Reducing Turnover Rates: The Impact of Pre-Employment Screening on Job Performance

Prediction of Early Voluntary Turnover and Job Performance: The Contribution of a Pre-Employment Screening Inventory

This study used a predictive validation design to examine turnover and job performance in a group of 78 security officers employed by a private Southeastern contract security company with approximately 2,600 employees. The Employee Reliability Inventory (ERI) was administered in four districts, but was not used for making selection decisions. Subsequent measures of job tenure were compared to scores on the Long Term Job Commitment scale.

• A relationship was found between an individual’s scores on the Long Term Job Commitment scale and the total number of days he or she worked after being hired. Better scores on the Long Term Job Commitment scale were associated with more time on the job.
• There was a significant difference in the number of days worked by individuals with “good” scores on the Long Term Job Commitment scale (scores in Zones 1 through 3) as compared to those with “poor” scores (those in Zone 4).
• Without using the ERI to make hiring decisions, the rate of terminations within 60 days was 23% of the total hires. If the ERI had been used as part of the selection process, the rate of terminations would have been only 13% of the total hires.

Using these figures, one can estimate the reduction in terminations that would result from using the Employee Reliability Inventory as part of the selection process. In this Company there are 2,000 new hires per year. If they are hired without using the inventory, the number of terminations within 60 days would be approximately 460 per year. If these individuals are hired using the inventory as part of the selection process, the number of terminations within 60 days would be approximately 260 per year. That is, using the ERI would result in an estimated reduction in turnover of 10% or 200 fewer terminations per year.

• The overall accuracy of Long Term Job Commitment scale scores in predicting length of job tenure was 75%.

Gerald L. Borofsky, Ph.D. (Harvard Medical School), Robert Watson (Charlotte, NC)