Date of Award

Fall 9-18-2014

Degree Type

Dissertation

Degree Name

EdD Education Leadership, Management and Policy

Department

Education Leadership, Management and Policy

Advisor

Anthony Colella

Committee Member

Barbara Strobert

Committee Member

Christopher Tienken

Committee Member

William Horn

Keywords

Student Achievement, Mathematics and English Language Arts State Standardized Tests and Assessments, Elementary and Middle School, Community Demographics, School Climate and Culture, Simultaneous and hierarchical regression

Abstract

Student achievement has been measured in the United States for decades through the use of standardized state assessments. The purpose of this study was to examine which combination of 15 out-of-school community demographic variables best predicted and accounted for the most variance in a Connecticut school district’s percentages of students scoring goal or above on the 2010 Connecticut Mastery Test (CMT) for the third through eighth grade in Mathematics (Math) and English Language Arts (ELA). Analyses were conducted using both a simultaneous regression model and a hierarchical regression model. This study looked at the entire population of districts that were not regional districts, charter schools, private schools or high schools and had at least 25 students in third through eighth grade who took the 2010 ELA and Math CMT. There were two research questions that guided the researcher in the study. The two research questions involved finding the right combination of the 15 out-of-school variables that best predicted how students actually performed on the 2010 CMT as well as accounted for the greatest amount of variance on the 2010 CMT in Mathematics and English Language Arts. The results of this study revealed that out-of-school community demographic factors greatly affect how students perform on state standardized assessments. This study predicted between 68% and 76% and accounted for between 67% and 79% of the variance in the 2010 CMT’s for third through eighth grade in ELA and Math. This study showed that each grade level produced a combination of out-of-school demographic data that was specific to each grade level and subject. Some of these variables (percent no HS diploma, percent married families, percent making less than $35,000) were common across grade level and subject area. The findings from this study corroborate and strongly support the findings from previous empirical studies on the impact of out-of-school factors on student achievement. This research study contributes to the limited but growing body of knowledge indicating inadequacy of the use of state standardized assessments as the sole measure of student achievement.

Share

COinS