Safety & Security

Best Practices in Emergency Communications

On April 16, 2007, a little more than a year ago a Virginia Tech student went on a rampage that killed 32 fellow students. And the carnage continued.  On February 14, 2008, a Northern Illinois University student gunned down five other students before killing himself.

Shortly after the Virginia Tech shootings I wrote in this column, "Even after this month's horrific events, our campuses are five times safer than the national average.  Stated differently, even if an event like the one at Virginia Tech were to happen every year, a student is far more likely to be murdered while home on summer vacation than on campus during the academic year."  

But risk assessment doesn't cut much weight in the world of public opinion.  In the aftermath of highly publicized violent incidents like those at Virginia Tech and Northern Illinois University, higher education has come under increased scrutiny.  In particular, students, their parents, and the general public want to know about the emergency notification procedures that campuses have deployed.

And emergency alerts are more than just good PR and doing the right thing.  Under the Clery Act campuses are required to provide "timely warning" of various criminal threats.   The emergencies faced by higher education, however, go far beyond those enumerated in the Clery Act and include severe weather, class cancellations owing to infrastructure failures, bomb threats, and hazardous material contamination incidents, as well as threats from deranged individuals.  

What Can a Campus Do?
Supported by funding from the United States Department of Homeland Security and the Office of Domestic Preparedness to the State of Florida, the University of Central Florida prepared a report on Emergency Communications Systems that provides a comprehensive discussion of the requirements for emergency communications and available solutions.  They identify the three requirements for an alert system as:

1. Alert as many people and as quickly as possible in a normal condition
2. Alert as many people and as quickly as possible without power and phone service
3. Constantly deliver alerts to specific groups of people in different locations

They also discuss why is difficult to design a single alert system that can respond to all hazards and all dynamic behaviors. For example, they note that campuses include a variety of facilities.  Some do not have fire alarm intercom systems whereas others have rooms that do not have telephones.  Human factors become important.  The Florida study found that 50 percent of students and faculty would not immediately pick up a ringing phone and thus limiting the effectiveness of reverse 911 strategies.   Similarly, while 95 percent of those surveyed prefer to be notified by mobile phones, cell phone are disruptive in classroom environments, and faculty frequently request that students turn them off.  

The Florida study considered the advantages and disadvantages, as well as cost, of a number of emergency alert systems, including sirens, reverse and broadcast 911 systems, Weather Alert and FM radio schemes, and Internet-based notification.  Their conclusion: There is no single alert system that meets the all the needs of a single campus, let alone the needs of diverse campuses.

So What Have Some Real Campuses Done in Practice?
As might be expected, Virginia Tech has implemented a comprehensive emergency alert system that includes a warning siren system, voice or text message to mobile devices such as cell phones (VT Alert), e-mails to all or portions of the university community, emergency announcements on Virginia Tech's main Web page, and messages sent through the departmental reporting chain.  They are also in the process of adding emergency alert capabilities to the cable TV system that serves their residence halls and are planning to make use of several TV channels to better inform the broader Blacksburg community and off-campus students.  

 VT Alert is based upon technology from 3n (National Notification Network), which has been selected by a number of other institutions, including the University of Miami, Rochester Institute of Technology, Pepperdine University, and the University of Alaska-Anchorage.

It should be noted that Virginia Tech began planning for emergencies long before the 2007 shootings.  Their May 2005 Emergency Response Plan still forms the basis for the campus' emergency response.

According to Judy Lilly and Pat Rogers, both in Virginia Tech's Communications and Network Services unit, the student participation rate in VT Alert, which is heavily promoted on campus, is 64 percent.  Brian Nichols, chief IT security and policy officer at Louisiana State University, has observed that 40 percent to 45 percent of LSU's population have signed up for a similar program at LSU.  To increase participation in VT Alert, Virginia Tech is considering making it a requirement for annual renewal of a network ID to explicitly either opt-in or opt-out of VT Alert.

With multiple notification systems, coordinating alerts can be a challenge.  Lilly and Rogers said that the Information Technology unit will be turning on an emergency alert portal this fall that will allow University Police and University Relations, the offices authorized to issue alerts, to access all of the campus emergency alert systems through a single front end.  

An Old Fashioned Siren and Crying Wolf
Ironically, the inability (and perhaps undesirability) of getting everyone to regularly monitor our various high tech devices has led to an increased interest a much older technology--a very loud siren.  Sirens cannot convey a lot of information.  Even two or three different siren codes are enough to confuse almost everyone.  They are, however, cheap (typically $50,000 per siren) and reliable. They get people to turn on their cell phones and check their e-mail.  Some of the more sophisticated systems can also broadcast a brief audio message as well as a siren tone. Some siren vendors are: American Signal Corp., ATI Systems, Federal Signal, Hormann America, Klaxon, Sentry Siren, and Whelen.

One of the things I learned living in Lincoln, NB in the 1980s was that when the tornado siren went off you took shelter!  Not in a few minutes, but right then.  The secret of their success was that the people at the controls were very sensitive about crying wolf.  The alarm wasn't sounded when there was a just possibility of a tornado, or even a tornado sighted in another part of town.  When the siren went off and you could hear it, it meant you, and it meant now.  Campuses need to remember that when they set policies about when they will use their siren.

According to Lilly and Rogers, before Virginia Tech installed VT Alert, they had extensive discussions with faculty and students that resulted in an agreement that the system would only be used for emergencies and would only be tested once a semester.  

Cell Phones  and Wireless Laptops Alternatives
As previously mentioned, cell phones and wireless-capable laptops can be disruptive in a classroom environment, and many instructors ask that they not be turned off during class.  That introduces a need for back-channel strategies, such as a siren.  Even though the campus has a siren system, the Information Technology unit at Virginia Tech is developing a Banner Notification system for classrooms.  Normally the 36-inch by 8-inch display unobtrusively shows the time of day.  In the event of an emergency, however, a brief message scrolls across the display.  The displays are hardwired into the campus network and are powered over the network.  The campus decided to develop the system in house after pricing commercially available systems.  

What About an Urban Campus?
Virginia Tech is a relatively rural campus.  What happens on a highly urban campus such as George Washington University located in the heart of DC?  They adopted three strategies.  The first was to make use of an already existing system, Alert DC, which delivers emergency alerts on e-mail accounts, cell phones, pagers, BlackBerrys, and wireless PDAs.  Alert DC is available free to citizens of the District of Columbia, as well as individuals traveling to or working in the District.  

As a frequent visitor to DC and as someone who always carries a cell phone when traveling I decided to sign up and see how the system works.  It only took a few minutes to sign up on the Web.  I was given a choice of  alerts to which I wanted to subscribe.  The weather alert was mandatory while the transportation, utilities, government office/school closings, amber, and breaking news alerts were optional.  Within moments of signing up I received a text message alert confirmation on my BlackBerry and an e-mail alert confirmation in my Web browser.  

In contrast to VT Alert, where the alerts were campus-focused, the Alert DC warnings were generic and covered a broad range of emergencies affecting a wide geographic area and large population.  Which takes me back to my experiences with the tornado sirens in Lincoln.  I have to wonder how many alerts, unrelated to a campus, it takes to cause a student to just "tune out?"  

One problem generic to all voluntary systems is participation.  Particularly if the system generates alerts not directly relevant to students and faculty.   I suspect that is what led GW to implement a second strategy, "GW Alert."  GW Alert makes use of a commercial personal communications product for Windows-based computers from BIA Information Network that allows institutional branding and management. To use GW Alert a student or faculty member installs a program on his or her computer that runs in the background unless there is an active alert. When activated over the network, a crawler runs across the bottom of the computer screen to inform users of emergency situations.  Although the program adds extra time to a computer's startup, it is not noticeable when running.  

Finally, reflecting their urban character where a large number of undergraduates live in university housing, GW is adopting a third strategy to drive alerts to individual residence hall rooms as part of an upgrade of the campus cable TV plant to integrated voice, video, and data.  

What Should a Campus Do?
While each campus is unique and must develop an emergency alert system that is consistent with campus demographics, geography, politics, and culture, every campus can benefit from the work of others.  Review what other campuses have done.  What worked? What didn't work?  Look at their studies and plans.  Talk to vendors.  Look through the emergency alert thread in the Educause April 2008 Security Discussion Group.  

"There is no such thing as luck. There is only adequate or inadequate preparation to cope with a statistical universe." Robert Heinlein.
comments powered by Disqus