Date of Award


Degree Type


Degree Name

Master of Science in Education (MSEd)


Education and Human Development

First Advisor

Dr. Arthur Smith


The purpose of this study was to determine if there was a statistically significant and important relationship between fifth grade students' writing products when self-selected-topic compositions were compared to assigned-topic compositions.

The study was conducted over an eight-week period of time with 88 fifth grade students from a suburban school in western New York. There were 48 boys and 40 girls in the study. Each student was requested to write two compositions; the first composition was based on a topic of the student's own choice, the second composition was based on an assigned topic selected from a previous New York State "Survey Test in Writing." Scores comparing self-selected-topic compositions to assigned-topic compositions were analyzed using the point biserial coeffecient of determination.

An analysis of the data revealed no statistically significant and important relationships between the scores of self-selected-topic compositions and assigned-topic compositions. While the statistics were not sufficient to reach the criteria deemed "educationally important", the general trend reflected higher mean scores for self-selected-topic compositions for both boys and girls. Girls acquired higher mean scores than the boys on both assigned and unassigned compositions. Students ranked as "satisfactory" writers by their classroom teachers exhibited the greatest increase in mean scores on unassigned-topic compositions.

Based on this study, further research in the area of process writing and topic selection would benefit the educational system. Skills, acquired while writing compositions on self-selected topics, will transfer to other educational and content areas.