Date of Award

8-6-2019

Document Type

Dissertation

Degree Name

Doctor of Education (EdD)

Department

Education

First Advisor

Ann Robinson

Abstract

This study investigated academically advanced middle school students’ sense of belonging as it pertains to the STEM disciplines. The study examined the relationship between STEM sense of belonging, STEM class/peer climate, amount of STEM exposure, and the strength of identification with STEM. A quantitative research design was used to examine factors that influence STEM sense of belonging. The researcher utilized a survey research design with data collection via a web survey. This was accomplished through the use of a descriptive rating, Likert-type sense of belonging survey, modified from the Sense of Belonging to Math Scale (PSOBS) created by Rattan, Good, and Dweck (2012). Additionally, demographics, subject area of interest, and perceptions of class climate data were collected (Goodenow, 1993). Participants for the study were selected randomly from the Duke University Talent Identification Program (Duke TIP). By using the Duke TIP 7th Grade Talent Search to obtain the participants for this study the researcher was able to obtain a representation from across the country. A sample-size calculation was utilized to determine the minimum sample size needed for the study. The sample consisted of 337 participants. Of the 337 complete responses, 191 students (57%) identified as female and 145 (43%) identified as male, there was one missing cases in which participants did not provide gender information. An independent t-test, a two-way ANOVA, and a multiple regression analysis were used to analyze the quantitative data. The results of these statistical analysis were used to examine differences in STEM sense of belonging of academically advanced middle school students in STEM. The results from the independent t-test indicated that there is not a significant difference in sense of belonging in STEM between females and males. The results from the ANOVA indicated that the two-way interaction of gender and self-identification with STEM proved to be statistically significant, but not the main effects of self-identification with STEM and gender. Finally, the regression model explained 46% of the variance in STEM sense of belonging. Key predictors of STEM sense of belonging identified by the regression model include class/peer climate and STEM self-identification.

Download

Share

COinS