Police officer physical ability testing – Re‐validating a selection criterion
Gregory S. Anderson, Darryl Plecas, Tim Segger
Gregory S. Anderson (Department of Kinesiology and Physical Education, University College of the Fraser
Valley, Mission, British Columbia, Canada)
Darryl Plecas (Department of Criminology and Criminal Justice, University College of the Fraser Valley,
Abbotsford, British Columbia, Canada)
Tim Segger (Department of Criminology and Criminal Justice, University College of the Fraser Valley,
Abbotsford, British Columbia, Canada)
Policing: An International Journal
ISSN: 1363-951X

Abstract

The aim of this study was to determine the bona fide occupational requirements of general duty police
work, and use this information to re‐validate a physical abilities test used in the police recruit selection
process. A systematic random sample (n = 267) of general duty police officers completed two
questionnaires: one concerning “average” duties, and one concerning the most physically demanding
critical incident occurring in the 12 months prior. Of those completing the surveys, observational data
were collected on every second officer, resulting in observational data collected for 121 officers,
involving the recording of all physical activities and movement patterns observed throughout a ten hour
shift. Data collected suggest there is a core of bona fide occupational requirements for general duty
police work – walking, climbing stairs, manipulating objects, twisting/turning, pulling/pushing, running,
bending, squatting and kneeling, and lifting and carrying. Many of these are involved in physical control
of suspects, and can be tested using a well designed physical abilities test that simulates getting to the
problem, controlling the problem, and removing the problem.
Like many occupations specializing in emergency services and crisis intervention, such as the armed
forces and firefighters, there is no doubt that the physical demands of police work are higher than those
occupations of a more sedentary nature. From the routines of shift‐work and uneventful patrol to the
physical responses and actions required in critical incidents, police officers must be physically capable of
performing all occupational requirements successfully, and in a way which maximizes the safety and
security of all those concerned (Anderson et al., 2000; Bonneau and Brown, 1995). Failure to screen out
candidates who cannot perform such duties can result in injury, long term disability, rapid employee
turnover and poor productivity, each having both a human and an economic cost (Brownlie et al., 1985;
Superko et al., 1988; Greenberg and Berger, 1983; Reilly et al., 1979; Wilmore and Davis, 1979; Trottier
and Brown, 1994).
In the early part of the last century, particularly in North America, police departments set arbitrary
standards of height and weight, presuming large males were best suited for the rigours of police work,
especially those aspects of the duty involving pursuit, confrontation and arrest. By the 1950s, however,
many agencies realized the arbitrary and discriminatory nature of these requirements and began to lift
the traditional height and weight restrictions which had prevented the employment of those who,
although small in stature, were capable of performing the job‐related duties. Eventually, in order to
identify some way of ensuring that recruits could indeed perform these duties, “it became necessary to
define and measure the physical abilities needed to do police work in a manner that was objective,
realistic and non‐discriminatory” (Bonneau and Brown, 1995, p. 157). This need provided the impetus
for the development of physical ability tests.

1

In both Canada and the USA, the development of such tests have been influenced by law. In the case of
Dorthard v. Rawlinson (1977: 433 US 321) heard in the US Supreme Court, the judge encouraged
occupations with specific physical demands that were related to satisfactory levels of performance to
develop physical ability or agility tests that were objective, realistic and nondiscriminatory. This is also
reflected in the Canadian Human Rights Act (CHRA, 1985) in the Bona Fide Occupational Requirements
Guideline (CHRA, 1985: S1/82‐83). Since these legislative landmarks, numerous tests have been
developed for physically demanding occupations (Farenholtz and Rhodes, 1986; Bonneau, 1994, 1996;
Jette et al., 1990; Gledhill and Jamnik, 1992a, 1992b; Stevenson et al., 1992; Jamnik and Gledhill, 1992;
Wilmore and Davis, 1979; Greenberg and Berger, 1983).
Many of the selection criteria used in the past by law enforcement agencies have been challenged in
court and dismissed as discriminatory. In particular, many of the physical abilities tests have been
questioned in regard to their validity and adverse impact on female applicants (Greenberg and Berger,
1983; Evans, 1980). Much of the problem, from the courts’ point of view, has surrounded the difficulty
in demonstrating the job‐relatedness of the tests (Anderson et al., 2000).
Since the mid‐1980s municipal police departments throughout Canada’s westernmost province, British
Columbia, have used the Police Officers’ Physical Abilities Test (POPAT) (Farenholtz and Rhodes, 1986) as
a recruit selection tool. Since that time issues related to workplace equity and discrimination in hiring
have brought such selection or screening devices under the legal microscope (Tinsley, 1999). For
example, a 1997 challenge of the fitness standard for forestry firefighters in British Columbia by a female
firefighter resulted in a 1999 ruling by the Supreme Court of Canada in favour of the firefighter. Citing
problems in test construction and the assignment of criterion scores, the decision was based, in part, on
factors related to the inability of the British Columbia Ministry of Forests to demonstrate conclusively
the validity of the test (Anderson et al., 2000).
Thus the issue of test validity has been brought to the fore in regard to defending the use of a physical
abilities test as a job qualification criterion. Furthermore, the issue needs to be addressed in both
general and specific terms. First, to what extent does the test constitute an accurate simulation of the
physical actions and responses required by the job; and, second, to what extent does the test reflect
changes to these actions and responses over time due, for example, to the introduction of new
technologies or law reform affecting the job description and duties contained therein?
This paper describes the methods used to “re‐validate” the POPAT, first developed and used as a
selection tool for municipal forces in British Columbia in 1985. These methods are compared to those
used previously in test development and the legal dimensions of test validation are also discussed.

Previous approaches to test development

Previous approaches to developing tests which purport to distinguish those physically capable of job
performance from those incapable have tended to focus either on determining a general occupational
fitness standard for the job or on determining each candidate’s ability to perform each of the physical
activities required for satisfactory job performance. While the process for determining a minimal fitness
standard is job‐specific in that it takes into account the job‐related physical demands of the occupation,
it is less job‐specific than the latter approach which involves a detailed analysis of the tasks which
comprise the job and a procedure for incorporating these very tasks, or simulations of them, into the
test itself. In this way, it is much easier to demonstrate that the test, as a selection criterion, is based on
bona fide occupational requirements (BFOR) or bona fide occupational qualifications (BFOQ).
Osborn (1976) describes a task analysis methodology for developing bona fide occupational
requirements using a self‐report questionnaire survey of current employees to collect information about
job‐related duties and which divided the process into three distinct phases. These phases include:
questionnaire development, test development and the establishment of clearly defined levels of
performance. Osborn’s goal was to construct a physical agility test consisting of components which
2

simulated actual field conditions as closely as possible. This methodology has been used by various
agencies, including the Justice Institute of British Columbia (Farenholtz and Rhodes, 1986), the Royal
Canadian Mounted Police (RCMP) (Bonneau, 1994, 1996; Gaul and Wenger, 1992), Canadian Infantry
(Jette et al., 1990), firefighters (Gledhill and Jamnick, 1992a, 1992b), Canadian Armed Forces (Stevenson
et al., 1992) and a large multi‐faceted gas company (Jamnik and Gledhill, 1992) and is now embedded in
the Bona Fide Occupational Requirements Guideline (CHRA, 1985: S1/82‐83).
In terms of police‐specific testing, several physical ability tests have been used in North America. One of
the earliest of these was developed for the Los Angeles County Sherriff’s Department (Osborn, 1976).
Using a questionnaire to collect information about essential job‐related duties, Osborn constructed a
physical agility test which resembled, “as closely as possible, conditions in the field” (Osborn, 1976, p.
44). The test included a six‐foot wall climb, a 440‐yard run, a body transport, a balance beam walk, a
vehicle push and a crawl. Time standards for each event were developed using 153 civilians and 89
deputies (although the exact procedure for deriving time standards was not identified).
Wilmore and Davis (1979) developed a job‐related physical abilities test for the selection of California
Highway Patrol officers. This test included a standard battery of field tests designed to assess the
“discrete components” of cardiovascular endurance, muscular strength, flexibility and body
composition, as well as two physical abilities tests: a barrier surmount and arrest simulation and a
dummy drag injury rescue simulation. The physical abilities testing was added to the protocol because of
“the State Personnel Board’s strong desire to correlate performance scores on the field test battery with
an officer’s ability to perform some critically important job‐related task that involves physical strength
and ability” (Wilmore and Davis, 1979, p. 35). The barrier surmount and arrest simulation mimicked the
work environment of highway patrol on a highway with a median divider and a perimeter fence. A foot
chase for a highway patrol officer would typically involve vaulting one highway divider, scaling one
perimeter fence, and arresting an uncooperative suspect, the latter of which was tested using an arrest
resistor device.
A model to assess the ability to apprehend and restrain a resisting suspect in police work was developed
by Greenberg and Berger (1983). In an effort to avoid unnecessary injury in testing, and due to the
administrative infeasibility of simulating the restraint and apprehension of a suspect, these authors
developed a regression equation that predicted the likelihood of success from the results of basic
anthropometric and strength tests. They found that those who performed best in a combative task
could be predicted from a composite strength score (maximal bench press, upright row and leg press),
left hand grip strength, weight, height and gender.
Farenholtz and Rhodes (1986) developed a physical abilities test (POPAT) using the methods of Osborn
(1976) and Wilmore and Davis (1979). The test was designed to “predict the potential physical ability of
the participant to resolve a critical incident involving the average male suspect” (Farenholtz and Rhodes,
1986, p. 46). As did these previous authors, Farenholtz and Rhodes divided their test into three distinct
portions: getting to the problem (pursuit), solving the problem (arrest) and removing the problem (lift
and carry). Getting to the problem consisted of a 400‐metre agility run, including changes in direction
and stride length, and stairs; solving the problem involved a pushing and pulling apparatus
demonstrating the ability to dynamically control 35kg. (80lb) of resistance using a machine similar to
that of Wilmore and Davis (1979) and a series of squat thrusts; removing the problem involved a lift and
carry of 45.5kg (100lb) over a 15.6m (50ft) distance.
The physical ability requirement evaluation (PARE) was developed by the RCMP and modelled after the
work of Farenholtz and Rhodes (1986). Because of the potential for adverse impact discrimination, with
65 per cent of females failing the POPAT, the RCMP re‐evaluated those sections of the test which were
most problematic: the push/pull, the vault and the time frame of the fight portion of the test (Bonneau,
1996). As a result of an independent evaluation, the push/pull resistance of 35kg was confirmed, but the
vault component was eliminated. The original premise was that officers need to perform activities under
3

maximal stress; however, since the run portion of the POPAT elicits a near maximal level of exertion the
vaulting was deemed unnecessary. Further, as the fight sequence of the POPAT lasted almost twice as
long as the average physical encounter in the field, the duration of this portion of the test was reduced.

The POPAT re‐validation study

As a “re‐validation” exercise, this study does not represent an attempt to develop, from scratch, a new
physical abilities test for use as a selection tool for police officer hiring. However, by using and building
on the methods for test development used by Osborn (1976) and Farenholtz and Rhodes (1986), the
methods described here could well be used in situations in which no current testing criteria exist.
Besides the survey of general duty police officers used in developing these previous tests, the present
researchers also included an observational component, involving ride‐alongs with officers on patrol. By
elevating this design to a multi‐method approach similar to that used by Leamon (1999) in her approach
to developing screening criteria for the selection of US Border Patrol agents, the addition of this
component provided an opportunity to assess the construct validity of self‐report measures of job
requirements and related levels of performance. Furthermore, the need to link physical requirements to
the actual activities in and demands of the field environment is critical to reducing the vulnerability of
test validity to discriminatory law suits (Copay and Charles, 1998, p. 418).
The survey component
A systematic random sample of officers listed as being assigned to “general duty” in June 1998 was
drawn from all 11 municipal police departments in British Columbia. Every third officer was selected
from a list provided by each department, yielding a total sample of 279 officers, of which 267 (96 per
cent) actually participated in the survey. Each officer received two questionnaires, one which asked
them to describe the physical aspects of their job “on average” (the Physical Work Record Survey Form)
and another which asked them to describe the most physically demanding critical incident that they
experienced in their most recent 12 months of work (the Critical Incident Survey Form). Both
questionnaires were presented in a package along with a letter of introduction explaining the general
purpose of the study, the voluntary nature of their participation, the confidentiality of all responses and
instructions for returning the questionnaire to the researchers in a sealed envelope. Questionnaire
packages were distributed to officers through the Training Officer in each department who also
collected them for forwarding to the researchers.
The observation component
The second component of the study involved asking a sub‐sample of every other officer surveyed to
participate in a researcher ride‐along exercise. In this component a research assistant rode with the
officer for a full shift during which a detailed record was made of all physical activities performed by the
officer. The research assistant recorded all instances of each of 30 different physical activities occurring
within each minute of each shift, beginning from the time the officer leaves the briefing at the start of
the shift through to attendance in the locker room at the end of the shift. As a rule, observations were
recorded immediately on minute‐by‐minute tracking sheets; on some occasions they were recorded on
audio‐tape as the activity occurred and later transcribed onto tracking sheets.
Data were collected on as many as 720 minutes per 12‐hour shift and as many as nine physical activities
per minute. These data were combined with those from the questionnaire surveys into a single data set.
Scheduling of ride‐alongs ensured that all shifts and all days of the week were proportionately
represented. The observed data were collected during 121 ride‐alongs over a 12‐month period spanning
dates in both 1998 and 1999. The observational period included, in total, 75,867 minutes representing
1,265 hours of observation.
4

The samples
The primary sample of officers is representative of all officers assigned to general duty/patrol within
municipal police departments in British Columbia. Table I shows there are wide ranges in the
characteristics of participating officers: some were young (e.g. 24 years) or had only a few months of
service (e.g. three months), while others were nearing retirement at 55 years old or had nearly 30 years
(e.g. 353 months) experience in patrol division and this distribution is mirrored in the ride along sub‐
sample. The average officer was 36 years of age and had ten years of experience, which means that they
were hired, on average, at 26 years of age, which is, in fact, the average age at which municipal police
officers are hired in British Columbia.
In considering the sample of officers in the ride‐along component it should be kept in mind that, while
the study began with a systematic random sub‐sample of officers, it was necessary to accommodate the
logistics of scheduling research assistants to complete a representative range of shifts over the course of
a one‐year period. Also, during that year, selected officers moved out of patrol or were otherwise
unavailable for scheduled shifts. Accordingly, some originally selected officers were replaced with others
from the larger sample, and not all shifts were covered exactly proportionate to their occurrence over
the days, weeks and months of the year. Nonetheless, the researchers are reasonably confident that the
resulting sample of officers and shifts constitutes a very good representation of the officers and work
activity involved in municipal policing in British Columbia.
General duty task analysis
Each of the 267 officers participating in the questionnaire survey completed the Physical Work Record,
an instrument designed to provide self‐reports regarding their assessments of the physical demands of
general duty police work. The questionnaire asked officers to describe:
• how necessary it is that they be able to perform 15 selected physical activities;
• how frequently they perform these selected activities;
• how much time they spend on each activity during an average shift; and
• how much effort they feel is required to perform each activity.
The activities listed on the questionnaire were not intended to represent a comprehensive task analysis
of general duty police work; however, they did include those proposed in earlier studies for
consideration as bona fide occupational requirements of police work.
Necessity. In terms of how necessary officers believed it was that they be able to perform each of the 15
listed activities, the vast majority of respondents felt all but two (crawling and lifting above the
shoulders) were necessary. The necessity of officers being able to perform these duties was confirmed
further by ride‐along observations. The research assistant necessity ratings were identical to those
assigned by the officers in all but three cases see Table II.
Frequency and duration. The results regarding frequency mirrored those of necessity. When ranked
from the most frequently to the least frequently performed, the ranking of activities is similar to that
obtained for necessity. Only two activities were cited by the majority of responding officers as being
“never” or “seldom” performed (lifting above the shoulders and crawling). Ride‐along observers
provided confirmation of the reliability of these self‐reports see Table III. In terms of the amount of time
spent engaged in various forms of job‐related physical activity, Table IV reveals that officers surveyed
reported spending, on average, more than half of each shift (373 minutes) sitting, just over two hours
standing and about one and a half hours walking.
Effort. Officers surveyed were asked about the amount of effort used to perform each of six physical
activities at work. As Table V shows, all but one (climbing stairs) were cited as requiring at least
5

“medium to maximum” or “maximum” effort. Ride‐along observers again assigned the same average
ratings as were assigned by the participating officers.
Recorded observations of physical activities. The ride‐along component of the study involved research
assistant observers accompanying 121 officers on a regular shift during which work activities were
observed and recorded on a minute by minute basis. Observation records were made of the minutes (or
times) in each shift that officers performed each of those activities listed in the officer survey. These
results are summarized in Table VI and are extrapolated to yield estimates of the average number of
shifts and minutes (or hours) per month that each officer will be required to perform these activities see
Table VII. Over all, there is no activity listed that officers should not expect to perform at least once per
month, and the majority will be performed at least once per shift.
Recorded vs. observed activities. Perhaps the most striking finding with regard to the observation data is
the extent to which it corresponds to the self‐report data (compare for instance the self‐report
estimates in Table IV to the data presented on sitting, standing, walking and bending in Table VI).
Further, the 12 most frequently occurring activities observed are the same activities that officers most
frequently cited as being “necessary” on the job, with at least 84 per cent of respondents describing
each of these as either “somewhat necessary” or “very necessary”. In addition, of the 12 most
frequently observed activities, 11 were self‐reported as being “often” or “constantly” performed (the
twelfth, running, was cited by 46 per cent of officers as being “often” or “constantly” performed).
Critical incident task analysis
As part of the survey component, officer respondents were asked to complete a Critical Incident Survey
questionnaire, providing information about their most physically demanding critical incident within their
most recent 12 months of work. A critical incident was defined to include all physical custody (arrests),
all vehicle and foot pursuits, all dispatched code responses (emergency), all motor vehicle accidents that
require physical work and all calls which present an active threat to life and/or property. These data
provide a measure of what the physical demands of police work can be in the most difficult of
circumstances.
The nature of critical incidents reported. Survey data indicated that critical incidents can occur at any
time and under a variety of circumstances. For instance, while a greater percentage of critical incidents
occurred at night and in May, June and July, significant amounts occur at all times of the day and
throughout the year. Further, these incidents commonly occurred without much forewarning. Indeed,
37 per cent were initiated from observations of an officer as opposed to being despatched to the
incident (50 per cent) or being called as backup (4 per cent). While many officers were able to describe
the nature of the incident as either relating to a motor vehicle accident (4 per cent), domestic violence
(12 per cent), social violence (10 per cent), or resistance resulting from an investigation (31 per cent), a
full 43 per cent of officers responding had to describe the nature of the incident as something else (the
list of which is too numerous to mention here). Even the locations of critical incidents were without a
pattern. About half (51 per cent) of them were reported to have occurred on a street, highway, road,
but they also occurred in many other locations.
In some respects then, there is no easy way to characterize critical incidents which responding officers
describe as the most demanding; officers have to assume that such incidents occur any time, almost
anywhere, and without warning. On the other hand, the results make it very clear that there are some
commonalities among those incidents which officers describe as the most demanding. Specifically, they
are likely to involve the officer having to deal with one or more subjects, each of whom is likely to be a
suspect as opposed to a victim or anyone else. Further, while 89 per cent of the time the incident will
6

involve at least one subject (some incidents have no subjects; e.g. bomb scares), 20 per cent of the time
it will involve two subjects, 12 per cent of the time it will involve three, and 6 per cent of the time it will
involve as many four. More significantly, the subject is likely to be a young male, have average or better
physical abilities, and be in a less than desirable mental state. Most of the time (60 per cent), at least
one of the subjects will be violent and often, as Table VIII shows, the subject will often be taller and
heavier than the officer involved.
Physical demands of critical incidents. Given the characteristics of the suspects involved in the critical
incidents reported, it is not surprising that these incidents are also reported to be very physically
demanding. Indeed, to gain control of such incidents officers are likely required to engage in a broad
range of physical activity, and in doing so they can expect to exert considerable effort see Table IX.
Obviously, the primary reason that the physical demands are so great is that the suspect is resisting
control in a variety of ways. The suspect is likely to pull or push on the officer to resist, and otherwise
fight during the incident. The circumstances can also be extremely dangerous as suspects may also use
(or threaten to use) a club, knife, or gun, and even attempt to take the officer’s weapon.
Once the officer has control of the critical incident, significant physical demands commonly continue in
removing the suspect. As Table X shows, the officer will often be required to lift, pull, drag, and push the
suspect and in doing so be required to exert considerable effort. It is also worth noting that half the time
(54 per cent) the officer is required to run to get to the incident, make sharp turns in the process, and do
all of this exerting considerable effort. As well, the officer may be required to climb, vault, or jump
objects although most officers do not report these as requiring considerable effort. Finally, the results
showed that the critical incidents reported were generally not over quickly; while 20 per cent were over
in less than five minutes, 65 per cent lasted ten or more minutes, and nearly 15 per cent lasted an hour
or more.
Recorded observations of physical activities. Over the course of the ride‐alongs 14 incidents involving 12
different officers were recorded as “incidents involving significant resistance from a suspect”. Each of
these was detailed using the same Critical Incident Survey Form as used in the survey component of the
study.
The incidents themselves lasted an average of ten minutes (median = eight minutes) with a range of two
to 29 minutes. They occurred in a variety of circumstances and involved male suspects who were most
likely under the influence of alcohol or drugs. One‐half of the suspects were violent and 79 per cent
were of average or better fitness. In 36 per cent of the incidents the suspects were heavier than the
officers, and in the same percentage of incidents (36 per cent) the suspects were taller than the officers
see Table XI.
Getting to the problem. In terms of “getting to the problem”, 50 per cent of the officers were required
to run, and ran, on average, 87 metres (with a range of 5 to 350 metres). Of those required to run, 43
per cent reported to the observer that they expended either “difficult” or “maximum” effort.
Controlling the problem. In “controlling the problem”, 93 per cent of officers were required to push and
pull the suspect and 86 per cent had to twist and turn and use control holds see Table XII. Other
frequent activities included applying a wrist or arm lock (72 per cent), wrestling (57 per cent) and using a
take‐down (43 per cent). Also, 36 per cent lifted and carried the suspect and, in 79 per cent of cases, the
officer handcuffed the suspect.

7

Removing the problem. Table XIII summarizes the physical activities observed as officers dealt with this
aspect of the critical incidents. A large percentage of officers were required to lift, pull, drag and push
objects of significant mass and to expend considerable effort in the process.
Recorded vs. reported activities. As was the case with the general duty survey, there was a remarkable
correspondence between the self‐reported and the observed activities involved in responding to critical
incidents. Keeping in mind the large difference in sample sizes between the survey (n = 267 respondents
reporting the one most demanding incident in the previous 12 months) and the observed critical
incidents (n = 14), the similarities between Table IX (reported) and Table XII (observed) are considerable,
as are those between Table VIII and Table XI. In fact, as noted earlier, the most striking finding emerging
from this re‐validation exercise is the evidence from the observation data and the independent observer
ratings of frequency, duration and effort with regard to job‐related activities, which tends to
substantiate the accuracy of self‐reported estimates of these dimensions of police work. In other words,
by using these additional measures, it has become possible to undertake a validation not only of the
POPAT as a bona fide discriminant criterion, but also of the self‐report measures on which it has been
based. The data from this study clearly suggest that the self‐report survey method can provide highly
accurate measures and results on which to construct job‐related physical abilities tests and simulations.

POPAT as a selection criterion

Selecting the right people for police work is not only important to the employer, but also in the best
interest of the public. Police work, in general, can be quite sedentary; however, in the interest of public
safety, police are expected to have the ability to apprehend (which may include running, tackling,
pushing, pulling and wrestling), arrest and contain criminals (perform take‐downs and handcuffing),
remove people from damaged vehicles (lifting, carrying, pulling), control large crowds, and separate
individuals who are arguing or fighting (pushing, pulling, restraining). Several of these tasks require
maximal effort, and are extremely physically challenging.
Farenholtz and Rhodes (1986) developed the POPAT using the methods developed by Osborn (1976)
and Wilmore and Davis (1979). Using a questionnaire survey to identify essential job‐related duties,
Osborn constructed a physical agility test with components which reflected conditions in the field.
Wilmore and Davis included two physical abilities tests: a barrier surmount and arrest simulation and a
dummy drag test. The POPAT was designed to “predict the potential physical ability of the participant to
resolve a critical incident involving the average male suspect” (Farenholtz and Rhodes, 1990, p. 46), and
was later used as the foundation of the PARE developed by the Royal Canadian Mounted Police
(Bonneau and Brown, 1995). As did the previous authors, Farenholtz and Rhodes divided their test into
three distinct portions: getting to the problem (pursuit), solving the problem (arrest) and removing the
problem (lift and carry).
Getting to the problem
During a critical incident “getting to the problem” typically involves a pursuit or “catching up to” the
suspect. In the present study getting to the problem involved walking, running, climbing over objects,
vaulting objects, jumping down from objects and over objects, with multiple tasks required in each
pursuit. Most often (56 per cent of the time) officers were required to run to the problem with officers
exerting maximal effort in 75 per cent of cited incidents.
In the POPAT, the ability to “get to the problem” is tested through the use of a 400 meter agility run
which includes changes in direction and stride length, and the climbing of stairs. As reported earlier, 88
per cent of the respondents in the present study considered running to be an essential task, 92 per cent
considered climbing stairs, and 67 per cent considered leaping and jumping to be essential tasks.
8

The existing data supports the applicability of the POPAT elements related to “getting to the problem”,
although the run distance and times are longer than those reported. Officers report having to chase
suspects on foot, changing direction and speed often, while avoiding or manoeuvering over or under
objects. A total of 75 per cent of the officers surveyed in the present study reported encountering
maximal exertion during their pursuits, and this is reflected in the POPAT – near maximal cardiovascular
stress occurs in 90 per cent of the participants by the fourth lap (Bonneau, 1996).
Rhodes and Farenholtz (1992) found the average time on the run portion of the POPAT to be two
minutes and 13 seconds. While the run is longer than the median distance encountered, the distance
covered in the POPAT (400m) would fall at the 85th percentile of distances reported by Farenholtz and
Rhodes (1986), and police officers should expect to run further distances at least once a year. Further,
the total number of stairs encountered represents real life situations, although in the POPAT the stairs
occur in blocks of six stairs rather than complete flights of stairs.
Controlling the problem
During a critical incident “controlling the problem” typically involves a physical struggle and arrest. In
the present study it was found that 57 per cent of the suspects pushed or pulled on the officer to resist
arrest, while 76 per cent of the officers cited pushing and pulling a suspect; 88 per cent of the police
officers surveyed considered pushing and pulling an essential task.
Simulation of “controlling the problem” in the POPAT involves a pushing and pulling apparatus
demonstrating the ability to dynamically control 35kg (80lb) of resistance such as that found when
fighting an averaged sized individual. The resistance encountered has been validated elsewhere (as
reported in Farenholtz and Rhodes (1990) and Bonneau (1996)), reflecting the resistance encountered
during the arrest of the average male suspect.
There are very few methods of assessing one’s ability to control a dynamic resistance which are
appropriate for a field test, and do not involve a large skill component. Standard fitness tests (push‐ups,
pull‐ups and grip strength) do not correlate well with the “fighting” component of the POPAT (Rhodes
and Farenholtz, 1992); however, data from the present study would support the notion that the pushing
and pulling portions of the POPAT do reflect actual field conditions. For this reason it can be assumed
that the pushing and pulling activities measured in the POPAT are essential job‐related physical
activities, and competency in these tasks is not merely a product of fitness.
The push and pull segment of the POPAT, however, while measuring unique physical abilities, uses a
static body alignment without twisting and turning at the waist, which is not realistic. A total of 76 per
cent of the officers in the present study reported the need to twist and turn at the waist during the
control and arrest of a suspect. There is, however, a safety issue concerning the lower back should
twisting and turning at the waist be implemented into the test. This safety concern would also be the
case should simulated “take‐downs” and other controlling manoeuvers be introduced into the test, even
though officers report these tasks to be essential.
A report by Loree (1995) clearly demonstrates that force is required to resolve many situations, and that
failure to use force may jeopardize the safety of the public or fellow police officers. However, 80 per
cent of the physical encounters last less than one minute, which is not in line with the time required to
complete the “fight component” of the POPAT. In its present form Rhodes and Farenholtz (1992) report
the average time to complete the “fight component” of the POPAT to be two minutes and 16 seconds.
The time required to complete this section involves a period of time immediately after the push/pull
during which subjects are required to vault over a height of 0.9 meters, perform a controlled landing,
and then fall alternately on their back and their stomach. This portion of the test takes, on average, 60
seconds and was designed to elicit a maximal response from the candidate, during which conscious
decisions to fall to the stomach or the back had to be incorporated. The movement patterns, however,
are not specific to police work. In light of Bonneau’s (1988) demonstration that the run portion of the
9

POPAT elicits a maximal cardiovascular response in 90 per cent of the participants after the completion
of the fourth of six laps, the authors concur with previous research that suggests that the vaulting
segment of the POPAT may not be required, or should be modified so as to reduce the length of the
“fighting” segment of the POPAT.

Removing the problem

During a critical incident “removing the problem” typically involves the tasks of lifting or dragging a
suspect, or carrying confiscated materials. In the present study 46 per cent of the officers surveyed
reported having to lift and/or carry an object below shoulder level, 40 per cent reported pulling a person
or object, 36 per cent reported pushing an object, 22 per cent reported dragging a person, and 6 per
cent reported having to carry an object above shoulder level. Of those objects manipulated, 80 per cent
of them involved moving a person.
Of the officers surveyed, 84 per cent in the present study considered lifting and carrying as essential
tasks. These tasks are often performed at the end of a critical incident (46 per cent of the time), soon
after maximal effort has been exerted. The POPAT simulates this scenario by having a lifting and carrying
component following the modified squat thrusts at the end of the fight portion. The subject is required
to lift and carry a 45.5kg sack a total of 15.25m, while negotiating one corner. Both the distance and
mass carried appear reasonable, and if anything, conservative.
Summary
Analysis of present and past data indicates that the POPAT is a job‐specific test which measures physical
abilities required in the course of duty of a general duty police officer see Table XIV. The present data, in
general, support the physical tasks encountered in the POPAT. Discrepancies in distances covered and
times of the events are evident however. That portion which is least supported would be the inclusion of
the vaults at the end of the fight segment, with the vaulting involved in the POPAT being far more
frequent than that reported, and out of sequence, disrupting the normal flow of pursuit, arrest, and
removal of a person and/or property.
Data from the present study are similar to those used to construct the POPAT and its more recent
iteration, the PARE (RCMP), especially when rank ordering as opposed to percentage of respondents
reporting an activity is examined see Table XV. It is important to keep in mind that the percentage
differences are likely due to differences in methodologies among the studies. For example, in the
present study, officers completed one survey reporting on their average shift and one reporting on the
most physically demanding critical incident experienced in the past year, while the previous studies
obtained multiple surveys from each officer, and include data concerning full shifts and not single
incidents. By focusing on the most demanding incidents, the present study is able to address the validity
of the POPAT from the point of view of its ability to reflect the higher levels of physical demand made on
police officers. This is especially important in considering the authors’ recommendation for removal of
the vaulting segment and the consequent reduction in the length of the fight simulation sequence of the
test.

Legal issues

Task analyses performed by police departments in developed nations all indicate that there is a core of
essential job‐related physical abilities that must be performed in the regular course of duty. In fact,
there is “marked similarity in the type and intensity of physical activities reported” (Bonneau and Brown,
1995, p. 158). Few departments, however, have mandatory fitness or physical abilities requirements (or
testing for same) past a probationary period (IACP, 1988).
Several reports have demonstrated that the majority of police work is essentially sedentary, with 80‐90
per cent of the job being devoted to tasks involving limited physical activity (Maher, 1984; Balkin, 1988;
10

Farenholtz and Rhodes, 1986). In fact, there is ample evidence supporting the notion that the physical
requirements of police work do not occur frequently enough to maintain officers’ fitness levels
(McGhee, 1976; Wilmore and Davis, 1979; Charles, 1982; Metivier et al., 1982; Maher, 1984; Bonneau
and Brown, 1995). Collingwood (1974), in comparing the average police officer to the average inmate,
found inmates to be more physically fit. Gaul and Wenger (1992), in examining the health habits of
RCMP members outside of their work, found that only 17 per cent of the police officers surveyed
engaged in physical activity at an appropriate intensity three times per week.
From a legal standpoint, this information opens the door to several lines of litigation. Task analyses
demonstrate that the physical nature of police work does not change, at a given rank, with years of
service (Farenholtz and Rhodes, 1986; Bonneau, 1988), yet there are no mandatory fitness or physical
abilities requirements of police officers who have been in service for longer than their probationary
period (IACP, 1988). Carter (1982, p. 15), legal advisor for the Waco, Texas police department, clearly
states that “it makes no sense to select police officers on the basis of their physical fitness and abilities,
and then have no requirement that minimum fitness and abilities are maintained”. If physical fitness is
truly job‐related, all law enforcement officers should be required to maintain the same levels of fitness.
Moving towards global standards
Assuming the tasks identified are essential to satisfactory job performance, all individuals in similar
positions should be required to reach the same standard. According to Bonneau and Brown (1995, p.
162) recent adjudication has suggested that “an employer cannot demand from applicants a level of
performance not asked for from incumbents”. If such a standard is not universally applied, litigation
directed towards the standards will be difficult to defend (McGhee, 1976; Ebel, 1977; Evans, 1980;
Maher, 1984; Blair, 1995; Bonneau and Brown, 1995).
Moving towards a universal application is difficult, however, without the measure appearing to be
“punitive” in nature. If the employer can demonstrate that the standards are critical to job‐
performance, reasonable to obtain, and it is “impossible to accommodate those who do not meet the
standard, since it is based on the very nature of the work” (Bonneau and Brown, 1995, p. 161), it would
be necessary to remove those from their positions who could not reach the standards set. This,
according to Canadian courts, is acceptable provided the employees affected have been reasonably
accommodated and provided a sufficient grace period to meet the standards set.
The RCMP have moved towards implementing a force‐wide job performance standard. Gaul and Wenger
(1992) describe the results of a two‐year longitudinal study concerning the implementation of these
standards, and the educational support they received during this period. Initially, of those attempting
the, PARE 69 per cent were able to complete the PARE in less than four minutes, and 83 per cent were
able to complete the PARE in less than 4:30. After an 18‐month period these numbers improved to 72
per cent completing the PARE in less than four minutes and 88 per cent in less than four minutes and 30
seconds. While there were several medical exemptions to performing the PARE (20 per cent), 78 per
cent of the constables were able to complete the PARE in under four minutes at the beginning of the
project, while 93 per cent of them completed the PARE in under 4:00 minutes at the end of the project
with fewer higher ranking officers passing the test.
“To justify a discriminatory workplace standard, an employer must now establish the following steps on
a balance of probabilities: (1) the standard was selected for legitimate purposes rationally connected to
job performance, which includes, but is not limited to safety and efficiency…; (2) the standard was
adopted honestly and in good faith, without an unlawful ulterior motive; and (3) the standard is
reasonably necessary to accomplish the employer’s legitimate work‐related purpose, which can only be
shown if it is ‘impossible to accommodate individual employees sharing the characteristic of the claimant
without imposing undue hardship on the employer’ [Public Service, 1999)] (Tinsley, 1999, pp. 41‐2).”
11

This ruling sets out a clearly defined precedent with components which must be addressed in a
successful defense of the POPAT; however, in combination with the findings of this and previous studies
defining the occupational requirements of police work (Bonneau, 1988; Farenholtz and Rhodes, 1990), a
defense of the POPAT as a pre‐employment screening device should be easier to mount and defend
successfully.

Notes

1. A third component of the study involved each of those officers in the ride‐along sub‐sample
being fitted with a heart monitor which records every heartbeat. The monitor was activated the
minute the researcher began to record observations of physical activity at the beginning of the
shift and, at the end of the shift, the heart rate data were downloaded into lap‐top computers.
Analysis of these data is still ongoing.
2. The reason officers assigned a lesser rating to each of the three, in the view of the research
assistant observers, is that they perhaps did not include: • getting in and out of the patrol car
and using the in‐car mobile data terminal (MDT) as instances of “twisting and turning the upper
body”.
3. • using the radio as “handling and manipulating objects”; and • partial bending, such as in
standing while working on report writing and in reaching for objects below the waist level.

The conclusions suggested that the PARE was a reasonable and achievable standard for “most able‐
bodied regular members… if graduated goals are provided for members to attain (p. 83)”.
Tinsley (1999, p. 31) discusses the use of pre‐employment tests as measures of bona fide occupational
requirements. As defined by Tinsley, a bona fide occupational requirement is a rational link between a
policy that “discriminates on a legislatively prohibited ground” and on occupation, and is defendable in a
court of law. For example, in a recent case (Public Service Employee Relations Committee v. British
Columbia Government and Service Employees Union, 1997) the British Columbia Court of Appeal
overturned an arbitrator’s decision that required an employer to accommodate an employee on the
basis that indirect discrimination had occurred. This case involved a female employed in a physically
demanding role as a crew member of a British Columbia Ministry of Forests first response firefighting
team. This role places large demands on the aerobic and muscular systems, and these demands have
been defined as bona fide occupational requirements, as failure to perform on the job may jeopardize
the safety of the crew member, the crew, or the general public. However, the fitness standard created
resulted in a passing rate of only 35 per cent for females as compared to 65 per cent for males.
While on its face the fitness standard appears to be neutral, with all employees having to reach the
same fitness standard, the large difference between the passing rates of males and females would
suggest the fitness standard indirectly discriminates against women. In the ruling by the British
Columbia Court of Appeal, the court reasoned that “because the distinction was not based on personal
characteristics attributed to an individual solely on the basis of association with a group (i.e. females),
but rather based on individual merits and capabilities relevant to the occupation, discrimination did not
occur” (Tinsley, 1999, p. 25).
This decision was later appealed to the Supreme Court of Canada and a decision was rendered in favour
of the firefighter. The decision was based on several factors, including: failure to demonstrate undue
hardship; failure to demonstrate a safety risk (either personal or other); and problems in test
construction and the assignment of criterion scores. According to Tinsley, this decision inaugurated a
new unified approach to determining whether workplace standards are discriminatory by providing a
three‐step test for assessing BFOR validity:
12

Table I.
Characteristics of officers sampled

Table II.
Observed vs. officer ratings of how necessary it is that they be able to perform various physical activities

13

Table III.
Observed vs. officer ratings of how frequently various physical activities are performed during duty

Table IV.
Average amount of time police officers usually spend during a shift performing various physical abilities

14

Table V.
Observed vs. officer ratings of how much effort they use in performing various physical activities during
duty

15

Table VI.
Percentage of officers observed in various forms of physical activity each shift and number of minutes
involved in each form

16

Table VII.
Estimated number of shifts and number of minutes (hours) per month each officer will be required to
perform various physical activities

17

Table VIII.
Relative size and age of subjects involved in critical incidents reported

18

Table IX.
Physical activities and effort required in controlling critical incidents reported

Table X.
Physical activities required in removing the problem in critical incidents reported
19

Table XI.
Relative size and age of subjects involved in an observed critical incident

20

Table XII.
Activities performed by officers in controlling the problem

Table XIII.
Physical activities performed by officers in removing the problem during observed critical incidents

21

Table XIV.
Most frequently performed physical tasks as found in various task analyses

Table XV.
Frequency and median values for tasks performed in “getting to the problem”

22

References

Anderson, G.S., Plecas, D. and Segger, T. (2000), The Physical Abilities Required for General Duty Work in
British Columbia’s Municipal Police Forces: Re‐validating the Police Officers’ Physical Abilities Test,
Justice Institute of British Columbia, New Westminster.
Balkin, J. (1988), “Why policemen don’t like policewomen”, Journal of Police Science and Administration,
Vol. 16, pp. 29‐37.
Blair, D. (1995), Legal Opinion: Rougeau v. Vancouver Police, Victoria Square Law Office, Vancouver.
Bonneau, J. (1988), Task Analysis RCMP: Surrey and Rural Studies, RCMP Health Services Directorate,
Ottawa.
Bonneau, J. (1994), Revision to the Doghandlers Task Analysis, RCMP Health Services Directorate,
Ottawa.
Bonneau, J. (1996), PARE Standard Summary, RCMP Health Services Directorate, Ottawa.
Bonneau, J. and Brown, J. (1995), “Physical ability, fitness and police work”, Journal of Clinical Forensic
Medicine, Vol. 2, pp. 157‐64.
Brownlie, L., Brown, S. and Diewert, G. et al. (1985), “Cost‐effective selection of fire fighter recruits”,
Medicine and Science in Sports and Exercise, Vol. 17 No. 6, pp. 661‐6.
Canadian Human Rights Act (1985), Canadian Human Rights Commission, Ottawa.
Carter, R.W. (1982), “The chief’s counsel: legal aspects of maintaining physical fitness”, The Police Chief,
March, p. 15.
Charles, M.T. (1982), “Women in policing: the physical aspect”, Journal of Police Science and
Administration, Vol. 10 No. 2, pp. 194‐205.
Collingwood, T.R. (1974), “A comparison of policemen versus offender fitness”, Monograph Series of
Fitness, Vol. 1, pp. 1‐5.
Copay, A.G. and Charles, M.T. (1998), “Police academy fitness training at the Police Training Institute,
University of Illinois”, Policing: an International Journal of Police Strategies and Management, Vol. 21
No. 3, pp. 416‐31.
Ebel, R.L. (1977), “Comments on some problems of employment testing”, Personnel Psychology, Vol. 30,
pp. 55‐63.
Evans, D.H. (1980), “Height, weight, and physical agility requirements – Title VII and public safety
employment”, Journal of Police Science and Administration, Vol. 8 No. 4, pp. 414‐36.
Farenholtz, D.W. and Rhodes, E.C. (1986), “Development of physical abilities test for municipal police
officers in British Columbia”, Canadian Journal of Applied Sport Sciences, Vol. 11 No. 3, abstract.

23

Farenholtz, D.W and Rhodes, E.C. (1990), “Recommended Canadian standards for police physical
abilities”, Canadian Police College Journal, Vol. 14 No. 1, pp. 37‐49.
Gaul, C.A. and Wenger, H.A. (1992), RCMP Physical Abilities Requirement Evaluation Demonstration
Project: Final Report, RCMP Health Services Directorate, Ottawa.
Gledhill, N. and Jamnik, V.K. (1992a), “Development and validation of a fitness screening protocol for
firefighter applicants”, Canadian Journal of Sport Science, Vol. 17 No. 3, pp. 199‐206.
Gledhill, N and Jamnik, V.K. (1992b), “Characterization of the physical demands of firefighting”,
Canadian Journal of Sport Science, Vol. 17 No. 3, pp. 207‐13.
Greenberg, G.J. and Berger, R.A. (1983), “A model to assess one’s ability to apprehend and restrain a
resisting suspect in police work”, Journal of Occupational Medicine, Vol. 25 No. 11, pp. 809‐13.
International Association of Chiefs of Police (1988), “IACP 1987 annual law enforcement survey:
executive summary”, The Police Chief, January, pp. 38‐43.
Jamnik, V.K. and Gledhill, N. (1992), “Development of fitness screening protocols for physically
demanding occupations”, Canadian Journal of Sport Science, Vol. 17 No. 3, pp. 222‐7.
Jette, M., Kimick, A. and Sidney, K. (1990), “Evaluation of an indoor standardized obstacle course for
Canadian infantry personnel”, Canadian Journal of Sport Science, Vol. 15 No. 1, pp. 59‐64.
Leamon, J.A. (1999), “Improving the selection of US Border Patrol agents: screening for productivity”,
The Police Chief, August, pp. 65‐73.
Loree, D.J. (1995), Violent Incidents, CCAP RCMP Internal Report, Ottawa
Maher, P.T. (1984), “Police physical ability tests: can they ever be valid”, Public Personnel Management
Journal, Vol. 13, pp. 173‐83.
McGhee, G.L. (1976), “Job‐related pre‐employment physical agility tests”, The Police Chief, January, pp.
42‐3.
Metivier, G., Gauthier, R. and Gaboriault, R. (1982), “A screening test for the selection of police officers”,
Canadian Police College Journal, Vol. 6 No. 1, pp. 1‐12.
Osborn, G.D. (1976), “Physical agility testing: validating physical agility tests”, The Police Chief, January,
pp. 43‐5
Reilly, R.R., Zedeck, S and Tenopyr, M.L. (1979), “Validity and fairness of physical ability tests for
predicting performance in craft jobs”, Journal of Applied Psychology, Vol. 64 No. 3, pp. 262‐74.
Rhodes, E.C. and Farenholtz, D.W. (1992), “Police officer’s physical abilities test compared to measures
of physical fitness”, Canadian Journal of Sport Science, Vol. 17 No. 3, pp. 228‐33.

24

Stevenson, J.M., Bryant, J.T., Andrew, G.M., Smith, J.T., French, S.L., Thomson, J.M. and Deakin, J.M.
(1992), “Development of physical fitness standards for Canadian armed forces younger personnel”,
Canadian Journal of Sport Science, Vol. 17 No. 3, pp. 214‐21.
Superko, R.H., Bernauer, E. and Voss, J (1988), “Effects of a mandatory health screening and physical
maintenance program for law enforcement officers”, The Physician and Sports Medicine, Vol. 16 No. 9,
pp. 99‐109.
Tinsley, P. (1999), “Discrimination: a legal concept in an employment context”, unpublished manuscript,
University of British Columbia, Vancouver
Trottier, A. and Brown, J. (1994), Police Health: A Physician’s Guide for the Assessment of Police Officers,
Canada Communications Group, Ottawa.
Wilmore, J.H. and Davis, J.A. (1979), “Validation of a physical abilities field test for the selection of state
traffic officers”, Journal of Occupational Medicine, Vol. 21 No. 1, pp. 33‐40.

25