Preschool Math Skill Predicts Success

Controversial Study Says Math, Reading Skills Matter More Than Behavior

From the WebMD Archives

Nov. 12, 2007 -- A hotly controversial study shows that preschool math and reading skills predict later academic success, but behavioral problems and social skills don't.

Northwestern University economist Greg Duncan, PhD, and colleagues analyzed data from six long-term studies of school readiness. The studies measured kids' math and reading skills and various aspects of behavior both before entering school at age 5 or 6 and later, during early or middle elementary school.

"The study was pretty surprising -- all six studies showed the importance first of math skills, and second of reading skills," Duncan tells WebMD. "But most surprising was that the association we expected between behavior problems and lack of social skills and later learning seems to be zero."

Duncan, now president-elect of the Society for Research on Child Development, was a member of a National Academies of Science panel that in 2000 reviewed the science of early childhood development. That panel came to a very different conclusion. It found that school readiness depends just as much on social and emotional skills as on thinking skills.

"I was never really convinced by the studies that show social and emotional behaviors to be more important than cognitive skills," Duncan says.

In their study of the kindergarten skills that predict later academic success, Duncan and colleagues found that math skills were by far the greatest predictor of success. Kids who had mastered basic math skills before entering kindergarten were much more likely than other kids to do well not only in math, but also in reading.

Early math skills were twice as strong a predictor of academic success as were reading skills. But like kids with good math skills, preschoolers with good reading skills later did well in both math and reading. Math skills were three times as strong a predictor of future success as ability to pay attention, the only behavioral or social skill to show an effect in the Duncan study.

"We don't really know why behavioral variables do not affect later achievement," Duncan says. "But for kids with a given set of reading and math skills, it just doesn't seem that behavior problems give them a net disadvantage."


Child Experts Disagree

WebMD asked two child development experts -- both of whom recently published studies on school readiness -- to comment on the Duncan study. Both were highly critical of the study and of Duncan's conclusions.

Psychologist Clancy Blair, PhD, is associate professor of human development and family studies at Pennsylvania State University.

"Duncan and his colleagues are brilliant people, but their conclusions are built on feet of clay," Blair tells WebMD. "Their finding that behavioral measures did not correlate with later academic success is contrary to other data. They mainly focused on behavior problems. They did not tap into behaviors more related to school readiness."

Psychologist Megan McClelland, PhD, associate professor of human development and family sciences at Oregon State University, agrees with Blair that the Duncan paper failed to measure important aspects of children's behavioral and social skills.

"Finding out that the skills you start out with predict the skills you end up with is not very interesting," McClelland tells WebMD. "Other studies, which find that children's improvement in self-regulation prior to kindergarten predicts later academic skills, are much more compelling."

Teaching School Readiness

Neither Blair nor McClelland has a problem with teaching preschoolers basic math and reading skills -- if it's done the right way.

"Drill 'em, kill 'em," Blair says. "Flash cards, the old this-is-a-square, this-is-a-triangle -- that didactic stuff is poison. Parents must manufacture situations where children take on challenges just at or above their ability -- puzzling out words or letters or drawing a picture. If parents make it fun, kids develop self-regulatory abilities from this sense of success."

McClelland argues that children can't learn math or reading if they can't sit still and can't remember.

"Parents can make sure their children can sit still when they need to, that they can work independently and also in a group. Those are the skills that are going to set you up to be successful in life, because you follow through," she says. "Can you work independently? Can people depend on you? To do well on a math test you have to have these skills. Parents should focus on whether their children can play well with other kids, and on whether they have some self-regulation and persistence on tasks."


Duncan does not advocate preschool calculus classes. But he insists that his study points to the need for research into the best ways to improve preschoolers' early math and reading skills.

"It would be very interesting to see if reading and math interventions affected math and reading much later on -- to see if more complicated skills in third grade are boosted as well," he says. "We need to have a cupboard-full of evaluations of the various curricula that extend beyond the end of the programs themselves to see if there are lingering improvements in what kids learn."

Meanwhile, Blair says parents should trust their children's innate curiosity -- while being sure to set responsible limits in order to "seed" them with the ability to self-regulate.

"Follow your child's lead in the way to make things interesting. Don't direct every little thing," he says. "Whatever it is they are doing, try to structure learning around their play. Anything that allows children to integrate their wants, needs, and desires with some kind of planning process, they are going to learn and be better for it."

The Duncan study appears in the November issue of Developmental Psychology.

WebMD Health News Reviewed by Louise Chang, MD on November 12, 2007


SOURCES: Duncan, G.J. Developmental Psychology, November 2007; vol 43: pp 1428-1446. Shonkoff, J.P.From Neurons to Neighborhoods: The Science of Early Childhood Development, National Academy of Sciences, 2000. McClelland, M.M. Developmental Psychology, November 2007; vol 43: pp 947-959. Blair, C. and Razza, R.P. Child Development, March/April 2007; vol 78: pp 647-663. Greg J. Duncan, PhD, professor, school of education and social policy, Northwestern University, Chicago; and president elect, Society for Research on Child Development. Megan McClelland, PhD, associate professor of human development and family sciences, Oregon State University, Corvallis, Ore. Clancy Blair, PhD, associate professor of human development and family studies, Pennsylvania State University, State College, Pa.

© 2007 WebMD, Inc. All rights reserved.