Introduction

Since the mid-1990s, mass victimization incidents at schools of all types, sizes, and locales have spurred colleges and universities to pursue a variety of initiatives and efforts intended to prevent and ameliorate the consequences of violent incidents and other critical events. The 2007 shooting on the campus of Virginia Tech University and the 2008 shooting at Northern Illinois University resulted in the formation of an assortment of state and federal task forces and review bodies. These entities promulgated a host of recommendations intended to help institutions enhance campus safety, with most offering suggestions that tended to strike similar themes. It was generally recommended that institutions focus on physical access to campus facilities, increase vigilance to better identify early warning signs, improve communication systems, and enhance the capacity of local law enforcement agencies to respond to critical events on campus (Campus Security Task Force, 2008; Randazzo & Plummer, 2009; Report of the Review Panel, 2007).

In Illinois, the Campus Security Task Force (2008) was created to examine current practices at insitutions across the state, identify recommendations for improving campus safety, and create a dialog and central resource for agencies sorting through the financial, legal, and procedural aspects of the recommended changes. The Task Force recommended colleges and universities approach critical incident planning and response from an all-hazards approach, which made the focus of the recommendations broader than just campus-based shootings. The Task Force noted that incidents related to mental health crises could be reduced through promoting awareness and providing mental health services to at-risk students. In response to these recommendations, the Illinois Legislature enacted the 2008 Campus Security Enhancement Act, a law that required colleges and universities to develop all-hazards emergency response plans, engage in violence prevention efforts, develop and implement threat assessment plans, and train key staff annually on incident reduction and response (Campus Security Enhancement Act of 2008, 2008).

Campus safety initiatives have been influenced by more than incidents of mass violence (Fisher, 1995). For example, the 2012 Penn State University sexual assault scandal renewed attention on the handling of sexual assault incidents by campus officials. The topic would surface again in 2014, spurred on, in part, by a now-discredited Rolling Stone article about a sexual assault at the University of Virginia. Incidents of both mass violence and campus handling of select forms of victimization, particularly sexual assaults, have cast light on other forms of campus crime policy. Among other outcomes were changes to the Clery Act through a 2013 reauthorization of the Violence Against Women Act (Violence Against Women Reauthorization Act of, 2013). Modifications have resulted in colleges and universities implementing new procedures for sexual assault training, investigation, and adjudication, while focusing attention on the reporting mandate found in the Clery Act (Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics Act, 1990).

The net result has been a decade-long push for colleges and universities to adopt at least some of a series of recommended practices intended to enhance campus safety, though evidence as to the efficacy of many of these practices remain elusive. Absent from the discourse concerning enhanced campus safety and security is consideration of how recommended practices are viewed by those they are primarily intended to protect—students, faculty, and staff. While such views might be of less relevance when considering legally mandated response and reporting requirements, they are important elements of the dialogue about preventing and responding to other types of critical incidents on college campuses. Implicit support has been presumed, but the level of actual support remains largely unvalidated, as do explanations for potential variation in levels of support expressed by students and university employees. This study uses survey data from a Midwestern university to examine the level of support expressed for common campus safety policies and procedures. Using a combination of independent t tests and ordered logistic regression, the analysis seeks to disentangle the variation in support for these measures. Such understanding holds important implications for policy makers seeking to implement campus safety measures.

Literature Review

Campus crime and associated concerns have become a centerpiece of legislation and policy initiatives in recent decades. The 1986 murder of Jeanne Clery, a student at Lehigh University, resulted in the passage of the Student Right-to-Know and Campus Security Act of 1990 (Fisher, 1995; Janosik, 2001; Student Right-to-Know and Campus Security Act of, 1990). The Act, which has been updated and amended a number of times in the intervening decades, mandates annual crime data reporting for colleges and universities participating in federal student loan programs (Higher Education Amendments Act of, 1998). The Act represents one of the first efforts to use legislation or administrative edicts to ensure transparency surrounding campus crime and to motivate institutions to implement more proactive crime prevention strategies. Colleges and universities are facing a range of formal requirements and informal expectations to take more actions to increase campus crime awareness, education, reporting, and safety. These include efforts concerning the reporting of sexual assault and bias crimes, the establishment and promulgation of emergency response protocols, the capacity to issue timely notifications about campus safety and critical incidents, and practices to increase education and prevention of dating violence and stalking. Some of these efforts are the result of formal mandates, while others arise from informal public expectations that institutions must do more to increase safety, prevention, and communication efforts.

The result of these legislative mandates, coupled with the increased attention given to campus-based critical incidents (Drysdale, Modzeleski, & Simons, 2010), culminated in the issuance of a number of reports making recommendations for how both K-12 schools and institutions of higher learning should secure their facilities, view and handle “at risk” individuals, and protect their students, faculties, and staffs (Leavitt, Spelling, & Gonzales, 2007). The common recommendations make sense on their face, but generally lack empirical validation or any evidence they are supported by those they are intended to protect. What anecdotal support that can be found for commonly discussed practices is often reported by advocacy groups, such as those pushing to allow concealed carry of firearms on college campuses. Beyond a dearth of information regarding descriptive levels of support expressed by various campus constitutency groups, there is an absence of scholarship examining whether and how conceptually relevant criminological concepts (i.e., prior victimzation, fear of crime, perceived risk, and perceived capacity) might shape levels of support across subsets of the campus community. The review of the literature that follows provides a general overview of common campus safety intitiatives, then examines conceptually relevant clusters of predictors that might be expected to condition the level of support expressed for policies and procedures by students, faculty, and staff.

Frequent Campus Safety Recommendations

Various review bodies, task forces, committees, and professional groups have proffered a range of recommendations that ostensibly enhance campus safety. Common recommendations to colleges and universities include: training campus public safety and local law enforcement to respond to critical events using an “all hazards” view of potential risks; making safety and security information widely available to the campus community; developing all hazard incident response plans; implementing communication systems to alert faculty, staff, students, and parents about emergencies on campus; increasing the visiblity of campus safety personnel and improving their interactions with the campus community through a number of measures, such as the use of foot patrol; restricting access control to campuses spaces and facilities; and, improving communications between campus faculty/staff, public safety, and mental health officials in matters concerning students who might be deemed high risk or at risk for engaging in violent or serious criminal conduct (Campus Security Task Force, 2008; Chancellor’s Task Force on Critical Incident Management, 2007; Davis, 2008; Gubernatorial Task Force on University Campus Safety, 2007; Leavitt et al., 2007; Northern Illinois University, n.d.; Report of the Review Panel, 2007). Schools across the country have opted to adopt at least some of these practices, with some implementation stemming from legislative mandate and others from voluntary adoption (Midwestern Higher Education Compact, 2008; Schafer, Heiple, Giblin, & Burruss, 2010).

The majority of the recommended security practices lacked clear empirical validation at the time they were endorsed and few have been validated in a campus setting in the intervening years. At best, institutions can draw on the belief the recommendations are “best practices” within the higher education community, are loosely supported by broader empirical scholarship, or offer presumptive institutional indemnification from liability and other legal risk. Public and political discourse concerning crime prevention and reduction often include police visibility and presence as a presumed deterrent (e.g., Peak, Barthe, & Garcia, 2008; Sloan, 1992) despite mixed empirical findings (Kelling, Pate, Dieckman, & Brown, 1974; Salmi, Gronroos, & Keskinen, 2004; Winkel, 1986). In college environments, increased police presence and interactions can have unintended outcomes in the form of reduced police-student relations since many encounters between these groups involve enforcement activities (Brady, Balmer, & Phenix, 2007; Miller & Pan, 1987).

The passage of the Clery Act in 1990 initiated an expansion of the responsibilities campuses have for addressing matters of law enforcement and crime prevention. This appears to create a tension between an institution’s mission to educate students and its obligation to respond appropriately to criminal acts. Campus public safety entities find themselves with a broader role than many of their municipal counterparts as they balance the need to: educate students (both proactively and in response to detected violations); respond within federal requirements regarding the handling of sexual assault, stalking, and intimate partner violence; comply with federal crime-reporting requirements; and engage in conventional law enforcement activities (Polensky, 2002). More recently, institutions have been advised to create formal linkages between campus public safety officials and other units, such as admissions, university housing, and counseling staff. Institutions have been encouraged to create campus-based groups tasked with violence prevention and threat assessment (Campus Security Task Force, 2008; Fox & Savage, 2009). These groups may play a variety of roles, including serving to identify and share information about students thought to be problematic or “at risk” in terms of criminal activity or mental health status. The groups create opportunities for information sharing and dialogue that are intended to improve institutions’ abilities to identify and intervene when concerning behavior is detected from a member of the university community.

Recent campus safety initiatives have moved beyond just the realm of conventional police crime-reduction strategies of increased presence, increased visibility, heightened staffing, and alternative operational strategies. Incidents of campus-based violence have resulted in consideration of how institutions manage facility access. This conversation includes a range of practices from rethinking admissions review decisions to how campus facilities are physically structured and designed (i.e., restricted access to campus buildings, rooms, and spaces with security checkpoints, locks, gates, and other physical barriers). One study (Midwestern Higher Education Compact, 2008) found 20% of institutions had or were considering implementation of criminal background checks of student applicants. Other recommended practices include screening student applicants for history of psychiatric treatments or psychological conditions and establishing policies for how to reintegrate students into campus environments after they have been temporarily removed from an institution for psychiatric or psychological reasons.

Perhaps most controversial is the issue of allowing weapon carrying on campuses, concealed-carry of firearms in particular. Although resistance to allowing firearms on campus is nearly universal among administrators of institutions of higher education (often citing the potential detrimental effects on the educational milieu and/or safety concerns), the issue is brought to the fore by political action committees such as the National Rifle Association lobbying on a Second Amendment right to self-defense basis (Arrigo & Acheson, 2016). Proponents argue that the presence of armed individuals would serve as a deterrent for would-be perpetrators of violence and would potentially allow for a more rapid neutralization of an active shooter should such a situation arise. While all fifty states have provisions that authorize concealed-carry of firearms by private citizens, the privilege granted by such licenses/permits is differentially restricted by each state. At present, eighteen states prohibit carrying firearms (including concealed-carry) on college campuses by statute, twenty-three states allow institutions to form their own policy, and in eight states (CO, ID, KS, MS, OR, TX, UT, and WI), as a result of either state legislation or court decisions, concealed-carry license/permit privileges may be exercised on college campuses by law with few restrictions (National Conference of State Legislatures, 2016).

Unlike most of the other proposed initiatives and normative safety practices, student and faculty attitudes and opinions regarding concealed-carry on campus have been the subject of recent research. At least four studies, which together have surveyed more than 8500 students at 23 colleges and universities, have found that while levels of support differ somewhat according to geographic setting, the majority of students do not support concealed-carry on their college campus (Eaves, Shoemaker, & Griego, 2016; Patten, Thomas, & Wada, 2013; Schafer, Lee, Burruss, & Giblin, 2016; Thompson et al., 2013b). Two studies that examined the attitudes of faculty at a variety of 4-year colleges and universities found that an overwhelming majority of the surveyed faculty were not supportive of concealed-carry of firearms on campus (Bennett, Kraft, & Grubb, 2012; Thompson, Price, Dake, & Teeple, 2013a). Another study assessed the opinions of a sample of community college faculty from 298 institutions across 18 states, and although somewhat less extreme, produced similar findings (Dahl, Bonham, & Reddington, 2016). Likewise, 95% of 401 college and university presidents surveyed did not support concealed-carry on their respective campuses (Price, Thompson, Khubchandani, Dake, Payton, & Teeple, 2014).

Strong empirical validation of dominant campus safety practices is still generally missing. It is understandable that institutions may feel compelled to act in the absence of such validation, but as time passes and such practices become potentially more common and institutionalized, this omission becomes more alarming. Rigorous assessments of the actual efficacy of these practices might be difficult and resource-intensive, particularly given the low frequency of some campus-based critical events. Regardless of whether a campus has experienced a critical incident, the views and opinions of those ostensibly protected by campus safety measures are relevant considerations. While policies might need to be adopted to comply with federal law or because they are validated, if unpopular, the views and support of students, faculty, and staff are significant considerations in the process of weighing the use of various campus safety efforts. The views of those campuses seek to protect are of importance in ensuring many initiatives might achieve optimal efficacy. Campuses can create mechanisms for faculty to report students with whom they are concerned, but if faculty do not believe that is just, appropriate, or within their role, that mechanism might be under utilized. Students who perceive campus safety efforts go beyond security and reach a point of inconvenience might circumvent those measures or consider transferring to another school or ceasing their education. Although there is limited assessment of student views (Schafer et al., 2016), even less is known about the contrast between the views of students and university employees. These two groups should not be presumed to share similar views, as they are generally at different life phases, are differentially affected by some measures, and may have different relationships with the campus environment (i.e., it is a workplace for faculty and staff, but a home for students, especially those residing in campus housing).

Notwithstanding the similarities noted in the concealed-carry attitude findings, concordant levels of support should not be presumed between these two groups. Research in other domains has found students express mixed or low levels of agreement with institutional practices, such as policies regarding institutional admissions (Kealy & Rockel, 1987), the adjudication and resolution of violations of academic integrity codes (Jordan, 2001; Thakkar & Weisfeld-Spolter, 2012), and handling student use of alcohol (Marshall, Roberts, Donnelly, & Rutledge, 2011; Oster-Aaland & Neighbors, 2007). Research contrasting students with faculty and staff have found the two groups are often different in their views concerning problem drinking (Leavy & Dunlosky, 1989) and estimations of the frequency of academic misconduct (Hard, Conway, & Moran, 2006). In addition to expressing appreciably different views regarding smoking on campus, faculty and staff were more supportive of a tobacco-free campus policy at one institution (Hall, Williams, & Hunt, 2015). The present study extends the spirit of these inquiries and contrasts of support into the domain of campus safety initiatives.

Hypothesized Predictors of Campus Safety Policies

Existing theoretical frameworks provide a structure for considering the level of support expressed by members of campus communities. The broader study of campus crime illuminates some factors that would be expected to have relevance in predicting perceptions of campus safety policies. These factors (victimization, fear of crime, protective measures, perceptions of disorder, and associated demographic variables including race, sex, and age) are drawn from extant scholarship examining perceptions of crime and protective behaviors, particularly in college and university settings and can reasonably be expected to shape views of common campus safety policies.

Victimization

Perceptions of crime and associated concepts are often linked to victimization experiences in criminological research (Ferguson & Mindel, 2007; Fisher & Sloan, 2003; Fox, Nobles, & Piquero, 2009; Garofalo, 1979; Gibson, Zhao, Lovrich, & Gaffney, 2002; Skogan & Maxfield, 1981); those with direct and vicarious victimization experiences subsequently report more fear of crime. While most studies find victimization experiences increase fear of crime, other studies fail to demonstrate this associaton, leading some to suggest the victimization-fear relationship is not conclusively established (Baumer, 1985; Hindelang, Gottfredson, & Garofalo, 1978; Kaminski, Koons-Witt, Thompson, & Weiss, 2010; McGarrelI, Giacomazzi, & Thurman, 1997; Nofzinger & Williams, 2005). Crime victims report higher levels of fear of crime (Jackson & Gray, 2010; Rader, 2004; Reese, 2009) and are at an increased risk of future victimization experiences (Ferraro, 1995; Fisher, Cullen, & Turner, 2000; Gidycz, Coble, Latham, & Layman, 1993; Lauritsen & Davis-QUinet, 1995). Although public discourse about campus safety protocols has been catalyzed by highly publicized and tragic incidents of mass violence, in reality the conversation surrounding these initiatives is concerned with broadly reducing victimization experiences. The use of education, target hardening, increased police presence, improved communication, and more thoughtful handling of students believed to pose a safety risk might all have an influence on the general risk of victimization in college settings. Members of the campus community with recent campus-based victimization experiences would be expected to offer more support for campus safety initiatives. It would be expected that victimization experiences would elevate fear, with that fear translating into greater support for practices and policies designed to increase security and control by campus officials.

Fear of Crime

Campus users understandably might develop emotional response to the possibility that they might become the victim of crime (Ferraro, 1995; Rader, May, & Goodrum, 2007; Rader, Cossman, & Allison, 2009). This emotional response, the fear of crime, has the potential to modify beliefs, perceptions, and behaviors. Among other responses, members of a campus community might become fearful, support the implementation of more safety strategies, or modify when and how they use campus spaces, such as avoiding certain areas or being out after dark. Although it has been noted that crime rates on campuses tend to be lower than the surrounding communities, campus users, especially students, are still fearful of crime (Baum & Klaus, 2005; Bromley, 1992). Fear has been observed to increase, at least in the immediate aftermath of high profile incidents of violent crime (Kaminski et al., 2010; Lee & DeHart, 2007; Stretesky & Hogan, 2001).

The limited extant research considering campus safety has not yet demonstrated if or how demographics are related with perceptions of such practices, but broader literature on demographics and fear offer some insight into expected patterns. A robust body of criminological research has found females express more fear of crime than males. This holds true in studies of college and university students (Fisher, 1995; Fox et al., 2009; Jennings, Gover, & Pudrzynska, 2007; Kaminski et al., 2010; Lane, Gover, & Dahod, 2009; Wilcox, Jordan, & Pritchard, 2007), although variation is noted when researchers examine fear of specific types of offenses (Barberet, Fisher, & Taylor, 2004; Fisher & Sloan, 2003; King, 2009). Female students express greater fear of sexual victimization (Dobbs, Waid, & Shelley, 2009; Ferraro, 1995; Fisher & Sloan, 2003; Kelly & Torres, 2006), while males are more fearful than females when asked to consider property offenses (Lane et al., 2009). Because it is expected that greater fear will translate into greater support for campus safety initiatives, female students and female faculty and staff members are expected to offer more support for policies and practices intended to enhance campus safety.

Research on the fear of crime tends to find age is a consistent predictor. More fear is expressed by older citizens, although this fear is not necessarily proportional to objective risk of victimization (Ferraro, 1995; Pain, 1997). The ages of faculty and staff at most institutions will have wide variation, while students are typically truncated to a restricted age range. Faculty and staff would be expected to mirror fear of crime research, with older respondents expressing more fear, thus we would expect them to report more support for the use of campus safety measures. In contrast, younger students would be expected to express more fear (Fisher & Sloan, 2003; Kaminski et al., 2010) and support for such policies and practices. Age serves as a proxy for tenure, the amount of time a student has attended a college or university; it is expected students who have spent more time in a given campus environment will have developed a level of comfort with that space and its risk, and will have more confidence in their ability to negate that risk (Ferraro, 1995). Evidence tends to suggest white students experience slightly higher levels of victimization for violent crimes (Baum & Klaus, 2005), yet non-white students express more fear of crime (Fox et al., 2009; Kaminski et al., 2010; Lane et al., 2009; Skogan, 1995). This finding mirrors fear of crime research with samples from the general population. It is expected younger students, older faculty and staff members, and non-white respondents will express more support for campus safety measures.Footnote 1

Protective Measures

Concerns about crime and safety can lead individuals to engage in a number of behaviors to protect themselves and their property, or to otherwise mitigate or limit their risk of victimization (Wollnough, 2009). At a subjective level, taking protective measures might be considered healthy behavior and a functional adaptation to one’s environment and circumstances (Jackson & Gray, 2010). Concerns about safety that lead citizens to take protective measures and/or constrain their behavior might be positive if they actually reduce risk. Taken to an extreme, however, concern about safety might become dysfunctional if it reduces quality of life, excessively constrains behavior, or leads one to engage in potentially unlawful conduct (i.e., weapon carrying) (Warr, 2000). Fear and associated fears about crime have been linked with citizens engaging in a variety of protective measures, some of which might be deemed healthy and others which might be deemed dysfunctional or dangerous (Jackson & Gray, 2010; Melde, Esbensen, & Taylor, 2009; Nasar, Hecht, & Wener, 2007). Protective measures have been identified as a form of “self-help” that one might engage in when one has low levels of confidence in law enforcement’s ability to prevent victimization (Black, 1980; Smith & Uchida, 1988). Faculty, staff, and students who engage in more protective measures would be expected to offer more support for campus safety initiatives.

Perceptions of Disorder

Social and physical environmental cues can be important in conditioning crime-related perceptions (Barberet & Fisher, 2009; Garofalo & Laub, 1978; Kohm, 2009; Skogan, 1990; Warr, 2000; Wyant, 2008). People tend to be visually sensitive to environmental conditions, so the presence of disorderly circumstances can generate fear, modify perceptions, and shift protective behaviors (Blobaum & Hunecke, 2005; Pain, 1997). Whether disorder begets crime is somewhat irrelevant to the extent that disorder might engender fear and modify behavior (McCrea, Shyy, Western, & Stimson, 2005). Faculty, staff, and students who perceive higher levels of disorder would be expected to support more campus safety initiatives. The presence of social or physical disorder would be expected to suggest that space is in need of additional formal social control.

Research Objectives

The purpose of this study is to explore the support for campus safety initiatives among students, faculty and staff at a large, public, Midwestern university. Specifically, the analysis assesses support for common campus safety policies, including the identification and reporting of potentially dangerous students, denial of admission or expulsion based on criminal history or potential threat, and allowing the carrying of firearms or other weapons on campus. In addition, the study seeks to determine whether support for these campus safety initiatives is associated with prior victimization, fear of crime, the degree to which one employs protective measures, perceptions of disorder, satisfaction with the public safety department, and/or demographics.

Data and Methods

Sample

The sample for the current study was generated from students, faculty and staff at a large, public, rural, Midwestern university (10,000+ students). The research team randomly selected on-campus, face-to-face courses that were scheduled to meet on two consecutive days in the spring of 2013. Course instructors were contacted by email and asked for permission to administer a survey to their students on a specified date. In cases in which an instructor declined or failed to respond to the request, other courses were randomly selected. As a result, the research team was granted access to forty courses.

Over the course of the two consecutive dates the research team administered the survey to the students in attendance in each of the aforementioned classes.Footnote 2 The students were informed that participation was voluntary and that declining to participate would have no adverse effect on their course grade. The instrument collected no unique identifying information and instructors were not involved in the administration or handling of the survey instruments. On the specified dates there were 887 students in attendance in the forty selected courses, of which 840 completed and returned the survey for a 94.7% response rate.

Of the 3400 individuals classified as full-time employees in January 2013, a random sample of 1000 faculty and staff was generated for participation in the research project. Sampled employees were sent a survey instrument via campus mail accompanied by a cover letter requesting their participation. Subsequent to mailing the surveys it was determined that some of those selected to receive the survey had retired, were on medical leave, or were actually assigned to off campus duties. This reduced the potential respondents to 990. Reminder emails were sent to the survey recipients who had not yet returned them two weeks after they had been mailed, again asking for their voluntary participation. Ultimately, 410 completed surveys were returned for a 41.4% response rate. The demographics of the two samples are presented in Table 1, which also contains the values for the actual campus population where available. Overall both samples are reasonably representative of their respective campus populations with the exception of an underrepresentation of graduate and professional students and a lower mean student age, with the latter difference likely being a function of the first.

Table 1 Descriptive statistics—sample demographics (N = 1250)
Full size table

Measures

Dependent Variables

The survey asked respondents to rate their level of support for three commonly recommended campus safety policies, which include reporting of potentially dangerous students, denial of admission or expulsion based on criminal history or potential threat, and allowing the carrying of firearms or other weapons on campus. These three policy themes were measured utilizing nine survey items (three questions for each of the three themes). A five-point Likert scale was utilized for responses to these nine survey items, which consisted of: 1 = Strongly disagree; 2 = Disagree; 3 = Neither agree nor disagree; 4 = Agree; 5 = Strongly agree. These responses were recoded as follows: 1 and 2 = 1 Disagree; 3 = 2 Neither agree nor disagree; and 4 and 5 = 3 Agree, for compatibility with ordered logistic regression analysis.Footnote 3

The first of the three themes, reporting of potentially dangerous students, was measured with the following three items: 1a) “faculty/staff have a responsibility to report information about potentially dangerous students to relevant authorities” (hereafter referred to as faculty report); 1b) “students have a responsibility to report information about potentially dangerous students to relevant authorities” (hereafter referred to as students report); and, 1c) “campus counseling officials should share information on potentially dangerous students with public safety officials” (hereafter referred to as counselors report). Denial of admission or expulsion based on criminal history or potential threat, was measured with the following three survey items: 2a) “if a student is considered by campus officials to be a threat or potential threat, the student should be removed from campus prior to any type of college hearing” (hereafter referred to as expel w/o hearing) ; 2b) “campuses should have the right to deny admissions to students with multiple criminal convictions” (hereafter referred to as deny adm w/convictions); and, 2c) “campuses should have the right to deny admissions to students with records of non-criminal misconduct at other educational institutions” (hereafter referred to as deny adm non-criminal bx). The third theme, permittance of concealed-carry firearms and/or other weapons on campus, was measured with these three survey items: 3a) “students should be allowed to carry concealed firearms on campus” (hereafter referred to as students carry firearm); 3b) “faculty/staff should be allowed to carry concealed firearms on campus” (hereafter referred to as faculty carry firearm); and, 3c) “faculty/staff should be encouraged to carry concealed weapons, not including firearms, such as Tasers, stun guns, or Mace” (hereafter referred to as faculty carry less-lethal).

Independent Variables

The factors that are expected to predict perceptions of campus safety policies, which include prior victimization, fear of crime, protective measures, perceptions of disorder, satisfaction with the public safety department, and associated demographic variables (race, sex, and age) were measured as follows.Footnote 4 Prior victimization was measured with a survey item that asked respondents, “in the past 12 months, has anyone physically attacked you on campus causing you to receive bruises, scratches, or some more serious injuries?” Fear of crime was measured with 4 survey questions. The respondents were asked to rate their level of fear of the following incidents happening to them while on campus during the day on a scale of 1 (not fearful at all) to 10 (very fearful): being raped or sexually assaulted; being robbed or mugged; being physically beaten up; and, being shot at while in a classroom/at work. The internal reliability of these four items was high (α = 0.912) and the scores were summed to form a composite fear of crime score ranging from 4 (not fearful at all) to 40 (very fearful).

The measurement of protective measures involved three survey items that asked respondents if they engaged in the following protective behaviors: consciously chose well-lit walking paths when traveling on campus after dark; avoiding certain parts of campus considered dangerous or unsafe; and, attempting to walk in groups as much as possible. The internal reliability of these items was acceptable (α = 0.716) and these items were also summed to create a composite score ranging from 0 (none of the protective measures employed) to 3 (all three protective measures employed). Perceptions of disorder was measured with five survey items that asked respondents to rank the following issues on a three-point Likert scale according to whether they believed they were (1) Not a problem; (2) Somewhat of a problem; or (3) Big problem. These issues included: graffiti and vandalism; inadequate or broken outdoor lighting; areas of campus that seem neglected or “run down”; noise; and, groups engaging in rowdy behavior in public areas or outside buildings. The internal reliability for these items was also acceptable (α = 0.710) and thus these five items were summed to form a composite score ranging from 5 (not a problem) to 25 (big problem).

Satisfaction with the public safety department was measured with two survey items, which asked respondents to rate their level of satisfaction with the university department of public safety on a 5 point Likert scale: (1) Don’t know; (2) Very dissatisfied; (3) Dissatisfied; (4) Satisfied; or (5) Very satisfied. Respondents rated their satisfaction with university public safety officers being visible on campus and the overall quality of the campus university public safety department. A composite score was created by summing these two items that ranged from 2 to 10, which also had an acceptable level of internal reliability (α 0.718). Finally, the measurement of the following demographic variables was also included: respondent status (coded 0 = student, 1 = faculty/staff); race (coded 0 = nonwhite, 1 = white); sex (coded 0 = female, 1 = male); and age (in years).

Data Analysis

The analysis consisted of independent samples t tests and ordered logistic regression to assess whether the variables listed above have an influence on levels of support for the aforementioned oft-recommended campus safety policies, and to attempt to detect and disentangle variation between students and faculty/staff support.

Results

The descriptive statistics and results of the independent samples t tests for the nine DV indicators are reported in Table 2. The t test results indicated significant differences in means between students and faculty/staff for each of the nine indicators. However, the frequencies of some of the items did not appear, at first glance, to be drastically different between students and faculty/staff. None of the three indicators of the reporting of potentially dangerous students DV (faculty report, students report, and counselors report) appeared to have markedly different levels of support between students and faculty/staff, but there were some more notable differences in a few of the indicators for the denial of admission/expulsion DV and the permittance of firearms and/or other weapons DV.

Table 2 Student and faculty/staff support for campus safety initiatives
Full size table

With regard to the denial of admission/expulsion DV, the frequencies of responses for indicator deny adm w/convictions appeared to be more similar with both students and faculty/staff expressing greater support than was the case for indicators expel w/o hearing and deny adm non-criminal bx. While a little more than half of the faculty/staff surveyed (54.39%) expressed support for expel w/o hearing, only 38.85% of the students expressed support for such a policy with the remaining students distributed nearly equally between disagree and neither agree nor disagree. Likewise, nearly half of the faculty/staff (48.66%) expressed support for deny adm non-criminal bx while less than one-third of the students (29.74%) agreed with this type of campus safety policy.

The levels of support expressed by students and faculty/staff for all three of the indicators for the third DV theme, permittance of concealed-carry of firearms and/or other weapons also differed. While the majority of both groups indicated that they disagreed with students carry firearm, faculty/staff disagreed to a greater degree at 83.13% compared to 62.90% of students. However, less than half of the students (47.69%) disagreed with faculty carry firearm compared to 62.93% of faculty/staff. Concerning weapons other than firearms, responses of students and faculty/staff to faculty carry less-lethal were nearly inverse. While 44.58% of students agreed with such a policy and 31.06% disagreed, 48.65% of faculty expressed disagreement and only 27.52% indicated that they agreed with this kind of policy.

Aside from the expected correlation between age and faculty/staff (r = 0.858) no multicollinearity issues were detected. In the case of age and faculty/staff, the variance inflation factor was <3.00 at 1.081. Thus the impact of this correlation on regression results would be negligible. Each of the nine indicators (three for each of the DV themes) was analyzed utilizing ordered logistic regression. The results for the DV theme reporting of potentially dangerous students are presented in Table 3, the DV theme denial of admission or expulsion in Table 4, and the DV theme permittance of firearms and/or other weapons on campus in Table 5 below.

Table 3 Ordered logistic regression: DV reporting of potentially dangerous students
Full size table
Table 4 Ordered logistic regression: DV denial of admission/expulsion
Full size table
Table 5 Ordered logistic regression: DV permittance of firearms & other weapons
Full size table

Among those variables conceptually thought to drive support for campus safety initiatives, only protective measures emerged as significant for the reporting of potentially dangerous students theme. However, engaging in protective measures was consistently predictive of supporting policies across the three items associated with this theme (faculty report, students report, and counselors report), with odds ratios of (1.513), (1.364), and (1.213) respectively. Of the demographic variables that were expected to be predictive, race was consistently significant across all three of the theme’s items with the odds ratios of (1.852) for faculty report, (2.248) for students report, and (1.443) for counselors report indicating that whites were more likely to support such policies than nonwhites. Sex and age were also predictive for faculty report and students report, but not counselors report. With regard to sex, females were 42% more likely to support faculty reporting of potentially dangerous students and 59% more likely to support student reporting of potentially dangerous fellow students to relevant authorities than males. Age was less impactful with odds ratios of (1.048) for faculty report, and (1.049) for students report indicating that support increased slightly with age. Faculty/staff emerged as a significant predictor only for counselors report with an odds ratio of (1.443) indicating that faculty/staff were 44.3% more likely to support reporting requirements for campus counseling officials than were students.

Similar to the first DV theme, only one of the five variables thought to drive support for campus safety policies was significant across all three items of the denial of admission/expulsion theme. In this case perceptions of disorder emerged as a significant predictor for all three with odds ratios of (1.052) for expel w/o hearing, (1.088) for deny adm w/convictions, and (1.067) for deny adm non-criminal bx indicating that those who perceived higher levels of disorder were more likely to support the associated campus safety policies. Protective measures was also a significant variable, but for deny adm w/convictions only. The odds ratio of (1.162) indicated that those who engaged in protective measures were 16% more likely to support policies in which campuses have the right to deny admission to students with multiple criminal convictions. The demographic variables were even less consistently predictive for this theme than the previous theme. Race and age were significant for deny adm w/convictions only with whites (OR = 1.947) and older respondents (OR = 1.022) more likely to support denial of admission to students with multiple criminal convictions. Sex was predictive for deny adm non-criminal bx with an odds ratio of (1.463) indicating that males were 46.3% more likely than females to support denial of admissions to students with records of non-criminal misconduct at other educational institutions. Faculty/staff also emerged as a significant predictor for deny adm non-criminal bx only with an odds ratio of (2.403) indicating that the odds of faculty/staff supporting such policies is 2.4 as great as students.

Unlike the first two themes, none of the variables conceptually thought to drive support for campus safety initiatives was consistently predictive of support across all three items of the permittance of firearms and other weapons DV theme. Fear of crime was a significant predictor for students carry firearm with an odds ratio of (1.021), and faculty carry firearm with an odds ratio of (1.023). Perceptions of disorder emerged as significant for faculty carry less-lethal with an odds ratio of (1.050). Protective measures was a significant predictor of students carry firearm and faculty carry less-lethal; however, those who indicated that they engaged in protective measures were 12.6% less likely to support student concealed-carry of firearms than those who did not, but 18.2% more likely to support policies that encouraged faculty/staff to carry concealed weapons other than firearms. Of the demographic variables, race and sex were both strong predictors across all three items. Whites (OR = 1.852, 2.550, and 1.753 for students carry firearms, faculty carry firearms, and faculty carry less-lethal respectively), and/or males (OR = 2.186, 2.114, and 1.500 for the aforementioned items respectively) were much more likely than non-whites and/or females to support the allowance of weapons on campus. Lastly, age emerged as a significant predictor for faculty carry firearm with an odds ratio of (0.981), indicating that, as age increases, support for these pro-weapons policies decreases.

Discussion & Conclusions

While many of the policies considered here are controversial, and they all impact student and faculty life on campus, the models did not reveal any strong evidence of factors that drove opinions for or against them. All the models had low pseudo-R2 values indicating a poor fit with the data, and few of the variables hypothesized to influence support for campus safety policies were predictive for any one of the individual items, consistently across items for a given DV theme, or across themes. Furthermore, the results indicate that support among campus community members for oft-recommended safety policies is tepid at best. However, this study is not without some limitations. The sample for the current study was drawn from a single Midwestern university and thus the generalizability of these findings is unclear. The attitudes of students, faculty, and staff might vary significantly between college and university campuses located in different regions. Keeping these limitations in mind, considering the earlier points regarding the importance of campus user support for such policies, these findings warrant some attention.

First, students and faculty/staff diverged in their attitudes toward some of these policies. While faculty and staff supported non-weapon policies like information sharing, student findings indicated the reverse. Students consistently exhibited greater support for policies that would permit students, faculty, and staff to carry concealed firearms and other weapons on campus, though it is worth mentioning that students and faculty/staff alike tended to disagree with such policies altogether. In contrast, faculty and staff were more likely to agree—though again, attitudes were mixed—with polices that facilitate campus safety through information sharing and enhanced restrictions on who can enroll or remain on campus. However, neither group supported a theme of comprehensive safety policies.

The apparent inconsistencies can be reconciled for both groups, however, if we consider broader views of campus safety and academic freedom for faculty and staff, as well as individual liberties for students. In recent years, state legislatures have discussed, introduced, or passed legislation allowing firearms on campuses in more than a dozen states in response to several high profile campus shootings (e.g., at Umpqua Community College in Oregon in 2015). By legally allowing guns on campus, “certain people on the scene can mount an armed response before the police arrive” in the event of an active shooter situation (Nagourney & Turkewitz, 2015, para. 2). Most notable among these efforts is Senate Bill 11, a Texas law that went into effect on August 1, 2016, expanding the right to carry concealed weapons into classrooms and other formerly prohibited places on state university campuses. While opinions differ on these laws (as demonstrated in these research findings), anecdotal evidence suggests many faculty members believe that increasing the number of firearms on campus or expanding the locations where they may be carried might actually increase danger. The Faculty Senate at the University of Houston recently cautioned faculty members about making “provocative statements” about the law and, as one faculty member stated, “We are worried that we have to change the way we teach to accommodate this minority of potentially dangerous students” (Moyer, 2016, para. 3–5). Setting aside the veracity of the claims (guns improve campus safety vs. guns threaten public safety/academic freedom), current debates show why faculty members might oppose efforts to strengthen the collective security apparatus by permitting concealed carry, while supporting efforts to limit campus access and increase information sharing.

The differences between students and faculty/staff may also reflect political leanings, unmeasured in the current study, separate from age. Attitudinal differences remained even after controlling for the age of the survey respondent, a common predictor of firearm ownership or firearm-related protective behavior (e.g., Luxenburg, Cullen, Langworthy, & Kopache, 1994; Schwaner, Furr, Negrey, & Seger, 1999). According to studies by the Pew Research Center, college-age students and young adults (18–29) were more likely to identify as libertarian and one in five were classified as “young outsiders” who tended to favor limited government yet express socially liberal attitudes (Desilver, 2014; Kiley, 2014). Consequently, they may disagree with attempts to restrict their rights, or the rights of others, to carry a weapon, pursue educational opportunities, or maintain their personal privacy. These concerns should not be interpreted as a denial of the need for campus public safety, only an indication of values. Of course, this is mere speculation and must be addressed in future research where political ideology is included as a predictor.

Second, and perhaps most notable, those factors that presumably would predict support for these types of campus safety policies failed to do so in this study. While engaging in protective measures emerged as a consistent predictor of support across the reporting of potentially dangerous students theme, as did perceptions of disorder across the denial of admission/expulsion theme, neither of these, nor any of the other independent variables were consistently predictive across all three themes. This is an important finding because, as mentioned earlier, implicit support among campus community members for these commonly recommended campus safety policies has been presumed, and these presumptions have likely been based on the theoretical notions examined in the current study. In fact, given the negatively skewed results across the permittance of firearms & other weapons theme and similar results across the extant research, it seems likely that allowing weapon carrying on campus might increase fear of victimization.

Although attitudes significantly differed (particularly by race and gender), substantively, student views were similar to those expressed by faculty and staff. This suggests that campus policymakers and state legislatures would be well served to consider the opinions of campus community members before imposing what may be unpopular policies. For instance, in Texas, private universities are exempt from the concealed carry laws. At Rice University, a moderately sized school in Houston, the school’s president indicated that concealed weapons will not be allowed on campus, citing campus community resistance: “Not a single constituency consulted has endorsed having guns on our campus; in fact, each overwhelming(ly) opposed it” (Wermund, 2015). The implications are clear. With little (if any) empirical validation for these commonly recommended campus safety policies, the academic demands of an institution of higher learning balanced with the opinions of the campus community are paramount. As demonstrated in the current study, the perceptions of students, faculty, and staff regarding campus disorder, crime, and even their level of fear of victimization may not be indicative of their support for any given campus safety policy. Specific inquiries may produce unanticipated results. Consideration of campus user attitudes and opinions in these matters is crucial as lack of support from those individuals that such policies are intended to protect might ultimately impact their efficacy if it results in noncompliance. Worse yet, implementing extremely unpopular policies may adversely affect the campus climate and the institution’s mission.