Human Organization, Vol. 73, No. 4, 2014 Copyright © 2014 by the Society for Applied Anthropology 0018-7259/14/040326-14$1.90/1

Lockdown: Applied Anthropology and the Study of Campus Emergencies Roberta D. Baer, Rebecca K. Zarger, Ernesto Ruiz, Charlotte Noble, and Susan C. Weller In October 2009, a large public university in central Florida experienced a “lockdown” in response to a reported bomb at the campus library and two other threats. We conducted an anthropological study of how students perceived the university administration’s response to these incidents. This study addresses the gaps in our understanding of university campus emergencies. We studied the situated vulnerabilities of students in different parts of the campus during the emergency. Our focus centered on student perceptions and agreement about their shared experiences using systematic research grounded in both qualitative and quantitative data collection methods. We combined a sequential technique of open-ended questions to elicit concerns more broadly (freelisting) and then used a structured interview (questionnaire) analyzed for cultural consensus to validate impressions from the open-ended interviews. Our results indicated a number of key issues for universities, including the need to focus on individuals closest to the danger, improvement of messaging systems and messages themselves, development of more comprehensive plans, and the need for university administrations to create a feeling that they are in control and concerned with student safety. This study, conducted immediately after the event, contributes new insights into the experiences of university students during emergencies and to the anthropological literatures on risk perception and disaster response. Key words: lockdown, campus emergencies, student perceptions, applied anthropology, cultural consensus analysis

A

Introduction

rmed attacks have become increasingly common in our society, ranging from gunmen in shopping malls, movie theaters, public schools, and on university campuses. We focus here on the issue of threatened attacks on university campuses, drawing upon the 2009 “lockdown” at Mid-State University (a pseudonym), a large public university in Central Florida, following a bomb report and two other threats. We conducted an anthropological study of student perceptions of the administration’s response to these incidents. Our study advocates attention to student perceptions through the use of a mixed methods design. In this study, we comRoberta D. Baer is a Professor and Rebecca Zarger an Associate Professor in the Department of Anthropology, University of South Florida (USF). Charlotte Noble and Ernesto Ruiz are current and recent Ph.D. students, respectively, in Applied Anthropology at USF. Susan C. Weller is a Professor in the Departments of Preventive Medicine and Community Health and Director of the Research Program in the Department of Family Medicine, University of Texas Medical Branch, Galveston. The authors appreciate the assistance and support of Brent Weisman, Kristina Baines, Jennifer Hunsecker, Gene Cowherd, Jennifer  Bosson, Karla Davis Salazar, Robert Tykot, Karen Smid, Chris Roos, Michael Bowen, Mark Hafen, Amy Stuart, Kathleen Heide, Michael LeVan, David Merkler, Edward Kellner, John Hodgson, Heide Castañeda, A. J. Faas, Mark Moberg, and Nancy White, who suggested the study.

326

bined a sequential technique of open-ended questions to elicit concerns more broadly (free-listing) followed by a structured interview (questionnaire) analyzed for cultural consensus to validate impressions from the open-ended interviews. To our knowledge, there have not been any applied anthropological studies of incidents of school violence. Yet, there exists a wide range of anthropologically based literatures of potential use in understanding these situations. These include work on risk in general, and more specifically, in relation to disasters and the response to them. We complement these perspectives with other social science research focusing specifically on violence on university campuses. Bringing together these previously disparate literatures not only makes possible an understanding of the particular event discussed here, but also creates a framework for future anthropological studies and contributions to understanding and preventing situations of this type.

The Incident The threats all occurred on October 5, 2009. At 1:36 p.m., a mentally disturbed former student called in a threat in the third person, reporting that a suspicious individual dressed all in black had a gun and a bomb near the library (Polson 2009). The call went to the country sheriff’s office which then contacted the university police. At 1:43 p.m., university HUMAN ORGANIZATION

police charged into the library with guns drawn, searched the area, evacuated students, and closed the entrance to the building. Other university police armed with rifles patrolled the exterior of the building. At 1:49 p.m., text messages were sent out to alert students of the gunman at the library, and sirens went off around the campus. The message was “Armed intruder on campus. Stay inside. Lock doors.” Students who tried to enter the library when the sirens sounded were told to leave. However, not all of the students in the library received the alerts. Those in the library basement did not have cell phone reception, so they had no idea anything was wrong. One student elsewhere in the library reported getting a message from a friend outside the library who relayed to her that “there’s a gunman in the library” (Polson 2009). Other students on the fourth floor of the library were also unaware that the armed person might be in the library. This information was released at 2:06 p.m.; a second text was sent, saying officers were on the scene, and to “Avoid the area and report anything suspicious” (Polson and Withrow 2009). The students on the fourth floor then locked the door to their study room and called 911. Police arrived and escorted these students out of the library (Polson 2009). The person who called in the threat went to the library and joined the crowd outside. Detectives located and arrested him 48 hours later through postings he made on Facebook. The next incident occurred around 2:30 p.m., when a passenger on a campus shuttle bus announced, “I’m the bomber, I have a bomb right here.” After university police were contacted, they halted all campus shuttles and removed everyone. A text message was sent warning of “A separate report of a suspect on a [campus shuttle bus] …possibly armed. Avoid area and entire campus on alert” (Polson 2009; Polson and Withrow 2009). Fearing a hostage situation, campus police also pulled over the particular bus where the problem had been reported. This suspect was taken into custody at 2:52 p.m., but when his backpack was detonated by a robot around 4:30 p.m., it was found to contain only school supplies (Polson and Withrow 2009). The third incident occurred at 3:20 p.m. when the following text message was sent to all on campus: “[City (a pseudonym)] Campus—White male subject seen in the Cooper Hall area in black tank top, cowboy hat carrying black puppy, and a large hunting knife—Officers en route.” Another text message followed: “[City] Campus—Clarification regarding multiple alerts. Entire [City] Campus remains on alert these for the incidents reported this date. Remain inside.” At 3:30 p.m., police took that suspect into custody but later released him. At 4:23 p.m., the lockdown was lifted for part of the campus; the message indicated that “students could resume normal activities.” By 5:12 p.m., the rest of the lockdown was lifted, and all people at the university were allowed to leave the buildings in which they had been all afternoon (Mid-State administrator, personal communication, Sept. 2011; Polson and Withrow 2009; Zayas et al. 2009). VOL. 73 NO. 4, WINTER 2014

In the wake of this event, we felt it was critical to gather information about student perceptions of the university’s response, their personal safety, and possible improvements in procedures. Our offer to conduct such a study was accepted by the university administration, which provided the researchers with funding for the study. Our mixed methods design began with open-ended interviews allowing us to capture student perceptions in their own words. Their verbatim quotes were the basis for the questionnaire for the second phase of data collection. This integrated approach allowed us to explore a broad range of experiences in the first phase but then verify the salience of themes in a more systematic second phase. Cultural consensus analysis (CCA) was used to verify theme importance to a diverse sample of students. CCA also allowed us to see if students or groups of students shared similar perceptions of the event. The findings of this study contribute new insights into the experiences of university students during emergency events and to the anthropological literatures on risk perception and disaster response.

Background There are a number of key issues related to understanding emergencies on university campuses. These include the complexity of multi-building campuses versus single building campuses, student perceptions of the handling of the situation, spatial and temporal dimensions of risk, as well as mobile phone and social media communication, their vulnerability to weak signals, and variable adoption. These factors predict differently-situated recipients of information—based on power (administration versus students) and location (near the risk, near signals that work in the emergency situation). The risk of an active shooter appearing on a university campus is actually considered very low (Burruss, Schafer, and Giblin 2010; Fox 2008). However, incidents on university campuses have several important differences from such events at elementary/high schools (Fox and Savage 2009). In contrast to the campuses of most universities, whose many structures are dispersed over a sizeable area, elementary/high schools are usually comprised of fewer buildings at a single site. Locking down buildings may be appropriate in the elementary/secondary situation, but at a university it runs the risk of leaving both the shooter and unprotected potential victims outside (Fox and Savage 2009). The multi-building context of universities may create a variety of spatially-situated vulnerabilities in an emergency that might be responded to with a variety of strategies. “Almost every incident that occurs on a university campus has a spatial component”; spatially oriented data are “ideal for allowing a university to address, analyze, and assist administrators in making correct decisions” (Murchison 2010:84). The location of the individual relative to the “problem” may be important in understanding perceptions of and responses to these emergencies. Since university students may be some distance from the specific danger (Fox and Savage 2009), we focus here only on the literature on emergency events at university/college campuses. 327

We know very little about student perceptions of university responses to these types of situations. Unlike the campus emergency studied here in which there was no actual attack, and no resulting deaths or injuries, many other incidents of this type have had such consequences. The most serious of these resulted in the death of 33 people on the campus of Virginia Tech (VT) in 2007. Among the criticisms of the handling of that event was that over two hours passed between the first and second attacks, during which students and others on the campus were not notified (Hauser and O’Connor 2007). Little of the literature on campus emergencies deals with perceptions of how universities handle such events, and even less is based on student perceptions (Davies 2008; Giblin, Burruss, and Schafer 2008; Piotrowski and Guyette 2009). Such omissions are critical, for as one researcher notes, “The attitudes, beliefs, and experiences of [this] population [have been] largely ignored in policy discussion occurring after the VT and NIU [Northern Illinois University] tragedies…. [Giving] a voice to this population is of importance for the purposes of both research and policy formulation” (Burruss, Schafer, and Giblin 2010:31). Where this has been done, it is often in the form of an individual’s perceptions (Figley and Jones 2008). And some studies of student reactions focus on topics other than the university’s reaction to the threat (Kaminshi et al. 2010; Vicary and Fraley 2010). Two studies of student perceptions of university response to a campus emergency are notable. One entailed a quantitative assessment of Northern Illinois University (NIU) student perceptions of crime, perceptions of risk, experiences with victimization, and attitudes toward campus safety initiatives (Burruss, Schafer, and Giblin 2010). But while there had been shootings at NIU in 2008, the study did not specifically ask students to evaluate university responses to those incidents. The study found that students were generally aware of university recommendations for responding to a campus shooting and felt they would be able to respond appropriately to incidents of this type. They were also “modestly confident in the ability of faculty and staff to recognize dangerous students and to take appropriate action in the event of a campus-based critical incident” (Burruss, Schafer, and Giblin 2010:vii). The authors note, however, that “whether these perceptions would actually prove true in real situations could not be determined” (Burruss, Schafer, and Giblin 2010:30). Another study addressing the campus response to a gunman at a large public university in the Midwest collected qualitative data from students, as well as others on campus, including administration, faculty, staff, media, etc. (Asmussen and Creswell 1995). Findings included denial that the situation was actually dangerous, fear of the situation reoccurring if the alleged assailant made bail, and concerns about student safety on campus. However, as no quantitative data were collected, we cannot be sure how widely these perceptions were shared. Perception of risk is critical in understanding responses to emergencies, and anthropologists have stressed that such perceptions must be situated in a community and cultural 328

context (Boholm 2003). Additionally, some groups may respond to some dangers and ignore others (Douglas and Wildavsky 1983). Having knowledge or information does not necessarily imply that appropriate actions will be taken (see the quick response literature on hurricane evacuations [Natural Hazards Center 2006]). A sense of being in control (or not) is also an important part of risk (Thompson and Dean 1996); fear of losing control or being powerless in a particular setting can affect interpretation and actions toward perceived risk. Related to control is vulnerability, often discussed in the context of social aspects of disasters. There are vulnerable situations “which people move into and out of over time” (Wisner et al 2004:15); as such, vulnerability is situated and related to time and space (Cutter, Boruff, and Shirley 2003). It follows that disasters are also experienced differently by different groups, related to demographic and other differences (Oliver-Smith 2011). Finally, local knowledge, often ignored by disaster management, may influence the extent to which the recommendations of experts are trusted and followed (Oliver-Smith 2011). A final issue related to emergency situations is electronic communication. Social networking sites (Facebook, Twitter) and mobile device text messaging have become integral to information sharing during emergencies of all types (Merchant, Elmer, and Lurie 2011; Palen et al. 2009). Text messaging systems are common ways for university administrations to communicate during crisis events, but opinions differ on the value of this technology (Sullivan, Hakkinen, and Piechocinski 2009). Concerns include false messages, messages leading students and faculty toward instead of away from danger, and administrators sending out alerts regardless of the severity of the danger, resulting in overuse (Bambenek and Klus 2008). Text message systems should be highly reliable, have “excellent access control,” and high-speed delivery, but they rarely meet these criteria (Bambenek and Klus 2008). Many students must turn off cell phones during classes, and basement classrooms in older buildings may not receive signals (Fox and Savage 2009). Additional challenges arise from communicating with students, staff, and faculty with disabilities, and language barriers that may affect international students (Sullivan, Hakkinen, and Piechocinski 2009). Many text and instant messaging systems as a means of notification depend on “opt-in for participation” (National Campus Safety and Security Project 2010), although it has been suggested that registration be mandatory, with use of social networking sites to educate about the system (Sullivan, Hakkinen, and Piechocinski 2009). In summary, there remain significant gaps in our understanding of campus emergencies. To address these, we studied the situated vulnerabilities of students in different parts of the campus during an emergency situation. Our focus was on student perceptions and agreement about their shared experiences using systematic research grounded in both qualitative and quantitative data collection techniques (free-listing and cultural consensus analysis). HUMAN ORGANIZATION

Methods This study employed a two-phase design of data collection. Qualitative data can uncover the range of opinions about and responses to such a situation, while systematic, representative quantitative data can determine how widely those opinions and responses are shared and are essential for policy recommendations. The first phase consisted of qualitative interviews capturing verbatim student perceptions and experiences of the emergency just after the event. The goal of the analysis of this phase was to create a structured questionnaire incorporating the qualitative data (the specific responses made in the qualitative interviews) into a true/false questionnaire. This enabled determination of whether other students agreed with a series of specific statements. Phase one allowed us to gain insight into individual experiences, while phase two provided an opportunity to assess agreement among students. Interviewing began immediately after obtaining Institutional Review Board approval (a week and a half after the incidents). Our goal was to capture students’ immediate responses and perceptions as soon as possible following the event. Since the literature strongly suggested the importance of distance from such an event, we incorporated that issue into our sampling design and analysis. Free-listing, a rapid and effective method used in qualitative research (Weller and Romney 1988), was chosen to collect initial qualitative information. Individuals are asked to list or provide as many answers to a particular question as they are able to think of at the time. The collection of extensive information from each person, rather than single responses, allows for topics to reach the saturation point quickly and for meaningful results even with small sample sizes. (Bernard 2006). Guest, Bunce, and Johnson (2006) suggests that often only a dozen or so people may be necessary to reach saturation. During October 15-22, 2009, five research assistants conducted in-person interviews and collected free-listed responses in multiple locations on the campus at different times of day and days of the week. A stratified convenience sample (n=35) of undergraduate and graduate students who were on campus during the lockdown focused on gaining representation of students across campus. They were asked: 1. 2. 3. 4. 5. 6. 7.

How did you find out what was going on? How did you react to the event, and what did you do? What kinds of things should you have done? What were you worried about? What should the university administration have done? What should students do when such an event happens? What are all the risks to the university of a situation like this? 8. What other issues do you think there are related to this incident? 9. What are your feelings about the incident and the university response? VOL. 73 NO. 4, WINTER 2014

Responses were recorded verbatim and collated into a cumulative list of responses, which were then grouped into thematic categories of shared responses for each question and ranked for frequency. From the open-ended responses, we created a structured questionnaire with 175 questions (Tables 2-10) staying as close to verbatim comments as possible. Demographic data (level of the class in which the questionnaire was completed, census category, gender, year in school, residence [on/off campus], age, major, and ethnic background, as well as where the respondent was when they first found out about the situation) were also collected. The questionnaire was administered in person in the three months immediately following the event (November 2009 to January 2010). A convenience sample of diverse undergraduates who were on campus in multiple departments during the lockdown completed the questionnaires in classroom settings. Graduate students were accessed through graduate student organizations. The questionnaire results were analyzed with the cultural consensus model (CCM) (Batchelder and Romney 1988; Romney, Weller, and Batchelder 1986; Weller 2007). The CCM provides statistical criteria for inferring if there is a shared pattern of responses in a sample of respondents to see if they had a shared perception of events. CCM is similar to simple aggregation techniques (e.g., examination of the modal responses) but extends those techniques by providing a confidence level for each answer and also estimates each person’s correspondence to the group answers. We analyzed the responses to all 175 questions for all respondents together and then by subgroups based on location of respondents during the incident (inside the library, outside near the library, inside near the library, inside far from the library, or outside far from the library). We used the formal CCM implemented in Anthropac (Borgatti 1992) and classified answers using a stringent level of confidence (p>.999).

Results Qualitative Data The sample (N=35) for the free-listing was 53 percent female, 43 percent “White,” with a mean age of 25 years. The most frequently mentioned responses from the open-ended interviews were used to create the structured questionnaire (Tables 2-10).

Quantitative Data Demographics Diversity within the sample (N=566, mean age=20 years) reflected that of the campus population overall (31,352 students). However, a comparison of majors between the sample and overall university enrollment indicated students majoring in Business were somewhat underrepresented as were White students, while Native American students were somewhat overrepresented (Table 1). Nevertheless, considering the 329

Table 1. Demographic Information on the Sample Relative to the University as a Whole Characteristic

Study Sample (n=566)

University Population (n=31,352)

Female Gender 51% 57% White 57% 64% Black 10% 13% Hispanic 12% 16% Asian 9% 7% Native American 11% 1% College of Business Major 6% 18%

logistics and constraints on sample selection, we consider the sample to be broadly representative of the student body. Responses for the questionnaire were analyzed using CCA. The consensus model fit the responses for each location (eigenvalue ratios range 3:1-6:1), although the average level of agreement was low (average competence range .44-.49). Because agreement was relatively low, we also checked to see if there were any meaningful agreement patterns in addition to the shared consensus pattern. The extent of agreement (competence levels) did not differ by gender, location at the time of the incidents, and whether students were inside or outside of university buildings. However, we did find differences in patterns of responses based on location at the time of the incidents. We present

results for the five subgroups representing the different areas on campus at the time of the lockdown (IL—in the library N=29, OL—outside near the library N=68, IN—inside near the library N=157, IF—inside far from the library N=138, OF—outside far from the library N=90). How did you find out what was going on? (Table 2) Most students received information about the situation through university text messages and by hearing the sirens and alarms. Other students talking about it and sharing texts were also important sources. However, those in the library were less likely to have received texts from friends or to have heard the announcements. Some students saw the police and/ or heard the helicopters; many also later saw it on the TV

Table 2. How did you find out what was going on? IL OL IN IF OF Y Y Y Y Y I got a text message from the university. Y Y Y Y Y Other students were talking about it. Y Y Y Y Y I heard sirens. Y Y Y Y Y I heard an alarm. Y Y Y Y Y I saw it on TV news later. Y Y Y Y Y I heard helicopters. Y Y Y N Y I saw cops. - Y Y Y Y I got texts from friends. Y Y Y Y - Other students shared the university texts sent out with me. - Y Y Y Y I heard announcements. Y Y Y Y N Later we checked a news webpage. - - Y - N I got a phone call from a friend. Y - - N - I could see lots of students standing around. - - N N N I got an email from the university about it. N N N - N I checked Facebook. N N N N N I went online to the university website. N N N N N Teachers talked about it in class. N N N N N I got an email from a friend. N N N N N I heard a car outside saying please stay inside. N N N N N I looked at those TV screens that are in most of the buildings. Y = agreement with the statement N = disagreement with the statement - = no consensus

330

HUMAN ORGANIZATION

news. Interestingly, no group reported their teachers discussing it in class, and Facebook and university emails were not noted as initial sources of information. Finally, there was no use of the TV screens in most of the buildings on campus nor did students consult the university website.

confused and did not feel safe but took the potential threat seriously. Reactions typically did not include consulting Facebook or major news outlets, and those near the library were not calm. All of the students felt inconvenienced, but none reported leaving campus.

How did you react to the event, and what did you do? (Table 3) The most common response was to call/text friends and to be concerned about whether “it was real.” Students were

What kinds of things should you have done? (Table 4) Most of the students felt like they did what they should have done under the circumstances. However, those in and near the library itself felt that “no one knew what

Table 3. How did you react to the event, and what did you do? IL OL IN IF OF Y Y Y Y Y I thought, “This is real.” Y Y Y Y Y I was inconvenienced by it. Y Y Y Y Y I called/texted friends. - Y Y Y Y It’s just part of the nature of this country; crime is everywhere. - Y Y Y - I called my family. N N N Y Y Everyone else was calm – the library was far away. - N Y N Y I was annoyed at the odd timings of text alerts and emails; emails came so much later than text messages. - N Y Y - I stayed where I was inside. - - Y N Y I went outside. N N N Y N When doors were locked, I felt safe. Y N N N N I was confused about what to do. - N Y N - I think we were so stressed that it made some things seem funny. N Y N - - I checked Fox News. - N N N N I was scared. N - N N N I left the building I was in; I had to get my work done. N N N - N I checked CNN. N N N - N I knew what the alarm meant. N N N N N I thought, “again?” – I thought it was another false alarm. N N N N N I had trouble understanding what was going on because my English is not good. N N N N N I thought it was a drill. N N N N N I was exasperated; the university is always blowing stuff out of proportion. N N N N N I locked up and closed everything in my on-campus apartment. N N N N N I continued with what I had to do and thought, “Ok fine, there’s a gunman, whatever, I have to finish what I am working on.” N N N N N I left the university. N N N N N My class continued, so I stayed in my class. N N N N N I ignored it. N N N N N I stayed glued to my Facebook page; lots of people were updating using Facebook mobile so it was much quicker than seeing what the news had. N N N N N I checked CSNBC. N N N N N I saw people running to get inside a classroom and followed them. N N N N N I turned my cell phone to vibrate mode (I was in class). N N N N N I told people they couldn’t leave the building; if they did it was their responsibility. N N N N N I locked the door. N N N N N I tried to stay away from windows. N N N N N I told friends to get out of the library. N N N N N I kept updating people without cell phones. N N N N N Our teacher told us to evacuate the building, so we did. N N N N N I was locked out, so I hung out around the library. N N N N N I made a plan of where to hide if it was like Virginia Tech. Y = agreement with the statement N = disagreement with the statement - = no consensus

VOL. 73 NO. 4, WINTER 2014

331

Table 4. What kinds of things should you have done? IL OL IN IF OF Y Y Y Y Y Nothing, other than what I did. Y Y Y Y Y I did what I should have done. Y Y Y N - No one knew what to do. - N N N N I needed clear directions as to what to do. N N - N N I should have stayed inside. N N N N N I should have worried more. N N N N N I should have gone inside a building. N N N N N I should have locked the door so people couldn’t have come in and out. N N N N N I should have known how to get home without the bus. N N N N N I should have signed up to receive text messages. N N N N N I need to buy a gun to protect myself. N N N N N I should have left campus. N N N N N I should have ignored it all. Y = agreement with the statement N = disagreement with the statement - = no consensus

to do,” and those in the library would have liked more direction as to what to do. What were you worried about? (Table 5) The students were concerned because they heard sirens and were indeed worried about what was going

on. There was also confusion about whether they should go to class. Those in and near the library were concerned that they might be directly affected and were particularly worried because they did not think the situation was under control. Students inside the library found the texts recommending staying inside to be confusing, and only

Table 5. What were you worried about? IL OL IN IF OF Y Y Y Y Y I heard the sirens. Y - Y Y - I worried because many students were showing a lack of common sense. N N Y Y Y I thought it was under control, so I wasn’t worried. N N N Y Y I figured they won’t come all the way to the part of campus where I was. - - Y N N I was concerned if it was a bomber or gunman – was staying inside a good idea? N N Y N N I came to campus because I thought my class would meet as usual. Y N N N N There were confusing texts – they said to stay inside, but they evacuated us. - Y N N N I heard the speakers and could understand them. N N N - N I knew that my class was cancelled. - N N N N I thought about Columbine. N N N N N Nothing, I wasn’t worried at all. N N N N N I felt relieved when I heard that this happens all the time. N N N N N There is no cell phone signal in the area where I was. N N N N N I was more scared of getting hit by a cop who was driving on the grass than of getting shot. N N N N N I was concerned because no one was checking IDs in the library. N N N N N Inside my class, you could hear the loud speaker. N N N N N I couldn’t get to a computer to take an online test. Now I have to take the test under make up rules. N N N N N The doors of many of the classrooms cannot be locked. Y = agreement with the statement N = disagreement with the statement - = no consensus

332

HUMAN ORGANIZATION

Table 6. What should the university administration have done? IL OL IN IF OF Y Y Y Y Y They should have let students know that they wouldn’t be penalized for being late or absent to class. Y Y Y Y Y The lockdown was too long without knowing what was going on – they should have kept updating information more continuously. Y Y Y Y Y They should have broadcast what to do over the loud speakers. Y Y Y Y Y They should block traffic coming in to campus. Y Y Y Y Y They should be more careful with message wording; some of it was ridiculous or looked like it was a prank. Y Y Y Y Y They should keep professors in the know, so they can take a leadership role. Y Y Y Y Y The building secretary or staff should provide more information about the situation. Y Y Y Y Y They should have a more comprehensive plan. Y Y Y Y Y It seems like they are just covering themselves with the alerts, not actually getting out important info. Y Y Y Y Y They should have sent out an email explaining everything after the fact. Y Y Y Y Y A big problem is that one of the main ways they notify us is through our phones, but if we’re in class we’re supposed to have them turned off, so how can they reach us? Y Y Y Y Y There should have been more information on the university website. Y Y Y Y Y I knew which library they were talking about. - Y Y Y Y The cops should have locked doors of buildings surrounding the library. Y - Y Y Y They should not lockdown all the buildings – bombs could have been inside. Y Y Y - Y We needed more text messages. N Y Y Y - They did their best/what they should have. - - Y Y - Every single text message should have said, “You should stay locked inside.” N - N Y - The text messages told us what to do – they didn’t just say there was a threat, but they also told us how to respond to it, so there was no confusion. N - N - - The alarms can be heard in all areas. - N N - N They should do more with email. - N N N N There should have been a phone chain set up. - N N N N The cops didn’t search the building thoroughly. N N N N N The text messages gave the impression that it was ok to walk around as long as it wasn’t near the library. N N N N N They should run drills. N N N N N There were too many text messages. Y = agreement with the statement N = disagreement with the statement - = no consensus

those outside the library could hear the speakers and understand them. What should the university administration have done? (Table 6) Many aspects of the university responses were criticized by the students, particularly the lack of specific direction as to what to do, the need for more updated information in general, and specifically, whether classes were cancelled. Students felt that traffic coming into campus should have been blocked and that faculty and staff should have had more information and provided more leadership. There were also concerns that bombs could have been inside buildings that were locked down. It was felt that the text messages were not clear enough about how to respond, their wording was not clear enough, and the number of text messages was insufficient. Overall, VOL. 73 NO. 4, WINTER 2014

students felt that “[the Administration was] just covering themselves with the alerts, not actually getting out important information.” It was also agreed that “[the Administration] should have sent out an email explaining everything after the fact.” Drills and phone chains were not felt to be useful by the students. Students also agreed that “a big problem is that one of the main ways they notify us is through our phones, but if we’re in class, we’re supposed to have them turned off, so how can they reach us?” And those in the library did not feel that the university “did its best/what it should have.” What should students do when such an event happens? (Table 7) Students felt that appropriate responses were to take the event seriously, follow instructions, and learn how to respond to events of this type in advance. However, they were 333

Table 7. What should students do when such an event happens? IL OL IN IF OF Y Y Y Y Y We’ll do what we’re told, but someone needs to tell us. Y Y Y Y Y Students should take it seriously. Y Y Y Y Y People in the library should have left. Y Y Y Y Y You should get as far away from the building/threat as possible. Y Y Y Y Y Prior to such an event occurring again, students should familiarize themselves with emergency plans. Y Y Y Y Y You should leave campus if possible. Y Y Y Y Y I’m not sure if you should let people into your class – you don’t want to put yourself at risk or leave the people outside in jeopardy. Y Y Y Y Y You should stay where you are, unless you are where the bomber or gunman is. Y Y Y Y Y You should be able to assume that all classes are cancelled. Y Y Y Y Y You should follow the instructions of people who are supposed to know – like someone from the university who you know you can trust. Y Y Y Y Y You should wait for an “all clear” message. Y N Y Y Y We can only do what we’re told to do. N Y Y Y Y Students should look to faculty/supervisor to see if they are taking it seriously, and that’s how they’ll regard it. - Y Y Y Y You should stay where you are when an alarm sounds – unless you are outside – I’m not sure where those people should go. N Y N Y - You should decide for yourself if you want to listen to the alert. - N N Y Y If people are outside, let them in but don’t let a gunman in, just other ok people. - - N Y - I had enough information to know what to do. N N N Y N I was clear on the latest information. N N N - N I actually learned more from Facebook and Twitter than I did from the university, because people on Facebook that were over at the library kept updating their status. N N N N N When something like this happens, students should ask for their tuition money back. N N N N N I knew who was in charge – whose directions to listen to. Y = agreement with the statement N = disagreement with the statement - = no consensus

not clear on who was in charge or what to do. And students in the library were confused as to whether they should stay where they were; “you should stay where you are unless you are where the bomber or gunman is.” They were also not sure of what people who were outside should do; “you should stay where you are when an alarm sounds—unless you are outside—I’m not sure where those people should go.” Nor were they clear on who was safe to let into a building. What are all the risks to the university of a situation like this? (Table 8) Regardless of location, all respondents agreed upon a number of important risks that threatened the university’s integrity. Many students felt that false alarms and overreaction might affect university reputation, credibility, and enrollment. There was also widespread sentiment that “a false alarm like this creates a ‘boy who cried wolf’ syndrome.” What other issues do you think there are related to this incident? (Table 9) Sub-optimal communications and issues of open access to campus were also noted by all of the students. Many felt 334

that public safety should be more of a priority for the university. They also agreed that “the university needs to have a plan and not just leave it up to the police.” What are your feelings about the incident and the university response? (Table 10) Students felt that the location of the university, “in the middle of the ghetto,” puts it at constant risk. Another issue was the lack of resolution given to students; they also wondered why “the school didn’t show us anything about safety.” Students also did not feel safe on campus at night. Among those in and close to the library, there was a perception that the administration had no idea what they were doing; “It was like this was the first time, but it wasn’t; this stuff has happened before.” All in all, those in and near the library did not feel that “those in charge had the situation under control.”

Discussion and Conclusions These findings contribute to the very small body of literature on this aspect of campus emergencies. There are few studies based on student perspectives of university responses HUMAN ORGANIZATION

Table 8. What are all the risks to the university of a situation like this? IL OL IN IF OF Y Y Y Y Y The bad press of a false alarm might affect the university’s image. Y Y Y Y Y A false alarm creates the perception that the administration has no idea what to do in a situation like this. Y Y Y Y Y The university can get a poor reputation for handling incidents poorly. Y Y Y Y Y If it had been real, there could have been injuries or loss of life or a hostage situation. Y Y Y Y Y If it had been real, the university could get sued if something happens to any students. Y Y Y Y Y There’s the risk of being “another Virginia Tech.” Y Y Y Y Y Students/parents could lose trust in the university because of false alarms. Y Y Y Y Y In a real situation, there could be damage to the buildings. Y Y Y Y Y A false alarm like this affects our ability to do our schoolwork. Y Y Y Y Y A false alarm like this wastes our time Y Y Y Y Y A false alarm like this creates a “boy who cried wolf” syndrome. Y Y Y Y Y There could be car accidents from panicking students trying to leave campus. Y - Y Y Y The university could lose students who wouldn’t want to come because of fake alerts. Y N Y Y N The credibility of the university is at stake; overreaction might cause students to leave the university. Y = agreement with the statement N = disagreement with the statement - = no consensus

Table 9. What other issues do you think there are related to this incident? IL OL IN IF OF Y Y Y Y Y I am concerned that we have old crappy buildings that don’t have sirens or any communication between us and the rest of the school. Y Y Y Y Y There needs to be communication devices inside the classrooms. Y Y Y Y Y There was bad communication/dissemination of information. Y Y Y Y Y Campus is too open; anyone can come on campus. Y Y Y Y Y The university really needs to keep incidents like this in check. Y Y Y Y Y The university needs to figure out how to keep this from happening in the first place. Y Y Y Y Y This research is a good idea; after an event you evaluate what happened and try to figure out if how it was handled can be improved. Y Y Y Y Y False alarms like these mean there’s money wasted on investigating false reports. Y Y Y Y Y False alarms like these cause students’ time to be wasted. Y Y Y Y Y Situations like these set up a mood of fear with all these false reports like the campus is full of crime or something. Y N Y N N One big issue is the lack of security on campus - - Y Y - It’s too easy for people to get guns; the Untied States needs gun law reform. - Y Y - N I got the texts a lot later than other people – the university should make sure we all get them promptly. N N N - N The TV screens that are in the buildings were helpful. N N N N N The fast communication may have led to overreaction. N N N N N I didn’t have my cell, so I had no idea what was happening. Y = agreement with the statement N = disagreement with the statement - = no consensus

VOL. 73 NO. 4, WINTER 2014

335

Table 10. What are your feelings about the incident and the university response? IL OL IN IF OF Y Y Y Y Y Y - N N

Y Y Y Y - Y - N N

Y Y Y Y Y Y - N N

Y Y Y Y - - Y Y N

Y Y Y Y Y - Y - N

Students need to feel that public safety is more of a priority for the school. We’re in the middle of the ghetto. I’m surprised it doesn’t happen more. There was no resolution given to students. The university needs to have a plan and not just leave it up to the police. I wondered why the school didn’t show us anything about safety. I felt like the administration had no idea what they were doing. It was like this was the first time, but it wasn’t; this stuff has happened before. I feel safe on campus at night. I felt like the people in charge had the situation under control. I felt like the university basically abandoned us.

Y = agreement with the statement N = disagreement with the statement - = no consensus

and none involving a mixed methods design. We used relatively carefully selected samples, drawing upon data collected as soon as possible after the actual event. Although no injuries or violence occurred during this incident, the experiences of students resonated with much of the existing literature on responses to campus emergencies. Students see themselves as well informed about what a “lockdown” means on a college campus. However, they expect faculty and staff to know about the specifics of their campus and how to respond to different types of events. In other campus emergencies, students reported confidence in faculty and staff (Burruss, Schafer, and Giblin 2010), but that confidence was not seen in the event studied here. This need for faculty/staff competence in how to respond in these situations is also expressed by other researchers (Fox and Savage 2009; National Campus Safety and Security Project 2010). It is also important that students in this study felt they had done exactly what they should, regardless of what that was: doing nothing, checking Facebook, or staying inside and locking doors. This is in contrast to the denial that the situation was dangerous, noted by the students in Asmussen and Creswell’s study (1995). But the concerns about student safety in general they described were present in the responses of the students we studied. Based on previous studies (Palen et al. 2009), we expected social media to play a larger role in the events than they did. Text messages, word of mouth between students on campus, and the loudspeaker system were much more important than social media. Although most students have cell phones, many instructors require that cell phones be off during class, as noted in other situations (Fox and Savage 2009), and there are areas of buildings where cell phone reception is minimal, at best. In this case, that included the basement of the library, the building in which the bomb was reported to have been located. Another finding, paralleling Bambenek and Klus (2008), is that the content of messages is just as important as 336

the ability to reach as many students and staff as possible. The content of messages in this incident were unclear and did not specify what students/faculty/staff should do, so they were not as effective as they should have been, regardless of who they reached. There were also a few items on the questionnaire that did not receive a majority consensus response but which may also represent important issues. Some students agreed that they “had trouble understanding what was going on because my English is not good.” If text messaging is the primary means of communication, accessibility for all needs to be a consideration (Sullivan, Hakkinen, and Piechocinski 2009). Another issue is the extent to which electronic media will be available under all emergency situations. Power failures may accompany emergencies (indeed one occurred just days before the lockdown), affecting electronic communication and door-locking systems. Finally, in many emergency situations, cell phone networks become clogged. Proximity to the actual events (Murchison 2010) (i.e., spatial aspects of risk [Cutter, Boruff, and Shirley 2003]) did shape student experiences. Students near the library felt that “no one knew what do”—perhaps the exact opposite that university administrators might hope for in this situation. Those students in and near the library were the most critical of the university’s handling of the situation. Location also played a role in whether students heard alarms, regardless of whether they knew what they meant or not. Finally, the difference between a “lockdown” in a an elementary/high school setting and a large university campus setting with multiple buildings—which may be impossible to effectively lock down—is important to consider (Fox and Savage 2009; Piotrowski and Guyette 2009). One building may be locked; many may be locked; but the question remains—what happens to those who are outside when the lockdown is called? Twenty eight percent of students who completed the questionnaire were outside a building at the time of the lockdown. Student responses indicate a lack of HUMAN ORGANIZATION

clarity about what to do under these circumstances. They were particularly unsure about how to handle people outside buildings—how should they decide who were safe to let in? The dilemma was balancing risk to themselves versus leaving others at risk. With regard to larger issues of social aspects of risk (Boholm 2003; Douglas and Wildavsky 1983; Vaughan and Nordenstam 1991), students felt that their own responses were appropriate to the risks they perceived. But they were very critical of the responses of the university and did not feel the institution understood the risks students faced. Students perceived those in control actually had little control, and this fed their anxiety (Thompson and Dean 1996). The students expected the administration and faculty to be able to regulate the situation and, as it became clear to them that this was not the case, they perceived even more risks. Our findings suggested a number of policy recommendations for universities in addressing campus emergencies. The focus needs to be on those individuals closest to the danger. Messaging systems should be improved, and messages must be more carefully crafted. All students and faculty should receive messages at the same time. Universities need a comprehensive plan for handling these situations, including issues of access to campus, exit plans, and cancellation of classes. University administrations must create a feeling that they are in control and concerned with student safety. Closure, informing students of what exactly happened and how it was dealt with, is necessary after an event of this type, and research on campus safety issues should be sponsored regularly. There are also additional new media tools that may be of importance, such as those which use GPS and allow those involved to share information about what is going on in a specific location (Merchant, Elmer, and Lurie 2011). This would be particularly important in events that affect large areas, such as university campuses. Our results were viewed as extremely valuable by members of the university administration that commissioned the study. Many of our recommendations appear to have been incorporated into university emergency policies. Current methods used to communicate in an emergency include: sirens, the university website, a 1-800 number, digital display boards throughout campus, emails to faculty/staff and students, social media (Facebook and Twitter), a text alert system to approximately 65,000 subscribers, and media outlets. Trainings for campus police and university leadership are conducted, and efforts have been made to improve messaging in emergency situations. Message templates were created for the most likely crisis scenarios, which allow the message sender to “fill in the blanks.” They assure that specific message elements are included: (1) where the issue is occurring (i.e., campus, building, etc.), (2) what action the recipient should take (i.e., take cover, evacuate, etc.), and (3) a website to go to for more information. Additionally, the university has integrated the multiple electronic communication channels so the message sender can compose and send from a single administration console. Campus digital displays, text messaging, and main VOL. 73 NO. 4, WINTER 2014

website notices can all be sent from this single console allowing for a consistent message (Mid-State administrator, personal communication, Aug. 2013). Our findings also contribute to larger anthropological ideas about perceived risk, conflict, and control. Hoffman and Oliver-Smith (1999) have suggested the possibility for diverse aspects of hazards and risk calculation; we noted the extent to which the position of relative powerlessness of student in the university social structure was important in their perception of the hazards and risk calculation of the campus emergency situation. University policies created by the powerful (administrators) may not take into account the different perceptions of students. Understanding student perceptions is important as they will influence students’ actions in subsequent events of this type. Administrators can mandate, but students must cooperate. One policy implication is that students should be included in university committees which design responses and policy for campus emergencies. Another implication is that in the case of campus emergencies, parallel to that of natural disasters, the focus must be on the root causes (Wisner et al. 2004). Universities must not merely fortify their defenses but should also work on the social issues which create these incidents: liberal gun access and ammunition laws, lack of attention to individuals with mental health issues, and the neighborhoods surrounding the campus. This study relates to research on natural disasters in that much of the focus in anthropological analysis of natural disasters has focused on production of risk, vulnerability, and disaster recovery. There has been a relative lack of focus on disaster response (A. J. Faas, personal communication, Oct. 2013; Dyer 2009). In our focus here on perceptions of the institutional response to an “almost” disaster, we find the issues of leadership and communication to be of major importance. Hess and Arendt (2006) found similar issues in their study of the response of New Orleans acute care hospitals to Hurricane Katrina. Dyer (1999) also notes the need for effective leadership as well as stakeholder empowerment. In contrast to the situation he studied after Hurricane Andrew, the students in our study were not empowered by the university’s handling of the emergency threats. Our findings, then, contribute not only to the specific literature on campus emergencies but also to that of disasters, particularly where the hazard is “social” rather than “natural.” As Dyer (1999: 279) notes, “A major gap in our knowledge of disaster phenomena and response is at the institutional level of the disaster response agency.” The findings also contribute to issues of spatial and temporal distribution of risk in emergencies and disasters. When disaster researchers (see e.g., Cutter, Boruff, and Shirley 2003) address this issue, their focus is often on historical developments that place some populations in greater proximity to hazards. This study illustrates how location, relative to a possible threat on a multi-building university campus, affects perceptions of students at that university. These findings contribute to discussion of spatial and temporal distribution of risk on a micro level perhaps not anticipated by Cutter, 337

Boruff, and Shirley (2003), whose model generally takes a broader view of time and space. There are, of course, many important differences between natural disasters (hurricanes, famines, etc.) and the type of threat represented by campus emergencies. The study of disasters has traditionally focused on the involvement of a “natural” factor (Oliver-Smith 1996) as opposed to other types of disasters (including epidemics, nuclear war, and terrorism). In campus emergencies, the threat and hazards emerge from the social rather than the natural environment. Nevertheless, the recommendation of Wisner et al. (2004:38) for a focus on “root causes and not…quick…fixes,” would seem to be appropriate in both types of situations. Just as land development policies can make a region less—or more—vulnerable to natural disasters (Wisner et al. 2004), laws and policies may affect vulnerability to threats from the social environment. In summary, the study of perceptions of responses to campus emergencies is an area in which applied anthropological research can make important contributions. Our applied anthropological perspective was evident in a number of aspects of the study design. We demonstrate the relevancy of documenting the situated vulnerabilities, perceptions, behaviors, and concerns of students—a group whose perspective has largely been ignored in previous studies. We used both qualitative and quantitative data to obtain reliable results in this time limited research. In addition, we met the needs of those who commissioned the study. We feel that our model is a useful one, certainly in the case of studies of perceptions of campus/school emergency response efforts. This methodological approach may also be appropriate in parallel situations (epidemics, evacuations due to natural disaster, etc.), where the goal is to collect reliable mixed methods data on perceptions of institutional response to a community threat, where the data must be collected immediately and efficiently, and be appropriate for formulation of policy recommendations.

Epilogue: 9/20/13 A report of an armed man jumping a fence near the Mid-State health complex resulted in a lockdown of that part of campus for about three hours. Two of the authors of this paper were on the campus during the incident. The one in the east section of campus received text alerts in a basement room famous for lack of reception. No unusual activity was noticed outside the building, and the electronic screens at the campus entrance carried no information about the alert. Another author was on the west side of campus in one of the health buildings. She reported, “Police with big guns at key points outside some…buildings. Little to no info in the few alerts we’ve received so far, helicopter circling. Police came in [my building] with assault weapons in hand, and told… staff to lock up…. There are 30 police cars on campus. One darn text alert, that’s it. I’m getting more info from Facebook.” The suspect was never found (WFLA News Channel 8 2013). 338

References Cited Asmussen, Kelly, and John Creswell 1995 Campus Response to a Student Gunman. Journal of Higher Education 66(5):575-591. Bambenek, John, and Agnieszka Klus 2008 Do Emergency Text Message Systems Put Students in More Danger? Educause Quarterly 31(3):12-16. Batchelder, William H., and A. Kimball Romney 1988 Test Theory Without an Answer Key. Psychometrika 53(1):71-92. Bernard, H. Russell 2006 Research Methods in Anthropology: Qualitative and Quantitative Approaches. 4th ed. Walnut Creek, Calif.: AltaMira Press. Boholm, Asa 2003 The Cultural Nature of Risk: Can There be Anthropology of Uncertainty? Ethos 68(2):159-178. Borgatti, Steve 1992 ANTHROPAC 4.98. Columbia: Analytic Technologies. URL: (April 14, 2014). Burruss, George, Joseph Schafer, and Matthew Giblin 2010 Student Perceptions of Campus Safety Initiatives: Assessing Views of Critical Incident Prevention and Response. Chicago: Illinois Criminal Justice Information Authority. Cutter, Susan L., Bryan J. Boruff, and W. Lynn Shirley 2003 Social Vulnerability to Environmental Hazards. Social Science Quarterly 84(2):242-260. Davies, Gordon 2008 Connecting the Dots: Lessons from the Virginia Tech Shootings. Change 40(1):8-15. Douglas Mary, and Aaron Wildavsky 1983 Risk and Culture: An Essay on the Selection of Technical and Environmental Dangers. Berkeley: University of California Press. Dyer, Christopher 1999 The Phoenix Effect in Post-Disaster Recovery. In The Angry Earth. Anthony Oliver-Smith and Susanna Hoffman, eds. Pp. 278-300. New York: Routledge. 2009 From the Phoenix Effect to Punctuated Entropy: The Culture of Response as a Unifying Paradigm of Disaster Mitigation and Recovery. In The Political Economy of Hazards and Disasters. Eric C. Jones and Arthur D. Murphy, eds. Pp. 313-336. New York: Altamira. Figley, Charles R., and Russell Jones 2008 The 2007 Virginia Tech Shootings: Identification and Application of Lessons Learned. Traumatology 14(1):4-7. Fox, James A. 2008 Fueling a Contagion of Campus Bloodshed. Chronicle of Higher Education 54(25):36. Fox, James A., and Jenna Savage 2009 Mass Murder Goes to College: An Examination of Change on College Campuses Following Virginia Tech. American Behavioral Scientist 52(10):1465-1485.

HUMAN ORGANIZATION

Giblin, Matthew J., George W. Burruss, and Joseph A. Schafer 2008 Critical Incident Preparedness and Response on Campus. Report submitted to the Illinois Criminal Justice Information Authority, Center for the Study of Crime, Delinquency, and Corrections. Carbondale: Southern Illinois University Carbondale. Guest, Greg, Arwen Bunce, and Laura Johnson 2006 How Many Interviews are Enough? An Experiment with Data Saturation and Variability. URL: (April 4, 2014). Hauser, Christine, and Anahad O’Connor 2007 Virginia Tech Shooting Leaves 33 Dead. URL: (April 16, 2007). Hess, Daniel, and Lucy Arendt 2006 Critical Care in Crisis. In Learning from Catastrophe: Quick Response Research in the Wake of Hurricane Katrina. Pp. 177213. Boulder: University of Colorado and Natural Hazards Center. Hoffman, Susanna, and Anthony Oliver-Smith 1999 Anthropology and the Angry Earth: An Overview. In The Angry Earth. Anthony Oliver-Smith and Susanna Hoffman, eds. Pp. 1-16. New York: Routledge. Kaminshi, Robert, Barbara Koons-Witt, Norma Thompson, and Douglas Weiss 2010 The Impacts of Virginia and Northern Illinois University Shootings on Fear of Crime on Campus. Journal of Criminal Justice 38(1):88-98. Merchant, Raina M., Stacy Elmer, and Nicole Lurie 2011 Integrating Social Media into Emergency-Preparedness Efforts. New England Journal of Medicine 365(4):289-291. Murchison, Stuart 2010 Uses of GIS for Homeland Security and Emergency Management for Higher Education Institutions. New Directions for Institutional Research 146(2):75-86. National Campus Safety and Security Project 2010 Results of the National Campus Safety and Security Project Survey. URL: (December 2010).

Piotrowski, Chris, and Roger W. Guyette, Jr. 2009 Lockdown: Reactions of University Faculty and Staff. Organization Development Journal 27(4):93-99. Polson, Kelli 2009 Questions Linger after Suspect Arrested. URL: (October 6, 2009). Polson, Kelli, and Jenna Withrow 2009 Suspect Arrested, All Clear Given at Mid-State in Gunman Threat. URL: (October 5, 2009). Romney A. Kimball, Susan C. Weller, and William H. Batchelder 1986 Culture as Consensus: A Theory of Culture and Informant Accuracy. American Anthropologist 88(2):313-338. Sullivan, Helen T., Markku T. Hakkinen, and Dana Piechocinski 2009 Improving Participation, Accessibility, and Compliance for Campus-Wide Mobile Emergency Alerting Systems. In Mobile Response. Jobst Loffler and Markus Klann, eds. Pp. 32-40. New York, Springer. Thompson, Paul B., and Wesley Dean 1996 Competing Conceptions of Risk. Risk: Health Safety and Environment 7(4):361-384. Vaughan, Elaine, and Brenda Nordenstam 1991 The Perception of Environmental Risks among Ethnically Diverse Groups. Journal of Cross-Cultural Psychology 22(29):2960. Vicary, Amanda, and R. Chris Fraley 2010 Student Reactions to the Shootings at Virginia Tech and Northern Illinois University. Personality and Social Psychology Bulletin 36(11):1555-1563. Weller, Susan 2007 Cultural Consensus Theory: Applications and Frequently Asked Questions. Field Methods 19(4):339-368. Weller, Susan C., and Antone Kimball Romney 1988 Systematic Data Collection. Newbury Park, Calif.: Sage Publications.

Natural Hazards Center 2006 Learning from Catastrophe: Quick Response Research in the Wake of Hurricane Katrina. Boulder: Institute of Behavioral Science, University of Colorado at Boulder.

WFLA News Channel 8 2013 Mid-State Police Give All Clear, No Armed Gunman Found. URL: (Sept. 20, 2013).

Oliver-Smith, Oliver 1996 Anthropological Research on Hazards and Disasters. Annual Review of Anthropology 25:303-328. 2011 Anthropology in Disasters: Local Knowledge, Knowledge of the Local, and Expert Knowledge. In Disasters, Development, and Humanitarian Aid: New Challenges for Anthropology. Mara Benadusi, Chiara Brambilla, and Bruno Riccio, eds. Pp. 25-38. Rimini, Italy: Guaraldi.

Wisner, Ben, Piers Blaikie, Terry Cannon, and Ian Davis 2004 At Risk: Natural Hazard, People’s Vulnerability, and Disasters. 2nd ed. London, United Kingdom: Routledge.

Palen, Leysia, Sarah Vieweg, Sophia B. Liu, and Amanda Lee Hughes 2009 Crisis in a Networked World: Features of Computer-Mediated Communication in the April 16, 2007 Virginia Tech Event. Social Science Computer Review (2):1-14.

VOL. 73 NO. 4, WINTER 2014

Zayas, Alexandra, Richard Danielson, Tom Marshall, and Kevin Graham 2009 One Arrested after False Report of Man Wielding Gun, Bomb, Knife Sends Scare through Mid-State in City. URL: (September 9, 2011).

339