search for




 

Longitudinal study of academic achievement change using latent growth models
Journal of the Korean Data & Information Science Society 2018;29:937-50
Published online July 31, 2018
© 2018 Korean Data and Information Science Society.

Hyun Seok Choi1 · Cheolyong Park2

1Center for Educational Performance, Keimyung University
2Major in Statistics, Keimyung University
Correspondence to: Professor, Major in Statistics, Keimyung University, Daegu 42601, Korea. E-mail: cypark1@kmu.ac.kr
Received May 29, 2018; Revised July 1, 2018; Accepted July 2, 2018.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
In this study we use latent growth models to analyze GPA (grade point average) of students who completed 6 consecutive semesters from 2014 at K University. The results show that the intercept, slope and quadratic term are all statistically significant, so that there is curve effect in addition to linear effect on GPA as the number of completed semesters accumulates. Also as the number of completed semesters accumulates, the GPA slope difference between Regular and Early admission types becomes smaller. Specifically, the starting GPA is ordered as Regular(na) > Early general > Regular(da) > Early potential talents > Early interview, but as the number of completed semesters accumulates, the GPA difference among them becomes smaller except for Early interview admission type. The GPA of Early interview admission type that does not require minimum scores for admission is much lower than those of other admission types. Thus special education programs might be necessary to stimulate the motivation of learning for the students from Early interview admission type.
Keywords : Academic achievement change, admissions types, grade point average, latent growth model.