Pundit Impact

[Pangloss]

I referenced this earlier in a thread but I don't think I ever started a full topic on it. Things are a bit slow so what the heck. This is from a Politico/George Washington University poll back in September:

 

The question: "I am going to read you a list of some of these people. For each one, please tell me if you think this person has a positive impact or a negative impact on political debate in this country. If you do not recognize the name, just say so. Here is the first one …"

 

The results:

 

Positive Impact:

Bill O’Reilly 49%

Glenn Beck 38%

Rush Limbaugh 36%

Sean Hannity 35%

Jon Stewart 34%

Keith Olbermann 23%

RachelMaddow 18%

Ed Schultz 11%

 

Negative Impact:

Rush Limbaugh 52%

Bill O’Reilly 32%

Glenn Beck 32%

Sean Hannity 25%

Keith Olbermann 25%

Jon Stewart 22%

RachelMaddow 18%

Ed Schultz 11%

 

Never Heard Of:

Rush Limbaugh 5%

Bill O’Reilly 12%

Glenn Beck 23%

Sean Hannity 34%

Jon Stewart 34%

Keith Olbermann 42%

RachelMaddow 55%

Ed Schultz 70%

 

Some surprises and non-surprises here for sure. No real surprise that the Fox News Channel analysts score well in positive impact, but the negative impact is interesting -- Rush Limbaugh really leaps to the fore. And it's no real surprise that everyone knows who Rush Limbaugh is, but I would have guessed that Jon Stewart would have scored higher than Bill O'Reilly. Go figure.

 

To some extent I think you have to write off the high positive impact of the FNC pundits because of popularity (who's gonna vote against the guy they watch?). But the high negative on Limbaugh and the "never heard of" data suggests to me that people tend to give pundits the benefit of the doubt if they've never heard of them. For that reason we also have to write off the low negative impact scores for the MSNBC crowd (Olbermann, Maddow and Schultz).

 

What do you all think?

 

Full results can be found here.

Story on the poll here.

[D H]

Not a very representative poll, is it? 61% of the respondents were 45 or older, and 77% were white. Given the demographics, the low numbers for Jon Stewart are not all that surprising, but the huge negative impact numbers for Limbaugh are "wow". He is indeed a train wreck.

[Pangloss]

Nice catch, I missed that.

 

"Train wreck" frequently comes to mind regarding Limbaugh. I'm sure he sees it differently, of course!

[mississippichem]

Everyone knows Rush Limbaugh is the political equivalent of Jerry Springer. I think Stephen Colbert should have designed his stage persona more in the style of Limbaugh rather than O'Reilly. It would've had more lasting funny power.

 

Surprising that O'Reilly beat Hannity in that demographic though. It must be an age related phenomenon. Old people identify with an old pundit more I guess.

 

I'm surprised Olberman can even stay on TV. I have no vendetta against him politically, but his show is so boring! Maybe we need more boring shows though. Boring seems to roughly equate objective, not that I would dare expect a pundit to be objective...

[Pangloss]

Interesting points.

[Mr Skeptic]

Pundit Impact Poll shows they have it

 

Well, polls don't show they have it, they show people in the poll demographic think they have it. I'd be rather interested in a poll that compared both whether people agree with the pundits and whether they think they have a positive or negative impact. I suspect that would agree very closely, especially for the choice of positive impact. I'll go on a limb and guess that there would be more disparity in "disagree with" and "negative impact" in one political alignment than the other.

 

And now, slight modification of the data:

 

Positive Impact/Negative Impact:

Jon Stewart 154%

Bill O'Reilly 153%

Sean Hannity 140%

Glenn Beck 119%

RachelMaddow 100%

Ed Schultz 100%

Keith Olbermann 92%

Rush Limbaugh 69%

 

Positive impact/(Positive impact + Negative impact):

Jon Stewart 34% -> 61%

Bill O'Reilly 49% -> 60%

Sean Hannity 35% -> 58%

Glenn Beck 38% -> 54%

RachelMaddow 18% -> 50%

Ed Schultz 11% -> 50%

Keith Olbermann 23% -> 48%

Rush Limbaugh 36% -> 41%

 

Negative Impact/(Positive impact + Negative impact):

Rush Limbaugh 52% -> 59%

Keith Olbermann 25% -> 52%

RachelMaddow 18% -> 50%

Ed Schultz 11% -> 50%

Glenn Beck 32% -> 46%

Sean Hannity 25% -> 42%

Bill O'Reilly 32% -> 40%

Jon Stewart 22% -> 39%

 

Note that it seems that not everyone voted in one of the three categories; there's percentage points missing. For these last two categories, I divided as in the category heading, disregarding the "never heard of" data.

 

I would have guessed that Jon Stewart would have scored higher than Bill O'Reilly.

 

You're right, Stewart is barely at the top in positive/negative.

 

To some extent I think you have to write off the high positive impact of the FNC pundits because of popularity (who's gonna vote against the guy they watch?). But the high negative on Limbaugh and the "never heard of" data suggests to me that people tend to give pundits the benefit of the doubt if they've never heard of them. For that reason we also have to write off the low negative impact scores for the MSNBC crowd (Olbermann, Maddow and Schultz).

 

I think the reasons for such a bias would be different: a mostly unknown person is probably likelier to be known by his fans. Certainly, 100% of the people who watch their show have heard of them, but watching their show correlates with agreement or other positive ratings (if only importance of what they say).

 

Not a very representative poll, is it? 61% of the respondents were 45 or older, and 77% were white. Given the demographics, the low numbers for Jon Stewart are not all that surprising, but the huge negative impact numbers for Limbaugh are "wow". He is indeed a train wreck.

 

Good catch.

[Pangloss]

Interesting analysis. I like the ideas there for determining an overall score. I'm not sure how well that accounts for the problem of observer bias by voters, but it does seem to reduce that impact (was that part of the point?).

 

Regarding this:

 

Well, polls don't show they have it, they show people in the poll demographic think they have it.

 

Doesn't what they think the impact is translate pretty well into what the impact really is?

[Mr Skeptic]

Interesting analysis. I like the ideas there for determining an overall score. I'm not sure how well that accounts for the problem of observer bias by voters, but it does seem to reduce that impact (was that part of the point?).

 

My modification was just to reduce the effects of pundits being poorly known.

 

Doesn't what they think the impact is translate pretty well into what the impact really is?

 

Oh, good point. Just like what random people think are the answers to physics problems correlate well to what the actual answers are, or how people's estimate of how likely people are to die a certain way correlates to the actual percentages (eg "What is likelier to kill someone, a peanut or a terrorist?"). Sure, there's a correlation, but is asking random people really the best way to get reliable answers?

[Pangloss]

My modification was just to reduce the effects of pundits being poorly known.

 

Right, but it does that at the cost of measuring exposure, which is really the point of the survey. By normalizing those variables we see that his positives-to-negatives ratio is more impressive than we see with other pundits, which is really valuable to know (IMO). But that doesn't mean that he has greater overall impact. For that we have to factor in exposure.

 

 

Doesn't what they think the impact is translate pretty well into what the impact really is?

Oh, good point. Just like what random people think are the answers to physics problems correlate well to what the actual answers are, or how people's estimate of how likely people are to die a certain way correlates to the actual percentages (eg "What is likelier to kill someone, a peanut or a terrorist?"). Sure, there's a correlation, but is asking random people really the best way to get reliable answers?

 

The poll is about the impact of pundits on politics. Politics in a democracy is all about the manipulation of public opinion. Comparing that with accurate answers in physics is incorrect -- there are no objective absolutes in public policy discourse. There are only operationalized social-science variables based generally on polling data. Take it or leave it.

 

But either way, please leave the ridicule at the door. I didn't attack you, and I don't deserve a sarcastic response.

[Sisyphus]

The poll is about the impact of pundits on politics. Politics in a democracy is all about the manipulation of public opinion.

 

Does the poll demonstrate successful manipulation of public opinion? I'd say at most, it shows manipulation of public opinion about pundits. They've managed to convince people that they have an impact.

 

Or maybe not! The only options were positive impact, negative impact, and never heard of 'em. I think the answer is more likely to reflect "do you agree with this person," which is not the same thing. I can agree with you and think you have no impact at all (not a choice!), because you don't change anyone's mind.

[Pangloss]

Well I agree that more research would be needed to really see the exact impact here. If this were a peer-reviewed journal we'd be casting a pretty jaundiced eye at the study's limitations and suggestions for future research.

 

(Reminds me of a paper I read the other day on whether outsourcing adds value to corporations. The authors, three scholars from a prominent business school writing for a major journal, stated that they had shown that outsourcing adds value. Their evidence? A simple analysis of immediate impact on a company's stock value when an announcement about outsourcing is made. Not even a perfunctory nod at the obvious question of long-term impact. Talk about limitations!)

[imatfaal]

For a different perspective, ie from the eastern side of the atlantic, John Stewart is well known (the daily show is broadcast nationwide free-to-view), Rush Limbaugh is pretty well known in a "have you heard of this american guy who says..." sort of way, and Glenn Beck was the subject of a few biogs in the serious news media during the run up to the mid-term elections. Personally, I find the influence of the regular TV/Radio political commentator within the United States quite fascinating; I am sure I will be corrected within a few minutes, but I cannot think of a good UK analogue in mainstream broadcast media. I am not sure if this is a good thing or a bad thing.

[Pangloss]

Jeremy Clarkson!

 

The Daily Show free to view in the States as well, by the way. Part of what they call "basic cable", and also available the next day on the Web.

[imatfaal]

Jeremy Clarkson!

 

 

Unfortunately he is as close as we can get - even if he is paid to talk about cars.

 

 

[Pangloss]

By the way, Jeremy Clarkson is the ONLY British commentator I'm aware of who regularly poo-poos global warming and nanny-state tendencies. (I'm not saying he is the only one, he's just the only one I'm aware of, in terms of the point about overseas perspective.)

 

Interestingly, his show is probably seen by more people on a weekly basis than every American political commentator combined, possibly by an order of magnitude (according to a recent story on 60 Minutes, Top Gear has 350 million weekly viewers!). Of course, they don't tune in for his political commentary, but I think in a funny sort of way it's part of the appeal -- the regular derision of road regulations and toll fees, anyway. But that kind of "soft politics" does have an impact on overall public ideological opinions, IMO.