Theses and Dissertations

Issuing Body

Mississippi State University

Advisor

Hare, R. Dwight

Committee Member

Jayroe, Teresa

Committee Member

Forde, Connie

Committee Member

Browning, Donna

Committee Member

Walker, Linda

Date of Degree

5-13-2006

Document Type

Dissertation - Open Access

Major

Elementary Education

Degree Name

Doctor of Philosophy

College

College of Education

Department

Department of Curriculum and Instruction

Abstract

The purpose of this research was to determine if the variables included in the Mississippi Report Card 2003-2004 utilized for the calculation of AYP can be used to predict with accuracy greater than that which can be attributed to chance, whether or not Mississippi LEAs will attain adequate yearly progress in reading and math using the logistic regression technique. An additional goal of this study is to identify whether the inclusion of a variable representing the proportion of teachers in each Mississippi LEA with a one-year teaching certificate can notably enhance the explanatory power of the logistic regression models. This study addressed two research questions: Research Question 1: Can variables (included in the Mississippi Report Card 2003-2004) required for the calculation of adequate yearly progress be used to successfully predict Adequate Yearly Progress using the Logistic Regression technique with an accuracy greater than that which can be attributed to chance? Research Question 2: Could the addition of another predictor variable (Percentage of Teachers with One-Year Educator Licenses) notably add to the predictive accuracy of the model? This study demonstrated that using the variables utilized for the calculation of AYP, a predictive model can be successfully utilized to classify Mississippi LEAs that will and will not attain AYP in reading and math with an accuracy greater than that which can be attributed to chance. This study also established that the inclusion of a variable corresponding to the percentage of teachers in a LEA with one-year educator licenses does not add to the predictive accuracy of the model.

URI

https://hdl.handle.net/11668/17330

Download

Share

COinS