[Editor's note:]
We received 29 letters on the research paper and editorial on slapping and spanking in childhood published in our Oct. 5, 1999, issue, more than for any other article in recent years. A selection of the letters appears below.
The article by Harriet MacMillan and colleagues1 presents interesting data from a large-scale epidemiological sample. Unfortunately, the authors interpret their results with insufficient caution, leading to seriously misleading conclusions. To claim "a linear association" between slapping and spanking children and adult psychopathology, they need to take into account other mechanisms that could explain their data.
As acknowledged in the paper, their self-report data on childhood had no independent validity. In view of the vast literature on recall bias,2 individuals with mental disorders may be more likely to remember negative early experiences.
The findings in this study are essentially correlational, and the authors acknowledge that confounding variables might account for the association they report. One major possibility involves common genes between parents who slap or spank and children who develop alcohol abuse or externalizing problems.3
It has been shown that children with impulsive personality traits are much more likely to receive physical punishment.4 Thus, some of these parents may have been responding, rightly or wrongly, to problematic temperamental characteristics in their children.
Given all these problems in interpretation, the accompanying editorial by Murray Straus5 borders on the outrageous. We need to be very careful about changing laws on the basis of epidemiological research. There is absolutely no evidence from this study, or from any other study, to support Straus' claim that "ending spanking will reduce the prevalence of mental health problems."
References
- 1.
- 2.
- 3.
- 4.
- 5.