Wall Street Journal, October 14, 2009, Wednesday

Wall Street Journal

October 14, 2009, Wednesday

Wall Street Journal

New Research on Spanking Might Need a Time Out

Studies Aim to Settle the Longstanding Debate Over the Disciplinary Practice's Effects, but Statistical Shortcomings Persist
By CARL BIALIK

Three recent, widely reported studies on spanking children claimed to show that the disciplinary practice impairs cognitive development in children. Together, they held out the promise of providing the latest, definitive word on a passionate debate.

Numbers Guy Blog
Why It's Hard to Measure Spanking's Effects Yet the three aren't likely to resolve anything. Many statisticians say they find in them less a firm conclusion than further proof of the difficulty of measuring spanking's impact.


Getty Images
Statistical analysis of spanking's effects on cognition are clouded by many complicating factors. Effects can be attributed to the wrong cause, statisticians say; rather than spanking causing problems in children, it is possible that their existing cognitive problems can make spanking more likely. Moreover, any effects of spanking are difficult to measure and probably small. And unlike, say, a study on prescription drugs that removes a misleading placebo effect, no ethical study can assign some children to be spanked. Instead, parents must be trusted to remember and share their disciplinary practices.

Spanking studies have a long history of fueling rather than settling the corporal punishment question. Earlier findings that spanking can contribute to aggressive behavior in children helped spur the American Academy of Pediatrics to study the issue and recommend against spanking in 1998 -- a conclusion that is still disputed.

For instance, the American College of Pediatricians, which split from the academy over its backing of adoption by same-sex couples, supports spanking in certain instances. Den Trumbull, vice president of the group, says studies need to distinguish between appropriate spanking -- following a warning and done in privacy following specific, proscribed misbehavior -- and reactive, anger-based spanking. "Spanking gets a bad name of late, because parents tend to use it reactively, when having a bad day," Dr. Trumbull says.

The most methodologically sound of the new studies, a number of researchers concur, was conducted by Lisa J. Berlin, a developmental psychologist at Duke University and published in the journal Child Development. She and her co-authors examined survey and observational data about 2,573 low-income toddlers, tracking the punishment they received and their scores on cognitive exams at ages 1, 2 and 3.

The researchers were able to trace relationships over time, for instance showing that spanking, but not verbal punishment, at age 1 predicted aggressive behavior at age 2 and lower mental development scores at age 3. Conversely, they found that kids with underlying problems such as aggressive behavior aren't later spanked more frequently. So the point of criticism of many studies -- that problems cause spanking and not the other way around -- is addressed in this research.

But researchers noted that the study didn't remove all possible explanations for why some children develop faster than others, such as their parents' intelligence. Dr. Berlin used maternal educational levels as an imperfect proxy for parents' intelligence.

Daniel Mundfrom, a statistician at New Mexico State University in Las Cruces, says that even without accounting for other factors, spanking at age 1 explained less than 1% of the variation in cognitive ability at age 3. In other words, maybe spanking does lower intelligence, but not by much.

Dr. Berlin, who says she isn't an advocate for or against spanking, says she conducted the study because prior studies raised alarms over spanking but also because she hoped to address their shortcomings. Prior studies may have been flawed, she says, but their conclusions may not have been, she says. Her results "are highly consistent with the large majority of the spanking research showing spanking's negative effects."

Two other studies released last month were too flawed to help settle the debate, some researchers say. They were conducted by Murray Straus, a sociologist at the University of New Hampshire and an outspoken antispanking advocate.

In one peer-reviewed study, published in the Journal of Aggression, Maltreatment, and Trauma, Prof. Straus and Prevention Research Center senior research scientist Mallie J. Paschall examined how 1,510 children were disciplined and how they scored on standardized cognition tests.

The researchers found that children aged 2 to 4 who weren't spanked gained an average of five points, equivalent to points on an IQ test, four years later compared with those who had been spanked three times or more per week.

While the researchers made efforts to assure that other explanations weren't possible, they didn't go as far as Dr. Berlin did. They didn't take into account whether the children exhibited signs of being difficult or oppositional, and whether children received physical punishment beyond spanking, which could create emotional problems that affect their development.

In addition, the study failed to dispense with an alternative explanation for reasons why some kids scored lower on cognitive tests. "It also could be that kids who got spanked more...were already developing at a slower rate," says Marjorie Gunnoe, a professor of psychology at Calvin College in Grand Rapids, Mich. The Berlin study addressed this concern by showing that kids who were developing slower at one age weren't getting spanked more later.

In an email, Prof. Straus defends the research, saying it is "a very solid study methodologically." A press release from his university about the research noted that the sample was "nationally representative."

A second study of his, which was widely reported in the press and presented at a conference about violence, shows that people in countries where spanking is most prevalent have lower average IQs.

Prof. Straus concedes that the methodology was flawed and that spanking may not account for the differences in average national IQs. He says he presented the data on national IQ in part because it corroborated his other study. "The questionable statistics are so consistent with the statistics in the other paper," he says, adding that his second study can provide "a field day writing about questionable statistics."

Some statisticians agree. For one thing, the results are skewed by a relatively small number of countries with high rates of spanking and especially low average IQs, particularly Tanzania and South Africa -- where about a third of university students reported being spanked a lot before age 12, and where average IQ rates stood at 72. Excluding these countries, "the line would be much closer to flat, indicating little or no relationship," says Dr. Mundfrom.

Attributing cognitive problems in children to spanking is hard enough. But then saying it is a major reason behind the lower IQ of a nation's entire population is even trickier because there could be a new raft of potential causes.

Martin Wells, a statistician at Cornell University, re-ran the statistical test to check whether regional variations in IQ -- which is lower in Latin America and Africa -- could account for the IQ differences Prof. Straus found. After accounting for regional variations, Dr. Wells found the effect of spanking vanished. Dr. Wells plans to use the Prof. Straus's research in the classroom to demonstrate why it is important to consider alternative explanations.