Date of Award

Spring 1992

Degree Type


Degree Name

Master of Science in Education (MSEd)


Education and Human Development

First Advisor

Robert Ribble

Second Advisor

Not legible

Third Advisor

Not legible


As part of New York State regulations designed to improve the quality of education, students must complete Pupil Evaluation Program (PEP) tests so that teachers can identify those students who are not learning at a rate that will allow them to pass the Regents Comprehensive Test (CT). This study investigates general comparisons between bilingual program students and non-bilingual program students on the New York Pupil Evaluation Program (NYPEP) tests for total Reading and total Mathematics. It seeks to move beyond comparisons of mean performance and to analyze the evaluation within each distribution of scores and investigate the explanatory power of the treatment effect size. Consequently, it can be viewed as a pilot exploration into the worth of Point Biserial Correlation techniques in furthering an understanding of the observed variation in the two sets of scores. The author compared the PEP scores of 80 sixth grade students, half of whom are taught in bilingual programs and half of whom are not. The results showed that Bilingual program students were performing slightly lower than the non-Bilingual group. However, the author cautions that the non-Bilingual group is more academically prepared in the English language in which the tests are given. That the Bilingual students were able to do as well as they did in a language and cultural-academic setting which is newer to them than non-Bilingual students should be taken into account when comparing groups. Finally, the use of Point Biserial Correlation allowed for the collection of more statistically supportive data than simply comparing differences between performance means.