This paper analyzes prevalent methods of comparing schools' performance and recommends alternative measures more helpful toward understanding how schools might be improved. The influence of two factors--resource availability and financial need--on the level of school programming is explained in light of Michigan's system of educational finance. Because of varying professional salaries and special costs for categorical programming, per pupil expenditure is judged an ineffective measure of quality. Instead, assessing services by identifying the number of professionals per 1,000 students is advocated, a method accounting for such quality indicators as lower teacher-pupil ratios and use of special staff. A discussion of measuring quality in education in terms of student achievement draws on effective schools research. Following Ronald Edmonds and Lawrence Lezotte, this paper argues for the fairness of determining the effectiveness of schools not on the basis of aggregate scores, but through test data broken down by social class. An effective school is thus defined as one in which the percentage of students from the lowest socioeconomic class in attendance who attain mastery of the basic skills taught at each grade is the same as the percentage of students from the highest socioeconomic class attending. (MJL)