A new study in Marriage & Family Review explores the long-debated effects of spanking on children’s development. The researchers found that spanking explained less than 1% of changes in child outcomes. This suggests that its negative effects may be overstated. The study recommends that if spanking is used, the most effective approach is two open-handed swats on the bottom for children aged 2 to 6 when they don’t respond to milder discipline like time-outs.
The question of whether spanking causes harm to children has been a topic of intense debate for years. Previous studies have shown strong correlations between physical punishment and negative child outcomes, but these studies often failed to account for children’s pre-existing behavioral problems. As a result, it has been difficult to establish whether spanking itself causes these problems or if it is simply used more frequently with children who already have behavior issues.
“I have been doing research for 40 years to try to answer two questions,” explained study author Robert E. Larzelere, an Endowed Professor for Parenting Research in the Department of Human Development and Family Science at Oklahoma State University and author of Authoritative Parenting.
“(1) What disciplinary responses are more effective than spanking, so that we can recommend them to replace spanking? Answer so far: none, although parents should prefer milder and verbal disciplinary responses as much as possible. (2) Although it is clear that spanking is correlated with adverse outcomes such as delinquency, does it cause those outcomes (like smoking) or not (like hospitalization, which is also associated with worse outcomes compared to those never hospitalized)? Answer so far: Spanking does not cause harmful outcomes unless it is used too often or too severely or out of meanness rather than out of concern for the child’s welfare.”
For their study, Larzelere and his colleagues conducted meta-analyses using data from earlier reviews of longitudinal studies, which track the same children over time. These studies allowed for more accurate comparisons because they controlled for children’s baseline behavior—meaning the children’s behavior before the spanking was taken into account. This method helped the researchers isolate spanking’s effects and avoid attributing pre-existing behavioral issues to spanking itself.
The meta-analysis included data from 12,727 participants across 47 studies, with children ranging in age from 18 months to 11 years. The researchers focused on four key child outcomes: externalizing problems (such as aggression and defiance), internalizing problems (such as anxiety and depression), cognitive performance, and social competence.
To better understand the impact of spanking, the researchers applied two different statistical methods: the “beta method” and the “slope method.” The beta method controls for initial behavior scores and compares changes over time, while the slope method directly examines changes in behavior over the study period. These two methods provided complementary perspectives on spanking’s potential effects and helped identify any biases in the findings.
In addition to general spanking, the researchers analyzed “back-up spanking,” a specific form of discipline involving two open-handed swats to a child’s bottom when they refuse to cooperate with time-out. This method had been evaluated in randomized controlled trials, which the researchers included in their analysis to compare the effectiveness of back-up spanking with other disciplinary strategies.
After controlling for baseline behavior, the researchers found that spanking explained less than 1% of the variation in child outcomes. Using the beta method, the researchers found small harmful-looking effects, but these were trivial in size. For example, spanking explained only 0.64% of the variance in externalizing problems and even less in internalizing problems, cognitive outcomes, and social competence.
In contrast, the slope method showed slightly more beneficial-looking effects. For instance, spanking was associated with a minor decrease in externalizing problems, but this effect was so small that it explained less than 0.1% of the variance.
This suggests that spanking has a minimal impact on externalizing problems, internalizing problems, cognitive performance, and social competence. The results indicate that previous reports of the harmful effects of spanking may have been overstated due to residual confounding—meaning other unmeasured factors may have influenced the negative outcomes rather than spanking itself.
The analysis of “back-up spanking,” which involves two open-handed swats to enforce cooperation with time-out, provided more compelling results. In randomized controlled trials, back-up spanking was shown to be significantly more effective than allowing children to leave time-out without consequences.
Children who received back-up spanking were more likely to comply with parental commands and cooperate with time-out procedures. The effects were particularly pronounced in terms of faster cooperation with time-out, suggesting that in certain situations, spanking can be an effective tool for reinforcing discipline.
However, the effectiveness of spanking varied depending on the child’s age. The researchers found that spanking had slightly better outcomes for younger children, particularly those aged 2 to 6 years. For children in this age group, spanking was associated with slight reductions in externalizing behaviors.
As children grew older, spanking became less effective and was linked to slightly worse outcomes, particularly for children aged 8 to 11 years. This suggests that spanking may be more appropriate for younger children but less beneficial, or even harmful, as they age.
“Although different things work for different children (or for the same child at different times), spanking can be an effective disciplinary tool under some circumstances,” Larzelere told PsyPost. “The most effective way to use spanking is two open-handed swats to the bottom of 2- to 6-year-olds when they refuse to cooperate with milder disciplinary responses, such as time-out.”
“When used that way, defiant children learn to cooperate with the milder disciplinary responses, so that spanking can be phased out. Similarly, timeout and privilege removal can be used to back up verbal explanations and negotiations, which should be preferred by all parents when appropriate.”
The new findings are in line with a previous study, published in Child Development in 2021, which found that occasional, mild spanking had minimal negative effects on child behavior. Importantly, the results suggested that the harmful effects of spanking may have been overestimated due to a failure to separate between-subject and within-subject variance. In other words, earlier studies may have conflated differences between individual children with the effects of spanking itself, leading to conclusions that were influenced by pre-existing behavioral traits.
Larzelere expressed surprise that research on parental discipline has “rarely tried to distinguish between more and less effective ways to use spanking or other disciplinary techniques.” He noted that, in contrast, medical research consistently defines the correct dosage and the conditions under which a treatment or medication is most effective. This level of precision is often missing in studies on discipline, where guidelines for appropriate use are rarely specified.
“The studies of two-swat back-up spanking were done by one research clinic in the 1980s to try to find an effective alternative for back-up spanking,” Larzelere said. “They showed that a brief room isolation was just as effective on average, although they questioned whether rooms in homes would be as suitable for room isolation as in their clinic.”
“But hardly any conclusive research has been done since then. The failure to specify precise implementations and appropriate situations in research since 1990 may help explain why clinical treatments for oppositional defiance in young children are half as effective now as they were 30 to 50 years ago when psychotherapists trained parents to use spanking (or a brief room isolation) to enforce cooperation with time-out.”
The findings suggest that spanking, when used in a controlled and limited manner, does not cause significant harm, particularly for younger children. However, this should not be interpreted as implying that all forms of spanking are harmless. Inappropriate or excessive use of physical punishment can still result in negative outcomes and should be avoided.
“Obviously, spanking can be misused and is then likely to have harmful effects on children,” Larzelere said. “Spanking will have harmful effects if it is used too severely, too frequently, or as part of a rejecting, neglecting, or chaotic approach to parenting.”
It is also important to note that the results are based on average effects, which means individual children may respond differently to spanking depending on their temperament, family environment, and other factors.
“Different things work for different children,” Larzelere said. “Parents should try disciplinary and parenting strategies they hear about to see if they help with their children. But they are justified in questioning the absolute opposition to all disciplinary consequences sometimes expressed online. Any disciplinary method can be harmful, but even spanking can be used appropriately in a very limited role within a positive parent-child relationship.”
Future research could focus on identifying the specific circumstances in which spanking might be harmful or beneficial. This includes looking at factors such as the frequency and severity of spanking, the relationship between the parent and child, and alternative disciplinary methods.
“I would like to collaborate with anti-spanking researchers to conduct more conclusive research on how parents should discipline their children, with or without spanking,” Larzelere said. “Many popular parenting books say that parents should avoid all negative disciplinary consequences. That may work for easily managed children, but it is likely to be unrealistic and harmful for children who are more oppositional and defiant. European researchers have shown that these unrealistic expectations may help explain why parental burnout is more common in our country than in many others.”
The study, “Resolving the Contradictory Conclusions from Three Reviews of Controlled Longitudinal Studies of Physical Punishment: A Meta-Analysis,” was authored by Robert E. Larzelere, Marjorie Lindner Gunnoe, Joshua Pritsker, and Christopher J. Ferguson.
