Exploring the Impact of Student Engagement on Dropout Warning Signs

Problem Statement: Students who drop out of school face a wide range of future issues, including higher rates of unemployment and lower annual salaries than their peers who graduated high school. Lower incomes impact where people live, typically resulting in houses in areas of higher crime and lower achieving schools. Low socioeconomic status and parents with low educational attainment are common among students that drop out, indicating a potential continuing cycle that may be challenging to break. Understanding what leads to dropping out and what warning signs exist is crucial for schools, teachers, parents, and anyone involved in the lives of students. Purpose: This study explored predictive factors of dropout warning signs in 4th, 6th, 8th, 9th, and 11th grades including 15,838 students to answer the following research questions: 1) Does student engagement predict the number of dropout warning signs a student exhibits? 2) Do any subconstructs of student engagement better predict exhibited dropout warning signs? 3) Is there a difference in the relationship between student engagement and dropout warning signs between grade levels? 4) Does the number of dropout warning signs differ significantly between schools that are rated “needs improvement,” “proficient,” and “distinguished”? Methodology: Participants included 15,838 students in 4th, 6th, 8th, 9th, and 11th grade from 105 schools in Kentucky participating in a grant program funded by the U.S. Department of Education. Data sources included schoolwide public data and the student responses on a survey implemented as a part of a program evaluation conducted by researchers at Western Kentucky University during the 2015-2016 school year. Negative binomial regression was used to determine how well student engagement and its subconstructs can predict the number of dropout warning signs exhibited for questions one and two. Question three was answered by calculating the correlation between student engagement and dropout warning signs. For question four, two-way ANOVA was run to compare dropout warning signs for different school ratings at each school type and to measure the interaction of the two factors. Findings: Student engagement and each subconstruct significantly predicted the total number of dropout warning signs across grades, p < .01. For each increase in engagement level, there was a 57.3% decrease in dropout warning signs. The relationship between engagement and dropout warning signs is statistically significant in all five grade levels studied, and strongest in the 6th grade, r = -.489. A two-way ANOVA found a statistically significant difference between dropout warning signs across the three school ratings and grades, p < .01, and the interaction between the two factors was significant, p = .039. Conclusion: Student engagement may be a potential factor in reducing the likelihood of dropping out of school. Schools, teachers, and stakeholders should seek to improve student engagement across all grades in order to help increase graduation rates. While focus on low performing schools and high schools, where dropout signs are the highest, should be of concern, there is evidence that student engagement is important in all environments from elementary through high school.

Name thesis:degree_name

Doctor of Philosophy

Level thesis:degree_level

Doctoral

Discipline thesis:degree_discipline

Curriculum and Instruction

Grantor

University of Houston

Year dc:date.issued

2018

Author dc:creator

• Kallman, Marguerite Patricia

Advisor dc:contributor.advisor

• Freiberg, H. Jerome

Committee members dc:contributor.committeemember

• Lee, Mimi Miyoung
• Zhang, Jie
• Horn, Catherine L.

• Student engagement
• Dropouts

Statement dc:rights

• The author of this work is the copyright owner. UH Libraries and the Texas Digital Library have their permission to store and provide access to this work. Further transmission, reproduction, or presentation of this work is prohibited except with permission of the author(s).

Language dc:language.iso

eng

Handle dc:identifier.uri

https://hdl.handle.net/10657/4448

OAI identifier oai:identifier

oai:uh-ir.tdl.org:10657/4448