So say some researchers, as reported in the New York Times.

Years ago, a study looked at kids who had been randomly assigned (within their schools) to kindergarten classes. Kids who attended good kindergarten classes (low class size, experienced teacher, etc.) did better in the next few years, but the effects eventually wore off as measured by test scores. More recently, researchers tracked down the participants in adulthood — and found that the effects of kindergarten seemed to have reasserted themselves, as measured by income, etc.

I’m skeptical. Specifically, in the PowerPoint report they provide (PDF), the scatterplots seem to support their conclusion, while the trend lines don’t. For example, look at pages 27 through 29, about class size, college attendance, and earnings. Or 32 and 33, about the effect of teacher experience and its effects on wages. Or 44 through 46, about “class quality” and earnings.

And not to mention the basic “sniff test” the results fail. Kindergarten is meant to teach academic skills, and the most direct measures of those skills reveal no difference after several years — and yet more indirect measures, decades later, somehow prove that kindergarten really, truly matters? I’m willing to follow the evidence where it leads, but I need more than this. I’m looking forward to a more thorough report, and some attempts to replicate these results with other data sets.