Monthly Archives: September 2013

Introduction to Cable News Twitter Ratings

Posted on by

All cable news shows (with a couple of odd-ball exceptions) engage with their audience on Twitter, either through official Twitter accounts, official hashtags, or the Twitter account of their hosts.  The Cable News Twitter Ratings measure the number of tweets that mention a show.  Because many shows are repeated in the overnight hours, there are two metrics. The first, “Tweets During”, counts the number of Tweets during the the particular showing that mention each show. The second, “Tweets Day”, measure the number of tweets that mention show during the calendar day (eastern time).

While Nielsen et al have the luxury accounting for the local time zone of each viewer, Twitter does not provide timezone information for each tweet and so we have no way of knowing the time zone for the sender.  When a show repeats, I sometimes list the repeat showing.  I will work, over time, to try to better handle these issues.  For now, though, the total number of tweets during a day is probably a better metric of a show’s popularity.

In some cases, however, it’s hard to distinguish between tweets that are in response to a show and those in response to some non-show activity of the host(s).  Sean Hannity has more hours on the radio than on TV, and so his daily mention counts are undoubtedly higher than they would be if he just had a TV show.  There is no decent way to account for this, unfortunately.

Here’s an example of a set of ratings:

Ratings for Weekdays 8pm

Network Show Tweets Hour Tweets Day % Male % Female
CNN Anderson Cooper 360 425 1745 48% 52%
Fox News The O’Reilly Factor 20 263 60% 40%
HLN Nancy Grace 115 416 33% 67%
MSNBC All In w/Chris Hayes 1109 2938 49% 51%

For each show, I give time of the showing (in eastern time), the network, the name of the show, and the above mentioned Tweets During and Tweets for the Day.  I also estimate the percentage of senders who are male or female.  Be aware, though, that these estimates are going to be very imprecise.  For a full explanation, see this page.

For each show, I filter the tweets based upon my best observation of what people use to talk to and about the show on Twitter.  When a show has one or two personalities who are strongly identified with it (almost exclusively), I include tweets that mention the hosts. Otherwise, I use the accounts and hashtags for the show.

For example, I use the following filter terms for MSNBC’s Morning Joe:

@Morning_Joe OR @morningmika OR @JoeNBC OR #morningjoe

This includes the accounts for Joe and Mika, since they are so closely associated with the show.  On the other hand, for MSNBC’s The Cycle I used:

#TheCycle OR @thecyclemsnbc

Because the show has four hosts with active Twitter lives of their own, I track the show by itself.  But since it’s interesting to see the aggregate counts including the host’s accounts, you’ll see a separate line item for counts including the hosts.  For The Cycle, then, the query that includes the hosts is:

#TheCycle OR @thecyclemsnbc OR @toure OR @krystalball OR @AriMelber OR @HuntsmanAbby

As you look through the ratings, you may see that the counts of mentions varies much more dramatically than the Nielsen ratings do; this is the influence of the shows themselves and how hard they try to engage on Twitter.