Correlation and Causation, Muslim Interwebs Edition

The Pew Forum on Religion & Public Life recently released the results of a survey, titled "Among Muslims, Internet Use Goes Hand-in-Hand With More Open Views Toward Western Culture." The Pew researchers, a group of folks who know their stuff, are careful to note that behavior goes "hand-in-hand" with attitudes; in other words, these two things are correlated. And, right on cue, an ill-informed reporter gets it all wrong: "Want to improve Muslim attitudes? Get them online." Sacirbey writes:

That [i.e. getting Muslims to use the internet will lead to an improvement of their attitudes toward the West] would seem to be the lesson from an analysis...by the Pew Forum....

This is simply wrong, and if a student in my undergraduate Research Methods course wrote this in a paper, s/he probably wouldn't pass. The Pew report does not make any causal claims. To do so, their study would have needed to have been longitudinal or, at the very least, had a qualitative component. Because of this, prescriptive statements, like both the title of the RNS article and the first line of the story, are incorrect. In fact, the likely causal story here is the opposite of what Omar Sacirbey claims in the RNS story: Muslims who start out with more tolerant views of Western culture and Christianity are less likely to view the internet suspiciously and, thus, more likely to use it.

Aside from my disappointment in the media coverage of the findings, the Pew results are intriguing because they run up against the common conception of the relationship between the Ummah and the internet. Fueled most recently by the revelation that Tamerlan Tsarnaev had a YouTube channel and that he followed an al Qaeda internet magazine, the assumption has been that the internet has the effect of making extremists of otherwise-moderate Muslims. We now have compelling evidence that the opposite relationship is true.