One Perspective: Don't mistake kindergarten claims for fact

Audrey Spalding, Mackinac Center for Public Policy

In a recent series of articles, Bridge Magazine suggests that kindergarten retention rates are not associated with student academic success, and may have more to do with districts pocketing extra tax dollars.
Author Ron French states his conclusion strongly:

“Taxpayers are in effect getting one school year for the price of two — paying an average of $7,000 per failed 5-year-old for an additional year of schooling that doesn’t produce better learners.”

Bridge suggests that the state could save money by spending more on Great Start, the preschool program administered by the state. The program, Bridge says, is cheaper than spending an extra $7,000 in state foundation allowance money to hold a kindergarten student back and produces significant benefits to taxpayers and students.

Though this is a possibility, the evidence on which Bridge is basing its conclusion is slim, if not nonexistent. French said in an email exchange that Bridge did not compare kindergarten retention rates of one class of students to how those students did a few years later on academic tests. Rather, French said Bridge looked for patterns between 2010-11 retention rates and 2011-12 district grades based on 4th, 8th and 11th grade student test scores. No pattern was found.

Comparing kindergarten retention rates to 2011-12 test scores of older students, some of whom were in kindergarten more than a decade ago, is not firm analytic ground for Bridge to stake its claim, especially since kindergarten retention rates have been changing over time.

According to state data provided by the Michigan Department of Education, during the 2010-11 school year, approximately 10 percent fewer kindergartners were held back than four years ago. Bridge’s failure to find a pattern between kindergarten retention rates and test scores could be simply due to the magazine attempting to compare mismatched data.

Moreover, Bridge cites “study after study finding no long-lasting academic benefit to holding students back in kindergarten…” though the link provided does not specifically address kindergarten retention. Rather, it is a nearly decade-old summary of retention research across K-12 grades.

Martin West, assistant professor of education at Harvard University, summarizes a criticism of much retention research in a paper for the Brookings Institution, noting that “…the disappointing outcomes of retained students may well reflect the reasons they were held back in the first place rather than the consequences of being retained.”

West notes that a program in Florida to retain third graders who aren’t reading at grade level has posted promising initial results, with retained students performing better than comparable students who were not retained on both math and reading tests. Florida students retained in third grade are also much less likely to be retained in later years.

In the case of Michigan, parents and school officials may be delaying students so that they are developmentally ready for first grade. Parents and school officials raised these points in comments posted on Bridge’s initial article.

Godfrey-Lee Superintendent Dave Britten posted some interesting analysis on his blog, showing that students born during the first half of the school year on average outscore students born during the second half of the school year. The gap is consistent across test subjects and student ethnicity. Britten’s analysis suggests that students who enter first grade at an older age might continue to do better throughout their public school career.

Readers should not mistake Bridge’s conclusion for fact. More research is needed to determine whether districts that are retaining more kindergartners are seeing positive or negative results. It is also a stretch to claim that Great Start is the solution, and should be expanded. Mackinac Center experts have noted before that the positive gains associated with Great Start might be due to selection bias, and that results from other preschool programs are not appropriate evidence for Great Start’s effectiveness.

Claims that expanding Great Start would save state taxpayers money are just claims — until concrete proposals are put forward. For example, French suggests expanding the part-time Great Start program could save money, while Phil Power, another Bridge author, has suggested expanding it to a full-time program, which would surely increase costs.

As with Great Start, it’s important to present sufficient evidence that a proposed policy solution has a chance of success and is fixing a real problem. There’s simply not enough to claim that district kindergarten retention practices are harmful.