Adventures with CNBC anchors’ statistics

By Felix Salmon
May 20, 2011
plugging his book, he has no time for baseless factoids. Oh, who am I kidding:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

CNBC’s Joe Kernen reports the news in the morning in a fast-paced environment where it’s difficult to be 100% accurate. If you write a book, by contrast, you have the time to make sure you get things right. So it’s good to see that now Kernen is plugging his book, he has no time for baseless factoids. Oh, who am I kidding:

Q: Give us some examples of how you see business people portrayed in the media.

A: There was a study done of TV which looked at hundreds of hours of prime time. You were four times more likely to commit a crime if you were a CEO than if you were a drug dealer or a gang leader.

Does he mean this study, by Ray Surette, author of Media, Crime, and Criminal Justice: Images, Realities and Policies? Here’s what it found:

table7.gif

Hm, no, he can’t mean that study, because it doesn’t show anything like what Kernen is talking about. Businessmen only accounted for 24% of crimes on TV even when you include crimes which were ordered by businessmen but carried out by professional-criminal flunkies. Meanwhile, it stands to reason that 100% of drug dealers on TV commit a crime — since by definition they’re being portrayed as illegally dealing drugs.

So maybe Kernen’s not quoting a study at all, and instead is quoting a half-remembered 1979 polemic by — wait for it — Ben Stein. Cecil Greek summarizes:

Ben Stein wondered why prime time crime consisted almost exclusively of white upper-class violence. After watching thousands of hours of TV crime shows, he reported never to have seen a major crime committed by a poor, teenage, black, Mexican, or Puerto Rican youth, even though in reality they account for a high percentage of violent crime.

Most likely, Kernen’s just quoting something somebody told him at dinner one evening, and which sounded too good to fact-check. After all, that’s the way that CNBC works. You just put a lot of shouty people on the TV saying things of dubious coherence, sit back, and enjoy the cacophony.

Kernen works in a world where ratings are more important than accuracy, and he’s just carrying that philosophy over to the book-publishing business. I doubt that his publishers, the “dedicated conservative imprint” Sentinel, mind in the slightest.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
5 comments so far

It’s not viewers who want cacophony. It’s the advertisers who want the cacophony. Because the cacophony leaves the viewers in a state of brainwashed delirium, a desirably impressionable state for when the commercials run. The commercial programming is really the point of television. Television “programs” are really just filler for the commercials.

Posted by EricVincent | Report as abusive

Apart from your point that he’s trivially incorrect (yes, drug dealers must be committing crimes unless they’re in some weird jurisdiction), your statistics don’t disprove him.

Those aren’t *conditional* probabilities, about which Joe was speaking – and at least on the shows I watch, there are far more drug dealers than CEOs onscreen. It very well may be that the CEO mostly shows up to commit a murder (let’s face it, they’re boring otherwise), while less than 25% of drug dealers kill people. (Also, have you seen The Wire? 25% is probably way too high.)

“Gang leader” really confuses me though, because I can’t think of a single depiction of a gang on TV where the gang abstains from murder. It’s reasonable to say that the leader is responsible in those cases, so that would make it 100%.

Posted by absinthe | Report as abusive

I see that the long-understood link between absinthe and decrease in reasoning ability is correct.
Disregarding that trivial point that Absinthe appears to be discussing a column other than the one you wrote, I find myself a little unclear about the question of the study you quote Kernen as referring to. It would seem clear that the study you cite is presumably not the study Kernen does, as the cited study shows no categories for CEOs, drug dealers or gang leaders, and the listed percentages (businessmen vs gangsters) are not 4:1, but rather about 2:3.
The Ben Stein blather is not really worth mentioning, as are most of his statements. The question is left hanging there, however – were these the only two examples you found? Were there no other studies besides that one which might have been the source for Mr. Kernen’s assertion? Remember – inquiring minds want to know!
Actually, I lean toward your explanation: for most of us, no story is so good that it can’t be improved, and if a number isn’t strong enough, doubling or tripling it to make your point is morally acceptable. More than quadrupling it might be pushing the point a little, but Kernen is, after all, a TV personality (and on CNBC at that). Only a fool expects such a person to have any more than the most tenuous relationship with accuracy.

Posted by mrtoads | Report as abusive

…but Kernen is, after all, a TV personality (and on CNBC at that). Only a fool expects such a person to have any more than the most tenuous relationship with accuracy.”

As a former stockbroker and small businessman, now retired, I watch CNBC regularly. Kernan, like Kudlow, has a narrative he uses to explain all that he sees.

The relationship between Kernan’s POV with what is actually going on out in the world of finance and investment is indeed tenuous.

Posted by Beezer | Report as abusive

“Related to the disproportionate emphasis on the most serious end of the crime spectrum is the portrayal of the demographic characteristics of offenders and victims presented by crime fiction. Offenders in fiction are primarily higher-status, white,
middle-aged males (Pandiani 1978: 442–7; Garofalo 1981: 326; Lichter et al. 1994: 290–5; Reiner et al. 2000a and 2000b). … Apart from gender, the demographic profile of offenders and victims in fiction is the polar opposite of criminal justice statistics (Surette 1998: 47 calls this ‘the law of opposites’). (See also Pandiani ibid.; Garofalo ibid.; Lichter et al. ibid.; Barclay and Tavares 1999: chapters 2 and 3. Sparks 1992: 140–45 offers a qualitative analysis.)”

Posted by Nameless | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/