Utility Analysis

Taylor-Russell Model

If we define utility in terms of the percentage of selected applicants who are successful (known as the success ratio), H. C. Taylor and J. T. Russell showed that it depends on consideration of more than just a validity coefficient. In the Taylor-Russell model, the overall utility of a selection device is a function of three parameters: the validity coefficient (the correlation between a predictor of job performance and a criterion measure of job performance), the selection ratio (the proportion of applicants selected), and the base rate (the proportion of applicants who would be successful without the selection procedure). This model convincingly demonstrates that even selection procedures with relatively low validities can increase substantially the percentage successful among those selected when the selection ratio is low.

Whenever there is a limit on the number of applicants that may be accepted, the selection ratio (SR) is a major concern. As the SR approaches 1.0 (all applicants must be selected), it becomes high or unfavorable from the organization's perspective. Conversely, as the SR approaches zero, it becomes low or favorable; the organization can afford to be selective. As noted earlier, if the SR is low and if an organization needs to choose only the cream of the crop, even predictors with very low validities can be useful. Conversely, given high selection ratios, a predictor must possess very high validity to increase the percentage successful among those selected.

It might appear, therefore, that given a particular validity, organizations should strive always to decrease the SR (become more selective). Unfortunately, the optimal strategy is not this simple, because lowering the SR forces recruiters to expand the recruiting and selection effort. Thus, to select 10 new hires, an SR of 0.5 means that 20 must be recruited. However, if the SR decreases to 0.1, 100 must be recruited. In practice, this strategy may be costly to implement.

...