Beware of “Evidence-Based” Preschool Curricula

Peter Gray, PhD :
If we care about the happiness and healthy development of children, we should be very concerned about what is happening with preschools. Increasingly, they are becoming what kindergartens have sadly become, that is, places for academic training of little children regardless of their interest in or readiness for it.
 This has occurred despite strong evidence that such training-whether in preschool or in kindergarten-produces no lasting benefit and sometimes produces lasting harm (see here, here, here, and here), and despite the reports of countless front-line early educators that these programs are stressing little children out and making many of them hate school even before they reach first grade (see here and here).
Now, with the Biden administration’s proposed program of universal preschool (White House, 2021), we should be especially concerned. The program, if approved, will make preschool available and free for all 3- and 4-year-olds. Enrollment will be optional, but we all know that, when it is universally available and publicly financed, families will feel pressured to enroll their children, just as they currently feel pressured to enroll children in kindergarten, and this pressure will be especially strong on families in poverty. So, it is more important than ever that we citizens, and through us the educational powers that be, come to grips with the problems that existing preschools and kindergartens are creating.
Plans Call for “Best-Practice Curricula” Supported by “Rigorous Evidence”
Everybody, of course, from Biden on down, says the new preschool programs will use “developmentally appropriate curricula” and will employ “evidence-based best practices.” But what do they mean by that? What kind of evidence leads to these labels? Here I address that question by examining a Child Development Brief, published last month by the Society for Research in Child Development (Weiland & Yoshikawa, 2021), which purports to inform us about best practices.
The brief begins by describing the Biden plan for universal preschool and its stipulation that the preschools “must use developmentally appropriate curricula.” The article then goes on to describe what the authors believe to be such curricula, based on what they refer to as “rigorous evidence.” Their primary conclusion is that curricula that involve the explicit teaching of math, language, or literacy are more effective than, and therefore preferable to, curricula that use more holistic methods.
But What Is that So-Called “Rigorous Evidence?”
What is this “rigorous evidence?” The authors cite roughly 20 research studies (more or less, depending on what you count as a study), and I looked them up. The evidence is this. If you do an experiment in preschools where children in some classes but not others are explicitly taught math skills, or literacy skills, and you test them on those skills at the end of the preschool year-or in a few cases at the beginning of kindergarten-those who were taught the skills score better on those skills than those who were not.
Well, of course they do! How could that not be true? Somewhat surprisingly, they don’t score much better, but sufficiently so to reach statistical significance. Does it lead them to do better later in school than they would have if they had not suffered the preschool training? If it doesn’t, then what’s the point of it? None of the studies cited in the brief provide evidence that academic training in preschool benefits children later in elementary school.
These researchers ignore the long-term studies, such as those I have reviewed here and here, showing that any advantage of early academic training, in either preschool or kindergarten or both, washes out and in some cases reverses itself as children go through the further grades. In one large, well-controlled study, for example, children in a state-sponsored academic preschool program for children from poverty performed significantly worse on academic tests in third grade and were 46% more likely to have been diagnosed with one or more learning disorders by third grade, than those children from poverty who, in the randomized design, had no preschool at all (Lipsey et al., 2018).
The Worthlessness of Short-Term Curriculum Research
Common sense should tell is that if academic programs for tots have no long-term benefit, make many children unhappy, and may even cause learning blocks that adversely affect their later academic development, then we should drop them like hot potatoes. That’s what the German government did years ago, in the 1970s, when they first experimented with academic kindergartens and discovered that children in those programs did worse, later in school, than those who were in play-based kindergartens (see Darling-Hammond & Snyder (1992). They paid attention to the science and responded to it reasonably.
Why has common sense not prevailed in the U.S.? The problem, I think, is that the self-interest of too many researchers and practitioners lies in short-term thinking. Researchers conduct short-term studies because they are easy to do and can be published in a timely fashion, which helps the researchers get promoted. Policy makers accept the results of the short-term studies on face value because they are too lazy, too busy, or too ignorant to dig deeper. Administrators and teachers (often reluctantly) accept the curricula promoted by the short-term studies because the higher-ups who evaluate them look mainly at short-term results, the immediate test scores, not at the long-term well-being or even the immediate happiness of the students. All this happens at the expense of our kids. Shame on us for allowing it.
Before closing, I must note that two of the studies cited by Weiland and Yoshikawa’s Brief did look for effects extending on through elementary school. One had the title, “Preschool mathematics intervention can significantly improve student learning trajectories through elementary school” (Dumas et al., 2019). The title is so misleading that it is remarkable to me that the journal editors let it through. In fact, the data presented in the article show that, by fifth grade, those in the experimental group (with math instruction) performed somewhat worse on math than those in the control group (though not significantly worse). By “trajectory,” the authors meant that the growth of measured math skills occurred a bit earlier (though barely so) for those trained in preschool than for the controls. This seems to me to be the result of really stretching far to find some way to say that the early training had a beneficial effect. All they really found was a short-term boost on math test performance that washed out over time, just what everyone else finds.
The other article cited in the brief as documenting a long-term effect is more interesting. It’s entitled “Sustained effects of a school readiness intervention: 5th grade outcomes in the Head Start REDI program” (Walsh et al., 2020). Unlike the other studies cited in the brief, this one looked at a program designed to promote social-emotional growth, not academic learning. The program involved some time spent talking about, hearing stories about, and acting out social relationships and emotional experiences.
Standard assessments showed that, on average, those in this program were better adjusted socially and emotionally than the control group by the end of preschool and that this advantage persisted all the way through the end of elementary school. Now that’s a program worth looking into. Preschool should be about learning how to get along well with others and how to keep your emotions under reasonable control. The best curriculum for that is social play, but some explicit discussion and practice of techniques for getting along with others and regulating one’s emotions may well be beneficial.

(Peter Gray, PhD is a research professor at Boston College, author of Free to Learn and the textbook Psychology (now in 8th edition), and founding member of the nonprofit Let Grow).