Effects of supplemental services on students' motivation, engagement and academic achievement
Open Access
- Author:
- Yang, Ming
- Graduate Program:
- Educational Psychology
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- November 01, 2022
- Committee Members:
- Matthew T Mc Crudden, Professor in Charge/Director of Graduate Studies
Matthew T Mc Crudden, Thesis Advisor/Co-Advisor
James Clyde Diperna, Committee Member
P Karen Murphy, Committee Member - Keywords:
- supplemental service
propensity score matching - Abstract:
- Students who enter elementary school with low academic achievement are provided supplemental services to help address skill deficits. However, the effect of supplemental services has not been shown to be robust according to previous research using state-wide standardized tests to measure students’ academic achievement. The probable explanations for these results are ineffectiveness of supplemental services, and interferences of potentially confounding variables demographic variables associated with the receipt of supplemental services. As such, the current research employed propensity score matching to statistically “control” for multiple demographic variables by across “control” (no supplemental services) and “treatment” (received serviced) groups. After matching first grade students’ gender, race/ethnicity, primary language, home language and learning disability (reading and mathematics), I used repeated measure ANOVA to compare student’s math and reading scores, motivation and engagement across groups (i.e., students receiving supplemental services and students not receiving services) at the three time points during the school year. The analysis showed that supplemental service did not significantly improve or reduce students’ academic achievement, motivation and engagement compared to the “control” group. Future research regarding the evaluation of supplemental services would benefit from using the propensity score matching method, including more potential confounding variables that were unavailable in the current data set (e.g., SES), and including a larger sample size of students and schools.