Tweeting Up a Storm

Hurricane Sandy on Oct. 25, 2012
Hurricane Sandy on Oct. 25, 2012

We are inundated daily with stories from the news media about the possible impact social media like Facebook and Twitter will have on our lives. When a storm like Hurricane Sandy hits the East Coast, can this technology actually help to save lives and reduce catastrophic damages? It’s possible.

For instance, mobile devices could allow emergency responders, affected communities, and volunteers to rapidly collect and share information as a disaster unfolds. Photos and videos provided through social media could help officials determine where people are located, assess the responses and needs of affected communities—such as water, food, shelter, power and medical care—and alert responders and citizens to changing conditions.

At least that is the promise. When Hurricane Irene barreled across the Eastern seaboard in August 2011, many in the news media cited it as a pivotal moment for social media for disasters. But research we conducted on the use of social media during Irene suggests otherwise. While some emergency management departments launched new social media outreach strategies during the storm, particularly to push information out to the public, many did not change their practices radically and overall use of the technology varied.

Why didn’t these technologies live up to the promises of social media evangelists? They fell short for at least four reasons.

The needle in the haystack: Data flowing from social media sources are messy. During a crisis, this information is loaded with non-essentials—opinions, jokes, and general conversation—and the volume and complexity of information can be overwhelming. On Saturday alone more than 335,000 Sandy-related “tweets” (i.e., posts) were generated in advance of the storm, increasing from 34,000 per hour during the day to 80,000 per hour by evening. Emergency management organizations typically don’t have the resources or staff to sift through the onslaught to extract actionable information.

Trust: Peoples’ live are at stake in crisis situations and trust plays a critical role in what data are used and why. Emergency response organizations face a challenge when dealing with data from the “crowd:” Who are these people?  Are they, the information they provide, and the technology used reliable? Emergency managers work to avoid risk and are understandably reluctant to use unverified information from unknown individuals.

Security: Using social media requires a greater degree of openness, and potentially exposes emergency response organizations and the public to inappropriate content, malware threats, and breaches of confidential information. Because data continue to live on the Internet long after a disaster, privacy and ethical issues persist. How do we protect the privacy of vulnerable people both during and after a crisis?

Liability: Providing or acting on information about disaster conditions via social media carries potential legal liability. In the last few years, self-organized “digital volunteers” have emerged to leverage social media for collaborative disaster response by monitoring social networks, aggregating data, and creating “crowd maps” to assist both affected communities and responders. Contrary to popular belief, so-called “Good Samaritan” laws offer little, if any, protection for these digital volunteers in the United States. Formal emergency response organizations, meanwhile, are acutely aware of the potential liability in using citizen-generated information to make critical decisions.

So what can be done to realize the promise of social media in disaster response? More tools are needed to identify patterns and make sense of the near-endless flow of information generated by social media users. One example of this work is TweetTracker, sponsored by the Office of Naval Research. It’s an algorithm that allows one to track, filter, analyze, and visualize social media feeds in real-time. There is also a need to develop methods and best practices for integrating the stream of data from social media with authoritative data from more-traditional sources.

Policies and regulations need to be updated to allow emergency management organizations to use social media not just as an outreach tool, but also for operations.  This will require flexibility as well as systems that provide actionable information without compromising privacy or security.

The digital volunteer community also needs training to better integrate the social media response efforts with those of the first responders. Models of vetted and trusted volunteers include the American Red Cross’ Digital Volunteer Program, the United Nation’s Digital Humanitarian Network, and newly forming volunteer Virtual Operations Support Teams.

Finally, the best use of social media may be in building social capital and resilient communities that will reduce the impact of disasters and speed recovery time. After all, that’s when we really need our friends.

4 thoughts on “Tweeting Up a Storm

  1. Although, I’m a little late to this post (as I just discovered the blog yesterday), but I definitely agree that receiving crowd info/knowledge is not necessarily some “walk in the park” without preparation and/or investment to make use of the knowledge. Along these lines, I think that readers of this post (and the blog in general) will likely appreciate some recent emerging IT/Management research. If you’re so inclined, have a look at “The Theory of Crowd Capital”, which attempts to elegantly explain such problems, and a number of related phenomena too. I’d love to hear your input about this work…?

    See here:


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 )

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