The impact of the Saxon Mathematics program on group achievement test scores

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Howard N. Anderson

Advisor Department



The Saxon Mathematics program has been reported to dramatically increase standardized test scores for students in public education. This study compared gains on the Stanford Achievement Test of students (N = 185) in two rural Mississippi counties between 1990-1992. During that period, the experimental group (not randomly assigned) were taught using Saxon textbooks, and the control group was instructed using non-Saxon texts. Gains for each student between 1990 and 1992 were compared using a one way analysis of covariance. When controlling for the ability of the student as measured by the Otis-Lennon Verbal School Ability Index, no significant differences were noted on Total Mathematics, on Concepts of Numbers or on Mathematics Application. There was a significant difference in gains on the Mathematics Computation subtest; however, gains were in favor of the control group. Gains are reported by school, gender, and race. Black girls in the experimental group achieved higher mean Normal Curve Equivalent (NCE) gains than did black girls in the control group. Two schools were identified as obtaining extremely high or low pretest means. These schools were also predominantly white. Those schools were removed from the study, and another analysis of covariance was conducted. The change resulted in students in the experimental group achieving higher actual mean NCE gains, but the differences still were not statistically significant.