This study investigated the relationship between a teacher's knowledge of algebra and student performance. Twenty-eight Algebra I teachers and their 807 students took part in the study. Regression analysis was used to predict for each teacher an “expected” score for teaching algebraic concepts and skills. The difference between the observed score and the expected score was called “teacher effect.” The effect scores (one for algebraic concepts, the other for skills) were correlated with the teacher's knowledge of the real number system and other related algebraic structures, length of service, college mathematics grade point average, and the number of postcalculus courses taken. The only significant correlation obtained was between the teacher's knowledge of algebraic structures and the number of postcalculus courses.