Critique of Afifi & Gershoff 2017 study
This spanking study suffers from the following common deficiencies (in red):
- Vague or inaccurate definition used for “spanking”
- Combines abusive physical punishment with ordinary spanking
- Includes physical punishment of adolescents, which is abusive
- Fails to control for frequency of spanking usage
- Fails to limit spanking to younger ages, especially under 6 years
- Relies upon statistical correlation rather than true causation
- Fails to control for, or define, child behavior prior to use of spanking
- Fails to compare spanking with other available correction measures
By Erik Lief — November 3, 2017
There are studies that provide fascinating insight, or perhaps unexpected and interesting findings. And then there are other studies that make you wonder why they were even conducted in the first place.
This is one that falls squarely in the second group.
The study’s primary conclusion was that spanking “can lead to a host of mental health problems in adulthood,” which include depression, suicide attempts, use of illegal drugs and moderate-to-heavy drinking.
Apparently, because this is an adverse physical act, there was reason for University of Michigan researchers to investigate whether it might be associated with problems later in life. Perhaps, they postulated, spanking psychically scars its victims, raising the chances of these future issues occurring.
Results of their study, which was published in the journal Child Abuse & Neglect, showed that nearly 55 percent of more than 8,300 people, ranging in age from 19 to 97, reported they were spanked as a child. And that those “reporting exposure to spanking had increased odds of depression and other mental health problems,” researchers said in a statement, with 37% of them having a greater chance of attempting suicide, 23% a greater chance of moderate-to-heavy drinking and 32% a greater chance of using street drugs, the study reported.
Spanking, the authors write, is typically defined as the “use of physical force with the intention of causing a child to experience pain, but not injury, for the purpose of correcting or controlling the child’s behavior.”
But the study’s biggest drawback is that there’s no way to know whether spanking is linked to any of these later-in-life problems. The study has no ability to connect this act to future ills, so all we’re left with is: “OK. So what?” – because the findings cannot show causation in any way.
Everybody knows this, it’s so basic: Just because there’s a correlation, that doesn’t prove a causality because confounders exist, as my colleague Alex Berezow just explained the other day: “For instance, one might be tempted to conclude that moving to Florida makes people develop Alzheimer’s. But this correlation has been confounded by age; in reality, old people both retire to Florida and develop Alzheimer’s. The Sunshine State is blameless.”
The respondents were also asked, “if an adult inflicted physical abuse (push, grab, slap or shoved) or emotional abuse (insulted or cursed).” And of course, this raises the question of how researchers can isolate these acts from later-in-life problems, and solely focus on the implications of spanking. If the study’s findings were questionable to start with, this information about other forms of abuse only makes it even more so.