In Brief

Should you spank your 4-year-old for kicking his little sister and refusing to sit for a "time out"?

No, says Columbia University psychologist Elizabeth Thompson Gershoff, PhD, in the light of her meta-analysis of corporal punishment research published this summer in Psychological Bulletin (Vol. 128, No. 4). In fact, she says, psychologists should suggest that parents avoid corporal punishment and try other discipline techniques--a recommendation some psychologists disagree with, saying there isn't enough evidence to make such a recommendation.

In an analysis that garnered national media attention, Gershoff examined 88 studies of 36,309 children for associations between corporal punishment--defined as using physical force that causes a child pain, but not injury, to correct or control behavior--and 11 outcomes in childhood and adulthood.

Corporal punishment was significantly associated with all 11 endpoints, including poorer moral internalization, quality of relationship with parent and mental health--as well as increased rates of abuse by a parent in childhood aggression, criminal or antisocial behavior, and abuse of own child or spouse in adulthood. What the results don't mean, says Gershoff, is that all spanked children will develop the negative outcomes she studied.

Gershoff did find one positive association with spanking: children were more likely to immediately comply with their parents' direction right after being spanked. However, there were also inconsistencies with this finding: Of the five studies that examined corporal punishment and a child's compliance immediately after, one found no relationship and one found a negative relationship.

Even if children do comply in the short-term--for example, by stopping kicking their little sister--Gershoff's analyses show they probably aren't internalizing the "moral message" of the spanking--that it's not appropriate to hurt people.

"By itself, it doesn't teach why children shouldn't engage in the misbehavior or what they should do instead," argues Gershoff. "Instead, it teaches that you don't want to misbehave when the parent is around."

But other psychologists object to her conclusions. In a commentary in the same journal, Diana Baumrind, PhD, Robert E. Larzelere, PhD, and Philip A Cowan, PhD, argue that the data are too unclear to warrant a blanket injunction against all corporal punishment.

They say there isn't enough research that shows nonabusive spanking's causal effects to draw conclusions about whether and when parents should spank because most of the studies in the meta-analysis are correlational. For example, three studies on the meta-analysis also investigated other disciplinary techniques for younger children's aggression, and all found stronger detrimental associations for the other disciplinary techniques.

However, none of the meta-analyses identified correlations between corporal punishment and positive long-term child outcomes, Gershoff notes.

"The first step in establishing causation," she adds, "is establishing that two constructs are correlated, and the meta-analyses overwhelming indicate significant correlations between parental corporal punishment and 10 negative long-term outcomes."

Baumrind and her colleagues counter that while the correlational evidence is a necessary first step, the crucial steps that could establish causation have yet to be made.

Moreover, they add, Gershoff's inclusion of some more severe forms of corporal punishment, such as slapping in the face, could have skewed the results. Gershoff says she included them because many parents use them.

The psychologists also differ on exactly what causes the negative associations in Gershoff's study. While Gershoff argues that spanking leads children to believe that violence achieves results, Baumrind and her colleagues counter that excessive child misbehavior, which can elicit increased disciplinary responses of all kinds, leads to the detrimental outcomes, not the spanking itself. In such circumstances, says Gershoff, excessive child misbehavior is likely the product of ineffective parenting strategies that a parent has used over time.

All agree that more research is needed to tease out how spanking compares to other discipline techniques and affects children in the short and long terms, as well as what could mediate those effects.

--D. SMITH