False Alarms

Colleges and universities across the United States are spending tens of millions dollars to create and operate emergency communications systems. Yet as officials at the University of Texas at Austin discovered recently during a deadly knife attack on the campus, students have their own informal system for communicating in an emergency.

As is usual in such emergencies, students in Texas pushed out warnings much faster than campus officials. Also as usual, that information was often inaccurate.

Or consider the recent incident at Colgate University. A student reported seeing a man with a gun enter a campus building. Following the same protocol that exists on many American campuses, an emergency alert was issued to thousands of people. And then another, this one reporting an active shooter and calling for a campus lockdown. As expected, panic ensued.

In reality, the alleged gunman was actually a student carrying a glue gun for a class project. The fact the student was black raises a whole other set of issues for the campus.

Even in situations where an institution generates a digital message to thousands of smartphones within minutes of an actual campus attack, like the one last November at Ohio State University, it can be too late. Ohio State pushed out a Buckeye Alert message just five minutes after that knife attack began on Nov. 28. By then, a campus police officer had already shot the assailant to death. Although the danger was over, the message still generated panic.

We have trained a digital generation to expect instant information on Facebook Live, Snapchat and Twitter. If accurate details are not available, people without accurate information quickly step in to fill the void.

Many of our students are so digitally adept they have thousands of followers on their campus and around the world. But while they may have a powerful $700 smartphone, they do not have training in journalism or emergency management.

Inaccurate distribution of emergency information by students on college campuses has been going on for more than a decade. On the evening of Dec. 13, 2007, no shooting occurred on Penn State’s University Park campus. No shooting at all. Tens of thousands of students and employees and visitors were all safe.

That’s exactly when all hell broke loose.

Campus police and the county 911 emergency center were overwhelmed with panicked callers. Armed officers were dispatched around the campus. Frightened students locked themselves in their rooms on and off campus. Some called their parents for reassurance that everything would be OK. Breathless news reporters called PR staff members at home wanting to know why a SWAT team rushed inside one of the residence halls.

Worried callers wanted to know about the gunman in Atherton Residence Hall. Some called about the gunman in the library. Others wanted to know about the student union building, and more called about a sighting of the elusive gunman a block off the campus.

The more police responded, the more the frantic calls increased.

After a few hours, it became clear that there was no gunman.

During those chaotic hours, I called police. I called administrators. I called my staff. But I resisted the idea of sending out an official text alert that a possible gunman was roaming campus. Years of experience as a police reporter and as a communications professional told me this emergency did not sound right.

If I made the wrong decision, people could die. I pictured myself sitting before a congressional committee a few months later testifying why I hesitated to push the send button on my BlackBerry with a message that might have saved lives. How would I face angry, grieving parents?

Up until that evening I believed, as vice president of university relations, that I oversaw one of the largest text alert systems in the country. I was shocked to discover students had their own system. And while the university could quickly communicate accurate information to 50,000 people, the students could reach one another, parents and news media just as easily with information nobody had vetted for accuracy.

What started out a few hours earlier that day as a well-meaning public warning from the local, off-campus police department about a suspected murderer spotted a few miles from campus at a motel on Atherton Street morphed into rumors among students of the alleged gunman being on the campus in Atherton Hall. And a bunch of other places.

That’s when police phone lines went crazy. Really, really crazy.

The next morning I visited the campus police station and listened to a recording of the calls that dozens of people, some nearly hysterical, made the night before. It was a classic, modern-day exercise in mass hysteria.

And, at that moment, I realized I did not have the biggest, most complete communications system on the campus. The students had a far bigger one. And unfortunately, they had absolutely no checks or balances on the claims they were sharing with one another.

When I arrived at the office the next day, I received a call from an angry woman. She rang me from London to tell me how stupid I was for not sending out emergency text alerts the night before.

She was the parent of a student who locked herself in her off-campus apartment during the fake crisis. She barricaded the door and, for more than an hour, talked with her mother 3,000 miles away for comfort.

I told the woman the only information I could have shared by text alert for a couple hours the night before was that police were receiving multiple reports about a dangerous gunman all over campus and downtown. I felt in my gut the information was wrong, but it was just my gut. Using the best information we had at the time would have put an official stamp on further panic and confusion.

She believed, after listening to the terror in her daughter’s voice, the best course of action was to issue a series of text alerts with information about each of the sightings and that, by the end of the night, it would all be resolved. A kind of digital three-act drama told in a frantic series of 160-character text messages -- all of them inaccurate except the final one.

But if I had given the all clear without facts, it could have been a bigger disaster if gunmen were actually roaming four different campus buildings.

I still think the imperfect but correct call was to not send a series of confusing, inaccurate text alerts throughout the evening as the rumors expanded in size and locations on student social media and text-message systems. The news media were already converging on the campus, and we didn’t need to attract more coverage and further hype the confusion.

In hindsight, I will say good luck and Godspeed to all the campus police and communications staff members who have to manage these situations in the future. There are many more possible wrong decisions in a crisis than there are right decisions. Indeed, with the growing number of mass shootings in our country and the expansion of civilian guns on campuses, the opportunity for mistakes has never been bigger.

In the clarity of the next day, everyone seems to know how they would handle a crisis more effectively. Just keep in mind that, in the best of scenarios, the official text-alert system will need a few minutes to distribute any information. And as we’ve seen in recent mass shootings around the country, the initial reports are often inaccurate.

The public’s informal text-communications system can jump into action immediately. Nobody will later need to apologize or be held accountable for passing on inaccurate information to a couple friends who then pass it on to 10,000 other friends.

What the police at Penn State a decade later still refer to as the “Invasion of Atherton Hall” taught me that, no matter how sophisticated official communications systems may be, the unofficial ones can have a bigger impact. In fact, long before the invention of Facebook, Twitter and text messaging, a quote sometimes attributed to Mark Twain appeared: “A lie can travel halfway around the world while the truth is putting on its shoes.”

Recommendations

So what, if anything, can institutions do? Accuracy in an emergency may always be a challenge, but they can take some steps to minimize the confusion:

• Teach students to ask themselves if the source of an emergency warning can be trusted. Campus police? Yes. That guy you met standing next to the keg at the illegal frat party Friday night? Maybe … but maybe not.
• Clean up any problem with emergency alert mistakes. They happen often. During a training session in February at the University of North Carolina at Wilmington, a false warning message circulated campuswide: “University Police have received a report of shots fired on main campus. Immediately take shelter in a secure location away from windows, secure any doors and wait for the all-clear message. Emergency personnel are responding.” Students ran for shelter and panicked, but it was a mistake.
• Campus officials need to get it right every time. If you are going to invest millions in an emergency alert system, make sure it works and works fast -- and that people trust it.
• Consider creating a continuing public-service campaign aimed at training students and staff to post information to a campus safety message board that is continuously monitored -- one digital location among hundreds that probably already exist that someone with authority can monitor and immediately and publicly respond when students and others post information that may not be accurate. Such a system allows for a digital conversation in real time with those who are panicked, upset or misinformed. If your university-branded Facebook page or Twitter feed already has a huge following, that may be the place for everyone to meet online.