Is academic tracking related to gains in learning competence? Using propensity score matching and differential item change functioning analysis for better understanding of tracking implications. (April 2020)

• Record Type: Journal Article
• Authors: Martinková, Patrícia; Hladká, Adéla; Potužníková, Eva
• Is Part Of: Learning and instruction. Volume 66(2020)
• Publication Date: 2020-04
• Subjects: Academic tracking -- Learning competence -- Propensity score matching -- Differential item functioning in change -- Instructional sensitivity; Academic school AS -- Basic school BS -- Differential item functioning DIF -- Differential item functioning in change DIF-C -- Learning to learn instrument LtL -- Propensity score matching PSM; Learning -- Periodicals; Teaching -- Periodicals; Apprentissage -- Périodiques; Enseignement -- Périodiques; Learning; Teaching; Periodicals; Electronic journals; 370.1
• DOI: 10.1016/j.learninstruc.2019.101286
• Languages: English
• ISSNs: 0959-4752

Abstract

Abstract: This study analyzes gains in cognitive components of learning competence with respect to cohorts based on ability tracking in a Czech longitudinal study. Propensity score matching is used to form parallelized samples of academic and non-academic track students and to eliminate the effect of selective school intake. We applied regression models on the total scores to test for the overall track effect. Furthermore, we analyze scores and gains on the subscores and check for differential item functioning in Grade 6 and in change to Grade 9. While after 3 years, no significant difference between the two tracks was apparent in the total learning competence score, we did, however, find significant differences in some subscores and in the functioning of some items. We argue that item-level analysis is important for deeper understanding of the tracking implications and may provide the basis for more precise evidence-based decisions regarding the tracking policy. Highlights: No significant track effect is apparent in the overall learning competence score. Significant differences are found in subscores and in the functioning of some items. Academic track has higher gains in Mathematical concepts subscore and items. Item-level analysis is important for deeper understanding of tracking implications. DIF-C analysis is proposed for item-level analysis of change in longitudinal data.