 9 (
7.626
-------
TtSTlNG OF  DATA  AND PKOGKAM




CHITtKiUN VARIABLE ICO
09/26/78
PAGE
VARIABLE
STATUS
STATUS)
STATUS
STATUS


i
1
CODE VALUt LAbEL
0. OK
30. TAMPtKED
31. ARGUABLE
33. MALFNCTND
WITHIN GROUPS TOTAL
ANALYSIS 0
SOUHCE SUM OF SQUARES
BETWEEN GROUPS 1099.238
WITHIN GROUPS 10843.988
ETA = 0.3034
*
SIS U r V A K I «
SUM
562.3999
1057.5999
2259.8998
65.5000
3945.3997
F VARIANCE
O.F. MEAN SQUARE
3 366.413
1945 5.575
ETA SQUARED = 0.0920
MEAN
0.9405
2,8661
2.3965
1.6795
2.0243

F SIG.
65.721 0.0000
STO DEV
1.6939
2.7055
2.5/26
2.3095
2.3612
*
*
*
*
ft
*
*
SUM OK SO N
1713.0605 ( 59b)
2693.6463 ( 369)
6234.5960 C 943)
202.oB.i6 ( 39)
10843.9664 ( 1949)


                                                           -98-

-------
Appendix VI-7
     -99-

-------
  TEST lNli Uh  DATA  AMU  PKOGHAM



  FILE    NUNAMfc.    (CHEATIUN  UATE  a  10/26/78)
                                                                      10/26/70
                                                               PAGE
      PbSTAI
LEAD CUNC STATUS
CHOSSFABULATIUN   OF   ********

                     BY  FILSTAT   FULEH NECK STATUS
  PBSTAT



    PASS








    FAIL
o
o
                                                                                                     PAGfc   1  UF   1
                      FILSTAT
COUNT I
NO* PCT IPASS FAIL
COL PCT 1
TOT PCT i o.i i.
0. I
I
I
I
1. I
I
I
I
_ f _
•*•
COLUMN
TOTAL
455
98.3
96.0
9 a ,'2
IS
7b.O
J.2
3.1
e
1.7
61.5
1.7
5
25.0
38,5
1.0
U6U 13
97.3 2.7
HUH
TUTAL
461
9S.B
20
a. 2
481
100.0
  COKRECTtO CHI SQUAKE 8    31.10187     KITH  1  DEGREE OF FREEDOM   SIGNIFICANCE »  0.0000

-------