Investigation gender/ethnicity heterogeneity in course management system use in higher education by utilizing the MIMIC model
Yi Li, Purdue University, United States
Purdue University . Awarded
This study focuses on the issue of learning equity in colleges and universities where teaching and learning have come to depend heavily on computer technologies. The study uses the Multiple Indicators Multiple Causes (MIMIC) latent variable model to quantitatively investigate whether there is a gender /ethnicity difference in using computer based course management system (CMS) for course tasks. The study further investigates whether there is an association between course management system use and academic achievement. The MIMIC model has many advantages for the current data analysis, including an allowance for simultaneous examination of multiple group differences in a latent factor framework.
In this study, data from over 993 students who enrolled in the three courses, Anthropology 201, Biology 206, and Chemistry 116, at a Midwest public university were collected in the spring of 2010. These three courses were selected because the instructors and students actively utilized a course management system for their teaching and learning activities. Data preparation involved the removal of extraneous records and transformation of the original data in order to satisfy the assumption of MIMIC model. The model building procedure began by checking the validity of the measurement model without covariates, followed by incorporating covariate variables – gender (males versus females) and ethnicity (African Americans, Asian Americans, Caucasians, Hispanic Americans) – into the fitted measurement model. A final model with the best model fit was selected for testing group invariance.
Results from the final model revealed that: (a) group invariance was present for gender and ethnic groups in use of the course management system. Male students had a significantly higher mean on their frequency of using the course management system than female students. Asian students had the significantly highest mean among all the ethnic groups. Additionally, (b) the amount of involvement in the course management system had positive association with student's academic outcome. When student's scholastic aptitude was controlled, Asian students who used the course management system the most frequently tended to have the highest academic outcome. These findings may imply that "digital divide" experienced in childhood might continue to impact course management system use in college to some extent; also, the quantity use of course management system was critical to student academic outcome, but quantity alone could not guarantee a dramatic improvement of academic achievement. The quality of course management system use needs to be defined. The study has significant implications for policy and practice. Specifically, the study discusses in detail practical methods for eliminating academic discrepancies, which may be due to different uses of course management system. Finally, the study points to new directions for research into course management system use in higher education.
Li, Y. Investigation gender/ethnicity heterogeneity in course management system use in higher education by utilizing the MIMIC model. Ph.D. thesis, Purdue University.
Citation reproduced with permission of ProQuest LLC.
For copies of dissertations and theses: (800) 521-0600/(734) 761-4700 or https://dissexpress.umi.com