Epidemiologists are stupid – say epidemiologists in NY Times Op Ed

General belief has to be revised in light of properly done study

Family dinners not as beneficial to kids social health as previously thought

Screamingly obvious controls were omitted

Remarkable Op Ed piece in the Times today (Jun 29 Fri 2012) corrects the wrong impression created by many epidemiologists in the past who did their studies on the effect of families observing dinners together often without correcting for other variables. Apparently if you do correct for factors such as whether the family is well off in other respects you find that the sharing dinner factor is much less influential than previously thought. Kids who share dinner with their families typically do not do that much better than kids whose families don’t eat dinner together very often.

The whole Op Ed amounts to a public condemnation of their colleagues for being stupid to the point of idiocy in designing their studies without controlling for the most screamingly obvious variables.

A paradigm falls in that group family dinners are not so powerful an influences on the social health of kids as has been mistakenly thought, owing to the abysmal work of other epidemiologists.

Another paradigm falls too – the general belief that researchers in epidemiology who publish studies in journals are more intelligent in designing studies than the average comedian, who could probably do a better job if the work wasn’t so dull.


June 29, 2012
Is the Family Dinner Overrated?
DOZENS of studies in the past decade have found that teenagers who regularly eat dinner with their families are healthier, happier, do better in school and engage in fewer risky behaviors than teenagers who don’t regularly eat family dinners. These findings have helped give dinnertime an almost magical aura and have led to no small amount of stress and guilt among busy moms and dads.

But does eating together really make for better-adjusted kids? Or is it just that families that can pull off a regular dinner also tend to have other things (perhaps more money, or more time) that themselves improve child well-being?

Our research, published last month in the peer-reviewed Journal of Marriage and Family, shows that the benefits of family dinners aren’t as strong or as lasting as previous studies suggest.

We considered a rich body of data: the National Longitudinal Survey of Adolescent Health. This is a nationally representative sample of about 18,000 adolescents who were interviewed twice, a year apart, in middle school or high school, and then again in young adulthood (between ages 18 and 26). They answered detailed questions about their lives and well-being, and their parents also answered questions on topics like income and living arrangements.

In our study, we analyzed how the frequency of family dinners was associated with three indicators of a young person’s well-being: depressive symptoms; drug and alcohol use; and delinquency (a tally of many behaviors, from petty shoplifting to physical assault).

First, we looked at the associations between family dinners and these measures of well-being at just a single point in time, in adolescence. Without controlling for any other factors, the associations between family dinners and well-being were quite strong and in line with past research. But the associations were far less striking after we accounted, with the help of the data, for the ways in which families who did and didn’t eat together tended to differ: for instance, in the quality of family relationships, in activities with a parent (a tally of things like moviegoing and helping with schoolwork), in parental monitoring (things like curfews and approving clothing) and in family resources (things like income and whether both parents were in the household).

To give an example: without controlling for such factors, we found that 73 percent of adolescents who seldom ate with their families (twice per week) reported drug and alcohol use, compared with 55 percent of those who ate with their families regularly (seven days a week). But controlling for these factors, the gap was cut in half, from 18 percentage points to 9.

Next, as a more stringent test of causality, we looked at adolescents over the course of a year and examined how changes in the frequency of family dinners related to changes in well-being. If adolescents were eating family dinners more often a year later, were they better off? We found that following teenagers over a year provided even weaker evidence for the causal effects of family dinners on adolescent well-being — only the effect of family dinners on teen depressive symptoms held up. There was no effect on drug and alcohol use or delinquency.

Finally, we looked at whether family dinners in the teenage years had effects that persisted into young adulthood. Again, evidence for benefits was thin. We found no direct, lasting effects of family dinners on mental health, drug and alcohol use or delinquency. (Of course, it may be that family dinners have a stronger or more lasting effect on behavior that we didn’t study, like eating habits.)

What, then, should you think about dinnertime? Though we are more cautious than other researchers about the unique benefits of family dinners, we don’t dismiss the possibility that they can matter for child well-being. Given that eating is universal and routine, family meals offer a natural opportunity for parental influence: there are few other contexts in family life that provide a regular window of focused time together. (A study by Columbia University’s Center on Addiction and Substance Use asked teens when, apart from dinner, they talked to parents about their lives: a vast majority said it was when driving in the car.)

But our findings suggest that the effects of family dinners on children depend on the extent to which parents use the time to engage with their children and learn about their day-to-day lives. So if you aren’t able to make the family meal happen on a regular basis, don’t beat yourself up: just find another way to connect with your kids.

Ann Meier is an associate professor of sociology at the University of Minnesota. Kelly Musick is an associate professor of policy analysis and management at Cornell University.

This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *