Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.

Department (Legacy)

Department of Leadership, Foundations, and Human Resource Education

Committee Chair

Choi, Namok

Author's Keywords

Homework; Mediation; SEM; Self-regulated learning; Science achievement


Homework; Personality and academic achievement; Academic achievement--Psychological aspects; Self-control


Homework has been shown to have a significant, positive effect on student achievement and grades, particularly at the secondary level (Cooper, Robinson, & Patall, 2006). However, homework completion and its effect on grades is controlled within the realm of the student and its success as a learning strategy depends on many things including the students’ interest in the subject, their confidence, the time they have to complete it, gender and other factors within the home such as parent education level and parent involvement (Trautwein & Ludtke, 2007; J. Xu, 2007; J. Xu, Corno, Lyn, 2006). Through the work of Eccles (2002) and Trautwein (2006), homework research has focused heavily on students’ affective beliefs and their relevance. Expectancy-Value theory (Eccles, 2002) and self-efficacy theory (Bandura, 1997) identify student self-confidence and valuation of the subject as integral components of academic motivation and Trautwein (2006) places them as antecedents to homework factors in his model of homework. However, causality among the variables has been unstudied and thus a method of analysis needed to be used that could confirm the placement of these variables in a causal sequence while at the same time demonstrate their importance to homework's relevance. It was the intent of this research to explore the causal relationship among the homework model factors of student positive affect towards science, student valuation of science, self-confidence in science, gender, parent education level, and self-confidence in science on science achievement in U.S. eighth-graders. With specific foci on those above variables involved indirectly in the relationship between gender, parent education level and time spent on homework, as well as the indirect effect of time spent on homework between the affective variables and achievement. Results revealed a significant indirect effect of gender on time spent on homework, reflecting complete mediation, through the affective variables of student positive affect towards science, student valuation of science, self-confidence in science. On the other hand, the affective variables act as partial mediators between parent education level and time spent on homework as both the indirect and direct effects were significant. Indications here are that there may be other unknown variables at work that are not studied in the current research. In addition, time spent on homework was a significant mediating variable between all of the affective variables and achievement. Conclusions and recommendations are made that homework does play a significant role in student achievement in science when the affective characteristics associated with self-regulatory learning are considered. Further research is recommended particularly involving better characterization of the homework construct.