Knowledge Structures II

A few days ago I wrote about a new study that examines the differences between actual knowledge and knowledge structures. I discussed those two concepts in the previous blog, but I do want to comment on an odd way of measuring news use.   Most of us would probably fall back on mere exposure as a way to measure "use" of the media, but the author instead uses two questions that deserve special attention.  To tap the concept news use, she does the following, quoted directly from the study:

Use of political news. Attention was measured on two items that assessed whether participants had ‘‘paid a lot of attention to news coverage’’ and whether people ‘‘felt informed’’ about political news in this election. The items were related (r=.66).

So on the one hand we have attention, which is fine.  There is a good body of work that examines the differences, conceptually and methodologically, between exposure and attention.  But then she adds an item that asks whether people "'felt informed' about political news in this election."

Can I have a huh from the audience?

I assume she miswrote -- feeling informed about political news rather than by political news, otherwise this doesn't make a lot of sense.  And we're mixing theoretical apples and oranges, combining attention to the media with what might essentially be political self-efficacy or, in political science terms, internal efficacy.  It's hard to say, but despite the high correlation between the two questions, I'm not particularly happy with lumping them together and would not have allowed it had I been a reviewer of this manuscript for Communication Research Reports, or I would at least have required a more vigorous defense.

Okay, enough with the nerdy methodological nonsense, what did she find?  Glad you asked.

Treating knowledge structure depth as the dependent variable (basically, how many different things respondents can name), we find attention to the news and reflective integration (a scale that measures how much you think about the news) to be positive predictors.  Factual knowledge was not a predictor of this structure variable, which is kinda odd and makes me wonder about the quality of the depth listings.  Mere venting?  Or is what people know versus what people say quite different?  Or is this simply a function of using college students as respondents because, as we all know, teenagers and college students -- while related to humans -- have not evolved to humanity yet.  In other words, they're not real people.  Yet.

Also a positive predictor is conservatism.  This also is odd, because ideology, once you control for education and the like, typically does not predict political knowledge.  Given these were college students, a fairly homogeneous sample, I'm not sure what to make of this one, or of the whole study.  There's a lot to like here, but also a lot to question in terms of guidance for other scholars.

Full study cite:  Rebecca M. L. Curnalia (2010), "Predictors of the development of applied knowledge structures for presidential candidates."  Communication Research Reports, 27, 80-89.