Writing in Psychology Today, Peter Gray reviewed the longitudinal study by Vanderbilt researchers of the effects of pre-kindergarten classes on low-income children.
He noted that the long-term effects were negative.
He usefully points out that the German government conducted a similar study in the 1970s:
The German government was trying to decide whether it would be a good idea, or not, to start teaching academic skills in kindergarten rather than maintain kindergarten as purely a place for play, stories, singing, and the like, as it had always been before. So, they conducted a controlled experiment involving 100 kindergarten classrooms. They introduced some academic training into 50 of them and not into the other 50.
The graduates of academic kindergartens performed better on academic tests in first grade than the others, but the difference subsequently faded, and by fourth grade they were performing worse than the others on every measure in the study. Specifically, they scored more poorly on tests of reading and arithmetic and were less well-adjusted socially and emotionally than the controls.
The Germans, unlike we Americans, paid attention to the science. They followed the data and abandoned plans for academic training in kindergarten. They have stuck with that decision ever since.
The newly reported Tennessee study of pre-K was carefully designed and focused on academic skills.
Yet the students in the academic-intensive pre-K program fell behind the control group in later years.
The major findings of the study are that this expensive, carefully planned pre-K program caused, by 6th grade, reduced performance on all academic achievement tests, a sharp increase in learning disorders, and much more rule violation and behavioral offenses than occurred in the control group….
The most striking finding in the study, to me, is the large increase in diagnosed learning disorders in the pre-K group. It seems possible that this increase is the central finding, though the authors of the report don’t make that claim. Previously I’ve discussed evidence that learning disorders can be produced by early academic pressure (here) and evidence that being labeled with a learning disorder can, through various means, become a self-fulfilling prophesy and result in poorer academic performance than would have occurred without the diagnosis (here). It would be interesting to know if the deficit in achievement test scores was entirely the result of poor performance by those diagnosed with a learning disorder.
A related possibility is that the early academic training resulted in shallow learning of the skills, sufficient to pass the pre-K and kindergarten tests but which interfered with subsequent deeper learning (an idea I discussed here). That could account for the finding that the deficit produced by pre-K grew over the years. As years go on, success on tests may depend increasingly on real understanding, so anything that blocks such understanding might show up more in later grades than earlier ones.
Another possibility is that the pre-K academic grind and pressure caused children to develop a hatred and rebellious attitude toward school. This might account for the increased rule-breaking and offensive behavior of the pre-K group as they went through elementary school. The same rebelliousness might also have caused the children to take their lessons less seriously, which could, over the years, result in an ever-greater gap between them and the controls in test scores.
Still another possibility is that the deficit shown by the pre-K group was caused not so much by what was done in pre-K as by what did not happen there. Four-year-olds need lots of time to play, create, socialize, take initiative, figure things out on their own, and learn to manage themselves. The time spent in academic training is time that they cannot spend on learning the much more important skills that come from self-directed activities. Perhaps the pre-K children were less prepared for school, especially the later grades of school, because they had not had the usual opportunities to learn how to manage themselves before starting school. This suggestion is consistent with previous research showing better long-term outcomes for play-based preschools and kindergartens than for those that have an academic component (here).
I suspect that all these hypotheses have some validity…Regardless of the mechanism, it is now abundantly clear that we should stop even thinking about teaching academics to tots. We should finally make the decision that the Germans made half a century ago and stop formal academic training for children below age 6.
How likely is it that our policymakers will learn from the science?
Diane Ravitch's blog 