Something about the debate over whether Metro public schools should switch to a balanced calendar — which would require $20 million and an earlier summer start date — gave Pith a sense of deja vu. Sure enough, Bruce Barry tackled the question in a series of Pith posts back in 2006, when MNPS conducted a survey similar to the one underway today over which calendar parents prefer: the traditional model used now, or the balanced. The results back then:
Among parents, 8,256 (45.3%) said they prefer a balanced calendar; 8,059 (44.3%) favored the existing "traditional" calendar; and 1,892 (10.4%) said they have no preference. Faculty supported the balanced calendar by a wide margin (71.2% to 25.2%). Staff members preferred the existing calendar by a small margin (47.7% to 45.0%).
But is a balanced calendar likely to have a positive impact on student achievement? Educational research on this issue indicates that gains from these calendar shifts fall somewhere between trivial and non-existent. Let's take a closer look. Perhaps the best one-stop source for meaningful academic research on this question is an article published in 2003 in Review of Education Research that carefully synthesizes and evaluates hundreds of studies of modified school calendars. The National Association for Year-Round Education (NAYRE), a group that advocates for balanced calendars, cites this article as showing an effect of balanced calendars on achievement that is "small, but nevertheless positive."
This is a somewhat optimistic reading of the research. The researchers do say that school districts have a "slightly better" than random chance of finding a performance benefit. However, they point out that statistical estimates of the amount of improvement (what statisticians call an "effect size") are around one-twentieth of a standard deviation. How tiny an effect size (d=.05) is this? Typically, educational or other interventions worth their salt yield effect sizes that are more like .50 than .05. The researchers concede that experts in program evaluation would regard the balanced calendar effect size as "trivial." They conclude:
"The possibility that the improvement may be nonexistent cannot be ruled out....It would be inappropriate to suggest that the current evidence indicates that modified calendars have a significant positive impact on achievement, in the practical sense."
Nevertheless, even if the margin in favor was as slim as a single percentage point, the survey five years ago showed that parents and teachers both favored the balanced calendar. It'll be interesting to see if that holds true today. If what I've heard from parents is any indication, I seriously doubt there'll be a 10 percent response of "no preference."