Adverse Impact Analysis of Objective Management Group’s Sales Candidate Assessment®
Objective Management Group’s Sales Candidate Assessment® is primarily aimed at distinguishing those individuals who have substantial potential for growth in sales from those with limited potential. The profile is based on a web based multiple-choice form completed by an applicant or employee. From the pattern of responses, a clear determination of potential to overcome selling limitations can be assessed.
Correlation Research, Inc. was engaged to perform a study of possible adverse impact on any “protected group” that might result from use of Objective Management Group’s Sales Candidate Assessment®. According to the Equal Employment Opportunities Commission (EEOC), the three primary protected groups are:
- Women (read this study on the difference between men and women)
- Age at least 40
According to the guidelines promulgated by the EEOC, written tests used to select employees must not discriminate against any of these protected classes of employees or applicants.
There are two main criteria that are routinely applied to decide whether more sophisticated analyses are warranted. Both of these standards are conservative and indicate only the possibility of adverse impact. Determining whether a particular test truly discriminates against a certain group can be extremely complex.
The first criterion is known as the “80% or 4/5ths rule” and is based on the ratio of the “pass rates” for the protected and majority classes. For example, suppose that 100 men and 100 women completed the assessment. Then the percentage of women who are rated as having substantial growth potential should be at least 4/5ths of the corresponding percentage for men. If 50 percent of the men have growth potential, then at least 40 percent of the women should achieve this outcome.
The second criterion is whether the difference in “pass rates” for the protected and majority groups reaches statistical significance. That is, could this difference simply be a matter of random variation or does it represent a real difference? A level of statistical significance of .05 or lower means that the difference is unlikely to be the result of random variation.
To apply these criteria to Objective Management Group’s Sales Candidate Assessment®, information on age, sex and minority status for a representative sample of 491 individuals was collected. For each individual, the Profile conclusion regarding whether or not there was substantial growth potential was also noted. The results were tabulated and analyzed by Correlation Research. A summary of the data is displayed in the following tables:
Table 1: Gender Comparison
Table 2: Ethnicity Comparison
Table 3: Age Comparison
|40 and over
For all three of these comparisons, the protected group’s percentage with substantial growth potential comfortably exceeded 4/5ths of the majority group’s percentage. Moreover, none of the three comparisons reached statistical significance at even the conservative .05 level required to raise questions. We conclude that there is no evidence of an adverse impact by the Objective Management Group’s Sales Candidate Assessment® on any protected group.
Herbert I. Weisberg, Ph.D.
July 12, 1996