Pre- and post-job offer physical ability tests (PATs) continue to face challenges from the Equal Employment Opportunity Commission (EEOC) and the Office of Federal Contract Compliance Programs (OFCCP) and occasionally private litigants.
For example, late last year, female employees of the Sheriff’s Office of Citrus County, Florida, filed a lawsuit alleging their employer used “unvalidated and unreliable” PAT to select employees for continued employment and promotion. The women claim that the PAT has a disparate impact based on sex, contending it has resulted in women receiving lower compensation, being denied transfer or promotion opportunities, and being discharged or forced into early retirement more frequently than men. As the allegations in the case suggest, the use of a PAT can carry significant liability if the test has a disparate or adverse impact.
Disparate impact occurs when a test appears neutral on its face but has a discriminatory effect in practice because it results in a “substantially different” and adverse rate of selection for members of a particular sex, age, race, or ethnic group.
Assessing Disparate Impact
To identify whether a test has a disparate impact, the EEOC, as well as other enforcement agencies and courts, have adopted the “four-fifths (or 80 percent) rule.” The rule is a nonstatistical comparison of selection ratios of protected and nonprotected groups of applicants. The rule is used to determine whether the selection rate for any group is substantially less (i.e., less than 80 percent) than the selection rate for the highest group. In the case of a PAT, where the pass rate for one group is less than 80 percent of the pass rate for the most successful group, a presumption of adverse impact generally arises. To illustrate, consider this hypothetical:
| PAT Results: | Job Applicants | Number Who Passed PAT | Test Selection Rate | Actual Hires |
| Male | 100 | 85 | .85 | 24 |
| Female | 20 | 8 | .40 | 8 |
| Comparison of female pass rate to male pass rate: .40 ÷ .85 = .47 | ||||
A disparate impact exists in this example under the four-fifths rule, because the female pass rate on the PAT (40 percent) is only 47 percent (not 80 percent or more) of the male pass rate on the PAT (85 percent). Therefore, even though all the females who passed the PAT were hired and 25 percent of all hires were female, the test may be deemed discriminatory by an enforcement agency—unless the test was job related and consistent with business necessity.
Proving the Validity of PATs
Proving job relatedness and business necessity generally comes in the form of validation evidence. The federal Uniform Guidelines on Employee Selection Procedures provides the framework for validating employment tests, including PATs.
Discussing PAT validation requires first mentioning that PATs commonly fall into one (or both) of two categories: (1) tests that simulate actual job tasks (e.g., lifting objects, pushing equipment, climbing stairs or ladders, dragging hoses, etc.) and (2) tests that more generally measure aspects of physical fitness (e.g., tests of muscle strength, balance, aerobic capacity, etc.). The second category, physical-fitness-type tests, have been a recent target of enforcement agencies because they are less intuitively job related and can be difficult to validate properly in accordance with the Uniform Guidelines.
PATs are commonly validated using content validity and/or criterion validity, both of which the Uniform Guidelines endorses. Content validation is a systematic way of accumulating evidence to demonstrate that the content of a test is representative of and relevant to the content of a job. A criterion validation is a study that gathers evidence to show that a test predicts or is related to job performance.
Either approach first requires a comprehensive, well-documented job analysis that focuses on the physical aspects of a job. A job analysis typically includes professional ergonomic observations of physical job tasks, such as lifting, carrying, bending, squatting, reaching, and climbing, as well as measurements of the frequency, weights, duration, and time-sensitivity requirements for all such tasks. In addition, a job analysis ordinarily includes sampling incumbent employees and their supervisors (these are the subject-matter experts) to collect their ratings of the importance, frequency, and time sensitivity of each physical task identified as an essential component of a job. If an employer will be using a PAT at multiple locations, the job analysis may also need to evaluate whether there are any variations in the physical requirements for the same job at different locations.
With a thorough job analysis in hand, content validation can demonstrate that a PAT is job related because it simulates essential physical job requirements. For example, a test that requires an applicant for a warehouse position to demonstrate that he or she can climb a ladder, squat to knee level, lift a 30-pound object and carry it 20 feet 10 times in 5 minutes, may be content validated by the accumulation of sufficient evidence to demonstrate that the job actually requires doing those tasks.
Criterion validation can demonstrate that the results of a PAT predict or are significantly correlated with employees’ actual proficiency on the job. Criterion validity typically requires that a test be administered to a group of current employees (“concurrent criterion validity”) or new hires (“predictive criterion validity”) and that performance data be collected and compared to test results for those same individuals. Criterion validation also requires a sufficient sample size of tested employees. It can be a costly and time-consuming process. Depending on the complexity of the job, it can also prove challenging to identify measurable job-performance data, particularly for entry-level manual jobs, for which performance measures may be limited to attendance or volume output.
Many PAT vendors market a quantifiable reduction in on-the-job injuries and workers’ compensation claims as the job performance criterion that correlates with their tests. In other words, vendors may represent that using a PAT on a prehire basis will result in better job performance in the form of a measurable reduction in workplace injuries.
Regardless of the validation methodology, analyzing required job duties is instrumental to determining whether a test reflects the essential physical requirements of a job. Employers may want to review their PATs to ensure they do not test for more difficult tasks than those normally encountered on the job. In addition, employers may consider revalidating PATs periodically, especially if employees’ job duties have changed (e.g., if automation is added). By taking steps to validate PATs in accordance with the Uniform Guidelines, employers may be able to screen for applicants who are physically qualified to perform jobs, while mitigating the risk of legal challenges based on disparate impact.
