Harnessing the Power of Social Media in Disaster Response

Harnessing the Power of Social Media in Disaster Response
28 SEPTEMBER 2010 353 VIEWS NO COMMENT
continuityinsights.com

Social media is transforming the Internet. Previously dominated by a handful of professional programmers pushing information, today’s Internet is ruled by social networks pulling contributions from millions of users. And while the emergency management community is utilizing this new medium to harness the collective knowledge of the public to direct response and save lives, the business continuity community remains largely on the sidelines of this movement.


Mostly, this is due to widespread lack of understanding within the continuity community about what social media is and how it has been used. Countless articles cajole business continuity practitioners to develop a “social media strategy” by adding Twitter or RSS feeds to their crisis communication plans. But that is not a social media strategy. Social media is not a channel for pushing information out to the public. Social media involves pulling information and resources from the public. It starts by listening, not talking, and it is revolutionizing disaster response.

We will look at how social media has been used in disaster response, some common misconceptions about social media, and ways that continuity professionals can leverage social media to improve the resiliency of their organization and community.

Social Media in Disasters

9/11
Immediately after the first plane hit the Twin Towers on September 11, 2001, many workers in Washington, D.C. were sent home. Andy Carvin was one of those workers. Feeling helpless at home, he decided to hack his own blog and allow anyone to post their thoughts (Carvin: 2010). Within hours, people around the country were discussing the attack to try to understand the events. Soon media reports came out claiming that the National Mall had been struck. Carvin, along with others, could see from their apartment windows that the Mall was fine, and immediately notified each other of that fact. Eventually, the regular media retracted the story (tedxnyed.com). In one of the earliest uses of social media, the public corrected false information distributed by the regular media, and today the regular media uses social media to gather much of its information.

Boxing Day Tsunami
In December, 2004, a tsunami devastated areas in the east Indian Ocean. Gathering information from remote regions was particularly difficult. To fill the information void, blogs soon appeared allowing people on the ground to post reports, including photos taken with cell phones, draw together information, correct misinformation, solicit aid, and inform rescue efforts. (tsunamihelpindia.blogspot.com/2004_12_01_archive.html)

Hurricane Katrina
Hurricane Katrina struck not long after the Boxing Day tsunami, and people outside of the Gulf Coast region were desperately looking for information on the status of loved ones in the affected areas. Soon the same bloggers who set up the tsunami sites, many from India, established a web database to gather information on the whereabouts of individuals. They then enlisted the help of hundreds of volunteers from around the world who scoured news items about victims from a variety of diffuse sources and input them into the database.

This developed into “PeopleFinder,” a website with two simple buttons: “I’m Looking for Someone” and “I Have Information about Someone.” That website was used to connect people in the affected areas with those looking for information about them. The website is built on a free, open source platform that can be used for any disaster. It has subsequently been used for disasters such as the Haitian earthquake, for which there are currently more than 55,000 entries (haiticrisis.appspot.com).

Virginia Tech Shooting
Within 90 minutes of the Virginia Tech shooting, a web page appeared on Wikipedia that accurately described the events (Bloxham: 2008). Only 20 minutes later, a Facebook page appeared called “I’m OK at VT” which allowed students, faculty and staff to notify others that they were safe, as well as correct false reports and distinguish those feared harmed from those confirmed harmed.

California Wildfires
One of the challenges to fighting a wildfire and evacuating people is getting accurate information about the fire’s movement. The smoke obscures aerial photos, and because wildfires generate their own wind, they can move quickly and change directions in an instant. Some fire departments solved this problem by establishing a text message line that allowed ordinary people to report the fire’s movement. If a fire marshal posted a line of firefighters along a road, and a barrage of text messages suddenly appeared warning that the fire had circled around the group, the fire marshal could move the firefighters out of harm’s way.

Subsequent studies of the public response to the fires found that many people relied on social media for their information because it was more relevant, more timely, and more accurate than what they were getting from the regular media (Sutton, Leysia Palen & Irina Shklovski: 2008). Regular media sources focus on the major areas, such as downtowns, not neighborhoods outside of those areas. Their reports may be hours old – an eternity in a fluid situation such as a wildfire.

As people affected by the 2007 fires reported:
• “Most of the news media … are utterly clueless about anything in rural areas. They constantly gave out bogus information, like locations and directions that made no sense at all.”
• “National news websites were completely worthless as they ignored everything except the comparatively minor Malibu fire which burned near some celebrity homes.”
• “The only way we all have to get good information here is for those who have it to share it. We relied on others to give us updates when they had info and we do the same for others.”

Contrary to the widespread assumption that social media will be inaccurate in an emergency, researchers found that “instead of rumor-mongering, we see socially produced accuracy.”

Haitian Earthquake
Once again, ordinary people were looking for a way to help others in an emergency. Within a day of the Haitian Earthquake, a “Crisis Camp” was called in Washington, DC. This camp had no set agenda. People showed up with computer in hand and simply mapped out ways they could help. Then they sat down and started aiding in the rescue efforts. Within a week, more than 40 other crisis camps appeared in cities around the world, and today the movement has been formalized into a response system for all new crises (crisiscommons.org).

Two of the systems created are particularly impressive. Within one day, an iPhone app was developed that translated Creole into English and back again. Then hundreds of volunteers hand-transcribed phrases into the system so that English speaking rescuers could talk with the victims to learn their needs or provide directions (Traduiapp.com).

Even more impressively, crisis camps created a software program called OpenStreetMap which took the Google map of the affected area, and using post-disaster satellite images, allowed volunteers to code on the map every building that was damaged and how badly, which streets were blocked, where hospitals had been set up, and other critical information for rescuers. Thousands of people hand-coded the information, which was converted into a format to display on GPS units. Rescuers on the street used the information and reported that it saved lives.

Three Misconceptions

If the emergency response community is using social media to save lives, then why hasn’t the business continuity community embraced this tool? Three common misconceptions have held it back.

1. Use vs. Cause
Nearly every article on social media for disaster response includes the obligatory “counter example” of the disgruntled worker blogging something embarrassing about a company. While this happens, it is an entirely different topic. These writers have confused social media used in disasters with social media that caused PR disasters. The former is a matter for disaster response, while the latter is a matter for PR and employee use policies outside of disasters. Confusing the two is the same as confusing “the use of cars to respond to disasters” with “responding to car accidents.” Unfortunately, most commentators slip from one to the other without noticing the difference, resulting in confusion about the issue and scaring away potential adopters.

2. Fallacy of Infallibility
The second major error is to dismiss the use of social media in disaster response on grounds that you cannot confirm that the information collected is accurate. “What if a social media report is wrong?” they ask.

But why is it assumed that regular media reports are infallible? As we have seen, the regular media gets information wrong, and social media often corrects misinformation from regular media reports.

In fact, it’s been proven that collaborative knowledge can be much more accurate than a single source. It’s been shown that if you ask farmers at a county fair to guess the weight of a cow, the average of all of the guesses will be more accurate than any particular guess (Snowdon: 2010). The “wisdom of crowds” can be used to correct false reports from a single source such as the regular media.

Even if we ignore the evidence, the mere fact that social media reports are not infallible does not tell us that we should ignore this resource. Should the Haitian rescuers have thrown away their OpenStreetMap system on grounds that some of the information might be wrong? People call in false alarms to 911 all the time, but that does not stop the fire department from responding to calls. The objection is not action-guiding – it doesn’t tell us what we should do.

3. Illusion of Control
Nearly every crisis communication plan is based on the assumption that information must be strictly controlled during a crisis. All communication should be formulated at the highest level and pushed out through a single source to control perceptions. But this model fails utterly in the world of social media, if it ever worked. People will talk about your event with or without you, and taking yourself out of that conversation does nothing to make it go away; it only makes you irrelevant.

BP’s response to the Gulf oil spill is a prime example. Only the CEO was allowed to speak about the event. Not only did that fail to prevent the company from embarrassing itself, but it sent the regular media to interview anonyomous sources within the company to get information. More importantly, BP set up a one-way information website that pushed out press releases. How many people used the BP website as their source of information on the disaster? A full 24 hours after every person in the free world knew that the “top kill” effort had failed, BP’s official website was still running its news release proclaiming that the process was going along as planned and BP was confident of its chances of working.

The top-down command and control model is based on what is perceived to be the military model. But that model was never used by successful militaries. Successful armies give lower-level commanders the freedom to respond to conditions. The U.S. war effort in Iraq stalled out when commanders insisted that units follow a prescribed method that wasn’t working. Eventually, lower-level officers tried new methods and shared the results with others via blogging. This turned the tide of the war, leading one top general to proclaim that “the only thing that worked was blogging; everyone ignored doctrine.” (Snowdon: 2010)

Much of the response to a disaster comes from the bottom up rather than top down. Nearly a million people needed to be evacuated from Manhattan after 9/11. With the subways and tunnels closed, how did they do it? Many were assisted by hundreds of boaters who arrived in lower Manhattan and organized the evacuation themselves. No central planning, just ordinary people organizing the response spontaneously. As one researcher noted:
“Studies of evacuation at times of crises have now been undertaken for the last 50 years. They have consistently shown that at times of great crises, much of the organized behavior is emergent rather than traditional. In addition, it is of a very decentralized nature, with the dominance of pluralistic decision making, and the appearance of imaginative and innovative new attempts to cope with the contingencies that typically appear in major disasters.” (Quarantelli: 2002)

Harnessing the Power of Social Media

So how can you and your community make use of this resource?
Here are just a few suggestions:

Tag your community: Andy Carvin notes that people can prepare their communities for a disaster by simply tagging local evacuation routes, shelters, hospitals, etc. on OpenStreetMap.com. This could be a school project, which will be a learning experience for the children. I plan to lead my son’s Boy Scout troop in such a coding exercise at a weekly meeting.

Prepare to use a crisis wiki: Learn how to use the crisis.org wiki so that you can quickly set up a site to gather information and volunteers in the event of a disaster. Let people in the community know that this will be the place to go to identify needs and ways that they can help.

Build a database for first responders: Bill Boyd, a fire chief in the state of Washington, points out that a fire department could use a password protected website with information on the hazards in local businesses, or even floor maps of those buildings, to consult during a fire (Boyd: 2010). Yes, firefighters have laptops and iPhones too, and businesses could enter relevant information on their own for firefighters to retrieve at the fire. Such a database could even be extended to homeowners who would provide information such as the number of people living in their home, the floorplan, hazards, or any other facts that could be relevant to firefighters.

How can you make use of social media in your organization’s crisis response plan? First off, give up the “illusion of control” that forms the basis of the locked-down, push theory of crisis response and prepare a crisis communication plan that includes listening as well as talking. In particular:

Monitor the traffic: You may have employees that are out of work during a disaster. These people can monitor the regular and social media traffic with applications such as TwitterDeck, which can be used to search or aggregate Twitter traffic. They can monitor public opinion and report news items that require immediate response.

Solicit input: Add any one of a number of free chat features to your crisis communication website to make it into a two-way conversation. Transocean, the company that owned the Deepwater Horizon rig that exploded, added a number of links to their website that allowed people to provide suggestions, volunteer to help, and report oiled wildlife. Make sure that there are people monitoring these channels and even (gasp!) respond to questions to demonstrate that people’s input is being heard.

Employee Networking: A social network is crucial to human resiliency during a crisis. Employees may be scared or have unmet needs. A simple social media site that allows employees to express their thoughts, or request help with housing, clothing, babysitting, etc. will go a long way towards helping your workforce to cope with the situation.

Transocean created a website about the 11 employees who were killed and invited those who knew them to share photos, videos, and other information about their lives (deepwaterhorizoncondolences.com). This acts as a cathartic device for the families and friends of the victims.

Ask for assistance: Common orthodoxy says that you should limit information about the true extent of your disaster to avoid spooking customers. But this can create the impression of having something to hide if your reports do not square with the information coming out of other channels. Plus, this policy can leave needed resources untapped if others do not know about your organization’s needs.

People are interested in helping, and governmental or private agencies can be bureaucratic and slow. Make it as easy as possible for people to identify your needs and offer assistance by posting them right on your crisis website. The results may surprise you.

Numerous researchers have noted that “when disasters have occurred, there has been an informal development of technology and communication that has self-organized during the event to provide coherent, relevant, information outside the traditional information providers” and because of this, emergency responders are “integrating local knowledge into disaster management activities” (Laituri: 2010).

These activities have been largely self-organized because systems have not yet been developed that can structure the use of social media in disasters. Adding structure will make them even more powerful. Studies have shown that the best gathering mechanisms are ones that provide structure to distributed submission nodes. Wikipedia, built by the contributions of thousands of people, has been proven to be essentially as accurate as the Encyclopedia Britannica because it provides a structure to submit its information (CNet: 2005).