In Choice and Trust: Amplifying an Event with Twitter I argued that it was not enough to simply promote and event hash tag and then expect the general topical tweeting to provide a clear representation of the actual content of your event to any remote participants. You need to provide a dedicated live blogger who can provide an official, detailed commentary of the event – thus adding context to the hash tagged tweets and choice for your remote participants.
To support this argument, I have been analysing the tweets generated by the audience at my last event: the Institutional Web Management Workshop 2010 (IWMW10). My aim was to determine the proportion of “natural” tweeting activity that focused on reporting the content of the event, compared to the more social uses of Twitter – including making comments about the event/content and networking with other delegates. A low level of reportage would support my assertion that a live blogger is necessary to engage with a remote audience, whilst a high level would suggest that event delegates are capable of supplying this need as a community without assistance.
The Evidence
I analysed a total of 2547 tweets tagged with the #iwmw10 event hash tag, which were recorded and archived by the brilliant Twapperkeeper service. These dated from 04:21 on Monday 12th July 2010 (the first day of the event) to 23:35 on Wednesday 14th July 2010 (the final day of the event). I should note that I omitted tweets from @iwmwlive and @iwmw from the data, as these were the official event accounts.
Only 17% of tweets from IWMW10 participants could be classified as containing original reported content, whether in the form of direct quotes, paraphrased summaries or a reported statement with an appended comment. Many included the same popular key points, but were issued as original (i.e. not re-tweeted) updates by the individuals involved.
Nearly 47% of tweets were purely personal comments on the content. This included opinion remarks, tweets containing links to related resources and conversations with other delegates about the issues raised.
Nearly 22% of tweets were purely social in nature, occurring mostly either pre- or post-event, or during the evenings when there were social activities organised for the delegates.
If we look at the reinforced effect of the tweets made, we find that in total, 14% of tweets from the event were re-tweets. Of these, 53% were repetitions of comments, whilst only 27% were re-tweets of delegate reporting (15% re-tweeted the live blogger). Re-tweets of reports do not necessarily give a fuller picture, as there may be multiple repetitions of the same quote. For example, the comment “The days of 2 year IT development projects are over says @cloggingchris at #iwmw10” from @briankelly attracted 4 re-tweets. However, reinforced messages can help the remote audience by replacing some of the emphasis used to identify key points in the live presentation.
During the two parallel sessions, there was no official Twitter commentary, but there was little shift in favour of reporting evident in the delegate tweets. There was less tweeting taking place during these sessions (each lasting 1 hour 30 mins, in which delegates were split into smaller groups for focused sessions). However, the amount of reporting was essentially the same, with 18% of tweets providing reported content (15% attaching a comment to the report). The number of comments and social tweets during this period shifted slightly to 30% comment, 37% social, but this may have been due to these sessions occurring towards the end of each day’s programme.
Again, 14% of tweets reinforced others. However, of this 70% were re-tweeting updates that included reported material and only 30% were re-tweeting pure comments, so reporting did take on a slightly stronger emphasis in terms of what was deemed important enough to re-tweet. This may be due to the disparate nature of the sessions, with fewer people appreciating the context of any comments.
These results show that the same group of delegates did not significantly change their tweeting habits to reflect the presence or absence of the official Twitter commentary provided by the event blogger.
Observations
There were two games played during the event which may have affected the balance of the results slightly. The first was a QR code game, which generated automatic tweets when a team answered a clue from a QR code correctly. I classified these tweets as “social”. The second was the Twitter Buzzword Bingo game created by Richard Pitkin, which involved delegates tweeting about Brian Kelly’s closing session in an attempt to guess one of the 16 buzzwords associated with his talk. I classified these tweets as “comments” as they were technically commenting on the content of the session (although not all delegates played the game in this way!). I felt it was important to include these tweets as they represented activities which took place at the event. However, it is worth noting that none of the delegates chose to tweet an explanation of these activities for their own followers.
I also observed that the quality of delegate reports can vary. Delegates sometimes gave only very brief quotes with no context. I classified these as reports, but on several occasions I had to consider them very carefully, as I did not feel that I would have been able to identify the statement as a report if I had not been in the room listening to the presentation myself. These may not necessarily provide enough high quality information to an online audience.
It is also worth noting that there was a live video stream provided for the main plenary sessions, so many of the delegates may have assumed the remote audience had access to this. However, we know that this was not always the case, as several online delegates were unable to access the live stream for various reasons, and were therefore observing the event live only through the lens of Twitter.
Conclusions
Without an official Twitter commentary provided by a dedicated event live blogger, there would have been a heavy bias towards comment at this event, with relatively low levels of reporting to provide context for these comments for those who are not physically present. Twitter was obviously more useful to this audience as a social tool for expressing opinions, exchanging links, making connections and building on personal relationships, rather than altruistically conveying a sense of the content to a wider audience. This is a perfectly valid and useful purpose for engaging with Twitter at an event, providing delegates with a whole host of additional networking benefits. However, it emphasises the need for the event organiser to provide an official commentary to provide context to their event hash tag if they want to effectively engage a remote audience or create a useful record of the content of the event.
[shareaholic app=”share_buttons” id=”7637501″]