Fox News and Death Panels: A Love Story

Remember death panels?  Freshly published research in Journalism and Mass Communication Quarterly examines the predictors of the death panel misperception and I'm gonna point out one main result:

Fox News, you did it again.

The research, by Patrick Meirick at the University of Oklahoma, used national survey data from 2009 to establish that watching Fox News, not other media outlets, led to belief in the myth.  Hardly shocking, but here's the kicker.  It do so -- only among highly educated respondents.

Huh?  Education tends to be associated with greater political knowledge -- not belief in political myths.  Enter into the fray a little thing called motivated reasoning.  I've written in detail about this theory before, and I've used it in my own research (which, oddly best I can tell is not cited in this study ... tsk tsk). Essentially, the theory argues people believe what they want to believe, and the more they care about an outcome, the more they'll believe stuff that fits their predispositions, even if it's obviously bogus.  Biased processing run amok.

The study includes a big fat multiple regression, which makes my data-crunching heart happy, but if you're not a number nerd let me translate.  The author statistically controls for a number of factors you'd expect to lead to belief in the myth and the combo of watching Fox News and education pops out quite nicely. 

Skip to the next graph if you're not into methodological quibbles: I'd argue he fails to account for political interest, but you could counter-argue that's captured by his "follow health reform news" variable, and then I'd counter-counter-argue that no, it doesn't -- that's a global media variable, not a motivational one, and if you included political interest, education might very well disappear.  Yes, methodologists argue about stuff like this, which explains why we rarely get invited to parties.

If you dig into the table, you'll see other media variables basically play no role in this screwy belief.  Only reading newspapers is statistically significant, and it's in the opposite direction of watching Fox News.  In other words, newspaper reading reduced your likelihood to believe in the death panel myth, and it did so regardless of education level.  Whew.  Newspapers still rule.

Skepticism is Not Cynicism

There's a very interesting study in the latest Journalism and Mass Communication Quarterly (and how often do you get to write those words?) that examines the differences between cynicism and skepticism, and how each plays a role the public's apathy and feelings that the government will respond to their needs (efficacy).

Simply put, skepticism good | cynicism bad.

So being skeptical leads to feeling government is more responsive, cynicism to feeling it's less responsive.  Being skeptical also leads to less apathy.  In other words, the cynics have tuned out, the skeptics are, well, skeptical, but in a good way -- that is, they're looking for more information and feel change can come.  Skeptics help fuel a democracy, cynics are a drag on it.

Okay, but what I want to get into here is the methodology.  Let's look at the two key variables.

Cynicism was measured by such questions as whether respondents agreed or disagreed with such statements as (1) candidates are interested only in votes, not people's opinions, and (2) government is run by a few big interests.  That sort of thing, all fairly generic.  No questions about media.

Skepticism was measured by such statements as (1) I think about news stories before I accept them and (2) I seek out additional information to confirm statements by politicians.  A mix of statements about media and politicians.

See the difference above?  One set of questions, measuring cynicism, doesn't mention the media at all.  The other set, measuring skepticism, has at least three of the six touch directly on the media and one indirectly does so.  Is this a problem?  Cynicism in this study is conceptualized as a lack of confidence in, or distrust of, the political system.  That's pretty global.  Here's a key quote from the study (p. 26, italics mine): "An alternative disposition to cynicism is skepticism, which also assesses individuals' critical evaluation of public affairs information sources, including elected officials and news media, but which serves a beneficial purpose rather than a debilitating one."  Somehow here is the suggestion that cynicism is also about information sources, but if you read the questions above used to measure cynicism, you see it ain't so, at least not as compared to the measures used to tap skepticism.  Why am I making a big deal about this?  In part because apparently the JMCQ reviewers didn't.  You can't treat them as so equivalent when the measures are so very different in the target.  One is global, the other is part global and about the system, part about the media.

Another useful line to examine in that quote above is that one of the concepts (skepticism) is thought to serve "a beneficial purpose" while the other (cynicism) a "debilitating one."  We generally see cynicism as bad, skepticism as good, with the latter suggesting further cognitive effort to resolve a problem.  The questions do get at this but let's be clear, only one set involves the media.  Not both.  Despite this, we have this line from the discussion section:  "Cynicism and skepticism both represent a negative posture toward public affairs and relevant media."  I believe you can say this about one but not the other, not based on the items used above to measure them.

Yeah, I'm nitpicking above.  It's a good study and includes an interesting path analysis to include satisfaction with the media.

Four Kinds of Knowledge?

The latest issue of Journalism and Mass Communication Quarterly includes a meta-analysis of the Knowledge Gap Hypothesis, which is a concept I've always liked, particularly when compared to the crud that is called agenda-setting research.  Rather than dig into the study findings, I want to focus deep in the article and its measurement section.  Relying heavily on a previous study, the author argues for four overlapping types of knowledge.  They are:

• Factual Knowledge -- which is exactly what it sounds like.
• Awareness Knowledge -- which is a softer version of above, though the article isn't exactly clear.
• Belief Knowledge -- Even more confusing, but apparently having to do methodologically with how many responses someone gives, and theoretically with how much information someone holds, regardless of its accuracy.
• And a combination category, for no good reason.

Most studies use the first two.  Factual is easy to measure.  Ask respondents to name the Vice President or Speaker of the House.  Awareness is more like asking people what they think they know or, a bit better, asking them how aware they are of something. In general, it's a sucky measure.  Belief has been used in political science for years -- often badly -- by merely counting how many likes and dislikes someone can generate about a political actor or party.

What's interesting, from the study, is the notion that belief is less correlated with education than the other kinds of knowledge.  This is viewed as a positive.  I'd argue it's evidence that how much information we hold, regardless of its accuracy, makes that conceptually distinct from actual knowledge -- the holding of correct information.  As a respondent or crazy uncle I may rattle off lots of information, none of it accurate, probably full of conspiracy theories about Barack Obama's religion or being the anti-Christ.  But is that knowledge?  Or a sort, certainly, and belief knowledge is probably as good a term as any, but I wouldn't treat that as political knowledge in the strictest sense.

And yeah, that image I put in there makes me want to have a seizure as well.

Sometimes you read a journal article and think, "Dammit, I should've done this study."  Such is the case in the latest issue of Journalism and Mass Communication Quarterly, fresh in my mailbox today.  Even worse, the first two two authors are doctoral students.  Humbling.

What's it about?  To simplify, the dependent variables are political knowledge and political participation.  No big deal there.  These concepts have launched a million scholarly studies.  But what they did is so obvious, and the results so interesting, that it kinda ruins my day.  The study examines the difference between exposure to citizen journalism and professional journalism -- again, no biggie there -- but also trust in those media.  Best of all is a two-way interaction I'll discuss in a minute.

The basic result is this: after statistically controlling for other factors, exposure to traditional news was positively associated with political knowledge while citizen journalism exposure was negatively associated with knowledge. Knowledge in this case was an index of four standard political actor recognition items.  The results for offline and online participation are a little messier, so let's set them aside and focus just on knowledge. 

Here's the interesting bit.  While the exposure items line up pretty much like I'd expect them to (pro news positive, citizen news negative, in their association with knowledge), the authors also included trust in pro and amateur news.  The citizen news media trust measure was unrelated with knowledge, but trust in professional news media was negatively associated with knowledge. 

Let me say this again.  Exposure to pro news is a positive predictor of political knowledge, but trust in such media is a negative predictor.  This isn't terribly surprising, given that those who have the most to do with the news media tend to trust it less, largely because they are stakeholders or have been misquoted or know a lot about stories that journalists are forced to simplify, thus losing some of the nuances that come with any issue.

To clear this up, the authors examine a two-way interaction between the trust and exposure measures to find that among those high in trust of professional news media, exposure has no real effect on knowledge.  But, among those low in trust, as exposure increases, so does knowledge. In other words, exposure to news increases knowledge, but only for those who don't trust the news.  I'm not sure the discussion quite clears this up, and to be frank I need to think on it more myself before reaching some kind of conclusion as to what it all means.  But it is fascinating stuff.

That's Hollander, Barry A.

So I'm reviewing a manuscript for an academic journal slightly outside my field but on a topic I'm more than familiar with.  Otherwise I wouldn't be reviewing it, right? 

As I read the literature review and theory section it comes to me.  Hey, I published something on this only a couple of years ago, and in a pretty damn good journal, yet I don't seem to be cited.  And get this, the author brags that this study the first to examine this subject.  Which is fine -- if you happen to exclude my article.

Okay, how do I handle this?  Vaguely suggest the author look further in the literature?  Suggest specifically he or she read the journal in which I published?  Or come right out and say, Yo Skippy, try this: Hollander, Barry A. (2008). Tuning out or tuning elsewhere? Partisanship, polarization, and media migration from 1998 to 2006. Journalism and Mass Communication Quarterly, 85, 23-40.