Opinion: Social media wrap: Glenn Beck, Jon Stewart, Rachel Maddow and the battle of TV’s talking heads

This article was originally on a blog post platform and may be missing photos, graphics or links. See About archive blog posts.

In social media, it rarely matters whether you’re loved or hated. In the old theatrical adage, if you’re entertaining, it means “bums on seats.”

The Ticket gleaned this from a fascinating new social-media study that examined the leading eight “talking heads” on TV and their effect on users of social-media networks including Facebook, Twitter, YouTube and Digg.

Perhaps unsurprisingly in light of his recent rally on the steps of the Lincoln Monument, Fox News host Glenn Beck is by far the most liked -- and disliked -- of TV’s talking heads on social-media networks, ranking ahead of Jon Stewart (‘The Daily Show’), Bill O’Reilly (Fox News) and Rachel Maddow (MSNBC) by a large margin.

Beck consistently generated the highest volume of “chatter” across social networks, or a plurality at more than 50% of user comments on public forums. Stewart placed second in both volume of ...

... chatter and like/dislike votes at about 16/10%. Wolf Blitzer of CNN, meanwhile, fared last at about 1% of either.


Other talking heads featured in the study compiled by Alterian, a social-media marketing firm, were Keith Olbermann (MSNBC), Anderson Cooper (CNN), and Stephen Colbert (Comedy Central).

(Chart credit: Alterian)

(Chart credit: Alterian)

O’Reilly had the biggest swing from like-dislike, registering in fifth place for volume of comments about his work and seventh on the “like” scale, but jumping to third place on the “dislike” scale.

In terms of chatter generated by TV networks, Fox News (largely driven by Beck’s show), leads the pack, with MSNBC placing second followed by Comedy Central (largely driven by Stewart and Colbert) and CNN finishing fourth.

Before we get to the complicated methodology behind the statistical analysis, which was compiled in the three-month period ending with Beck’s rally and examined social-media comments from U.S.-based English-speaking users, we’d like to offer our own (relatively) unscientific conclusions from the data:

+ Love him or hate him, as social-media users clearly do, Beck is by far the most polarizing -- but more importantly entertaining -- of all the talking heads mentioned above. The more outlandish Beck’s theories, the more chatter and traffic he seems to generate. His D.C. rally is a case in point. On the day of his rally, he swamped his combined competition in traffic and chatter by more than a factor of ten (as the Ticket also noted, Beck and his rally partner Sarah Palin also lit up the newly launched Google RealTime search engine on that day).

+ We base the conclusion above on the “entertainment” factor of CNN anchors Cooper and Blitzer. CNN famously a few years back purged outlandish political demagogues from many of its shows, preferring to concentrate their programs on straight -- or at least semi-straight -- news gathering and presentation. No matter what you choose to call Cooper or Blitzer, neither has the “wow-did-he-really-say-that?” factor that Beck possesses in spades. Stewart, also an entertainer, consistently ranked highly also.

+ We asked Alterian to run its analysis with and without the “spike” effect of Beck’s rally. This resulted in a slight drop in Beck’s average volume of chatter, but not a significant swing either way for his like/dislike ratio.

+ The study didn’t take into account social-media chatter in relation to TV viewership. Beck and O’Reilly at Fox News are far ahead in viewership figures than both Cooper and Blitzer, for example, which undoubtedly affects the amount of chatter they generate on social networks. The study’s authors have pledged to examine this in more detail.

+ Interesting to note is the positive/negative aspects of the “like/dislike” factor. For one, it’s not broken down into “strongly like,” “like,” or “strongly dislike,” as users did not fill out a survey (this is key). The research is based on people’s public social-media posts, not their solicited answers. It’s a straight like or dislike choice that is automatically generated from a keyword search of users’ postings. Ah, we hear you say, what if a teenager in the modern parlance says something is “sick” -- is that a like or dislike? Which leads us into the methodology bit.

(Chart credit: Alterian)

[The number of searches here is skewed by Beck’s rally in D.C. We asked Alterian to rerun the results discounting the effect of Beck’s rally.]

Michael Fisher, a senior vice president at Alterian, says the data collected by his researchers on TV’s talking heads indicated a “large volume of conversation,” a “weekly pattern of conversation” and conversation “primarily driven by Twitter and social networks.’

As the data are harvested by a variety of computer programs from public comments on social networking sites, Fisher says the data are not therefore affected by concerns over how a particular survey question has been posed. “It’s unsolicited feedback that describes consumer sentiment and commentator popularity; it absolutely shares and describes the nature of those conversations. No one has been prompted, no one has said, ‘Fill out this survey.’”

Therefore, Fisher says, the data have no statistical margin for error (although it does not take into account multiple postings on the same subject from the same user or group of users, or the possibility of computer-generated “spamming” of notice boards, or the political leanings of the Facebook/Twitter audience in general, or the average age of respondents).

Alterian has compiled similar social-media trending studies for the Super Bowl, soccer World Cup and the UK’s general election earlier this year and researches the effects of social-media usage in areas such as company branding and consumer loyalty.

In terms of user comments examined, Fisher says, “Anything private is out of bounds. If it is publicly available, we’re happy to share.” He says resources include Twitter, Facebook, newspaper and blog notice boards, Digg, Reddit, Delicious, YouTube and Flickr.

He says the data are broken down by a series of algorithms and language-processing tools by keywords, which are cross-referenced against a customizable dictionary of terms or topics.

He says the analysts “sometimes look at conversation directly. If there is commentary that doesn’t quite make sense or isn’t abundantly clear, you need to manually go in there and take a look. Language and meaning evolve. We consistently refine our dictionary.”

On the “Beck effect,” Fisher says: “Results were incredibly favorable for him. His commentary is polarizing, resulting in a lot of positive and negative sentiment and more communication volume across different channels. Absolutely the one most people like or hate. He’s equalized.”

On social media’s influence and effects in general, Fisher says: “Because consumers are involved in these social networks with groups of people, one person can have a significant influence in forming opinion. It exposes the power of these networks and the power of social communities.”

-- Craig Howie

You don’t need a talking head to click here for Twitter alerts of each new Ticket item. Or follow us @latimestot or @craighowieLAT. Our Facebook Like page is over here.Also available on Kindle now with a two-week free trial.

Photo credits: Associated Press; Reuters; EPA; CNN; Fox News; MSNBC