Date of Award

1-1-2017

Document Type

Masters Thesis

Degree Name

M.A.

Organizational Unit

Morgridge College of Education, Research Methods and Information Science, Research Methods and Statistics

First Advisor

Duan Zhang, Ph.D.

Second Advisor

Kathy Green

Third Advisor

Ron DeLyser

Keywords

Factor invariance, Admission tests, Gender

Abstract

This study examined the factorial structure of the SAAT with exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). Specifically, a CFA of a four-factor model (the hypothesized model) was tested to see if the model had good fit as well as a model generated from an EFA. The EFA showed that the SAAT test only measured two factors (biology and chemistry), not the four posited by the test developers. The CFA provided good fit to a two-factor model; however, a CFA showed that the hypothesized four-factor model also fit the data well and so the hypothesized model was selected as the most appropriate. Based on the CFA four-factor model, measurement invariance (configural, metric and scalar) were examined across school type (public versus private) and gender (male and female) on the test. The results revealed that the metric invariance model fit the data best compared to the other models. Finally, latent means differences were tested by using two-way ANOVA on all the four-factor subjects (biology, chemistry, physics, and mathematics). The results revealed that female students in high schools did better than males on all four sections of the SAAT test. On the other hand, male students in public schools did not achieve well on the test compared to males in private high schools. Also, male students in public school did not achieve well compared to females in both schools.

Publication Statement

Copyright is held by the author. User is responsible for all copyright compliance.

Rights Holder

Mohammed Alqabbaa

Provenance

Received from ProQuest

File Format

application/pdf

Language

en

File Size

56 p.

Discipline

Statistics, Education

Download



Share

COinS