On 20 Nov 2009 at 17:40, Rick Froman wrote, in relation to Lae and Crano's (2009) meta-analysis of parental monitoring of marijuana use:
> Would this be a meta-analysis of nothing but flawed studies > of the impact of parents on adolescent behaviors > that don't take into account the impact of genetics? I don't know whether this is serious, or whether the question is a joke, tongue-in-cheek, or (heaven forfend!) a comment to mock my previous posts. Nevertheless, it's a good question and here's the answer. Yes, definitely. Through the magic of the Internet, I obtained a reprint from the author in under an hour. It's a meta-analysis of 25 correlational studies based on adolescent self-report of parental monitoring (how much they know) and marijuana use. It seems, according to the authors, that there are no experimental studies (i.e. with a control group and randomized assignment). They found a modest significant effect, r = -.21, meaning that more monitoring was associated with less reported marijuana use. As these studies are correlational, they had no business in drawing causal conclusions, but (surprise!), they did anyway. They suggested, as indicated by their title "Monitoring Matters", that monitoring reduces adolescent marijuana use. They also said: "Our review suggests that parents are far from irrelevant, even when it comes to an illegal and often secretive behavior on the part of their adolescent children. Information derived from this quantitative synthesis may prove useful in marijuana-based prevention programs and campaigns targeting parents, and might offer insight on how to alleviate a risky behaviour". Clearly, they think that they have shown that parental monitoring decreases toking up. Yet (again, surprise!) they know they really shouldn't do this, as a small disclaimer appears elsewhere, right after they say there are no experimental studies. They say: "As such, we are unable to state that parental monitoring lowers marijuana use in adolescents". So why do they say it anyway, including in their title? They should not, because the adolescents are not randomized to groups, and therefore those who are highly monitored by their parents undoubtedly differ in substantial ways from those who are not, including (gasp!) differences in the genes that parents- who-monitor pass on to their children. So yes, these are nothing but flawed studies of the impact of parents on the adolescent behaviour of toking up, if you want to use the studies to leap to a causal conclusion. (Sings: When will they ever learn? When will they ever learn?). (Interestingly, they do cite as an alternative that "in efforts to avoid parental detection, adolescent marijuana users might go out of their way to evade their parents and consequently might assume (and report) low parental monitoring". They then say "This perspective is consistent with the controversial view that parenting behaviour is largely evoked in reaction to children's behavior" (Harris, 1995, her Psych Review paper)) But not in their conclusion. Stephen Lae, A., and Crano, W (2009). Monitoring matters. Meta-analytic review reveals the reliable linkage of parental monitoring with adolescent marijuana use. Perspectives on Psychological Science, 4, 578-- ----------------------------------------------------------------- Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University e-mail: [email protected] 2600 College St. Sherbrooke QC J1M 1Z7 Canada ----------------------------------------------------------------------- --- To make changes to your subscription contact: Bill Southerly ([email protected])
