Advisor(s)

Emanuel J. Mason

Contributor(s)

Jessica A. Hoffman, Theresa L. Osypuk, Karin N. Lifter

Date of Award

2011

Date Accepted

3-2011

Degree Grantor

Northeastern University

Degree Level

Ph.D.

Degree Name

Doctor of Philosophy

Department or Academic Unit

Bouve College of Health Sciences, Department of Counseling and Applied Educational Psychology

Keywords

counseling psychology, ECLS-K, elementary, gender differences, mathematics, science

Disciplines

Counseling Psychology

Abstract

The purpose of this study was to investigate three potential explanations for children's early developmental cognitive and educational experiences that could contribute to differences in the representation of women compared to men in science, technology, engineering, and mathematics (STEM) careers. This study utilized data from the Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 (ECLS-K), a project conducted by the United States Department of Education's National Center for Education Statistics (NCES). Longitudinal data from a nationally representative cohort of children from the 1998-1999 kindergarten class in the American schools were collected beginning in kindergarten and continuing through eighth grade. The study systematically explored the contribution of Bandura's social learning (social cognitive) theory and Vygotsky's sociocultural theory, to explain the gender disparity in performance on mathematics, science, and general knowledge test scores. The study hypothesized that boys would have stronger associations of child-guided learning with achievement and girls would have stronger associations of teacher-guided learning with achievement, based on the theories of Bandura (for child-guided learning) and for Vygotksy (for teacher-guided learning). In addition, the study examined the role of engagement on children's explorations and activities. Predictor variables were refined using exploratory and confirmatory factor analyses and a moderator variable, school engagement, was constructed from of a sum of items from the ECLS-K longitudinal database kindergarten teacher questionnaire, to predict performance in mathematics, science, and general knowledge from kindergarten to eighth grade. Analyses revealed a second moderator variable, student's prior learning. Path analysis revealed that the proposed models were not supported. However, results suggested that prior learning could have an effect on later school performance. In addition, girls showed a slightly higher level of engagement compared to boys. Lastly, the results demonstrated that boys had greater variability in their scores than girls, and boys had higher means across each testing period. The poor fit of the models was hypothesized to be affected by the quality and variety of variables in the database and how each construct was defined. Future studies seem warranted addressing the basis for differences in achievement and participation in STEM fields between boys and girls.

Document Type

Dissertation

Rights Information

copyright 2011

Rights Holder

Jannon C. Farkis,


View Full Text

Click button above to open, or right-click to save.

Share

COinS