Communicating Criterion-Related Validity Using Expectancy Charts: A New Approach

Authors

Jeffrey M. Cucina, U.S. Customs and Border ProtectionFollow
Julia L. Berger, Bowling Green State University, U.S. Customs and Border Protection, & ProMedicaFollow
Henry H. Busciglio, U.S. Customs and Border ProtectionFollow

DOI

https://doi.org/10.25035/pad.2017.001

Abstract

Often, personnel selection practitioners present the results of their criterion-related validity studies to their senior leaders and other stakeholders when trying to either implement a new test or validate an existing test. It is sometimes challenging to present complex, statistical results to non-statistical audiences in a way that enables intuitive decision making. Therefore, practitioners often turn to expectancy charts to depict criterion-related validity. There are two main approaches for constructing expectancy charts (i.e., use of Taylor-Russell tables or splitting a raw dataset), both of which have considerable limitations. We propose a new approach for creating expectancy charts based on the bivariate-normal distribution. The new method overcomes the limitations inherent in the other two methods and offers a statistically sound and user-friendly approach for constructing expectancy charts.

Recommended Citation

Cucina, Jeffrey M.; Berger, Julia L.; and Busciglio, Henry H. (2017) "Communicating Criterion-Related Validity Using Expectancy Charts: A New Approach," Personnel Assessment and Decisions: Number 3 : Iss. 1 , Article 1.
DOI: https://doi.org/10.25035/pad.2017.001
Available at: https://scholarworks.bgsu.edu/pad/vol3/iss1/1

Corresponding Author Information

Jeffrey M. Cucina

jcucina@gmail.com

202-863-6298

1400 L Street, NW 7S39 Washington, DC 20229-1145