Twitter Analytics

As a demonstration of how security information spreads on government networks, we undertook an analysis of the Secretaría de Seguridad Pública (SSP)’s twitter account.

The SSP represents Mexico’s Federal Police force, which have been heavily dispatched by the Calderon administration to combat Mexican drug trafficking organizations, as the FP are considered to be more professional and less corruptible than state and local police. At this time, using NodeXL to map the network has been challenging given Twitter data limits and the large size of SSP’s network (over 40,000 followers). We will pursue alternative mapping strategies moving forward. However, the network has a high number of followers, but only follows a small number of users. Additionally, the tweets we examined showed no mentions of users in SSP’s tweets, with the exception of its leaders and other prominent government officials, indicating that the network is not very interactive. We might expect a looser organization by which information is disseminated out and spread to other users, but interaction between SSP and its followers is limited. This stands in contrast to digital journalist group Periodistas de a Pie’s network, in which the second stage consists of interactions between the group and its followers.

Tweets about @SSP_mx over the past month show 7,308 individual tweets where people mentioned the SSP since February 26, 2012. This includes few tweets by @SSP_mx itself, which number only 715 total, and only 28 since February 26, 2012. Given that the account has 40,478 followers as of March 27, 2012, we infer that the high volume of SSP-related tweets is the result of people communicating about the SSP, rather than in direct communication with its twitter account. The account is only following 43 people, most of which are government officials or government agencies, supporting this assumption. Content analysis of 1,813 tweets archived from February 20, 2012 (11:09am) through March 2, 2012 (11:13pm) shows that this discussion includes retweets of security reports, discussion about security events, and both criticism and praise of the SSP. Sample content tweets are included below.

Figure 1: @ssp_mx tweets by volume 2/26/2012 - 3/28/2012

As shown in Figure 1, @ssp_mx receives a high cumulative volume of tweets, but with dips and peaks of volumes per day. This indicates that tweet volume mirrors a news cycle centering around new information related to the account, rather than sustained interest in the entity itself. On the first significant peak between March 7 and March 9, the SSP reported that a leader in the La Familia criminal organization had been captured. However, despite an increase in tweet volume, the message was only retweeted three times. The data shows a significant ratio of tweets to retweets (Figure 2), but a glance at selected tweets indicates that posts duplicating the information may not be marked as retweets (Figure 3), obscuring the ratio of original to unoriginal content in this sphere. However, users appear to be more consistent at crediting other users representing individuals and media organizations with the RT designation.

Figure 2: @ssp_mx tweets to retweets 2/26/2012 - 3/28/2012

Figure 3: @ssp_mx sample tweets. The first two show duplicate information without the retweet “RT” designation, while the third and fourth show the RT-style credit to an individual user and a large newspaper, respectively.

Finally, data on top users indicates who the high volume, high frequency tweeters in the space are. Quantity is not necessarily quality, however, and a large volume of tweets is not necessarily a clear indicator for the amount of influence a user has in the network. For this reason, we place more emphasis on a user’s position in the network as an indication of thought leaders. However, Figure 4 does show us that top users make up relatively equal shares of the tweets, indicating a diversity of opinion on the topic, that is, the conversation is not being dominated by a few key users.

Figure 4: Top users in @ssp_mx twittersphere


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: