From missing data to informative GPA predictions: Navigating selection process beliefs with the partial identifiability approach
British Journal of Mathematical and Statistical Psychology · 2025
DOI: 10.1111/bmsp.12377 ↗Kurz gefasst
Fachartikel in der Zeitschrift British Journal of Mathematical and Statistical Psychology, 2025. Verfasst von Eduardo Alarcón-Bustamante und 3 weiteren Autor:innen. Beteiligte Einrichtung: Escuela de Psicología, Pontificia Universidad Católica de Chile, Santiago de Chile, Chile. Hauptthemen laut MEDLINE: School Admission Criteria, College Admission Test, Educational Measurement. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 57 Literaturangaben. Gefördert von: Agencia Nacional de Investigación y Desarrollo, Chile, Grant Fondecyt, Grant/Award Number, Agencia Nacional de Investigación y Desarrollo, Chile, Grant Millennium Nucleus on Intergenerational Mobility MOVI, Grant/Award Number:, Fondo de Fomento al Desarrollo Científico y Tecnológico, FONDEF IDeA I+D.
Abstract (Originalsprache)
The extent to which college admissions test scores can forecast college grade point average (GPA) is often evaluated in predictive validity studies using regression analyses. A problem in college admissions processes is that we observe test scores for all the applicants; however, we cannot observe the GPA of applicants who were not selected. The standard solution to tackle this problem has relied upon strong assumptions to identify the exact value of the regression function in the presence of missing data. In this paper, we present an alternative approach based on the theory of partial identifiability that considers a variety of milder assumptions to learn about the regression function. Using a university admissions dataset we illustrate how results can vary as a function of the assumptions that one is willing to make about the selection process.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Escuela de Psicología, Pontificia Universidad Católica de Chile, Santiago de Chile, Chile.
- Departamento de Evaluación Medición y Registro Educacional (DEMRE), Universidad de Chile, Santiago de Chile, Chile.
- Millennium Nucleus on Intergenerational Mobility: From Modelling to Policy (MOVI), Santiago de Chile, Chile.
- Interdisciplinary Laboratory of Social Statistics, Santiago de Chile, Chile.
- Faculty of Mathematics, Pontificia Universidad Católica de Chile, Santiago de Chile, Chile.
- The Economics School of Louvain, Université Catholique de Louvain, Brussels, Belgium.
ORCID-Profile: Eduardo Alarcón-Bustamante, Jorge González, Ernesto San Martín
Förderung
- Agencia Nacional de Investigación y Desarrollo, Chile, Grant Fondecyt, Grant/Award Number
- Agencia Nacional de Investigación y Desarrollo, Chile, Grant Millennium Nucleus on Intergenerational Mobility MOVI, Grant/Award Number:
- Fondo de Fomento al Desarrollo Científico y Tecnológico, FONDEF IDeA I+D
- Agencia Nacional de Investigación y Desarrollo
- Fondo de Fomento al Desarrollo Científico y Tecnológico
Zitieren
APA
Alarcón-Bustamante, E., González, J., Torres Irribarra, D., & San Martín, E. (2025). From missing data to informative GPA predictions: Navigating selection process beliefs with the partial identifiability approach. British Journal of Mathematical and Statistical Psychology. https://doi.org/10.1111/bmsp.12377
Vancouver
Alarcón-Bustamante E, González J, Torres Irribarra D, San Martín E. From missing data to informative GPA predictions: Navigating selection process beliefs with the partial identifiability approach. British Journal of Mathematical and Statistical Psychology. 2025. doi:10.1111/bmsp.12377
BibTeX
@article{AlarcnBustamante2025,
author = {Alarcón-Bustamante, Eduardo and González, Jorge and Torres Irribarra, David and San Martín, Ernesto},
title = {From missing data to informative GPA predictions: Navigating selection process beliefs with the partial identifiability approach},
journal = {British Journal of Mathematical and Statistical Psychology},
year = {2025},
doi = {10.1111/bmsp.12377}
}
EUVIMED