Leave a comment

“Beirut, Blood and Twitter” by Mohamad Jouni


Lebanon experienced a horrific bomb explosion on Friday October 19, 2012 that took the lives of 3 people and injured 110. You can read more about this tragic event here. This was a nationwide shock, but amidst the furor, one event drew my attention. This event can be summarized in one tweet:

After the explosion Twitter was set ablaze by tweets asking people to donate blood to help the victims of the explosion. A short while after, the above tweet was sent. A large number of blood donors got turned down due to overcrowding.

All of this got me thinking; Can we analyze Twitter to understand how the blood donation movement grew? Can the analysis provide any insight that would help us mobilize large numbers of blood donors in case of similar events?

To answer these questions I rolled up my sleeves and started tinkering with the Twitter data.

The Data Set

All the data analysis that you will see in this post was done on tweets that contained English keywords related to blooddonate (or donor, donors, donation, give, etc…) and Lebanon (or Beirut, Sassine, Achrafieh, Ashrafieh, etc…), and were tweeted betweenFriday 19 October 2:50 PM (time of the explosion) and Friday 19 October 7:25 PM (time at which blood donation tweets became sporadic).

The resultant data set contained 2650 tweets generated by 1933 Twitter accounts. Out of these 2650 tweets; 2280 were retweets and 370 were original messages.Let us now see what those 2650 tweets can tell us.

Post Explosion Tweets

First Tweet

As we can see from the above graph; the first tweet to mention blood donation in context of the explosion was around 3:50 PM (3:47 PM to be exact)

First tweet telling people to donate blood at 3:47PM
Aya -an Egyptian national- was the first person to tweet and tell people to head to a hospital and donate. In a bid to understand what would trigger the idea of asking people to donate, I reached out to Aya:
Now even though Aya’s message was not retweeted, her initiative is a great example of how to nullify the Bystander Effect:

The bystander effect or Genovese syndrome is a social psychological phenomenon that refers to cases where individuals do not offer any means of help in an emergency situation to the victim when other people are present. The probability of help has often appeared to be inversely related to the number of bystanders; in other words, the greater the number of bystanders, the less likely it is that any one of them will help. The mere presence of other bystanders greatly decreases intervention. In general, this is believed to happen because as the number of bystanders increases, any given bystander is less likely to notice the situation, interpret the incident as a problem, and less likely to assume responsibility for taking action.

Aya explicitly told people what to do (“Donate blood”) and how to do that (“Go to Hotel Dieu hospital”).

Reaction Time

We can glean from the graph that the number of tweets related to blood donation peaked approximately 1 hour and 15 minutesafter the explosion took place. Then, 40 minutes later, tweets telling people to stop donating started streaming in. All in all, it took approximately 1 hour and 55 minutes from the instant the bomb exploded to the moment the first tweet told people to stop donating blood.
The other interesting piece of information that we can extract from this graph; is how fast the tweet numbers peaked. The first tweet mentioning blood donations started at 3:47 PM 18 minutes later– the tweet numbers peaked at 4:05 PM. This tells us that a large number of people were glued to their Twitter accounts and tweeting directly after the explosion. This is not a big discovery when we factor in the idea that people were interested in knowing the latest updates on the explosion, but this tells us that a large number of Lebanese people are depending on Twitter as a real-time source of information, and that this dependence helped spread the blood donation messages.

Most Retweeted Accounts

To understand which Twitter accounts were the most influential in spreading the blood donation message around, we can look at the accounts that generated the most retweeted blood donation messages during the time period under the microscope:
From this chart, we can see that singer Haifa Wehbe’s account had a huge effect on spreading the blood donation messages around (At the time of writing, @HaifaWehbe’s account had a total of 390,092 followers). But the most retweeted account was @DSCLebanon, an NGO that promotes blood donations in Lebanon. @DSCLebanon had its messages retweeted a total of 465 times; more than twice as much as the second most retweeted account. The interesting fact is that @DSCLebanon had 3,856 followers at the time of writing. Number of followers is by no means a competition in the context of this article, but understanding how a follower count of 3,856 can yield more retweets than an account with 390,092 followers can give us an interesting insight into the network effects at play in a social network. I lack the skill set to accurately interpret this observation so I will leave this to someone with a proper understanding of network theory.

Top Retweeted Messages

Dissecting the most retweeted messages yields the following chart:
Now an interesting question to answer is; how fast did these messages spread across Twitter?

Most Retweeted Message

The most retweeted message belonged to @DSCLebanon and we can see how fast the message got retweeted across Twitter.

Second Most Retweeted Message

The second most retweeted message also belonged to @DSCLebanon, but this tweet was instructing people to stop heading to Hotel Dieu Hospital due to overcrowding.

Third Most Retweeted Message

The third most retweeted message belonged to singer Haifa Wehbe.

Tweet Sources

In case you are curious to know which applications were used to send out these tweets:

Conclusion

So after going through this data we can see that the success of this blood donation campaign was due to two main factors:
  1. Influential Twitter accounts that acted quickly and asked people to go donate.
  2. Large number of users retweeting blood donation tweets as fast as possible to their followers.
To help replicate this serendipitous blood drive in the future, we must do three things:
  1. As soon as a crisis takes place, influential Twitter accounts (@DSCLebanon, celebrities, etc …) should tweet to instruct people what to do, or retweet any messages they see to their followers as quickly as possible. These tweets should explicitly mention in details what to do to help.
  2. Twitter users should follow key Twitter accounts to contribute in spreading this message (So please go ahead and follow@DSCLebanon).
  3. Twitter users should retweet any blood donation messages to their followers as fast as they can. If enough of us react fast, the message will quickly reach a large number of users, which will result in more people heading out and helping.
In theory, the three above mentioned guidelines should apply to a large number of emergencies and not just blood donation requests. Who knows how many lives we could save if we follow these three basic guidelines ?
Posted on October 19th by Mohammad Jouni
The original article can be read at the following location.
Thank you for the awesome post-Achrafieh explosion analysis!
Advertisements

About Fred Bteich

A metal fan, a blogger and a news/sports-obsessed medical student.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: