Do you chatterbox while you watch TV?

A new term has become popular - chatterboxing, defined  here as watching a TV programme while talking to others about it online.  A lot of people are doing it, mostly via Twitter

Many TV programmes encourage you to tweet about them by displaying their hashtag at the start.  Some presenters draw your attention to it, while the latest trend is to display the hashtag discretely onscreen throughout the programme.  The aim is to attract and retain viewers by allowing them to share their experience of watching the show.  As yet there don't seem to have been many public studies of doing this.  In this post we're giving some details of a very recent analysis of one use of hashtags by a TV channel.

The hashtag related to the coverage of the World Indoor Athletics championships by the UK's Channel 4 during the period 9-11 March 2012.  This is a significant event in the athletics calendar and it was the first time that Channel 4 had broadcast it.   The channel promoted the hashtag #c4athletics.  Tweets containing the text c4athletics were collected and analysed for a four day period covering the championships themselves and the day prior to it.   The collected tweets contained both those from and mentioning the programme's Twitter account, C4Athletics, and also the hashtag #c4athletics.


Overall the numbers were not large.  The totals of tweets and of people tweeting on each day was as follows:

Date	Tweets	People tweeting
8-Mar	74	51
9-Mar	887	613
10-Mar	441	195
11-Mar	715	395

The total audience figures are not available to us, but will undoubtedly have been at least several hundred thousand.  So we see immediately that quite a small proportion of viewers actually tweeted.

We  also looked at the number of days each person tweeted; chatterboxing for these viewers seems to be a one-off activity, perhaps a reaction to something on screen that they respond strongly to. 

Number of days tweeted	1	2	3	4
Tweeters	955	108	24	3

In more detail the following table gives the number of people who posted various numbers of tweets on each of the four days.

Number of tweets	>=4	3	2	1
8-Mar	2	1	2	46
9-Mar	18	14	51	530
10-Mar	14	8	32	141
11-Mar	23	33	69	268

Again the picture that emerges is of people only tweeting fairly rarely, or at least only using this hashtag rarely. Of those who tweeted most frequently several were members of the production team or the presenters.  Others were tweeting very frequently about the athletics but using several  hashtags including #c4athletics.  But the vast majority posted only once or at most twice during the day.  They felt a need to make a one-off comment but not to keep up a stream of posts nor to get involved in a lengthy conversation.  A very few people posted on the day before the event, reflecting some knowledge of what was coming, but it was not until the event was underway that most people thought of posting.


To get an idea of what happens during an event, in the following charts we have also plotted the number of tweets per hour containing the text c4athletics.  The columns in brown are for times when the channel was broadcasting the athletics, those in blue for other times.  

 

These charts give a picture of how viewers used Twitter for this one event.  No doubt other events would give a different pattern, but there are several point of interest.  As we might expect, most activity was during the transmission.  There were also significant numbers of postings at other times especially in the lead up to transmission.  On the 10 March in the morning the athletics was broadcast from 7 am to 10.30 am, but from 8-9 it was interrupted by a programme on another sport, regularly broad cast in that slot.  To maintain the other sport in its regular slot may well have ensured the satisfaction of its viewers, but, judged by the numbers of tweets, it lowered interest in the athletics, an effect that continued throughout the day.    

What conclusions can we draw from these numbers?

Although the numbers of tweets using the programme's hashtag was comparatively  small, their influence was potentially much wider, as that hashtag will have been propagated to the followers of those tweeting.  More details on this in a later post.

It is arguable that those numbers would have been larger if the hashtag had been actively promoted before the programme in all pre-programme publicity such as trails.  The downside of doing that might have been to discourage those who do not tweet from watching the programmes.

The pattern of tweets, such as that for 10 March above, gives information about viewer response to programme scheduling.



An analysis of the tweet contents will be produced later since they contain much of interest such as  reactions to individual presenters and the programme content.  Automatic analysis doesn't necessarily give a good picture.  In practice such analysis would best be done in conjunction with the analysis of audience data since those tweeting are only a small (perhaps atypical) part of the audience.