Reducing Race and Sex Subgroup Differences in Selection
Traditional Selection Methods and Trade-offsTraditionally, tests of cognitive ability have been the most widely used and valid selection measures. However, these measures often lead to significantly lower scores for females and non-white minorities than their male or white counterparts. By relying on measures of cognitive ability alone to select employees, organizations will likely hire a disproportionate number of white males relative to minority and female applicants. There are many alternative methods to cognitive ability that can be used for employee selection. However, these alternative methods often involve trade-offs in the form of decreased effectiveness or other negative consequences. In order to minimize these problems, organizations should:
- Ensure that alternative selection measures are valid predictors of job performance.
- C hoose selection methods that reduce the likelihood of applicant faking.
- Analyze time and cost investments of alternative selection methods.
- Consider the practicality of implementation and fit with your organization.
Best PracticesThe most effective strategies for reducing race and sex subgroup differences include:
- Using alternative methods such as interviews and assessment centers as selection measures.
- Assessing the entire range of knowledge, skills, and other abilities needed to perform the job effectively.
- Test banding – Grouping applicant scores rather than looking at them on a continuum. Applicants in the same “band” or group are considered to have the same score.
- Minimizing the verbal ability requirements of the predictor measure so that they meet, but do not exceed, the needs of the job.
The DeGarmo GroupThis was a summary of the research and practice implications from: Ployhart, R.E., Holtz, B.C. (2008). The diversity-validity dilemma: Strategies for reducing racioethnic and sex subgroup differences and adverse impact in selection, Personnel Psychology, 61, 153-172.