Friday, October 7, 2011

Friday Fun: Breaking News! Correlation Does Not Equal Causation

Man and IPhone pictured in loving embrace (source)
Right now I am teaching a personality psychology class and we are talking about research methods. Invariably, anytime I teach psychological methods I always end up talking about correlations--specifically, that a correlation is an association between two variables and nothing more. The important point is that correlations--even those that come from fancy associations between behavior and brain images--do not mean causation. Students are typically quite receptive to this information.

It's too bad that some journalists (and to be fair, even some scientists) forget this lesson. 

Just for fun, I gave out extra credit this week to any student who could find a news article claiming causation from correlation. I gave my students a 6 hour time window to complete this assignment. Not surprisingly, half my class of 60 students came back with a unique example (for those keeping score at home, that's 30 news articles that inappropriately infer cause from correlations)! Below, I summarize my three favorites:

(1) New York Times: Apparently, Science Says You Love Your IPhone
Others have posted about this here and here, and virtually the entire neuroscience community has denounced this finding here, so I won't beat a dead horse. Basically the study reported in this piece involves taking 16 people and showing them pictures of an IPhone while their brain is imaged using FMRI technology. FMRI technology can tell you where your brain is activated, via blood flow. The researcher--who also wrote this Opinion piece--found that the insular cortex was activated while viewing the IPhone. The article then claims that since the insula is activated during experiences of love, then ipso facto, we love our IPhone!!

Actually, the insular cortex is activated during a number of positive, negative, and attentional experiences and as such, it is plausible, given the correlational nature of these results, that the insular cortex is being activated because participants are simply paying attention to the IPhone that the experimenter has presented to them. To summarize in the words of 1960's Robin, "Holy third variables, Batman!"

Dear dating, stop making me drink! (source)
(2) CNN: Danger! Danger! Dating Spreads Drinking Among Teenagers
In a study reported by CNN, a group of researchers examined the social networks of teens and found that dating teens tended to be exposed to mixed-gender peer groups (oh the dread of mixed gender groups!). The article then claims that the strongest influence on the drinking behavior was the drinking patterns of the second degree friends that came along with the peer group. It is through these murky waters that we arrive at the conclusion: Dating spreads drinking in teens!

What the CNN article should say is that the strongest association with individual drinking behavior occurred with that of the second degree friends. Given that these variables were all simply measured based on friend and self reports, it is unfair to say that the second-degree friends caused the drinking behavior of teens to increase (this association could be due to some other unmeasured variable). It is an additional leap to say that the dating relationship is the underlying cause of these patterns, though it is a plausible explanation. In lieu of definitive causal evidence, it is probably a good idea for parents to hold off on their grounded-for-life strategies.

Photo taken just before attempted murder by TV (source)
(3) Time Healthland: Don't Look Now, But Your TV Is Trying To Murder You!
This article refers to an Australian study which found that survey reports of TV watching time were associated with higher rates of mortality. This is correlation in its simplest form, but that didn't stop the article from claiming: "spending too much time in front of the TV may take years off your life."

To be clear, the authors of this study are not trying to say TV, in and of itself will kill you, just that watching TV is a marker for a host of other behaviors that are unhealthy, and it is these factors that are likely to take years off one's life. So for those of you that are into science fiction, we have not uncovered a secret plot to take over the world, hatched years ago by TVs everywhere!

Many people in the science media do exceptional work, and this article is not meant to be a call for all researchers to stop talking to the media. To the contrary, we started this blog to do what many great journalists do: Make science accessible to everyday people in a responsible way! Of course this means remembering that correlation does not, in fact, mean causation (even when the correlation comes from fancy pictures in the brain!).

What are some of your favorite examples of over-claiming in media reports of science research? I'd love to hear about them in the comments!

Veerman JL, Healy GN, Cobiac LJ, Vos T, Winkler EA, Owen N, & Dunstan DW (2011). Television viewing time and reduced life expectancy: a life table analysis. British journal of sports medicine PMID: 21844603

No comments:

Post a Comment