Date of Award

May 2014

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Department

Educational Psychology

First Advisor

Timothy J. Cleary

Committee Members

Cindy M. Walker, John R. Surber, Jay H. Beder, Bonnie P. Klein-Tassman

Keywords

Mathematical Problem Solving, Metacognition, Microanalysis, Motivation, Self-regulated Learning (SRL), Self-regulation

Abstract

The current dissertation examined the validity of a context-specific assessment tool, called Self-regulated learning (SRL) microanalysis, for measuring self-regulated learning (SRL) during mathematical problem solving. SRL microanalysis is a structured interview that entails assessing respondents' regulatory processes as they engage with a task of interest.

Participants for this dissertation consisted of 83 eighth grade students attending a large urban school district in Midwestern USA. Students were administered the SRL microanalytic interview while completing a set of mathematical word problems to provide a measure of their real-time thoughts and regulatory behaviors. The SRL microanalytic interview targeted the SRL processes of goal-setting, strategic planning, strategy use, metacognitive monitoring, attributions, and adaptive inferences. In addition, students completed two questionnaires measuring SRL strategy use, and one questionnaire measuring self-esteem. The participant's mathematics teacher completed a teacher rating scale of SRL for each participant. Mathematical skill was measured with three measures including a three item measure of mathematical problem solving skill completed during the SRL microanalytic interview, a fifteen item posttest of mathematical problem solving skill completed two weeks after the SRL microanalytic interview, and a standardized test of mathematics skill.

The primary objectives of this dissertation were to compare the newly developed SRL microanalytic interview to more traditional measures of SRL including two self-report questionnaires measuring adaptive and maladaptive SRL and a teacher rating scale of SRL. In addition, the current dissertation examined whether SRL microanalysis would diverge from a theoretically unrelated construct such as self-esteem. Finally, the primary interest of the current dissertation was to examine the relative predictive validity of SRL microanalysis and SRL questionnaires. The predictive validity was compared across three related but distinct mathematics outcomes including a short set of mathematical problem solving items, a more comprehensive posttest of MPS problem solving skill, and performance on a standardized mathematics test.

The results of this study revealed that SRL microanalysis did not relate to self-report questionnaires measuring adaptive or maladaptive SRL or teacher ratings of SRL. The SRL microanalytic interview diverged from the theoretically unrelated measure of self-esteem. Finally, after controlling for prior achievement and SRL questionnaires, the SRL microanalytic interview explained a significant amount of unique variation for all three mathematics outcomes. Furthermore, the SRL microanalytic protocol emerged as a superior predictor of all three mathematics outcomes compared to SRL questionnaires.

Share

COinS