How to Reduce Cheating on Pre-Employment Testing

abstract woman on computer
Topic(s): assessment, Faking, selection, workplace deviance
Publication: Journal of Business and Psychology
Article: How secure are unproctored pre-employment tests? Analysis of inconsistent test scores
Authors: T. Kantrowitz, A.M. Dainis
Reviewed by: Alexandra Rechlin

Most companies that use pre-employment testing include at least some of it as unproctored internet testing (UIT). UIT may include biodata responses (e.g., where an applicant attended college), personality tests, and cognitive ability tests. It is harder to cheat on biodata or personality tests, as the “correct answers” are not always apparent. However, outright cheating on cognitive ability tests could be a problem.

PRE-EMPLOYMENT COGNITIVE TESTS

There are numerous empirically-based suggestions for reducing the likelihood of cheating on cognitive ability UITs. One of these is using computer adaptive testing, which utilizes a large test bank and selects subsequent questions based on the test-taker’s previous responses; in essence, the test adapts to the knowledge or skill level of the test-taker. Another common method is to use confirmation testing, during which test-takers in the first stage take a UIT, and in the second stage prove their identity and take a shorter proctored test.

THE CURRENT STUDY

In the current study, the authors (Kantrowitz & Dainis, 2014) analyzed data from applicants’ cognitive ability tests as part of a two-stage confirmation testing process that utilized computer adaptive testing. The data included test results from over 4000 applicants in numerous types of jobs, companies, and industries. The researchers were interested in inconsistent test scores, which could indicate cheating, and found that there wasn’t much of a difference between test scores in the unproctored versus proctored conditions. In other words, there wasn’t much cheating going on.

IMPLICATIONS FOR ORGANIZATIONS

This research indicates that cheating on cognitive ability tests is not much of an issue if the testing process is conducted well. If you plan to use a cognitive ability UIT, you should utilize computer adaptive testing, protect your test questions, and enforce a time limit on the test. You should also warn candidates about the consequences of cheating, use single-use links to the UIT, and use confirmation testing (include a proctored exam after the initial UIT). This will help deter cheaters, and ensure that your pre-employment testing system is truly identifying the right employees for your organization.

Kantrowitz, T., & Dainis, A. M. (2014). How Secure are Unproctored Pre- Employment Tests? Analysis of Inconsistent Test Scores. Journal of Business and Psychology, 29, 605-616.