Theses and Dissertations

Issuing Body

Mississippi State University

Advisor

Forde, Connie M.

Committee Member

Jayroe, Teresa

Committee Member

Cornelious, Linda

Committee Member

Davis, James E.

Committee Member

Brandenburg, Teri

Date of Degree

1-1-2005

Document Type

Dissertation - Open Access

Degree Name

Doctor of Philosophy

College

College of Education

Department

Department of Instructional Systems, Leadership and Workforce Development

Abstract

The purpose of this study was to determine if Problem-Based Learning (PBL) is as effective an instructional method at the elementary level as traditional instruction in learning content. This study also is a contribution to the literature on PBL in the elementary classroom. The research design was quasi-experimental with a non-equivalent control group. A pilot study was conducted in science classes prior to the commencement of the research project in social studies. Eighty-eight students participated in the two studies. The control groups received instruction in a traditional format, and the experimental groups received instruction through the use of PBL. The research question dealt with whether or not PBL was as effective an instruction method as traditional instruction in student achievement. T-tests were run at the conclusion of each study to compare the means of posttest scores and presentation assessment scores. Analysis of Variance (ANOVA) was used to determine if the differences in means were because of treatment effect or by chance. Analysis of Covariance (ANCOVA) was used to determine if prior knowledge had an impact on the student achievement scores. After the science data were collected and analyzed, the researcher determined that there was a statistically significant difference in the student achievement scores between those involved in the PBL class and those taught traditionally on both the posttest scores and the group presentation scores. Students enrolled in the traditional class scored significantly higher than those enrolled in the PBL class. The researcher noted, however, that both groups made gains in achievement. Assumptions for normality and homogeneity for t-test, ANOVA and ANCOVA were not met for the social studies classes. Transformation of the data took place using arcsine because of a negative skew of the data. After the social studies data were collected and analyzed, the researcher determined that there was no statistically significant difference in the posttest scores for the PBL and traditional classes. The group presentation grades produced conflicting results. Transformed data indicated a significant difference in student achievement while non-transformed data indicated that there was no statistically significant difference in the scores. The researcher noted that both groups made gains in achievement.

URI

https://hdl.handle.net/11668/19158

Comments

elementary education||traditional instruction||problem-based learning

Share

COinS