Physical Ability Test Validation Research for Entry-Level Corrections Officers

Charles Sproule & Michael Epoca

In 1996 the Pennsylvania State Civil Service Commission (SCSC) and the Pennsylvania Department of Corrections (DOC) began a project to develop new physical tests and medical guidelines to evaluate the physical capability of persons applying for Corrections Officer Trainee positions. Doctor Deborah Gebhardt, President, Human Performance Systems, Inc. was hired as project consultant. The project was conducted from mid-1996 to early 1998 by staff of the SCSC and the DOC.

Previous research on physical tests was reviewed. This included a survey of consultants and other public jurisdictions. Following is a summary of the study and the research results. An ergonomic job analysis of Corrections Officer positions was conducted (Schneider, Sproule, Nelson & Marcinko, 1997). The job analysis study organized task information by job function and by physical ability, sensory and perceptual categories. Twenty-seven (27) essential physically demanding tasks were identified and were used as the basis for determining the physical abilities needed for the work. The findings of the ergonomic job analysis were summarized in the April 1998 issue of Assessment Council News.

Physical ability test battery selection was based on the findings of the 1997 job analysis, test transportability comparisons of the Pennsylvania job analysis study with results from other jurisdictions, test validation evidence from other studies, and a concurrent criterion-related validation study of an initial test battery selected for tryout in Pennsylvania.

The following table shows the initial test battery which included eight tests selected to assess five physical abilities determined to be important in the job analysis and ability analysis phases of the project.

Tests used to assess required physical abilities in the test validation study

Physical Ability	Tests
Static Strength	Hand Grip Arm Lift
Dynamic Strength	Arm Endurance
Trunk Strength	Sit-ups Trunk Pull
Anaerobic Power	Leg Ergometer
Flexibility	Sit & Reach Twist & Touch

The tests were tried out on a sample of 234 Pennsylvania Corrections Officer 1s in the fall of 1997. Each officer was rated, for research purposes, by two higher-level officers (usually Training Lieutenants) on performance of physically demanding job tasks. The most recent weapons qualifications scores were collected for the officers for whom these data were available. Injury, time-off, and cost data for medical treatments were also collected for the previous two-year period. The physical ability test scores of the officers were then compared with ratings on performance of physically demanding tasks, firearms scores, and the injury, time-off, and medical cost data.

The 234 test tryout participants were representative of Correction Officer 1s in Pennsylvania, except their level of physical fitness was probably higher than the typical Pennsylvania Corrections Officer 1. This characteristic of the participant group was likely to result in an underestimate of true test validity.

The test try-out results provided strong support for the use of physical ability tests. The physical ability tests were deemed much safer for participants than work simulation tests reported in another research study on physical tests for Corrections Officers.

The tests used in this research worked well, except for the Twist & Touch test. The number of tests in the final test battery will be reduced, since the tests overlapped in what they assessed.

The variation in supervisory ratings provided sufficient data to select a final test battery and to establish meaningful test passing scores. There was sufficient variation in firearms scores, the secondary criterion measure, for research purposes, but useful firearms proficiency data (sidearm and shotgun) were only available on 53% of the participants.

The validity results were very positive. The Arm Lift and Sit-ups tests correlated .76 (.57 uncorrected for criterion unreliability) with supervisory ratings on "Overall Performance of Physically Demanding Tasks." The Arm Lift test was the best single predictor of performance of physically demanding tasks. These results are similar to and supported by the results of research on the validity of physical ability tests in other public jurisdictions.

The Arm Endurance, Handgrip, and Twist & Touch tests correlated .52 (.49 without ideal regression weights) with firearms qualifications scores. The two tests that best predicted supervisory ratings (Arm Lift and Sit-ups) correlated .42 (uncorrected zero order correlation) with firearms proficiency.

The two criteria (supervisory ratings on "Overall Performance of Physically Demanding Tasks" and firearms proficiency) were combined to form an overall measure of job success. The Arm Lift and Sit-ups tests were most predictive of this overall job success measure (.66 uncorrected zero order correlation).

Differences in physical ability test performance and differences in performance of physically demanding job tasks were fairly parallel by race, gender, and age. That is, groups that did less well on the physical ability tests also tended to do less well on performance of physically demanding tasks and firearms proficiency, or there were only trivial or small differences.

The following expectancy chart illustrates the validity of the tests for predicting job performance ratings collected for research.

Expectancy Chart - Physical Ability Test Scores and Job Performance: Arm Lift + Sit-ups Scores compared with "Overall Performance of Physically Demanding Tasks"

[chart]

Percent receiving ratings "above average" on supervisory ratings of "Overall Performance of Physically Demanding Tasks"

The bar chart shows that 95% of persons who scored in the top 20% on the Arm Lift test plus the Sit-ups test were rated above "average" on their performance of physically demanding tasks, while only thirty-six percent (36%) of those who scored the lowest on the two physical ability tests were rated above "average" on their performance of physically demanding tasks.

Test fairness analysis was conducted by comparing regression lines for subgroups by race, gender and age using Arm Lift plus Sit-ups test scores as predictors and supervisory ratings on "Overall Performance on Physically Demanding Tasks" as the criterion. The tests were founds to be fair by race and gender. The tests overpredict for older persons (age 40 and over). The test will not discriminate unfairly by race, by gender, or against older persons. Similar results were found for a supplementary test battery consisting of the Leg Ergometer Test, the Sit & Reach Test, and the Arm Endurance Test.

Using a compensatory scoring model was effective in reducing adverse impact by gender. A test passing score in the range of the 10th to 20th percentile on the Arm Lift plus Sit-ups tests would screen out many persons who do not have the ability to effectively perform physically demanding job tasks.

Data on injuries, time-off for injuries, and medical and other costs associated with injuries and accidents were reviewed. About 19% of the officers in the study had an injury on the job in the previous two years. A few physical ability tests were significantly related, at a very modest level, to days-off and medical cost data. There were no significant relationships between the injury, time-off, and cost data and the Arm Lift plus Sit-ups test combination.

A more comprehensive study of accidents, injuries, medical costs, time-off, and their relationship to physical ability test scores is being considered for the future to properly determine the costs and benefits of using physical ability tests. The potential cost savings are enormous since the costs for lost wages and workers compensation insurance for Pennsylvania Corrections Officers were approximately eleven million dollars in 1997.

Based on the research results, the Pennsylvania Department of Corrections has decided to use a physical ability test battery containing the Arm Lift, Sit-ups, Leg Ergometer, Sit & Reach, and Arm Endurance tests. This battery of five tests will assess all of the physical abilities found to be important by job analysis for the performance of essential physically demanding tasks, and includes tests found to be valid in the Pennsylvania research as well as in other research.

The job analysis data are also being used by a team of physician specialists to develop medical employment guidelines. Information on the medical guidelines is expected to be available in a few months.

A copy of the "Physical Ability Test Development and Validation Report" for entry-level Corrections Officers can be obtained by writing to: Charles F. Sproule, Chief, Research Division, Pennsylvania State Civil Service Commission, P.O. Box 569, Harrisburg, PA 17108-0569, or to: Michael Epoca, Bureau of Human Resources, Pennsylvania Department of Corrections, 2520 Lisburn Road, P.O. Box 598, Camp Hill, PA 17001-0598.