[This article was originally published on The Conversation. Read the original article.]
Elizabeth Gershoff, University of Texas at Austin and Andrew Grogan-Kaylor, University of Michigan
Whether you are a parent who has occasionally spanked your child, an adult with recollections of childhood spanking or even an observer who has witnessed spanking in a public place, you likely have personal experience with the spanking of children for misbehavior.
Spanking remains a common parenting practice in the U.S. A nationally representative, long-term study of children beginning in kindergarten found that 80 percent of children had been spanked by the time they reached third grade.
While we all may have our opinions about whether spanking “works” as a method of disciplining children, what does science tell us? And what are the average outcomes for children who are spanked?
Recently, we conducted a “meta-analysis” – a review of existing research on spanking – to answer these questions. We found that spanking significantly increases the risk of detrimental outcomes for children. While this finding does not mean that every child will experience problems as a result of spanking, it does mean that a large body of research has shown it significantly increases the risks for problems.
Is there any evidence that spanking is good for children?
We included data from 75 studies from the U.S. and 12 other countries that were conducted over a period of 50 years, which included over 160,000 children. We looked at the associations between spanking and several different child outcomes.
Spanking was not linked with better child behavior. Instead, we found spanking was linked with worse child behavior. Spanking was associated with 13 of the 17 outcomes we examined, and all showed spanking was linked with detrimental outcomes.
The more children were spanked, the more aggressive and antisocial they were. We also found that children who were spanked were more likely to have mental health problems, problematic relationships with their parents and lower cognitive ability.
Most troubling were our findings that children who were spanked were at significantly increased risk of being physically abused by their parents. In addition, the link between spanking and negative outcomes for children was two-thirds the size of the link between physical abuse and the same negative outcomes.
This could well mean that spanking and physical abuse are not categorically different behaviors but rather are points along a continuum of hitting children.
The findings of our study were incredibly consistent. Nearly all (99 percent) of the statistically significant effect sizes indicated a link between spanking and negative outcomes for children.
Which comes first? Spanking or behavior problems?
Our findings that spanking is linked with more problem behavior cannot prove that spanking causes the problems. It could well be that children with more problems elicit more spanking from their parents.
Clearly, we cannot do “randomized controlled trials” – that is, have a control group where children are spanked and another where they are not – to determine if spanking is the cause of children’s problem behaviors. So we depend upon statistics to know whether these outcomes are a result of spanking or a child’s behavior.
To tease apart this chicken-and-egg problem, several studies have looked to see if the link is entirely from children’s behavior to parents' spanking rather than from parents to children. Studies with large samples and sophisticated statistics have found that when children are spanked, their behavior worsens over time, even when we account for many other factors, including the extent to which parents' use of spanking is a reaction to children’s earlier problem behavior.
One such study, conducted by one of the authors (Gershoff), found that kindergartners with problem behaviors did elicit more spanking from parents over time. However, spanking still predicted increases in problem behavior by third grade, over and above the extent to which children elicited spanking from their parents.
The second author of this article (Grogan-Kaylor) has worked with data from a large and national sample of children. He too found that corporal punishment was related to increases in children’s behavior problems even when accounting for multiple other factors.
Results from other long-term studies also indicate that spanking predicts a worsening of child behaviors over time, regardless of how problematic they are to start with. For example, researcher Lisa Berlin from the University of Maryland and her colleagues found that the more one-year-olds were spanked, the more aggression they showed one year later and the lower cognitive ability they showed two years later.
When the parent is affectionate
A common question is whether spanking is effective if used by an otherwise warm and loving parent.
One of the authors (Gershoff) and her colleagues analyzed data from a large study of parents and children and found that spanking increases problem behavior regardless of how loving mothers are. Love and affection did lead to increases in children’s positive behaviors over time, such as being caring, polite and responsible.
However, spanking did not lead to increases in these positive behaviors. Instead, spanking predicted increases in problem behavior over time, proving that hugs, not hits, help children become caring and responsible individuals.
Some researchers, such as Kirby Deater-Deckard and Kenneth A. Dodge, have argued that spanking could have positive outcomes when it is considered a culturally accepted practice. However, research in the ensuing two decades has not supported this idea.
Using the same long-term study of kindergartners cited above, one of us (Gershoff) found that spanking was used more often (which suggests that it was more acceptable culturally) in black families than in white, Latino, or Asian-American families, but spanking predicted increases in behavior problems equally across all four race and ethnic groups. Spanking was not “better” for children among groups that use it frequently.
A study we collaborated on using data from families in six different countries found spanking to be linked with higher aggression and anxiety. This was true even when parents and children believed spanking was an accepted way of disciplining in their communities. Spanking is not linked with positive child behaviors, even when it is supported by the family’s culture.
No reason to spank
The evidence against spanking is so strong that the American Academy of Pediatrics has recommended that parents not spank their children. The U.S. Centers for Disease Control and Prevention also has recommended that education campaigns and legislation be used with a goal of reducing spanking.
The United Nations considers all forms of physical punishment to be violence and has called for an end to the practice. A total of 49 countries have already banned all physical punishment of children, including spanking by parents.
The message from academic, medical, public health and human rights organizations is one and the same: spanking is ineffective and potentially harmful to children. It should be avoided for the good of all children.
Elizabeth Gershoff, Professor of Psychology, University of Texas at Austin and Andrew Grogan-Kaylor, Associate Professor of Social Work, University of Michigan