Retention: A History of Failure

K-12 Testing

In the 1980s, as minimum competency testing took hold, city after city implemented retention policies in which low scores on tests forced students to repeat a grade. In city after city, the policies failed: after a few years, students who were held back did no better than comparably-scoring students who were not retained, but those held back were up to twice as likely to leave school. As a result, many retention programs were dropped in the early 1990s or scaled way back. Still, a 1998 report from the National Academy of Sciences, High Stakes, pointed out that millions of children, disproportionately boys and children of color, were being retained – despite the evidence that the policy is counter-productive.


With the onset of test-based “accountability” in late 1990s, grade retention has expanded again. Chicago was one of the first cities to implement such a policy, and the harmful consequences soon appeared (Examiner, Fall 2000). Several recent reports from the Consortium on Chicago School Research have thoroughly demolished claims of the practice’s value (see p. 16).


New York City education officials maintain that this time they will implement retention properly by providing extra assistance to students, such as summer school. However, the Chicago program promised the same. As the Sun Times explained in 1997, “Some experts say Chicago’s promotion policy – with its multi-layered safety nets – is a whole new animal.”


It turns out that summer school programs focusing on boosting test scores provide at best a one-to-two-year modest gain that then disappears, whether or not participating students are retained. Indeed, New York’s own “promotional gates” program of the 1980s also purported to provide extra help but was dropped when it became clear it was an educational disaster.


Independent educators and parents in New York argue that New York should focus on helping students in grades K-3 before they fail, rather than jumping in with retention programs in grade 3.