Date of Award

1-1-2013

Document Type

Dissertation

Degree Name

Ph.D.

Department

Quantitative Research Methods

First Advisor

Antonio Olmos-Gallo

Keywords

data use, school climate, student perceptions, survey

Abstract

Over the past 25 years, researchers have consistently reported that students' perceptions of their school's climate can have a measurable impact on their level of engagement in school, motivation to learn, social development, and, ultimately, their academic achievement. In light of the continued emphasis on education reform and school accountability, the ability to accurately measure the learning environment, interpret the results from that measure, and then determine the appropriate course of action in addressing areas of concern has become increasingly important for school leaders. This dissertation used an embedded mixed methods design to examine one district's self-developed measure of their students' perceptions of school climate in order to determine if there was sufficient validity and reliability evidence for results to be used by school leadership (i.e., principals) to make data-driven decisions regarding implementing initiatives and interventions for improving or enhancing the school climate. Additionally, the measure was examined for variations in results for specific groups of students based on their grade level, gender, and ethnicity. Both exploratory and confirmatory factor analyses were conducted, along with reliability analysis and invariance testing. Because the purpose of the survey was to obtain data on school climate that could then be examined and used to make decisions, interviews with school leadership team members were conducted to provide insight into how they had used previous data reports and how a new reporting structure (i.e., by factor) would impact their use of the data.

Overall, results from the quantitative analyses found the EFA results indicated over half of the survey items were not functioning the way in which they were intended. With these items removed, a three factor CFA model was conducted to determine if there was appropriate model fit as well as invariance across gender, grade level, and ethnicity. The CFA model fit statistics were acceptable and the invariance tests held across each group however, further revisions of the instrument are recommended in order to develop a measure that will address the needs expressed during the school leadership interviews and will also accurately reflect students' experiences in their school. Interviews revealed limited use of the data reports from the survey due to lack of time, length of the reports, and absence of district guidance on how to use student results to make program decisions. Recommendations for survey and reporting structure revisions are included.

Provenance

Recieved from ProQuest

Rights holder

Mya L. Martin-Glenn

File size

156 p.

File format

application/pdf

Language

en

Discipline

Educational tests & measurements, Educational psychology, Sociology of education

Share

COinS