This report explores the student background characteristics that might be associated with success or failure in calculus and evaluates the effectiveness of remedial mathematics education. The sample consisted of two groups at the University of California, Santa Cruz: (1) all students (105) who took first quarter calculus in spring, 1985; and (2) all students (138) who enrolled in a remedial course during 1983-84 and 1984-85 academic years. A path analytic approach was used to evaluate the background data of sex, ethnicity, and high school grade point average (HSGPA); scores on the mathematics section of the Scholastic Aptitude Test; and performance in the calculus course. Neither the path from sex to HSGPA nor the path from ethnicity to HSGPA was found to be significant. The path from sex to Scholastic Aptitude Test was highly significant. The path from ethnicity to performance in calculus (CALCTOT) was not significant. The highest path coefficient in the model was from HSGPA to CALCTOT. A survey and interviews with a small sample also gathered information on study habits, use of support services, perception of adquacy of instruction, self-efficacy, and learning modality preference. Larger samples are needed to determine if the survey instrument might be a useful predictor of student mathematics performance. Survey instruments are appended. (BAE)