LEGAL CONCERNS IN THE USE OF WEIGHTED APPLICATIONS¹

Abstract

The use of the standard weighted application procedure relies on strictly empirically derived keys for validity. While weighted applications are generally valid in terms of the correlation between score on the application and position on the criterion, the empirical scheme may weight items that can not be shown to be relevant to the job for which they predict. To the extent that employment decisions are based on non‐job‐relevant weighted applications, such decisions may be contrary to the letter or the spirit of the Civil Rights Act. Alternative methods incorporating job relevance into the weighting procedure via job analysis inject the “hand of reason” into the process. Such rational‐empirical approaches ensure job‐relatedness as well as validity.

Differential Item Functioning in Biodata: Opportunity Access as an Explanation of Gender- and Race-Related DIF

“…One example of such inadvertent "discrimination" is an empirically keyed item used in research by Pace and Schoenfeldt (1977). Having a Detroit address as opposed to a suburban address was related to the criterion, but it was also highly correlated with race.…”

“…The bulk of the previous biodata research has examined gender and race differences at the scale level, though consideration of specific item content might be more easily interpretable (Pace & Schoenfeldt, 1977) as a more refined approach …”

Content/Construct Approaches in Life History Form Development for Selection

“…Biodata inventories are appealing in selection situations because research has indicated that they tend to have low adverse impact when items are screened in advance (Mumford and Stokes 1992;Pace and Schoenfeldt 1977;Reilly and Chao 1982;Stokes and Reddy 1992;Van Rijn 1992) and because they tend to provide incremental validity over that obtained with cognitive measures (Hunter and Hunter 1984;Mumford and Stokes 1992;Owens 1976;Stokes, Toth, Searcy, Stroupe and Carter 1999;Wise, McHenry and Campbell 1990). In comparison to many alternatives to cognitive ability, biodata forms are feasible for group administration and scoring at a lower cost.…”

Construct‐oriented Biodata Development for Selection to a Differentiated Performance Domain

“…Irrespective of the unit at which weighting is carried out, empirical keying capitalizes on chance, or sample specific factors. It is this capitalization on potentially unstable or spurious factors that has resulted in empirical keying attracting the term`dustbowl empiricism' (Pace and Schoenfeldt 1977). Whereas such capitalization has enabled this strategy to produce higher validity levels than rational procedures (Fuentes, Sawyer, and Greener 1989;Mitchell and Klimoski 1982), proportionally larger shrinkages on cross-validity can occur compared to those shown by the rational methods (Mitchell and Klimoski 1982).…”