Presenter Information

Kyle MarksFollow

Department

University of Tennessee at Chattanooga. Dept. of Psychology

Publisher

University of Tennessee at Chattanooga

Place of Publication

Chattanooga (Tenn.)

Abstract

The current study aims to compare the accuracy of decision trees to logistic regression in a personnel selection context. To make this comparison, two studies are proposed. The first study uses simulated applicant data. For each applicant, there will be a cognitive ability score, conscientiousness rating, and a structured interview score. Job performance will be simulated as a function of the simulated scores. Additionally, different selection ratios will be applied to the simulated data to mimic how organizations select applicants and to determine whether the selection ratio has an impact on the accuracy of each analytic approach. A second purpose of this study is to examine whether the decision strategies used by decision makers in a real selection context (graduate school admission decisions) reflect the strategies that the decision makers should be using. For each graduate school applicant, the predictors of undergraduate grade point average (GPA) and graduate record examination (GRE) scores will be used. To measure performance, final graduate GPA will be used. The performance data will be used to determine the accuracy of the decision tool in a real selection context. Additionally, there will be information on the decision of whether the applicant was admitted to the program. Whether an applicant was admitted will be matched to the prediction made by the regression analysis and the decision tree analysis to determine if what was done matches what the regression and decision tree models. This is meant to determine which analysis more closely matches how individuals actually make decisions.

Date

October 2017

Subject

Industrial and organizational psychology

Document Type

posters

Language

English

Rights

http://rightsstatements.org/vocab/InC/1.0/

License

http://creativecommons.org/licenses/by-nc-nd/3.0/

Share

COinS
 
Oct 28th, 10:00 AM Oct 28th, 10:55 AM

Comparing the accuracy of decision trees and logistic regression in personnel selection

The current study aims to compare the accuracy of decision trees to logistic regression in a personnel selection context. To make this comparison, two studies are proposed. The first study uses simulated applicant data. For each applicant, there will be a cognitive ability score, conscientiousness rating, and a structured interview score. Job performance will be simulated as a function of the simulated scores. Additionally, different selection ratios will be applied to the simulated data to mimic how organizations select applicants and to determine whether the selection ratio has an impact on the accuracy of each analytic approach. A second purpose of this study is to examine whether the decision strategies used by decision makers in a real selection context (graduate school admission decisions) reflect the strategies that the decision makers should be using. For each graduate school applicant, the predictors of undergraduate grade point average (GPA) and graduate record examination (GRE) scores will be used. To measure performance, final graduate GPA will be used. The performance data will be used to determine the accuracy of the decision tool in a real selection context. Additionally, there will be information on the decision of whether the applicant was admitted to the program. Whether an applicant was admitted will be matched to the prediction made by the regression analysis and the decision tree analysis to determine if what was done matches what the regression and decision tree models. This is meant to determine which analysis more closely matches how individuals actually make decisions.