Colbert, Take 2

"By attempting to mock conservative commentators, Colbert may unintentionally be helping these commentators sway potential voters to the right."

That's one of the conclusions to a study I discussed in detail yesterday. Published in the latest JOBEM, the research suggests that channel-surfing young people may not quite get what Stephen Colbert is up to as he plays a far-right, self-obsessed, TV talking head, or at least the effects of Colbert's comedy may have unintended consequences.

Watching Jon Stewart's faux news show? No such effect, suggesting something about playing an Archie Bunker-like role can have the opposite effect than you hoped for. Or maybe this is actually some kind of secret Colbert plot? A conservative, posing as a liberal, playing a conservative, and ... er, never mind. Getting a headache.

The mixed message is no doubt the root of the problem.

The authors hypothesized that Colbert's mixed message would lead to a decrease in internal efficacy, the idea that one is capable of participating in politics. The hypothesis makes sense. But Table 4 reports the results on "Internal Efficacy" and the standard question "Agree that politics and government seems too complicated." High scores mean low efficacy and there is a positive relationship between watching Colbert and this measure, meaning viewing it leads to less efficacy. But the authors say their hypothesis was not supported. It's confusing as hell, or I'm just confused. Clarification needed. They would have been better served by flipping the measure, recoding it in such a way that high meant greater internal efficacy, otherwise their table title is misleading.

Enough quibbles. It's a fascinating study, thus getting two days of discussion from me. Hell, might get a third if nothing better shows up.

Recognition and Hyperlinks

Neat little study in the latest Journal of Broadcasting & Electronic Media shows that more headline/photo hyperlinks leads to greater processing and recognition of information in a news story.

There are some differences in this set of experiments.

First, they use heartbeat and other psychophysiological measures, which is cool. Nothing beats strapping a few undergraduates into a chair and measuring their response to media content. I prefer electric shocks, but that's just me. The "increased allocation of cognitive resources" from a larger display of hyperlinks, the authors say, leads to greater recognition.

Makes sense. I do wish the authors (Wise, Bolls, and Schaefer out of the j-school at Missouri) had reported exactly what questions they used to measure recall and what the Cronbach's alpha was for their 4-item index. I'm getting PhDweebish, and maybe it's in the study and I missed it, but the alpha tells us how well the four questions hung together, their internal consistency. And recognition is different than recall, both methodologically and theoretically. Not much on this.

There's some other cool stuff here, such as the literature on why we attend more to negative versus positive images, and some use of previously rated and coded images as part of the experiment. It is important to note the power of "unpleasant" images to increase attention and, we assume, recognition. That's key here, and something journalists need to pay attention to themselves.