The purpose of this paper is to review the relative contribution of selected school-related variables to the output of the Arkansas public schools, as measured by Science Reading Associates (SRA) achievement test scores. Multiple step-wise regression and Pearson correlation were used to analyze the data. The most important finding was that very little of the variance in student achievement scores was related to variables commonly considered to be important in the educational process. Average teachers' salaries, class size, school district size, average expenditures per pupil, amount spent on transportation, and local effort taken either individually, or as a group, never account for more than 12 percent of the total variation in student achievement scores. (BW)