More Research on Boston Marathon Official Twitter Activity #SMEM

Post by: Kim Stephens

Project Hazards Emergency Response and Online Informal Communication (HEROIC)* has posted two more reports that describe their research around the use of Twitter by public safety organizations during the aftermath of the Boston Marathon bombing (see the complete citation at the end of this post).  You can read them in their entirety by clicking the hyperlinked titles and then choosing “Research Highlights.”

• Tweeting Boston: The Influence of Micro-structure in Broadcasting Messages through Twitter
• Tweeting What Matters: Information, Advisories, and Alerts Following the Boston Marathon Events

Below, I briefly describe both of these reports and provide the most significant findings.

Emergency services working after the 2013 Boston Marathon Bombings

 Micro-structure in Broadcasting Messages

The Microstructure report reviews  “conversational microstructures.”  They specifically examined whether or not Federal, State or local entities took part in or were the focus of the conversation on Twitter.  For the purposes of the study,  conversational Tweets were identified if they included a hyperlink, a hashtag, an @ message  or were ReTweeted, each of which  “…signify varying dimensions of online engagement…. Aspects of conversational microstructure use are of interest because they provide insights into which Twitter messages are amplified and why.”

Findings

By studying the data in a systematic fashion (which they describe) they found something very interesting regarding the use of hashtags–by now a common item  in public Tweets, especially for advance notice events.  They discovered that even though this crisis lasted a week, there was NOT a consistent use of one particular hashtag by public safety organizations. They state:

While there were a series of events throughout the week, including the detonation of improvised explosive devices at the beginning of the week, the killing of a police officer at MIT, and the lockdowns of Boston and Watertown, there was no indication that a consistent hashtag emerged or trended among official organizations to organize their content into a traceable stream.

That finding, in my opinion, can be turned into a simple take-away lesson:  agencies that are part of the Joint Information System should immediately determine which hashtags will be used throughout an event to ensure the broadest possible message distribution (of note, the public initially used the tag #BostonMarathon, which was rarely used by public safety organizations). Furthermore, this could have easily been something decided upon when planning for the marathon. The researchers note:

Hashtags that were utilized varied by sector, such as #tweetfromthebeat, #WANTED, and #CommunityAlert by law enforcement, and #oneboston from local government, indicating different aspects of the response.  However, a single hashtag, related to the weeklong investigation and subsequent manhunt and capture, did not emerge.

Tweeting What Matters

Increasingly there are a variety of systems or channels in place to notify the public about what protective action measures they should take before, during or immediately following a crisis event. These systems include everything from the Emergency Alert System–which should reach almost everyone, to targeted text messages from local Universities,  to reverse phone calls from local government, to opt-in mobile applications from the State, etc., all of which were used at some point during the manhunt stage of the crisis. The Project HEROIC report, however, specifically examined Tweets that were posted that contained guidance  to  shelter-in-place, therefore “discussing the role of Twitter as a redundant channel for risk communications.”

Findings

Interestingly, they found that during the immediate aftermath of the bombing (on Monday) there were not a lot of official Tweets providing direction to the public. They speculate why: “…few [protective action] guidance-related tweets were posted, possibly signifying the lack of certainty about the event, the speed at which it unfolded, and having little information regarding what people should do in response.  However, at the end of the week, guidance tweets became more prevalent and focused on sheltering in place.”

They found that during the manhunt stage of the event Twitter was “definitely” used as a redundant channel to provide protective action guidance to the public; however, the public did not necessary repeat (ReTweet) these message as much as other content posted by official organizations. Again, this finding required the researchers  to speculate. They conclude that since the information about the protective action measure was provided  in so many different formats, it is quite possible that people did not feel the information required repeating–particularly not during the day once it was widely distributed and repeated on all forms of media.

Mission Hill Capture Celebrations

The take-aways for the researchers:

“With this in mind, it is becoming prudent for organizations to consider the kinds of information that is most desired by an online audience, at different points in time, and for different sectors of the public.

Messages can be crafted for both locally affected community members in need of advisories and guidance, as well as distant observers intent on serving as information conduits.  Future disaster communicators ought to learn from these detailed observations about public retweeting practices in order to determine how to more effectively focus, shape, and share messages that make a difference.”

Let me know what you think. Does their analysis fit with your own experience?

Sources:

Sutton, J., Johnson, B.,  Spiro, E.,  and Butts, C. (2013). “Tweeting What Matters: Information, Advisories, and Alerts Following the Boston Marathon Events.” Online Research Highlight. http://heroicproject.org

Sutton, J., Spiro, E., Johnson, B., Fitzhugh, S., and Butts, C. (2013). “Tweeting Boston: The Influence of Microstructure in Broadcasting Messages through Twitter.” Online Research Highlight. http://heroicproject.org

**Project HEROIC is a collaborative, NSF funded effort by researchers at the University of Colorado at Colorado Springs and the University of California-Irvine to better understand the dynamics of informal online communication in response to extreme events.

Related articles

• Researchers Review Boston Bombing Social Media Activity (idisaster.wordpress.com)
• 3 Observations: Social Media and the Boston PD #BostonMarathon (idisaster.wordpress.com)
• Boston Marathon: Social Key in Response (cnbc.com)
• Research in to use of hashtags on Twitter (protohub.net)

Researchers Review Boston Bombing Social Media Activity

Post by: Kim Stephens

Project HEROIC–which stands for Hazards, Emergency Response, and Online Informal Communications (see footnote)–took a close look at the online activity of official organizations during the recent domestic terrorist event in Boston and the ensuing suspect chase–that seemed like a marathon in itself.  They released a report today (May 10) titled “Following the Bombing” which I have summarize below.

Their Methodology and Findings

In order to understand  what types of information was provided to the public and how broadly it was distributed, the project team reviewed 29 different government agency or related Twitter accounts. The first question might be: why only Twitter? Researchers like Twitter–the data is easy to grab and analyze.

The project team reviewed two main items: 1. Rate of posting by the selected organizations and elected officials;  and 2. The percent change in followers  (spoiler alert: Boston PD had a 500% increase and the Boston PD PIO Cheryl Fiandaca had a 2291% increase).  The rate that these organizations posted was tied to their increase in followers, which is no surprise, however, there was a notable exception–Boston Fire Department.

Boston FD gained a 25% increase in followers without posting once the day of the attack. Their absence  was not lost on the Twittersphere, and the Boston FD even felt it necessary to defend their decision the next day.  They Tweeted that they deliberately did not post any Tweets from the scene because it is their policy not to “…show any injured person or discuss our treatment.” Quite a few people, however, thought their decision was unfortunate; at a minimum they could have simply ReTweeted the Boston PD account. As the researchers pointed out:  “…organizations that have increased their network size must provide information of value and to be aware that the public is watching.” Honestly, its about trust. People who follow official accounts do so because they know they can trust the content. The public followers also have a notion that they will provided information in timely manner-especially during incidents such as this one where everyone was looking for any tidbit  they could find in order to make sense out of the chaos.  It is not a stretch to see why people were upset.

Read the whole report here. I like some of their questions they pose at the end:  What can organizations do to ensure their newfound followers stick around? and What educational preparedness-type information should organizations provide to take advantage of the narrow window of attention they have? Let me know your thoughts.

Footnote: “Project HEROIC is a collaborative, NSF funded effort by researchers at the University of Colorado at Colorado Springs and the University of California-Irvine to better understand the dynamics of informal online communication in response to extreme events. Through a combination of data collection and modeling of conversation dynamics, the project team aims to understand the relationship between hazard events, informal communication and emergency response.” (via: http://heroicproject.org/)

Related articles

• 3 Observations: Social Media and the Boston PD #BostonMarathon (idisaster.wordpress.com)
• How Social Media was used During the Boston Marathon Bombings (abarbuto3.wordpress.com)

The Work of Disaster Reporting in the Age of Digital Distortion

Post by: Kim Stephens

Academics from IBM Research Labs in India; Indraprastha Institute for Information Technology, Delhi, India; and the University of Maryland, Baltimore  County collaborated on an article titled: “Faking Sandy: Characterizing and Identifying Fake Images on Twitter during Hurricane Sandy.” This article is interesting in light of the events in Boston and the debate about the veracity of content on social media. Although there hasn’t been time to do complete quantitative data analyses of the Marathon bombing social media feeds, this research adds to the increasing collection of academic studies that can help us better understand how misinformation is distributed on social platforms, specifically Twitter, and how  it can be easily and quickly identified as false.

Abstract:

In today’s world, online social media plays a vital role during real world events, especially crisis events. There are both positive and negative effects of social media coverage of events, it can be used by authorities for effective disaster management or by malicious entities to spread rumors and fake news.

The aim of this paper, is to highlight the role of Twitter, during Hurricane Sandy (2012) to spread fake images about the disaster. We identified 10,350 unique tweets containing fake images that were circulated on Twitter, during Hurricane Sandy. We performed a characterization analysis, to understand the temporal, social reputation and influence patterns for the spread of fake images. Eighty-six percent of tweets spreading the fake images were retweets, hence very few were original tweets. Our results showed that top thirty users out of 10,215 users (0.3%) resulted in 90% of the retweets of fake images; also network links such as follower relationships of Twitter, contributed very less (only 11%) to the spread of these fake photos URLs. Next, we used classification models, to distinguish fake images from real images of Hurricane Sandy. Best results were obtained from Decision Tree classifier, we got 97% accuracy in predicting fake images from real. Also, tweet based features were very effective in distinguishing fake images tweets from real, while the performance of user based features was very poor. Our results, showed that, automated techniques can be used in identifying real images from fake images posted on Twitter.

The last sentence is the most important: “Our results, showed that, automated techniques can be used in identifying real images from fake images posted on Twitter.” Hopefully, those automated approaches will be available to use quickly and intuitively without needing to know how to write an algorithm on the fly. There are people working with that goal in mind. See this great post “Automatically Extracting Disaster-Relevant Information from Social Media” by Patrick Meier–where he describes his efforts  “to develop open source and freely available next generation humanitarian technologies to better manage Big (Crisis) Data.” A software solution is on the horizon.

Why do people post false information? That interesting psychological question was not addressed in this study. Maybe people think an image of a shark swimming the street is funny; maybe they are out for a minute or more of fame.  I, for one, am increasingly leery to ReTweet any photo when an event is unfolding unless I see it on several sources (e.g. on Twitter and also streamed live from a “traditional” local news station). Tell me, how comfortable do you have to be with a source before you hit “ReTweet?”

Thank you @janehenrici for the title! I had posted “Are those photos for real?” I hear the phrase “Is that for real”  in my house full of teenagers and was just trying to be “witty.” I like Jane’s title much better though. Let’s call this crowdsourced blogging!

Related articles

• Fake tweet reveals a nation on edge (redtape.nbcnews.com)
• Research Post #5: Creating the Site and Fake Story (nmrsspring2013.wordpress.com)
• The Birth Of Fake Viral Images Hurricane #Sandy Style (digitalhighrise.com)
• How journalists can avoid getting fooled by fake Hurricane #Sandy photos (nextlevelofnews.com)