Do you think your teenager will drink alcohol?
On a scale from 1 to 10, how likely do you think it your child will drink alcohol regularly as a teenager?
Moms who underestimate the chances may actually prevent kids from drinking. That's good news, believing your kid won't drink can discourage your teen from drinking.
But the opposite is also true...
Moms who overestimate their kid's risk can end up promoting more drinking.
We're talking self-fulfilling prophecy here. That's fancy talk for a false belief that becomes true according to my good friend, fellow psychologist, and darn good researcher Stephanie Madon. The queen of self-fulfilling prophecies, Madon's rooted them out in different contexts, including the classroom where teacher's false poor expectations about students actually dampens their academic achievement.
How does a false belief come true?
"Not by magic" says Madon. Somehow the mother transmits this belief, communicates it to the teen who then confirms the belief. How do you study it? With lots of data, almost 800 mothers (and their children) who completed surveys at several points in time over several years.
How do you measure the accuracy of mom's beliefs?
At the beginning of each study the mothers answered questions related to their child's future drinking, factors that predict drinking behavior. For example, the parents drinking behavior, their attitudes towards alcohol, towards adolescent/teen drinking, their parenting skills/style, their child's access to alcohol. Basically, these predictors were compared to the mother's expectations about their child's drinking. Voila. Some parents were pretty accurate, others not.
How do the false beliefs actually turn into behavior?
The researchers found it goes like this:
Mom expects kid to drink.
Kid starts to believe he's the kind of kid that drinks.
Kid, being the "drinking type", acts in line with his self-image.
Thus kid drinks and drinks.
It's an example of self-verification - a cherished concept in social psychology holding that people seek behavior that confirms their identity, that's fits their sense of self. It's a primary motivation to act in accordance with your self-image, even when it's not so attractive.
Madon and her team, including her husband, tested two other theories that didn't end up explaining the self-fulfilling effect of mom's belief so well. One, the mothers influenced children's beliefs about the acceptability of teen drinking in general, and then the kids adjusted their behavior to reflect their new beliefs. The other, that the mothers beliefs influenced their own drinking which in turn the teens "modeled" to conform to their parent's. But the self-image won out.
How does the study measure up?
Gotta say my friends covered all the bases here. The data comes from two independent large surveys in which parents and kids responded multiple times over several years. True, the families all come from Iowa and didn't include many minorities but there's no reason to expect a different pattern in African American or Latino families.
Sure, it's a correlation design - it's not like they could do an experiment and randomly assign some moms to make their kids believe they were at high risk for drinking. The main issue with correlational results - it's difficult to infer causation. Sometimes it's hard to even say which is the predictor and which the outcome variable. Here, the mom's false beliefs pre-date the kid's drinking so we can be sure we have the right order. The kid's drinking didn't influence the mom's beliefs. The other great aspect, the team ruled out many other factors that might account for the link. They had access to and controlled for a slew of possible confounding variables. That's what a well-designed survey will get you.
The study design was so strong, the authors included an entire section just on correlational design, its weaknesses, and in their case, its strengths. It's something you don't often see. Of course they also had at their disposal some of the newer powerful statistical software to do their mind-boggling path analyses.
Here's what I love about the paper, something else many researchers don't do and it's a shame. This crew tells us the magnitude and importance of their results, first in technical language, which is sometimes done, then plain English, a rarity.
What's the magnitude? It's the size of the self-fulfilling prophecy, that is, how big of an effect the mothers' false beliefs have on their teen's drinking. The authors tell us the "effect size", a technical term, which we nerds can compare across other studies to get some idea how it rates.
Turns out, the self-fulfilling effect here is pretty small, and the researchers admit it's even small compared to other self-fulfilling prophecy studies.
The portion of the self-fulfilling effect accounted for by self-verification? "It's important to keep in mind that it is 40% of a small effect." If only we had this kind of honesty from people studying breastfeeding, bisphenol-a, pthalates and other worries plaguing parents.
Better still, Madon and all put their results into perspective. Remarkably, they yank their result out of the lab and tell us what it looks like in the real world (my translations of the technical stuff is in parantheses):
Another way to understand the magnitude of the (self-fulfilling) effects we observed is to translate them back into the alcohol use scale that we used to assess children’s alcohol use... The self-fulfilling prophecy effects that we observed translate into 16% of children, on average, reporting a (1-point change on a scale from 0 to 5 measuring alcohol use) for every (approximate 1.5 point change in mothers’ beliefs on a 10-point scale). In other words, for every (1.5 increase in mothers’ beliefs on a 10 pt scale), 16% of children reported a “yes” response to one additional question on the (5-pt) alcohol use scale. For instance, in addition to what they would have otherwise reported, these children might now also have reported that they had consumed alcohol without a parent’s permission (or that they had now also gotten drunk or that they had now also consumed three or more drinks in a row within the past month). Likewise, for every (1.5 pt) decrease in mothers’ beliefs, 16% of children reported a “no” response to one additional question on the alcohol use scale.They even remind us that the self-verification process (the kid acting in line with his self-image) only accounts for 40% of the above effect, so like 6% of kids versus 16%. It's a lot to take in, I know, but it's good to hear what these effects mean in real life, in our daily behavior.
Even though it's a small effect, it could accumulate over time. One drink here, another there, next, getting drunk, getting into a car with another teen who's been drinking, each step spells plenty of trouble. Underage drinking is a major public health issue, even a little bit can be a problem.
How do we make our kids believe they're not the kind of kids who hit the bottle? Not clear from this study. It didn't look at how parent's beliefs (or behavior) influence their children's identities. That's another study altogether...maybe a second stage of research.
Yet it's no wonder the present study, with its complicated yet strong design, rigourous analyses, and keen attention to detail, not to mention pertinent message, landed in the Journal of Personality and Social Psychology, the hands down top journal in psychology and why I can brag on it with only a modicum of bias.
Too bad parents don't often hear or read about this kind of work. Especially because most of us worry about our kids getting involved with alcohol, especially when it comes to driving. But this study suggests worrying too much, if it gets translated into overestimating the risks, might be counterproductive. Might push your kid to the bottle.
I suspect the same might happen with our other concerns, like smoking, sex, maybe eating disorders, getting their tongues pierced. It also shows that parents should try to instill a positive self-image in their kids, even if this means being overly optimistic about some dangerous or disagreeable behaviors. Now I'm not saying we should tell our kids they are the most brilliant, athletic, saintliest, most gifted people we know. But there's something to be said for expecting the best of them. There's a difference between coddling and expecting them to do their homework, be polite, and stay away from booze.
Now, go finish those frightfully depressing drug-addiction father-son memoirs Beautiful Boy and Tweaked, but however tragic and cautinary their stories, do not panic.
Madon, S., Guyll, M., Buller, A., Scherr, K., Willard, J., & Spoth, R. (2008). The mediation of mothers' self-fulfilling effects on their children's alcohol use: Self-verification, informational conformity, and modeling processes. Journal of Personality and Social Psychology, 95 (2), 369-384 DOI: 10.1037/0022-35184.108.40.2069