This paper addresses three questions policymakers may have concerning the results of the Project STAR data. First, when reducing class size, is there a "turning point" at which class size reductions lead to increased gains in student achievement? Second, do small classes help low socioeconomic status "at risk" students more than they help high socioeconomic status students? Third, what is a useful way of determining whether the gains from reducing class size are worth the costs involved? Appended are four tables summarizing standard achievement test scores. (SI)