“Workplace Theft: An Analysis of Student-Employee Offenders and Job Attributes” is an article authored by Elizabeth Ehrhardt Mustaine (University of Central Florida) and Richard Tewksbury (University of Louisville) and published in the American Journal of Criminal Justice 27:1 (pages 111 – 127, 2002). This employee theft study surveyed a large population of college students attending a number of major universities. Since existing research suggests that many dishonest employees are younger, part-time, untenured, and dissatisfied, these two researchers concluded that college students would make an ideal sample of employees to survey about their occupational criminal behavior.
They conducted a self-report survey of 1,531 students in the fall of 1996 asking them to report personal demographics, opportunity, and previous theft activities. The findings are consistent with a number of other studies (including mine), but with rather unique results. The authors found that three factors differentiate between those who admitted stealing from an employer from those who did not. Some of these predictors included theft behaviors that occurred in other settings.
For example, most impressive was the fact that students who have admitted that they recently have broken into a motor vehicle were almost 14 (13.87) times more likely to steal from their employers. Moreover, students who have recently stolen something from a stranger were over four times (4.35) more likely to commit employee theft. Also of significant interest was the fact that ex-convicts were nearly four times (3.59) more likely to admit stealing from their place of work than those respondents who have never been sent to prison.
There were a few other findings of interest. Alcohol use was related to admitting stealing at work. Public intoxication, but not drug use, predicted admitted employee theft. College students who reported that they have been drunk in public were 1.56 times more likely to admit stealing while at work. Finally, the more jobs that a student has had in the past and the more often these jobs involved cash handling was also related to workplace theft, but at a lower level of predictive power.
We must remember that this study was conducted with college students and used self-reported indicators of workplace theft. Nevertheless, even with this caveat about the sample, the policy implications are significant.
First, drug testing may be a good indicator of current and future drug use, but may not be the best indicator of theft behavior.
Second, criminal background checks that screen out applicants with prior convictions that resulted in incarceration are obviously supported by this screening practice.
Third, as we know in social science, the best indicators of future behavior is past behavior, especially when we consider that stealing in non-employment situations is a good predictor of workplace theft.
The principal paradox of this study is the finding that with the exception of the above factors, most of which retailers screen for already, the average college student who does not steal is not dramatically different from the one who does. Since we rely on these young people for a substantial proportion of the retail industry workforce, there is clearly no silver bullet that can distinguish those who will steal at work from those who will not.
This article was excerpted from “Research Findings on Employee Theft.”