Jay P. Greene and Marcus A. Winters
Manhattan Institute
September 2006
This report continues last year's research, which studied the introduction of a test-based retention policy in Florida schools. That policy, enacted in May 2002, requires students in third grade to score at or above the Level-2 benchmark on the reading portion of the state's FCAT assessment in order to be promoted. (Only students who score at Level 3 or above are considered to be proficient for the purposes of NCLB evaluations.) The first study showed that social promotion is not as beneficial to student achievement as making low-performers repeat grades and found (like the comprehensive report on Chicago's elimination of social promotion) significant student academic improvement in the year following retention. Unlike the Chicago report, however, this new study finds that students subject to Florida's retention policy continue to perform markedly better than their peers even in the second year after being held back. The authors write that, "students lacking in basic skills who are socially promoted appear to fall farther behind over time, whereas retained students appear to be able to catch up on the skills they are lacking." The authors used two analytical models, first focusing on across-year comparisons, and then focusing on comparisons of same-age students who were either just above or just below the state's cut-off for promotion. The two models are intended to be redundant to account for each other's inherent statistical weaknesses--thus, Greene and Winters strengthen their overall conclusions. This report is a thought-provoking nugget, which should encourage educators to reevaluate softer criteria for retention and hammer home the benefits of making students try, try again.
