In the early days of Twitter, one of the biggest and most rampant uses of the platform was to spread false rumors of celebrity deaths. I’ve seen several of them in my day, fallen for a few, and in the years since, I’ve managed to determine the difference between the real ones and the fake ones.
With the rapid pace and sheer volume of information that’s available via social media, pieces of that information can be altered easily. Content can go “viral” at any moment, regardless of whether or not it’s accurate, which is how many people find false celebrity death news crossing their feeds.
Social media is essentially a digital version of the game of Telephone, where information shared, if altered in even the slightest of ways, can misinform the reader or recipient.
People turn to social media for everything these days. It’s got just about anything you can think of: from book and doctor recommendations to traffic and transit updates to breaking news. With all of its readily available and accessible information, social media has become a huge resource for Americans.
According to the Pew Research Center, almost seven out of every ten Americans use social media.1 To put that in perspective, the Census Bureau reports that the United States population was 324,309,805 as of January 1, 2017.2 Seventy percent of that is over 227 million people who use some sort of social media platform. (The survey focused on the top five social media platforms in the United States: Facebook, Twitter, Instagram, Pinterest, and LinkedIn1.)
With so many Americans taking advantage of social media’s usability and readily accessible information, it only makes sense for government agencies like us to tap into it.
News can and often does break faster on social media than it does anywhere else.
Trending topics on Twitter and Facebook are constantly being updated with hordes of different information, whether it’s a new hashtag game, a celebrity death (real or otherwise), or a weather event—just to name a few.
Disaster news, just like standard news, also breaks on social media, and platforms like Facebook and Twitter have worked to help disseminate relevant, timely, and important information in disaster scenarios. Many meteorologists have taken to using hashtags containing “wx” to denote weather in various areas, so seeing something like #okwx accompanying content pertaining to storms or severe weather in Oklahoma is now relatively commonplace.
With all of the various information floating around, how do we actually work to verify it all?
It starts with a series of questions:
Is the information plausible?
This question is potentially the most crucial. Does the information actually make sense in the scenario? Sticking with the celebrity death scenario, many people may click the retweet or share button on a post without double-checking it, particularly if it is a prominent celebrity or public figure. (The most common of these rumors as of late would be about Betty White—which has tripped up many social media users over the last few years.)
After all, some of these posts make a lot of logical sense. In the case of older celebrities a death may make sense. Sharing one of the posts immediately upon seeing it could have negative consequences, like causing concern and disruptions for family members.
In a disaster scenario, the stakes are much higher.
One of the commonly shared items out of Hurricane Sandy were photos of sharks floating about in New York City. While with a storm that strong it may have seemed somewhat possible, the photos were later debunked.3
However, one very important note is that plausible information isn’t necessarily the truth. It may just seem sensible and that’s why we ask multiple questions instead of just stopping at this one.
Is the content original?
Sharing content is one of the cornerstones of social media. Its roots trace back to the ages of chain emails passed along with “FW:FW:FW:FW:FW” subject lines back in the days of Netscape and Hotmail. So where’s the source? Where did this come from? Following the thread of shares or retweets can help find out if the information truly is genuine.
And then there are the photos. Not all information from disasters is distributed via 140 characters of text (or more if we’re talking Facebook). Many times, photos are included. With photo editing software becoming more prevalent and accessible to everyone, it is easy to alter a photo in order to make it eye-catching or shocking enough to go viral, potentially misinforming many people.
Thankfully, there is a way to double check photos before sharing them: reverse image search. Google has an excellent way to actually perform these searches both on desktop and mobile, where you can either save a photo to upload it to their search or use an image’s URL–and Google’s support pages describe exactly how to do it.
And there has been another new development in reverse image searching; Tweetdeck has now integrated a reverse image search function into its streams, making it almost a one-click process. (Keep in mind that Tweetdeck is only available on desktop.)
Is the information timely?
Often, articles about old floods or other events get re-shared, particularly in instances where events are repeating—like if one particular area were to see another bout of heavy rain and flooding.
Many articles will have date and timestamps in a much smaller font or fainter color, which can make them a bit hard to spot. It may take the use of Sherlock Holmes’ magnifying glass to find, but it’s worth it.
Is the source credible?
If the answers to the previous three questions are all satisfactory, the next step is to determine if the source itself is credible. Examining and evaluating your sources is one of the most important pieces of verifying information. How do you know what sources are credible?
One of the quickest ways is to see whether or not a tweet or post that is being analyzed has been geotagged to a certain location. Is that location in, near, or around the impacted area? Is it somewhere else not even close? If the post is coming from the area, it could be considered to be more reliable.
Profile information can be a big help in trying to verify information. Has the user filled out the section of profile information? Does it include a location? Is that location near or around the currently impacted area? Is there a photo uploaded to their profile? (Back in the not-too-far-away days where Twitter used egg photos as placeholders, this was a big thing to look out for. Now it’s a faint, ghost-like silhouette instead.)
Another big thing to look at is what people have published previously. Have they only shared content or have they also penned content of their own?
If most of their content is made up of retweets, it could be a sign of a bot. Bots have become more and more prevalent and sophisticated as of late. These bots can serve a variety of purposes, from trying to hack accounts to trying to get certain topics or hashtags to trend, or even to cause chaos or confusion in an emergency.
Then there’s the question of whether or not recent content supports any current shared content.
One example is a woman tweeting about a power outage at a hospital. Has she previously shared content that would indicate that she would be in or around the impacted hospital? Has she shared something that would discuss whether she would be a patient (or know a patient) at the hospital or be employed there? Or is this completely out of place?
There is also the option to look for verified accounts.
Both Facebook and Twitter have methods of verifying accounts of a variety of different sources: news outlets, celebrities and other public figures, and government agencies. Both platforms denote their verified accounts with blue checkmarks on the profile or listing in search function. Twitter also has a “filter:verified” function in its search options that will only bring up tweets published by these verified accounts.
Of course, it’s also important to corroborate the information as it appears. Corroborating stories, like police officers do in investigations, is key. If multiple sources are reporting the same information, it is more likely to be accurate.
Once we have all this information and have analyzed it, what do we do with it?
The information we gather from social media serves a very important purpose—it helps inform us during our response and recovery operations. Over the years it’s helped us answer a lot of questions—questions that have enabled us to keep our fingers on the pulse of exactly what’s happening in a response situation, like what impacts are being felt, what gaps in information or operations there are, what issues are coming up (here we tend to call those “limiting factors”). Social media helps us fill in a lot of blanks, making sure that we are fully able to help disaster survivors, and provides a fuller picture of any given disaster’s impact.
While social media is an incredibly valuable asset, it’s only one of many ways we gather information; just one of the tools at our disposal.
One last very important thing to remember is that the right information is not always the first information and the first information is not always right and as emergency managers, it’s our job to keep up with it all, for the sake of our local and state counterparts as well as impacted communities and the citizens who inhabit them—especially when all the information seems to be right one minute and wrong the next.
Resources:
Sources:
- Pew Research Center stats on social media usage in the United States: http://www.pewinternet.org/fact-sheet/social-media/
- Census Bureau Population Clock: https://www.census.gov/popclock/
- Mental Floss article: "10 Fake Photos of Hurricane Sandy that Went Viral" http://mentalfloss.com/article/12943/10-fake-photos-hurricane-sandy-went-viral