This is now time for assessing the amount of the vast damages suffered by New York, New Jersey and other places in the East Coast as a consequence of the unprecedented violence of the hurricane called Sandy. Number of deaths directly caused by the hurricane increases hour by hour.
Emergency situation is not yet overcome right now and difficulties and problems linked to shortages, energy blackout, floods and public transport restoration is just on the way.
The crisis is thus not solved and closed, and the storyline of this disaster is still developing.
In spite the fact that new relevant events will emerge in the next hours and days around the tragedy, we want to share with the readers of our blog on Reputation Metrics some of the results we have already gathered about this disaster.
Disasters, the role of social media and social conversation… and November 6, 2012 US Presidential Elections
As the vast magnitude of the tragedy was already apparent some days before it actually happened, analysts discussed about special issues linked to this crisis.
Some feared that social media could play a highly distorting role in creating ‘noise’ and bad quality communications and information, making harder to correctly channel the valuable and sometimes critical emergency information by emergency agencies, local authorities or news providers counting with means and experience for filtering correctly the information to be published. This fear has apparently confirmed at least to some extent, as many fake disaster pics have been widely disseminated, thus creating unjustified panic reactions.
This nationwide crisis counted with a very special and unusual ingredient: it is happening just one weak before the US Presidential elections. The end of the electoral race show an extraordinary close and uncertain electoral output (Average polls collected by RealClearPolitics by October 30 identify a voting intention Romney 48% Vs Obama 47.1%). The hurricane and the responsibilities that President Barack Obama, one of the candidates is required to assume in managing the crisis, may become a fact influencing the vote of undecided voters. Both Republicans and Democrats have strong interests in using the disaster in advantage of their political aspirations.
Who Is Leading the Social Conversation About Hurricane Sandy In Twitter?
All these considerations pushed us to monitor the social conversation around Hurricane Sandy in Twitter.
A now traditional way to proceed in measuring the impact of a crisis is to count Twitter traffic around the selected topic. An additional step is to undergo a Twitter content analysis in order to identify main elements of the storyline of the crisis and its evolution in time. We have run some this approach in some of our previous crisis analysis published in this blog. It is more than probable that such kind of analysis will be published soon.
We have chosen this time a different strategy.
This time we have looked for the tweets that are playing a more influential role in shaping global and USA public opinion and knowledge about the disaster. Those are the ones that appear and stay in the Twitter timeline in English when someone is searching for the word ‘Sandy’. This is a query that has been performed million of times in Twitter as well as in all other sources of information and conversation.
Let’s call ‘Top tweets list’ to tweets that stay in Twitter timeline corresponding not to all tweets but to ‘Top’ ones as so chosen by Twitter.com. Those are the tweets that people will see hour after hour if they scroll roll down the timeline in order to read more tweets about the disaster.
Our initial analysis corresponds to ‘top tweets’ in Twitter about ‘Sandy’ published in English between 1am 29 October 2012 and 2pm 30 October 2012 ET. This makes a total amount of 4410 different tweets.
We have organized them and we have counted how many different tweets the Twitter accounts have succeed in keeping them as ‘surviving tweet’.
We present in the following table the list of the top 20 Twitter accounts by number of tweets in the ‘top tweet’ list.
Top references in Twitter are Time, Hurricane Central, Anonymous, all of them with 67 tweets. We find thus among the leaders a well established top media reference like Time Magazine, a social activism group and a specialized source of information about hurricanes.
Top ten Twitter account include some top traditional news providers like Wall Street Journal and The Washington Post, but also new powerful actors in the news that have created their business model in pure online structures, like The Huffinton Post and Mashable (Pete Cashmore). We find also some few individuals, who are all of them professionals in the field of emergencies and disasters.
There are more than two thousand contributors entering in the ‘Top tweet list’. But just top 40 contributors generate 27.2% Twitter selected conversation in Twitter. They are really social media influencers in the coverage and reactions to Sandy disaster.
First figure shows the degree of concentration of social media conversation in Twitter around hurricane Sandy. Top 5% contributors generate 38% of all Twitter conversation about Sandy. Top 10% Twitter accounts produced 48% of all tweets. Top 20% produced 59% of all tweets about Sandy.
Next figure refers to the profile of top 40 contributors concerning their affiliation. We find that traditional news providers are the main source of selected social media conversation. They produce 52% of all content: 42% by official corporate accounts, and another 10% by top journalists pertaining to traditional social media counting with personal Twitter account. Another 16% of the tweets como from pure online media. Considering that it refers to top online media references like Mashable and Huffington Post, it can be assumed that these sources follow similar quality standards than traditional media. So, more than two thirds of all selected social media conversation come from well established sources of information. This result comes probably quite in contradiction with perceptions and intuitions of some analsis concerning the quality of information disseminated in the social media.
There are two other quite authoritative sources of information who receive a notable exposure among top influencers in Twitter: local public authorities (NYC Major, NJ Governor) and Twitter accounts from emergency and relief agencies or sources of information.
Summing all these qualified sources of information, we find that they represent 86% of all tweets produced by the top 40 Twitter accounts more successful in the ‘surviving list’.
We consider in the following figure the profile of top 50 (instead of top 40) contributors concerning their engagement in the social media conversation. We measure the median number of total tweets of top 50 contributors in hurricane Sandy storyline in Twitter. Median value of top 50 contributors is 26,089 tweets. This corresponds to experienced and heavy Twitter users. This result is influenced by that fact that more than 50% of all top 50 accounts pertain to news providers.
The figure shows the 10-data moving median values. It allows us to evaluate the relationship between their degree of influence in this disaster story and their all time record of activity in Twitter. We find that most prominent Twitter accounts are even more active than the values for all top 50 Twitter accounts. It can be observed that top contributors have published as median value more than 35,000 tweets. Values morve around 20,000 for Twitter accounts un the second half of the top 50 list.
Our result suggests that only by exception newcomers in Twitter can create content that will stay in the influential ‘top tweet list’.
Another characteristic to consider about top 50 Twitter contributors is their number of followers. This result is of course related with our previous analysis. Number of followers is a clear measure of social media influence. We find that the median value of top 50 contributors concerning hurricane Sandy is 350,000 followers. This amount of followers corresponds to very powerful Twitter accounts. It is telling us that ‘controlling’ the social media conversation is really not a matter of ‘aficionados’. Only well established and already influential Twitter accounts can succeed in shaping the social media conversation by the accumulation of tweets in the ‘top tweet list’ in a hot and sensitive topic.
Next figure shows the results concerning the median number of followers by groups of ten accounts. We find again a positive relationship between the actual degree of influence in the social media conversation and their previous degree of social media influence measures by number of followers. The positive relation with followers is even stronger than the one observed for the leel of Twitter activity. Top contributors count with more than 1.5 million followers. This degree of influence decreases continually and quickly to a first floor of some 400,000 followers. Top contributors in the second half of the list have a median number of 200,000 followers.
This value, the median number of followers of top Twitter contributors, could probably be satisfactorily used as a way to caliber how hot a topic is in a social media conversation or debate, besides the traditional measure of total number of tweets per minute or per hour.
The table below presents the list of top 50 Twitter accounts by number of appearances in ‘Top tweet list’ concerning hurricane Sandy. We include information about the profile of the account, its activity (total number of tweets) and reach (number of followers).