Premium
Investigating College Learning Gain: Exploring a Propensity Score Weighting Approach
Author(s) -
Liu Ou Lydia,
Liu Huili,
Roohr Katrina Crotts,
McCaffrey Daniel F.
Publication year - 2016
Publication title -
journal of educational measurement
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.917
H-Index - 47
eISSN - 1745-3984
pISSN - 0022-0655
DOI - 10.1111/jedm.12112
Subject(s) - propensity score matching , weighting , psychology , sample (material) , ethnic group , regression analysis , academic achievement , higher education , regression , linear regression , mathematics education , statistics , mathematics , medicine , sociology , political science , chemistry , chromatography , anthropology , psychoanalysis , law , radiology
Learning outcomes assessment has been widely used by higher education institutions both nationally and internationally. One of its popular uses is to document learning gains of students. Prior studies have recognized the potential imbalance between freshmen and seniors in terms of their background characteristics and their prior academic performance and have used linear regression adjustments for these differences, which some researchers have argued are not fully adequate. We explored an alternative adjustment via propensity score weighting to balance the samples on background variables including SAT score, gender, and ethnicity. Results involving a cross‐sectional sample of freshmen and seniors from seven groups of majors within a large research university showed that students in most of the majors demonstrated significant learning gain. Additionally, there was a slight difference in learning gain rankings across major groupings when compared to multiple regression results.