This item is only available as the following downloads:
1 VIOLENCE IN CAMPUS COMMUNITIES By MICHAEL LEE SULKOWSKI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Michael Lee Sulkowski
3 ACKNOWLEDGMENTS I thank my advisor, Dr. Thomas Oakland, for his support, guidance, and commitme nt to my growth and development. His dedication to the empirical pursuit of knowledge, scholarly rigor, and academic productivity has had an indelible influence on me as well as this project. I also warmly acknowledge my committee members, Drs. Diana Joyce Walter Leit, and Sherry Benton for their helpful suggestions, feedback, and respective commitments. Input from each committee member has helped to bring this dissertation to fruition. This study would not have been possible without the help of key mem bers in the University of Florida Administration. In particular, I am grateful to Drs. Jeanna Mastrodicasa and Eugene Zdziarski for their help in orchestrating institutional support for this study. Additionally, I thank Chief Linda Stump and Major Brad Bar ber for their expert feedback and frank discussions on the state of violence prevention efforts in campus communities. I am grateful to members of the Melissa Institute for Violence Prevention and Treatment for their generous finical support. I also st rongly appreciate the time that hundreds of participants have invested in this study. Their responses have been overwhelmingly positive and encouraging. On a personal note, I would like to express my gratitude to my friends and family members for their s upport, encouragement, patience, and perspective. I am grateful to Amy Mariaskin for her unwavering support that grounds me and makes me optimistic about the future. Last, but certainly not least, I am grateful to my parents, Marge and Lee Sulkowski. Their belief in me has been steadfast over the years. They have supported me during the good times and the bad. Through their loving guidance, my parents have influenced my professional and personal commitment to scholarship, the pursuit of knowledge, and servi ce.
4 TABLE OF CONTENTS P age ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 3 LIST OF TABLES ................................ ................................ ................................ ........................... 6 LIST OF FIGURES ................................ ................................ ................................ ......................... 7 CHAPTER 1 REVIEW OF LITERATURE ................................ ................................ ................................ .... 10 Weapon Carrying Students on Campus ................................ ................................ .................. 10 V ulnerability of College Campuses to Attack ................................ ................................ ........ 11 Effects of Violent Attacks on College Campuses ................................ ................................ ... 12 Contemporary Responses to Violen t Attacks on School Campuses ................................ ...... 14 The Campus Security Act ................................ ................................ ................................ 14 Zero Tolerance Policies Toward Violence ................................ ................................ ...... 15 Security Technology ................................ ................................ ................................ ........ 16 Concealed Weapons Permits ................................ ................................ ........................... 17 Profiling of School Attackers ................................ ................................ .......................... 17 Threat Assessment Approach to Preventing Violent Attacks ................................ ......... 19 Student Threat Reporting ................................ ................................ ................................ ........ 22 Threat Reporting in College Students ................................ ................................ ..................... 24 Trust in the College Support System ................................ ................................ ............... 25 Campus Connectedness ................................ ................................ ................................ ... 26 Self E fficacy Toward Service ................................ ................................ ......................... 27 Fear of Negative Evaluation ................................ ................................ ............................ 28 Delinqu ency ................................ ................................ ................................ ..................... 29 Research Questions for Investigation ................................ ................................ ..................... 31 2 METHODS ................................ ................................ ................................ ................................ 33 Pa rticipants ................................ ................................ ................................ ............................. 33 Procedure ................................ ................................ ................................ ................................ 34 Measures ................................ ................................ ................................ ................................ 35 Campus Connectedness Scal e ................................ ................................ ......................... 35 Self Efficacy Toward Service Scale ................................ ................................ ................ 36 Self Report Delinquency Scale ................................ ................................ ....................... 36 Brief Fear of Negative Evaluation Scale Revised ................................ ........................... 37 Marlowe Crowne Social Desirability Scale Short Form X2 ................................ ........... 37 Pilot S tudy ................................ ................................ ................................ .............................. 38 Instrument Development ................................ ................................ ................................ 38 Trust In College Support System Scale ................................ ................................ ........... 38 Threat Reporting Decision Passages ................................ ................................ ............... 39 Instrument Modifications ................................ ................................ ................................ 41
5 Additional Provisions to Address Pilot Study Limita tions ................................ .............. 42 Statistical Analyses ................................ ................................ ................................ ................. 42 Measurement Model ................................ ................................ ................................ ........ 43 Structural Regression Models ................................ ................................ .......................... 44 3 RESULTS ................................ ................................ ................................ ................................ .. 49 Preliminary Analyses ................................ ................................ ................................ .............. 49 Meas urement Model ................................ ................................ ................................ ............... 50 Structural Equation Models ................................ ................................ ................................ .... 51 Model 1 ................................ ................................ ................................ ............................ 51 Model 2 ................................ ................................ ................................ ............................ 52 Model 3 ................................ ................................ ................................ ............................ 54 4 DISCUSSION ................................ ................................ ................................ ............................ 61 Trust in the College Support Sy stem ................................ ................................ ...................... 62 Self Efficacy Toward Service ................................ ................................ ................................ 63 Campus Connectedness ................................ ................................ ................................ .......... 64 Delinquency ................................ ................................ ................................ ............................ 65 Fear of Negative Evaluation ................................ ................................ ................................ ... 66 Other Findings ................................ ................................ ................................ ........................ 67 Limitations ................................ ................................ ................................ .............................. 68 Conclusions ................................ ................................ ................................ ............................. 70 APPENDIX A: INFORMED CONSENT FORM ................................ ................................ ......... 72 APPENDIX B: PARTICIPANT MEASURES ................................ ................................ .............. 74 LIST OF REFRENCES ................................ ................................ ................................ ................. 91 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 101
6 LIST OF T ABLES Table P age 2 1. Results of independent samples t tests for study variables ................................ .................. 46 2 3. Percentages of school shoo ters displaying specific characteristics prior to carrying out attacks. ................................ ................................ ................................ ............................... 46 3 1. Descriptive statistics, internal consistency estimates, and interfactor correlations between study variables ................................ ................................ ................................ ..... 56 3 2. Goodness of fit test and index values for tested models ................................ ....................... 56 3 3. Factor loadings and factor variances by construct for the measurement model ................... 57 3 4. Effects decomposition for study variables ................................ ................................ ............ 58
7 LIST OF FIGURES Figure P age 2 1. Model 1: Hypothesized structural relations among variables. ................................ ............. 47 2 2. Model 2: Allowing direct effects of campus connectedness and fear of negative evaluation on trust in the college support system. ................................ ............................. 48 2 3. Model 3: Allowing a direct effect from self efficacy toward service to campus connectedness. ................................ ................................ ................................ ................... 48 3 1. Model 1: The hypothesized structural equation model. ................................ ....................... 59 3 2. Model 2: Allowing direct effects of campus connectedness and fear of negative evaluation on trust in the c ollege support system. ................................ ............................. 60 3 3. Model 3: Allowing a direct effect from self efficacy toward service to campus connectedness. ................................ ................................ ................................ ................... 60
8 Abstract of Dissertation Pre sented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy VIOLENCE IN CAMPUS COMMUNITIES By Michael Lee Sul kowski August 2011 Chair: Thomas Oakland Major: School Psychology The safety of colleges and universities has been questioned in the wake of several high profile shootings (i.e., Virginia Polytechnic Institute and Northern Illinois University). Internat ionally, since 1966, at least 114 persons have been killed on college campuses. Furthermore, between 2001 and 2009, 71 deaths have occurred on college campuses in the United States during violent attacks. In response to these attacks, many universities hav e adopted threat assessment plans to mitigate threats of violence. However, this approach does not address dissemination of important information that may help to prev ent an attack. Therefore, this study examined relationships among several variables that were expected to facilitate or inhibit college hundred and twenty students participated and structural equat ion modeling was used to test associations between latent variables. As expected, trust in the college support system (i.e., trust in police, administrators, hand, students who reported a history of delinquency displayed a low willingness to report.
9 report threats directly and indirectly through trust in the college s upport system. Lastly, self efficacy toward service (i.e., the belief that one can have a prosocial impact) only was related to willingness to report in the presence of trust in the college support system and campus connectedness. Thus, the belief that one can have a positive impact in the campus community does not increase threat reporting by itself. Instead, to facilitate threat reporting, it is important for students to feel connected to the campus community and to trust in members of the college support associated with their willingness to report threats of violence.
10 CHAPTER 1 REVIEW OF LITERATURE The shootings that occurred at Virginia Polytechnic Institute an d Northern Illinois University left 37 persons dead and an additional 39 persons severely injured (Fox & Savage, 2009). Despite being two of the most lethal attacks committed on college campuses, these shootings are not isolated events. For more than 50 ye ars, violent attacks have occurred at various colleges with diverse student body sizes and in different geographical locations. Based on data gathered from records provided by the U.S. Department of Education as well as media reports that were gathered fro m searching electronic newspaper databases, at least 114 shooting related deaths occurred on college campuses around the world b etween 1966 and 2008. Moreover, the potential for violent attacks on college campuses still exists as 76 deaths occurred during violent attacks on U.S. college campuses over the previous eight years (U.S. Department of Education Office of Postsecondary Education, 2009). Weapon Carrying Students on Campus Many college students report carrying weapons on campus, resulting in the pos sibility for violent attacks to occur unexpectedly. In a sample of more than 10,000 students attending 119 U.S. colleges, 4.3% reported having a working firearm at college and 1.6% reported having been threatened with a gun while on a college campus (Mille r, Hemenway, & Wechsler, 2002). A survey of 5,652 students in 29 colleges in California found that 6% reported having carried a weapon (e.g., gun, knife) on campus within the previous thirty days (Patrick, Covin, Fulop, Calfas & Lovato 1997). Finally, a survey of 26,225 students from 61 U.S. colleges found a similar percentage of weapons carrying students (7%) on school grounds (Presley, Meilman & Cashin 1997). Higher rates of weapon carrying behaviors have been identified in males and students attendin g colleges located in the Southern United States in a reanalysis of the data from
11 the latter study (Meilman, Leichliter & Presley 1998). Specifically, 15% of the mal e students in the South reportedly carried a weapon on campus within the previous 30 days Furthermore, the presence of weapons on college campuses may be problematic for individual weapon carriers in addition to being a concern for members of the campus community (e.g., administrators, law enforcement officers). Weapon carrying students are at risk for being threatened with a weapon by another individual (Miller et al., 2002), being involved in physical fights or altercations (Barrios, Everett, Simon, & Brener, 2000), and to engage in criminal activities (e.g., vandalism, assault) (Presley et al., 1997). Weapon carrying students also have been found to be at risk to abuse alcohol and illicit substances (e.g., cannabis sativa, cocaine), engage in high risk behaviors (i.e., having unprotected sexual intercourse, driving while intoxicated) (Mill er et al., 2002), and have suicidal ideations (Barrios et al., 2000). Thus, although a direct relationship has not been established between carrying weapons and socially undesirable behavior, individuals who carry weapons on college campuses are more likel y to experience psychosocial difficulties and engage in risky or impulsive behaviors than are their non weapons carrying counterparts. Vulnerability of College Campuses to Attack s by the director of the Federal Bureau of Investigation (Boynton, 2003) since they are densely populated and allow access to high numbers of potential victims. Furthermore, both within and across institutions, the availability of campus security resources (e.g., police officers, metal detectors), safety procedures, and administrative roles often are limited. Together, these conditions may decrease the effectiveness of crisis response efforts (Greenberg, 2007). Violent campus attacks often are highly chaoti c and end tragically before law enforcement officials have the opportunity
12 violent attacks last only a few seconds, thus providing bystanders little time to coordi nate a response to an attack. Once underway, despite the best efforts of university personnel, survival instincts are likely to determine how individuals respond to an attack (Greenberg, 2007). Effects of Violent Attacks on College C ampuses The effects o f violent attacks on college campuses often are diffuse and may affect the entire ecology of an institution (Fox & Savage, 2009; Utterback & Caldwell, 1989). Similar to other traumatic events (e.g., natural disasters, terrorist attacks), college students, faculty, staff, parents/caregivers, college alumni) often experience psychosocial distress after a violent campus attack (Zdziarski, Dunkel, & Rollo, 2007). Survivors of campus shootings frequently experience symptoms of acute stress or post traumatic stress disorder (PTSD) if their stress symptoms persist for more than a month (Norris, 2007). These symptoms include the persistent reexperiencing of the traumatic event, lasting av oidance of sti muli associated with the trauma, emotional numbing and persistent symptoms of increased arousal that cause significant distress or impairment in social, occupational, or other important areas of functioning (American Psychiatric Association, 2000) Prevalence rates for post shooting PTSD among individuals who directly witness a shooting tend to range from 10% to 36% ( North, Smith, & Spitznagel, 1994; Schwarz & Kowalski, 1991 ; Trappler & Friedman, 1996) However, even if not exposed directly t o a violent attack, individuals with personal connections to victims or communities in which they occur also may experience negative outcomes associated with an attack. For example, students in the process of grieving a personal loss (e.g., the death of a peer, colleague) may cope with grief through using or abusing drugs or alcohol, isolating themselves from others and social situations (Balk, 1997; Balk, 2001), decreasing their involvement in college activities (Tinto,
13 1993), or even challenging the autho rity of community or college authority figures (e.g., police officers, college administrators) ( Ephraim 1997). Violent attacks also inspire fear in individuals who have no direct relationship to the victims. For example, in August 1990, the gruesome kill ings of five college students in Gainesville, Florida by Daniel Harold Rawlings caused many individuals to arm themselves and evacuate the community adjacent to the University of Florida where the attacks occurred. Although the killings did not occur on ca mpus, many students stopped attending classes and other college functions due to fear and safety concerns. At the time, some media personnel even speculated that the University of Florida would have to close for the semester due to low student attendance ( Mercer, 1996) Almost universally, violent attacks in campus communities inspire a sense of vulnerability in students that may be alien to them when faced with other risks. In general, students display an optimistic bias about not contracting a sexually t ransmitted disease (Ellen, Boyer, Tschann, & Shafer, 1996 ), developing a substance abuse disorder (Hansen, Raynor, & Wolkenstein 1991 ), becoming pregnant (Smith, Gerrard, & Gibbons, 1997) or being affected by a natural disaster (Helweg Larsen, 1999 ). Howe ver, less than a third of college students (32%) believe they are less likely than their peers are to b e the victim of a violent attack at school. Instead, the majority of students (68%) exhibit either a pessimistic bias (e.g., believing they are more like l y than others to be victimized) or no bias, thus indicating that they perceive no difference between themselves and others in their likelihood to be a victim of a violent attack at school (Chaplin, 2001)
14 Contemporary R esponses to Violent Attacks on Scho ol C ampuses The Campus Security Act Various attempts have been made to protect campus communities from violent attacks. The Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics Act (Public Law 105 244) 1998 Higher Education Amendments Act to increase knowledge of crimes perpetrated on college campuses (Janosik & Gregory, 2002). Named after Jeanne Clery, a 19 year old Lehigh University student who was raped and murde red on campus, the Campus Security Act requires all colleges and universities that receive federal funds to collect and disclose information about campus crime (Lipka, 2009) Despite being monitored by the United States Department of Education, w hich can impose civil penalties against institutions for each infraction (up to $27,500 per violation) and can suspend the receipt of federal student financial aid, institutional compliance with this Act has been poor (Harshmen, Puro, & Wolff, 2001). A stu dy of the willingness of 4 year higher educational institutions to provide campus crime data to prospective students found that only 25% of institutions complied with regulations imposed by the Act (Gehring & Callaway, 1997). A similar study that explored the provision of campus security information in 2 year colleges found an even lower percentage of institutions (22%) to provide information that is required for compliance with the Act (Callaway, Gehring, & Douthett, 2000). However, whether compliance with the Act actually would improve safety of college campuses is not clear (Lipka, 2009). Most college students are unaware of this Act (71%) and among those who are, few (7%) reported that campus crime statistics influenced their decision to matriculate at t he institution of their choice (Janosik, 2001).
15 Zero Tolerance Policies Toward Violence In addition to increasing awareness of threats, many educational institutions have adopted zero tolerance policies toward bringing weapons on campuses. The goal of these policies is to improve school safety by prohibiting weapons on campuses and increasing the severity of consequences for policy violations (e.g., an automatic yearlong expulsion for a first offense). Zero tolerance policies became more prevalent in se condary educational institutions following the passage of the 1994 Gun Free Schools Act. As of 1999, 94% of U.S. elementary and secondary schools reported having zero tolerance policies toward the possession of firearms on school campuses and 91% reported having similar policies for the possession of weapons other than firearms (i.e., knifes) (Skiba & Peterson, 1999). However, despite the widespread implementation of these policies, a national review by the American Psychological Association Zero Tolerance Task Force (Skiba et al., 2006) found no evidence that zero tolerance policies are effective at preventing adolescents from bringing weapons to school. Moreover, these policies have been criticized for being implemented either too rigidly or flexibly (Skib a & Knesting, 2001); being overly punitive towards minority students, males, and students receiving special education services (Skiba, Michael, Nardo, & Peterson, 2003); and failing to be applied to adolescents who may pose a significant threat to the scho ol environment (Tobin, Sugai, & Colvin, 1996). For example, case reports indicate that many school shooters are actively suicidal during the time of the attack and unconcerned with the consequences for their actions. Even if implemented fairly and with fi delity, it is not clear whether students who pose significant threats to campus safety can be identified accurately and punished effectively by zero tolerance policies. An investigation by the U.S. Secret Service and U.S. Department of Education ( Vossekuil Reddy, Fein, Borum, & Modzeleski, 2000 ) found that nearly two thirds (63%) of school shooters never had been in trouble or rarely were in trouble at school. On the other hand,
16 zero tolerance policies tend to identify and punish students who engage in chr onic but less extreme forms of delinquent behavior (e.g., stealing, fighting) (Tobin et al., 1996 ). Thus, important differences seem to exist between students who commit violent crimes (e.g., homicide, school shootings) and those who engage in nonviolent d elinquent behaviors (Cornell, 1990; Cornell, Benedek, & Benedek, 1987). For example, compared to students who engage in non violent forms of delinquent behavior, students who had committed homicide tend to be less likely to have had a history of mental ill ness, to have prior arrests, and to be placed in a juvenile detention facility (Cornell et al., 1987). Therefore, zero tolerance policies are unlikely to prevent violent attacks and instead they may discriminate against students who engage in nonviolent de linquent behaviors. Security Technology Along with zero tolerance policies, many schools have attempted to increase campus security in an attempt to deal with school violence. During the past 20 years, countless school districts have spent hundreds of tho usands of dollars on increasing security measures. Some of these measures include hiring security guards and installing video surveillance cameras, emergency phones, metal detectors, X ray baggage scanners, and duress alarms. However, little is known about the effectiveness of these measures (Garcia, 2003). Given an absence of data supporting their efficacy, the implementation of security technologies may be too costly for educational institutions to install and maintain (Reddy et al., 2001). Furthermore, r ecent increases in campus security and security technology pose new challenges for college administrators who must balance the need to maintain campus safety with a desire to uphold a college environment that fosters the open sharing of ideas and informati on (Schuh, 1999; Sewell & Mendelsohn, 2000).
17 Concealed Weapons Permits Lawmakers in several states are debating whether citizens with concealed weapons permits should be allowed to carry weapons on campuses. As of January 2010, only the state of Utah allo ws such practices. Whether additional states will pass legislation that allows weapons on campuses in 2010 is unclear. Proponents of easing weapons restrictions on campuses claim that ves during violent campus attacks (Harnisch, 2008). However, the belief that increasing the number of weapons on campuses will increase campus safety or decrease the probability of a campus attack is not supported by data. Additionally, most students, law enforcement personnel, and college administrators strongly oppose provisions to allow concealed weapons on campus. The presence of lethal weapons on campuses could escalate the lethality of conflicts among students, enable suicidal students increased acces s to lethal means to commit suicide, increase the chance for police to mistake an armed student for an attacker, and increase collateral shootings among attackers and bystanders during attacks (Harnisch, 2008). Additionally, whether bystanders would be abl e to intervene effectively during attacks without extensive training, how this training would be provided, and if educational institutions would be responsible for providing it are unknown. Profiling of School Attackers Various profiles have been developed that attempt to identify potential school shooters before attacks occur. An offender profile developed by the Federal Bureau of Investigation (FBI) (1999) exemplif y this effort. However, the use of profiling techniques to identify future school shooters have been criticized as concerns have been raised about the selectivity and specificity of shooter profiles (Reddy et al., 2001). Most individuals who fit a specifie d profile will not commit
18 acts of violence and many school shooters do not embody characteristics described in the extant shooter profiles (e.g., having fascination with violence, being socially isolated) (Fein et al., 2002; Sewell & Mendelsohn, 2000). Fo r example, Steven Kazmierczak displayed few indicators that he was a threat to campus safety before he killed 5 individuals and injured 18 others in a mass shooting at Northern Illinois University (NIU). Kazmierczak was a high achieving and highly regarded student (Heinzmann & St. Clair 2008) who recently had co author Self in jury in pu blished in Criminology & Public Policy prior to this shooting. Kazmie rczak also was the Vice President of the NIU chapter of the American Correctional Association had expressed an interest in becoming a social worker, and had worked at a correctional facility for women. After the attack, Kazmierczak was described by the NI U police as having been normal and unstressed (Friedmen, 2008). Northern Illinois University President John G. Peters was recorded to have said that very good academ The questionable a ccuracy of shooter profiles constitutes a major concern about using profiling techniques to identify school shooters (Fein et al., 2002; Reddy et al., 2001). The offender profile developed by the FBI is based on six school shootings (Band & Harpold, 1999) and these six shootings are not representative of the more than 30 school shootings that occurred in the U.S. over the past 20 years (Reddy et al., 2001) Additionally, the Classroom Avenger profile (McGee & DeBernardo, 1999) has been criticized for assumi ng that all school shooters are Caucasian when students from diverse backgrounds also have perpetrated violent attacks in schools (Reddy et al., 2001).
19 School shootings are infrequent and considerable variability exists in the nature of attacks (e.g., tar geted students vs. random victims) and in shooter characteristics (e.g., single shooter vs. multiple shooters, dominant vs. minority group status, presence/absence of psychopathology) (Langman, 2009). Thus, it currently is impossible to predict potential s chool shooters or ways in which they may act (Greenberg, 2007). In this vein, a key finding from the Final Report and Findings of the Safe School Initiative is that no accurate or useful profile exists to identify school shooters or students who may engage in violent attacks (Vossekuil et al., 2000 ) The authors of the report further caution that two substantial risks emerge from overly relying on profiles to p rofile never will perpetrate an act of severe violence. Second, the use of profiles will inevitably fail to identify some students who do in fact pose a risk of violence but do not share some characteristics with prior school shooters (Vossekuil et al., 20 01). Moreover, the use of profiling techniques to identify school shooters may have negative consequences for students and other members of the school community (Morse, 2000). In addition to being unpopular, profiling techniques may stigmatize and label ce rtain students as threats unfairly. For example, contrary to popular perceptions, few school shooters have histories of abuse (Langman, 2009), psychotic video games (F erguson, 2008). Threat Assessment Approach to Preventing Violent Attacks The School Shooter: A Threat Assessment Perspective for the Analysis of Violent Crime. The purpose of this report was to provide schools with a blueprint for conducting threat assessments to diffuse threats. It encourages schools to use a multidisciplinary team of professionals to evaluate school threats using a four pronged
20 assessment model, including 1) the personality and behavior of the student who made a threat, 2) dynamics of the larger community. However, despite th ese recommendations, this FBI report does not describe how the aforementioned evaluations should be conducted. In this vein, Cornell and colleagues (2004a) refined and field tested the procedures outlined in The School Shooter drawing on information from 188 student threats in 35 primary (grades K 5) and secondary (grades 6 12) schools over the course of one school year. The study began with the organization of multidisciplinary threat response teams that consisted of school administrators (e.g., superint endent, principles), school psychologists, and school resource/law enforcement officers. Preliminary meetings with school administrators revealed concerns about a lack of clear guidelines for evaluating threats. Additionally, school psychologists expressed concerns about their limited training in conducting psychological evaluations on students who had made threats of violence. A decision tree model was adopted to guide school officials through the threat assessment process to address these concerns. School officials then were trained in conducting threat assessment procedures. The first step of the threat assessment process was for a school administrator to gather information about a threat for triage purposes and to determine if additional information was needed. Thus, a school administrator was responsible for determining if he or she would be able to resolve the threat or whether additional members of the multidisciplinary team needed to be consulted. While interviewing a student to assess the severity of a threat, school administrators were advised to assess what had happened to instigate the threat, what the student meant by the making the threat, and whether the student intended to carry out the threat. After compiling
21 transient (i.e., could be resolved quickly and easily with an apology or explanation) or substantive (i.e., lasting and enduring threats of causing serious harm to others). Transient threat statements often dissipate. Substantive threats represent a sustained intent to harm someone beyond the immediate situation in which a threat was made (Cornell et al., 2004b). Therefore, although transient threats warrant a prompt disciplinary response (Cornell et al., 2004a), school administrators are strongly advised to consult wit h the multidisciplinary team on substantive threats in order to take protective action to prevent emerging threats of violence and to develop an intervention plan to address any issues that appear to have precipitated the threat (e.g., bullying). The use resolve the majority (70%) of threats quickly and efficiently (Cornell et al., 2004a). The remaining 30% of threats required a more extensive evaluation and intervention plan, incl uding conducting a comprehensive psychological evaluation, designing interventions that involve caregivers or law enforcement officers, and notifying potential victims of targeted violence. During this study, only 3 students were expelled, and no acts of v iolence were carried out. Cornell and colleagues (2004a) concluded that student threat assessment is a feasible and practical approach for schools that deserves further research. Although the results of this study support the use of threat assessment proc edures delineated by the FBI, they do not inform how threats are reported to school officials or potential barriers to threat reporting (Stueve et al., 2006). Consequently, Cornell and colleagues (2004a; p. tudents respond to threats, how they determine
22 how many students do not report being victimized until it becomes unbearable (Unnever & Cornell, 2004), some students may fail to report threats or threatening statements when they are initially made and when a quick response could obviate an act of violence. Student Threat R eporting Threatening messages and statements made by students should be considered caref ully. Results from the U.S. Secret Service and U.S. Department of Education Final Report and Findings of the Safe School Initiative suggest that at least one individual has prior knowledge of an attack before it occurs in the majority (81%) of school shoot ings (Vossekuil et al., 2000 ) Furthermore, another study found that all school shooters that were studied in nine shootings had communicated their violent intentions to others before the attack, often including the time and place of the future attack (Ver linden, Hersen, & Thomas, 2000). Lastly, a New York Times study on rampage killings (i.e., violent attacks in public places that involve multiple victims) found that 63 of the 100 attackers had made threatening statements publicly before initiating an atta ck. Collectively, these data suggest that, in most cases, some individuals are aware that an attack is imminent or being planned before it is carried out. In fact, several individuals may have knowledge of an attack before it occurs. For example, more than half (59%) of school shooters tell more than one individual of their violent intentions before carrying them out (Vossekuil et al., 2000 ) A follow up study to the Final Report and Findings of the Safe School Initiative explored relationships between bys tanders (i.e., individual with prior knowledge of attacks) and attackers as well as how bystanders obtain information about potential attacks (Pollack, Modzeleski, & Rooney, 2008). Results of this study indicate that 34% of the bystanders are friends of at tackers, 29% are peers (i.e., acquaintances, co workers, classmates), 6% are family members, and 31% are other acquaintances or do not have close relationships with the attackers. Additionally, 82%
23 of bystanders appear to receive information about an attac k directly from an assailant, and only 13% receive this information secondhand, suggesting that most individuals with prior knowledge of school shootings often have close relations with attackers. Unfortunately, however, these findings may fail to capture words, it is not clear whether the cases investigated in the Final Report and Findings of the Safe School Initiative are indicative of low student willingness to report threats in genera l or represent report threats of violence. Although no studies have directly explored college students willingness to report threats of violence, four studies on this topic have been conducted with adolescent students. One study reported willingness to tell an adult if they heard another student state that he/she was going to shoot someone. Upon questioning, slightly more than half (5 4%) of the students reported that they would be willing to tell an adult if they had prior knowledge of a shooting (Gaughan, Cerio, & Myers, 2001). In another study, about three fourth (74%) of students reported that they would tell an adult if they heard a student threaten to hurt someone actions: intervene directly, tell a teacher or principal, discuss the situation with a friend but not an adult, and do nothing (Syvertsen, Flanagan, & Stout, 2009). Results of this study indicate that students favored taking action on their own over the other options, which may suggest low willingness to report threats. Lastly, a fourth study found that majority of adolescents (70%) were willing to report if they knew that another student had a weapon at school (Brank et al., 2007). However, few students reported they definitely would report (5%) a student with a deadly
24 weapon. Also in this study, a higher percentage of adolescents responded that they would report peers if they could do so anonymously (83%), and fewer students (58%) responded that they would report their friends. Threat Repo rting in College S tudents Compared to adolescent students, limited research exists on threat reporting behaviors among college students. One study investigated reporting behaviors in victims of violent crimes in an effort to assess the reliability and val idity of campus crime statistics generated by the Crime Awareness and Campus Security Act (Sloan III, Fisher, & Cullen, 1997). Results from this study suggest that a minority (25%) of campus crime victims report their victimization to campus authorities. F found that 10% of college students report being victims of campus crime; and among these students, 43% actually report their victimization to an authority figure (Revels, 1999). Students who do not report being victims of crime mot commonly respond that their victimizations are willingness to report threats, thus seriousness of threats and how reporting threats may affect them socially. Why do an alarming percentage of students display a low willingness to report threats? To answer this question, research will be reviewed as it relates to a variety of reporting behaviors (e.g., tattling, whistleblowing) that have been investigated with different populations (e.g., adolescent students, non college populations of adults). Although the research base on these topics is evolving, it is sufficiently developed to suggest that some specific qualities influence
25 Trust in the College Support System Trust involves an expectancy held by an individual that the words, promises, and intentions of others can be relied on (Rotter, 1971). Trust is important for interpersonal and interorganizational functioning in industrial/organizat ional settings and it provides a basis for clear and effective communications between group members (Zaheer, McEvily, & Perrone, 1998). Interorganizational trust improves intuitional performance, negotiation, and conflict resolution. However, interpersonal trust does not exert a direct influence on the former three variables, suggesting that efforts to improve organizational harmony require individuals in organizations to trust the organizations more than to trust the individuals in the organizations (Zahee compared to trust in specific individuals within the college (e.g., the college president, security officers, professors), may be critical to establishing harm ony in the institution. report threats to campus safety (Brank et al., 20 07; Gaughan et al., 2001; Greenberg, 2007; Pollack et al., 2008; Stueve et al., 2006). Adolescents who report threatening peers generally have positive relationships with adults, believe that their reports will be taken seriously, and think that threats wo uld be addressed appropriately by figures of authority. Moreover, adolescents who feel they will be believed, have strong relationships with adults, and think that the information they provide will be protected are more likely to report threats (Brank et a l., 2007). In contrast, adolescents who do not report threatening peers tend to anticipate negative responses from authority figures resulting from their sharing information about potential threats (Pollack et al., 2008).
26 A study that explored crime repo rting behaviors in college students found about a quarter (24%) of students to not report being victimized due to a lack of trust that campus police would vi ctimizations because they believed the police would not recover the lost property or find an offender. Furthermore, 7% did not report because they thought the police could not solve the crime and a small percentage (1%) of students did not report due to fe eling harassed or threatened by campus police. Thus, even if the majority of college students trust campus police to respond effectively to crime reports, the lack of trust exhibited in a somewhat large minority of students (24%) still may impede the repor ting of some threats of violence. Campus Connectedness Cam pus connectedness, an aspect of social connectedness, relates to feelings of closeness to others in a campus community (Lee, Sexton, & Keough, 2002). Research on social connectedness among col lege students indicates that students who feel strongly connected to their campus climate tend to become more involved in social activities and demonstrate higher levels of academic achievement (Lee et al., 2002). Conversely, students who lack a sense of c onnectedness tend to feel socially isolated and are prone to feelings of loneliness, low self esteem, and higher rates of anxiety and depression (Hagerty & Williams, 1999; Lee & Robins, 1995). Additionally, being perceived as socially disconnected on campu s may be socially stigmatizing (Lau & Gruen, 1992). The stigma associated with low social connectedness may explain why socially disconnected individuals often appraise social environments negatively (Pinel, 1999) and have negative perceptions about thei r friends, roommates, and people in general (Jones, Sansone, & Helm, 1983; Rotenberg, 1994). Furthermore, socially disconnected individuals may eventually develop beliefs that the college campus is uninviting and these beliefs may prevent them from
27 engagin g in social behaviors, thus perpetuating their social isolation as they avoid social interactions (Lee et al., 2002). Some socially disconnected students even may blame their lack of connectedness on others whom are seen as unfriendly instead of (or in add ition to) internalizing their feelings of isolation (Hurtado & Carter, 1997). In a similar vein, some students who feel disconnected and alienated from the campus climate may be less inclined to report threatening students because they either may be unawar e of where to make reports, distrust authority figures, feel awkward around peers, or be unconcerned with maintaining the safety of the campus climate (Brinkley & Saarnio, 2006; Pollack et al., 2008; Syvertsen et al., 2009). Self E fficacy Toward Service Self efficacy toward service (i.e., the belief that one can have a prosocial impact on his or (Weber, Weber, Sleeper, & Schneider, 2004). Thus, individuals with st rong self efficacy toward service may be likely to report threatening students, given their expected commitment to maintaining a safe campus environment. One study found general self efficacy to influence after contr olling for age, gender, and work experience (MacNab & Worthley, 2008). Specifically, individuals with higher levels of self efficacy were more likely to report unethical actives than were individuals with low self efficacy. Trust in the college support sy efficacy toward service and their willingness to report threatening students. Although no research that directly explored these relationships could be located, research on virtual communities suggests that trust must be established between individuals before self efficacious individuals (i.e., individuals who are competent in their abilities to share information) willingly share information with others electronically (Hsu, Ju, Yen, & Chang, 2007). Perc eived risks associated with sharing knowledge electronically, including social threats (e.g., being perceived negatively
28 by others in the virtual community) (Wasko & Faraj, 2005), the potential for lost opportunities work), or being vulnerable to the influence of others in the virtual community decrease willingness of individuals to share information. Thus, a similar relationship may exist between trust, self efficacy toward service, and threat reporting. For example, a student who feels confident in his or her judgment and ability to inform an authority figure about a threat to campus safety but does not trust this individual to act reasonably when informed of a threat may not report the threat. Fear of Negative Evalu ation Fear of negative evaluation, a central component of social anxiety, involves fear of being judged negatively by others (Watson & Friend, 1969). This fear often is maintained by nyone see me make a true without question (Clark & Wells, 199 5; Rapee & Heimberg, 1997). Students tend to become increasingly concerned with peer evaluations when parental influence decreases in adolescence and peer relations increase in importance (Beck & Treiman, 1996; Bro wn, Dolcini, & Leventhal, 1997; Perkins, 2002). In response to negative peer norms, some students engage in high risk behaviors (e.g., binge drinking, sexual promiscuity) to avoid negative peer evaluations (Borsari & Carey, 2001; Scholly, Katz, Gascoigne & Holck, 2005). Similarly, individuals wh o fear negative social evaluations may withhold information about warrants censure. Social costs and negative perceptions associated with tattling or other f orms of peer reporting also may exist among adults. For example, attorneys tend to eschew turning in a fellow
29 attorney for misbehavior due to possible negative social consequences (Toomey, 2004). Furthermore, peer reporting in military and law enforcement settings also is influenced by similar social factors. Despite having organizational policies that require U.S. Naval Academy cadets to report other cadets who engage in inappropriate behaviors, few midshipmen report their peers due to informal norms about peer loyalty, norms that override the organizational polices that cadets expected to follow. In fact, loyalty to a friend is the most influential factor determining whether cadets will report inappropriate behaviors of other cadets (Pershing, 2002). Simil arly, police officers who report other officers for misconduct are seen as violating an Similar to these settings, fear of negative evaluation also may curtail th e reporting of behaviors in college students. An investigation of reporting practices in college students found the most common response students gave for not reporting a serious threat was that the threat was a ndicating that students may eschew reporting out of a fear of negative social consequences. Delinquency Delinquent behavior is considered unlawful and against prevailing social norms and conventions. Adolescent students who report a history of delinquen cy appear to display a low willingness to report treats of violence (Brank et al., 2007; Brinkley & Saarnio, 2006). This phenomenon appears to be even more exaggerated in adolescents who carry weapons. For example, among students who reported having carrie d a weapon to school, 74% responded that they would not report another weapon carrying student even if the student threatened to harm other students (Brinkley & Saarnio, 2006). behaviors have not been explored directly, research on other adult populations suggests that delinquency
30 may inhibit their willingness to report threats. For example, a study of active street offenders Jacobs, & Wright, 2003). All interviewed street offenders denied reporting criminal activity to police or other authority figures and displayed contempt for those who had snitched on others. Furthermore, this study a lso explored how reporting behaviors may be influenced by street officers despite acknowledging their contributions to maintaining public order. Individuals who have had negative experiences with authority figures are less likely to trust and rely on these individuals for protection (Hurst & Frank, 2000). Mutual distrust characterizes relationships between police officers and delinquent police informants, wit h the latter perhaps somewhat motivated to report criminal activity to receive a financial reward or a more lenient criminal charge (Dodge, 2006; Rosenfeld et al., 2003). Police officers often question the veracity of information reported by informants (Po grebin & Poole, 1993) and informants report being exploited or mistreated by police officers (Marx, 1998). Furthermore, individuals in delinquent social groups may be disinclined to report threatening behaviors because of social pressures not to cooperate with authority figures. Delinquent students tend to belong to social networks that condone delinquency and share similar values, norms, and beliefs. Thus, the fear of social exclusion may prevent some individuals from reporting if they think that in their social network other individuals will perceive such acts negatively (Haynie, 2001) Additionally, in delinquent groups, beliefs about informing authority figures about delinquent behaviors appear to be well established. Cooperation with authority figures m ay cause delinquent individuals to be considered weak, ineffective, or untrustworthy by their delinquent peers (Rosenfeld et al., 2003). Lastly, cooperating with authority figures may place individuals at risk
31 for being victimized if they are members of de linquent social groups or live in communities with high crime rates (Rosenfeld et al., 2003). This risk may further dissuade these individuals from reporting individuals who may engage in violent crimes and could potentially retaliate. Research Ques tions for Investigation This literature review suggests that several variables could be reasonably expected to affect ystem, fear of negative evaluation, history of delinquency, feelings of campus connectedness, and self efficacy toward service. A primary goal of this study is to test a comprehensive model that includes these variables. Additionally, specific relationship s between variables will be tested to determine paths of influence between variables. The primary hypotheses are as follows: 1. report threats to campus safety. 2. Campus campus safety. 3. Self efficacy toward service threats to campus safety. 4. Fear of negative evaluation will b threats to campus safety 5. safety. Trust may be an important mediator of relationships between several of the aforementioned variables. Delinquent adolescents and adults often distrust authority figures (Dodge, 2006; Hurst & Frank, 2000; Rosenfeld et al., 2003; Stoutland, 2001) and are less likely than non delinquent students are to report threats (Brank e efficacy toward service beliefs may be influenced by their trust in campus authorities. Thus, trust in the college
32 support system was expected to partially mediate relationships between delinquency and illingness to report threats as well as self efficacy toward service and willingness to report. Secondary hypotheses are as follows: 6. Trust in the college support system will partially mediate relationships between delinquency eport threats. 7. Trust in the college support system will partially mediate relationships between self efficacy Lastly, there may be relationships between other study variables (i.e., campus conne ctedness, fear of negative evaluation) and trust in the college support system even though research could not be located that establishes these relationships. Therefore, in addition to testing the aforementioned hypotheses, alternative models will test the se putative relationships.
33 CHAPTER 2 METHODS Participants Data collection was conducted in two prongs at the University of Florida. During the spring and fall of 2009, 160 participants were solicited through an undergraduate research pool (prong 1). Tw o participants were excluded from analyses due to scoring high on a measure of social desirability. During the fall of 2009, the second prong of data collection was conducted. Four thousand undergraduate students were solicited to participate in this study via email. Among these students, 678 participated (17% response rate). An additional 15 participants were excluded due to scoring in the extreme range on a measure of social desirability. Thus, valid data were obtained from 663 participants. Data collecte d at prongs 1 and 2 were combined since no significant differences were found between participants on willingness to report threats of violence, t (818) = 1.67, p = .11, and trust in the college support system, t (818) = 1.43, p = .15 Therefore, the total s ample included 820 participants. Table 2 1 presented means and standard deviations for study variables for both participant groups (i.e., participants from prongs 1 and 2 of data collection). Approximately 70% of the total participants were female (30% ma le). Sixty five percent identified as White/Caucasian, 9% as Black/African American, 14% as Hispanic/Latino, 7% as Asian, and 5% as Mixed Race. Twenty six percent of participants listed their class standing as sophomore, 33% as junior, and 41% as senior. E ighty two percent of participants lived off campus and 18% lived on campus. Approximately, 57% of participants participated in at least on extra curricular activity, 4% were student athletes, 16% were members of a fraternity or sorority, 24% were members o f an honor society, and 30% were not affiliated with any of the
34 aforementioned groups. Overall, undergraduate students from 86 different academic majors were sampled. The most common major was psychology with 44 students. Procedure Data first were collected from an undergraduate research pool (prong 1). These students were enrolled in classes offered by the Department of Educational Psychology (e.g., Human Growth and Development, The Adolescent, Teaching Diverse Populations) and were required to par ticipate in a research study to meet a class requirement. Students in the research pool generally have diverse majors, as 42 different academic majors were represented in this first sample. Students who provided data in prong 1 accessed the study questions using an electronic portal provided by the Department of Educational Psychology. Data were collected using SurveyMonkey, a widely used survey generator. All participants answered all survey items. Thus, there were no missing data. The second prong of dat a collection involved a large sample of undergraduate students who provided data on their own volition. These students (N = 4,000) were selected randomly by the Office of the Institutional Research at the University of Florida. All students were enrolled a s sophomores, juniors, and seniors. The Assistant Vice President for Student Affairs first contacted students to encourage their participation in the study. Each student then receive d one to four emails that included a link to the study. Students were cont acted weekly (on Monday or Tuesday) for a period of four consecutive weeks. Participants ac knowledged their agreement to the terms of the study by signing an electronic c onsent f orm and providing data (Appendix A). Participants were excluded from receiving follow up contacts and having access the survey a second time after they completed all survey questions. Thus, responses were independent and participants provided data only on one occasion. Data only were collected from participants who completed all sur vey items to ensure equal cases for each variable. Students were excluded if
35 they scored high (i.e., total score above 17) on a measure of social desirability. Student Voice, an online survey generator and manager, was used to collect data and email studen ts Representatives at Student Voice copied the SurveyMonkey protocol that was previously used in the educational psychology research pool to ensure that the data collection processes were identical for both participant groups. Student Voice and SurveyMonk ey do not allow third parties to access or view collected data. They also hold provided materials ( e.g., images, email addresses ) securely, in confidence and prevent the collection of personally identifiable information that could be used to identify part hese provisions help ensure compliance with guidelines specified by the American Psychological Association (APA, 2002) for data collection and researc h participation, the Health Insurance Portability Accountability Act and the Common Rule (45 CFR 46) r egarding human subject research. At the completion of the study, students were provided with a secure email address to responses remained anonymous and independent from their survey responses. Incentives included one free Apple ipod Nano 8GB digital music player valued at $200 and gift cards for Outback Steakhouse valued at $50.00 (quantity = 1), Sports Authority valued at $40 (quantit Grill valued at $25 (quantity = 1), TGI Fridays Restaurant valued at $25 (quantity = 2), the Regal Entertainment Group valued at $10 (quantity = 5), and Starbucks v alued at $5.00 (quantity = 12). Measures Campus Connectedness Scale The Campus Connectedness Scale (Summers, Beretvas, Svinicki, & Gorin, 2005) is a self report 14 item adaptation of the 20 item Social Connectedness Scale (Lee & Robins, 1995).
36 The measure connectedness to the campus community. Campus connectedness items are rated on a 6 point S cale convergent validity (e.g., with a measure of classroom community [ r = .55]), and discriminant validity (e.g., with a measure of group processing [ r = .17]; Summers et al., 200 5). Self Efficacy Toward Service Scale The Self Efficacy Toward Service Scale (Weber et al., 2004) is a five item self report measure that measures the strength of beliefs that one can have an impact on their college community. Items on the Self Effi cacy Toward Service Scale are rated on a five point scale, Efficacy Toward Servi ce Scale demonstrates good internal consistency ( = .80). Positive associations have been found between the scale and measures of civic participation ( r = .59) and attitudes toward helping others ( r = .72). The Self Efficacy Toward Service Scale is uncorrelated with a theoretically unrelated measure of fashion trends ( r = .05 ; Weber et al., 2004). Self Report Delinquency Scale The Self Report Delinquency Scale (Piquero, MacIntosh, & Hickman, 2002) is a 9 item self report measure that examines the frequency of minor and seriously delinquent behaviors com mitted over the last 12 months. Nine categories are used to assess the frequency of assessment of the frequency of delinquent behaviors. The Self Reported Delinquency Scale has acceptable reliability for research purposes ( = .76) (Piquero et al., 2002). The correlation between self reports of delinquency and official criminal records generally are positive and statistically significant (e.g., Elliott, Ageton, & Huizin ga, 1985; Farrington, Loeber, Stouthamer
37 Loeber, Van Kammen, & Schmidt, 1996). Five items were excluded from analyses and 4 were retained due to limited variability in responses across items and categories. Retaine d items then were dichotomized into being either positive or negative for delinquency. Brief Fear of Negative Evaluation Scale Revised The Brief Fear of Negative Evaluation Scale Revised (Carleton, McCreary, Norton, & Admundson, 2006) is a 12 item measure of fear of negative social evaluation. I tems on the Brief Fear of Negative Evaluation Scale Revised are rated on a 5 Brief Fear of Negative Evaluation Scale Revised has excellent internal c onsistency ( s = .95 to .97) (Carleton, Collimore, & Asmundson, 2007; Carleton et al., 2006). The scale also displays convergent validity with measures of social anxiety ( r = .60) and fear of socially observable anxiety reactions ( r =.55), as well as divergent validit y with measures of fear of illness ( r = .38) and fear of injury ( r = .28) (Carleton et al., 2007; Carleton et al., 2006). Marlowe Crowne Social Desirability Scale Short Form X2 The Marlowe Crowne Social Desirability Scale Short Form X2 (Strahan & Gerbasi, 1972) is a 10 item adaptation of the commonly used 33 item Marlowe Crowne Social Desirability Scale (Crowne & Marlowe, 1960). These measures purport to measure the tendency to reply in a manner that will be viewed favorably by others. Items on the Marlowe Crowne Social Desirability Scale total scores range from 0 to 10. Factoral validity has been found for a single factor model of the Marlowe Crowne Social Desirability Scale Short Form X2 (Leite & Beretvas, 2005). Mixed internal consistency estimates have been reported for the measure, ranging from the low to acceptable range (Beretvas, Meyers, & Leite, 2002). In the current study, the Marlowe Crowne
38 Social Desirability Scale Short Form X2 demonstrated low reliability ( = .35). The measure was used only to exclude participants who scored in the upper extremes for social desirability. Pilot Study Instrument Development A pilot study was conducted to assess the psychometric characteristics of previously unstudied measures. Participants included 93 undergraduate students enrolled at the University of Florida. Participants were solicited using an undergraduate subject pool, and data were collected in the spring of 2008. Participants completed a questionnaire containing the following measures: The Campus Connectedness Scale, the Self Report Delinquency Scale, the Brief Fear of Negative Evaluation Scale Revised, the General Self Efficacy Scale (Schwarzer, 1993), and a newly developed Trust In Col lege Support System Scale. The first three threat reporting passages (passages 1, 2, and 3) also were included ( Appendix B). Trust In College Support System Scale The Trust in the College Support System Scale (Sulkowski, unpublished) is a self report mea safety and integrity, establish channels of communication between students and college staff, and t in higher educational institutions were not located. However, measures have been developed to assess trust in other organizations, including school systems and business organizations ( Tschannen Moran & Hoy, 2000). A review of research conducted in organi zational systems identified the variables of openness, competence, beneficence, reliability, honesty, and willingness to be vulnerable as the key facets of trust ( Tschannen Moran & Hoy, 2000). Additionally, while acknowledging the data on assessing trust i n schools are limited, these authors suggest that openness, competence, reliability, and beneficence may be relevant to trust in educational settings. Therefore, particular
39 attention was paid to developing items that purport to measure these qualities when designing the Trust in the College Support System Scale. The Trust in the College Support System Scale initially included 17 items designed to assess openness (N = 5), competence (N = 5), beneficence (N = 3), and reliability (N = 4). However, 11 items we re excluded to improve ease of scale interpretability and model fit to the data. The 6 item Trust in the College Support System Scale displayed excellent fit to the data as evidenced by a non significant chi square test value, [ 2 (9, N = 93) = 15.56, p = .08] and values on fit indices: CFI = .99, NNFI = .99, SRMR = .04, RMSEA = .06. Therefore, the hypothesis that the tested model did not fit the data could not be rejected. Factor loadings for the 6 items ive or non significant residual variances were (N = 1), and competence (N = 3) as determined by their content validity and factor loadings. The Trust in the College Support System Scale employs a self report format, and items are rated on a 4 item scale used in this study demonstrated good internal consistency in the pilot st udy ( = .85). The Trust in the College Support System Scale also displayed low to modest convergent validity with the Campus Connectedness Scale ( r = .32) and General Self Efficacy ( r = .21). Support for the Self Reported Delinquency Scale ( r = .43) and the Brief Fear of Negative Evaluation Scale Revised ( r = .21). Threat Reporting Decision Passages threatening stu dents could not be located. The use of such passages allow a researcher to present a hypothetical situation to participants to which they can respond in order to assess what they may do in a specific situation. Pilot study participants read and responded t o three passages that
40 used this format. The passages described common characteristics of school shooters identified in a report from U.S. Secret Service and U.S. Department of Education (Vossekuil et al., 2000). Participants were asked to read background i nformation about a student, to assume they had overheard them make a specific threat statement, and then to rate their willingness to report this person (Appendix B). n the would be much better with a few of these stuck up s each type of threat and suggests that direct threats are most likely to pose imminent and serious danger to the sa fety of others. However, given an absence of data that suggest that the type of a threat (i.e., direct vs. indirect) influences students threat reporting decisions, each passage was followed up with a direct, indirect, veiled, and conditional threat to con trol for the influence that Efforts were made to create passages that accurately portray students who may pose a significant risk to campus safety. Thus, common characteristics of school attack ers identified in the U.S. Secret Service and U.S. Department of Education report (Vossekuil et al., 2000) were ordered by the degree to which they were shared by attackers. For example, all characteristics that were common to at least 80% of attackers wer e categorized in category 1, all characteristics shared by 60% to 79% of attackers were categorized in category 2, all characteristics shared by 40% to 59% of attackers were categorized in category 3, and all characteristics shared by less than 40% of atta ckers were categorized in category 4. Percentages of characteristics shared by
41 attackers in the U.S. Secret Service and U.S. Department of Education report (Vossekuil et al., 2000) by each respective category are listed in Table 2 3. Passage 1 included th e following characteristics from category 1. Chris, the student in question, recently had experienced a major loss with which he was having difficulty coping. Passage 2 was designed to contain fewer severe characteristics. In fact, it contained only one ca tegory 1 characteristic (e.g., Thomas felt persecuted). Passage 3 described characteristics found in the first two categories. Drawing from category 1, Ray was described as experiencing difficulty coping with a major perceived loss; drawing from category 2 he also was described as experiencing periods of major depression and had experienced a major loss to his social status. After reading the passages, participants rated their willingness to report on a 4 point scale analyses yielded good internal consistency on the 12 threat statements ( = .89). Instrument M odifications Linear regression was used to assess whether participants scores on the campus connectednes s, delinquency, fear of negative evaluation, general self efficacy, and trust in the college support system predicted their responses to threat statements. No significant association was identified between the aforementioned independent variables and parti cipants responses to threat statements, R 2 = .03, F (1, 84) = 1.05, p = .40. The lack of association between variables may have been due to the relatively small sample size (N = 93), a somewhat high percentage of missing data (10%), and limited variability Regarding the latter, many participants may have responded in a socially desirable manner, as a point scale was used.
42 Additional Provisions to Address Pilot Study Limitations In the current study, the following measures were taken to address limitations found in the pilot study. First, electronic data collection prevented selective answering and missing data. Second, additional threat reporting passages were designed to increase the number of threats to which participants would responded. A female gender passage also was included to reflect the possibility that school attackers could be male or female. In fact, the first campus shooting perpetrated by a female occurred at Louisiana Technical C ollege on February 8, 2008 (Supino, 2008), a few months after the design of the original three passages. Third, two additional response categories were added to increase the variability of answers (Appendix B). Forth, a commonly used measure of social desi rability was included (i.e., Marlowe Crowne Social Desirability Scale Short Form X2), which allowed for the exclusion of participants who respond in an unrealistically favorable manner. Lastly, two senior police offers (i.e., Chief of Police, a Police Majo r) and two other police officers from a critical incident response team reviewed the threat passages and statements. Independently, the police officers all reported that the proposed threats were germane, realistic, and would warrant immediate response fro m the department. No substantive changes to the threat passages or items were suggested during this review. Statistical Analyses Preliminary analyses were conducted with Statistical Package for the Soci al Sciences (SPSS) version 16.0. Dummy coding was applied to categorical variables with more than two levels (e.g., race/ethnicity). Linear regression was used to test the associations among demographic characteristics and variables related directly to research questions (e.g., trust in the college suppor t system, willingness to report). Variables representing demographic characteristics were entered into the model simultaneously to control for the influence of these variables on each other.
43 This study used a two step analysis approach to structural equat ion modeling (SEM). Step 1 involved specifying a measurement model using first order CFA with categorical factor indicators Step 2 involved specifying a series of structural regression models that tested the associations among latent variables. Mplus vers ion 5.21 (Muthn & Muthn, 1998 ) with weighted least squares estimation was used to speci fy and test all models. W eighted least squares estimation is a theoretically appropriate method for fitting CFA models to polychoric correlations (i.e., a correlation that estimates the linear relationship between two unobserved continuous variables in the presence of observed ordinal data ; Muthn, du Toit, & Spisic, in press ) T he chi square statistic and standard errors are adjusted using a full weight matrix with wei ghted least squares. A diagonal weight matrix is use d for parameter estimation. Further, the goodness of fit test associated with this estimation method involves multiplying the usual chi square test statistic used with maximum likelihood estimation by a r obust adjusted chi square statistic, resulting in estimated model degrees of f reedom (Flora & Curran, 2004). Thus, degrees of freedom are estimat ed rather than being determined directly from the specification of the model ( Muthn et al., in press; Muthn & Muthn 1998) Data simulation studies indicate asymptotically unbiased, consistent, and efficient parameter estimates are provided with weighted least squares. This estimation method also correct s chi square test of fit with dichotomous or ordinal observ ed variables ( Muthn & Satorra, 1995). Further, weighted least squares estimation is appropriate for use with non normal data (Quiroga, 1992) and provides stable estimates fo r large sample sizes (N > 500; Flora & Curran, 2004). Measurement Model All item s were expected to load highly on their related factor and not on other factors. Residuals for observed variables were expected to be weakly associated with latent factors. Furthermore, latent factors were expected to account for the associations among the observed
44 variables that load on each respective factor. Model fit for the measurement model was assessed using the chi square test of independence, goodness of fit indices, standardized factor loadings, and factor variances. The following goodness of fit indices were assessed: the comparative fit index (CFI), the non normed fit index (NNFI), and the root mean square error of approximation (RMSEA). A non significant ( p > .05) chi square test value indicates that a model fits the data exactly (i.e., no signi ficant discrepancy exists between the model implied and unrestricted correlation matrix) and cannot be rejected by the data. The CFI index indicates the proportion of possible improvement f rom a null model (i.e., a model that assum es the indicator variable s in the model are uncorrelated ) to a saturated model (i.e., a model in which all parameters are estimated) that is achievable by using a targeted model. The non normed fit index also reflects the proportion by which the target model improves fi t compared to the null model yet the measure does not require making chi square assumptions associated with the CFI. Lastly, t he RMSEA estimates the discrepancy p er degree of freedom between an unrestricted correlation matrix and the model implied matrix. Suggested m odel criteria include a non significant model chi square statistic, CFI .95, NNFI .95, and RMSEA Hu & Bentler, 199 9). However, results from data simulation studies with weighted least squares suggest that the traditional cutoff value for models, e specially if data are non normal (Yu, 2000). Therefore, the historically prefe rred Schumacker & Lomax, 1996 ) was used to interpret model fit. Structural Regression Models Structural regression models were compared using robust chi square difference tests with mean and variance adjusted statistics, an approach suitable for chi square difference testing with weighted least squares (Muthn & Asparouhov 2002). Significant chi square difference test
45 values indicate a significant change in model fit from the less restrictive comparison model resulting from the estimation of the test model. Exogenous (independent) variables in the model included trust in the college support system, campus connectedness, delinquency, fear or negative evalu ation, and self efficacy toward service. The endogenous (dependent) latent variables in the model included willingness to report as well as trust in college support system. In total, three structural regression models were tested. Model 1 tested all expect ed associations among study variables. Model 2 included direct effects from campus connectedness and fear of negative evaluation to trust in the campus support system. Lastly, model 3 allowed for a direct effect from self efficacy toward service to campus connectedness. Model respecifications were made in light of appropriate empirical (e.g., modification index values, expected parameter change statistics) and theoretical criteria (Kline, 2005).
46 Table 2 1. Results of independent samples t tests for study variables Sample 1 (N = 158) Sample 2 (N = 662) Variable M SD M SD t (818) 1. Willingness to report 81.96 24.89 85.56 25.01 1.67 2. Trust in the college support system 17.1 7 2.33 17.46 2.25 1.43 5 .79 ** .94 7.58 ** .42 3. Campus connectedness 57.73 10.6 0 64.28 13.26 4. Fear of negative evaluation 26.07 7.7 0 25.4 0 7.97 5. Delinquency 5.5 0 2.52 4.56 .9 0 6. S elf efficacy toward service 21.06 3.27 20.94 3.48 Note: Sample 1 = data collected during prong 1; Sample 2 = data collected during prong 2; p <.05, ** p <.01 Table 2 3 Percentages of school shooters displaying specific characteristics prior to carrying out attacks. Cate gory level Characteristics of school shooters Percentage of shooters with characteristic 1 Male 100 1 Recently experienced or perceived a major loss 98 1 Planned an attack in advance of carrying it out 93 1 Had difficulty coping with a recent l oss/perceived loss 83 2 Exhibited a history of suicidal thoughts/attempts 78 2 Had a grievance against at least one victim 73 2 Felt persecuted, bullied, or vulnerable 71 2 Recently experienced a loss of social status or major failure 66 2 Had a hist ory of feeling extremely depressed 61 3 Demonstrated excessive interest in violence 59 3 Experienced loss of a romantic relationship 51 3 Demonstrated excessive interest with weapons 44 3 Experienced a change in academic performance 44 3 Experienced a change in friendship patterns 41 4 Were frequently disciplined in school 37 4 Had previously received a mental health evaluation 34 4 Demonstrated excessive interest with explosives 32 4 Motive for attack included an attempt to gain notoriety 24 4 Had a diagnosis of a psychiatric disorder 17 Note: Data are reported by the U.S. Secret Service and U.S. Department of Education (Vossekuil et al., 2000)
47 Fig ure 2 1 Model 1: Hypothesized structural relations among var iables Note: FNE = fear of negative evaluation; DEL = delinquency; SETS = self efficacy toward service; CC = campus connectedness; TICSS = trust in the college support system; REP = willingness to report FNE DEL SETS CC TICSS REP
48 Fig ure 2 2 Model 2: Allowing direct effects of campus connectedness and fear of negative evaluation on trust in the college support system Note: FNE = fear of negative evaluation; DEL = delinquency; SETS = self efficacy toward service; CC = campus connectedness ; TICSS = trust in the college support system; REP = willingness to report Fig ure 2 3 Model 3: Allowing a direct effect from self efficacy toward service to campus connectedness Note: FNE = fear of negative evaluation; DEL = delinquency; SETS = self efficacy toward service; CC = campus connectedness; TICSS = trust in the college support system; REP = willingness to report FNE DEL SETS CC REP TICSS FNE DEL SETS CC REP TICSS
49 CHAPTER 3 RESULTS Preliminary Analyses r correlations between latent variables are presented in Table 3 1. Linear regression analyses were conducted to test for associations between demographic characteristics and variables pertinent to research questions. Specifically, associations among gende r, race/ethnicity, grade, place of residence, willingness to report, and trust in the college support system were explored No significant association was identified between willingness t b = .01, t (819) = .16, p = .88 place of re sidence (i.e., on/off campus), b = .03, t (819) = .90, p = .37 and race/ethn icity. Black/African American, b = .08, t (819) = 2.03, p = .12, Hispanic/Latino, b = .05, t (819) = 1.36, p = .17, Asian, b = .03, t (819) = .90, p = .37, racia lly/ethnicall y mixed students, b = .00, t (819) = .08, p = .90 did not differ from White/Caucasian students in willingne ss to report. College juniors, b = .06, t (819) = 1.10, p = .14, and seniors, b = .07, t (819) = 1.39, p = .09, did not differ from freshmen in willi ngness to report. As a result, grade level was not associated with willingness to report. system and place of residence, b = .02, t (819) = .44, p = .66, or race/e thnicity. Black/African American, b = .02, t (819) = .50, p = .62, Hispanic/Latino, b = .08, t (819) = 2.02, p = .10, Asian, b = .01, t (819) = .26, p = .79, racia lly/ethnically mixed students, b = .02, t (819) = .53, p = .60, did not differ from White/C aucasian students in trust in the college su pport system. College juniors, b = .02, t (819) = .56, p = .58, and seniors, b = .07, t (819) = 1.60, p = .12, did not differ from freshmen in trust in the college support system. However, a small yet significant a ssociation was identified between gender and trust in the college support system, b = .07, t (819)
50 = 2.23, p = .03, d = .17, as females ( M = 12.70, SD = 2.72) displayed greater trust than did their male counterparts ( M = 12.28, SD = 2.06). Measurement M odel polychoric correlations ) was assessed using multiple criteria. The finding of a significant chi square statistic value, [ 2 (176, N = 820) = 1782.88, p < .01 ] indicated that the measurement model did not fit the data exactly. However, goodness of fit index values were examined since any non saturated model will only fit data approximately. Additionally, non saturated models generally are rejected by the goodness of fit test (i.e., evidence a non significant m inimum fit chi square statistic; Kline, 2005). Table 3 2 presents g oodness of fit test an d index values for each model. According to fit indices, the measurement model demonstrated adequate data fit: comparative fit index (CFI) = .95, non normed fit index (NNFI) = .97, and root mean square error of approximation (RMSEA) = .10. Obtained CFI and NNFI values were above acceptable limits, and the RMSEA value approximated it s historically preferred level. No negative or non significant variances were significant variances may suggest poor model specification or insufficient data. All indicators displayed significant loadings on the ir latent constructs. High standardized factor loadings were found for items that were specified to load on self Standardized factor loadings for items specified to variances are presented in Table 3 3. No respecifications were attempted to the measurement model due to limited statistical and theoretical criteria to inform such respecifications.
51 Structural Equation Models Model 1 Table 3 4 presents total, direct, and indirect effects for stu dy variables across tested models. A structural equation model (model 1) was specified to investigate the effects of latent variables. Figure 3 1 presents the model with standardized parameter estimates. This model was fully recursive and demonstrated marg inal data fit: 2 (176, N = 820) = 1815.38, p < .01, CFI = .93, NNFI = .97, and RMSEA = .11. Furthermore, estimating model 1 resulted in a significant decrease in data fit as compared to the measurement model: 2 (2, N = 820) diff = 41.06, p < .01. Overall structural paths in Model 1 accounted for a gre willingness to report threats ( R = .40 ) than they did for trust in the college support system ( R = .13 ). All exogenous variables in model 1 were expected to be in tercorrelated and have direct effects on willingness to report. Direct effects of trust in the college support system, campus connectedness, and delinquency on willingness to report were significant. However, direct effects of self efficacy toward service and fear of negative evaluation on willingness to report were not significant. As specified in model 1, delinquency and self efficacy toward service were expected to have direct effects on trust in the college support system in addition to willingness to r eport. The direct effects of delinquency on trust in the college support system and self efficacy toward service on trust in the college support system were significant. Trust in the college support system was expected to partially mediate the relationship between self efficacy toward service and willingness to report. Trust in the college support system also was expected to partially mediate the relationship between delinquency and willingness to report. Two conditions described by Kenny, Kashy, and Bolger (1998) were used to evaluate mediation. First, an association must be established with an initial and proposed
52 mediating variable. Second, an association must be established between the mediating variable and the outcome variable. Partial mediation is det ermined by meeting these conditions and by establishing an indirect effect between an initial and outcome variable through a mediating variable. Further, full mediation exists if the relationship between the initial variable and the outcome variable does n ot differ significantly from zero with the mediating variable in the model (Baron & Kenny, 1986). In other words, full mediation implies that the indirect path through the mediating variable determines the relationship between the initial and outcome varia bles completely. Considering the aforementioned criteria, the significant indirect effect from delinquency to willingness to report through trust in the college support system indicates that trust in the college support system partially mediates the relat ionship between delinquency and willingness to report. Similarly, a significant indirect effect was found from self efficacy toward service to willingness to report through trust in the college support system. In this case, trust in the college support sys tem almost completely mediates the relationship between self efficacy toward service and willingness to report since total and indirect effects of self efficacy toward service on willingness to report are significant while the direct effect is close to zer o. Model 2 A second structural equation model (model 2) was tested to explore the aforementioned relationships as well as the effects of campus connectedness and fear of negative evaluation on trust in the college support system. Figure 3 2 pres ents the model with standardized parameter estimates. A conceptual link between campus connectedness and fear of negative evaluation to trust in the campus support system may exist even though research was not located that directly establishes this associa tion. Further, the presence of a large modification index and expected parameter change statistic values in the previous model (model 1) supports an association
53 between these variables. Model 2 was fully recursive and displayed better data fit than model 1 did: 2 (2, N = 820) diff = 41.60, p < .01, CFI = .93, NNFI = .97, and RMSEA = .10. Structural paths in Model 2 accounted for a gre threats ( R = .43 ) than they did for trust in the college sup port system ( R = .20 ). All exogenous variables were expected to be intercorrelated and were specified to have direct effects on trust in the college support system and willingness to report in model 2. Similar to first model, direct effects of trust in th e college support system, campus connectedness, and delinquency on willingness to report were significant. Direct effects of delinquency, self efficacy toward service, campus connectedness, and fear of negative evaluation on trust in the college support sy stem also were significant. Thus, all exogenous variables were related directly to trust in the college support system. The indirect effect of self efficacy toward service on willingness to report was significant while the direct effect was not. Thus, the relationship between self efficacy toward service and However, trust in the college support system did not mediate the relationship between delinquency and will ingness to report in model 2. Therefore, the relationship of delinquency to willingness to report appears to depend on the influence of other variables (e.g., campus connectedness). Furthermore, the magnitude of the direct effect of delinquency on willingn ess to report also decreased with the inclusion of the two additional paths in model 2. The inclusion of the two additional paths in model 2 allowed for the testing of indirect effects between campus connectedness and fear of negative evaluation on willing ness to report. With the inclusion of these paths, trust in the college support system partially mediated the relationship between
54 campus connectedness and willingness to report. However, similar to model 1, no significant associations were identified betw een negative evaluation and willingness to report. Model 3 Model 3 includes a direct path from self efficacy toward service to campus connectedness in addition to the paths that were estimated in the previous two models. Figure 3 3 presents the model with standardized parameter estimates. This respecification was made due to a relatively large modification index value and theoretical criteria (i.e., individuals who believe they can be successful in public service endeavors are likely to be connected to and involved with members of the campus community). This modification resulted in a significant improvement in model fit, 2 (2, N = 820) diff = 46.02, p < .01, and slight improvements in fit index values, CFI = .95, NNFI = .98, and RMSEA = .09. Structural paths in Model 3 accounted for the greatest percentage of R = .4 1) followed by trust in the college support system ( R = .18) and campus connectedness ( R = .14). Similar to the previous two models, the direct effects of trust in the college support system, campus connectedness, and delinquency on willingness to repor t were significant. Further, the direct effects of campus connectedness, delinquency, fear of negative evaluation, and self efficacy toward service on trust in the college support system were significant. Thus, all exogenous variables remained casually rel ated to trust in the college support system. Additionally, the direct effect of self efficacy toward service on campus connectedness was significant. Similar to the previous two models, the effect of self efficacy toward service on willingness to report re mained entirely indirect through trust in the college support system in model 3. However, two new indirect paths were found from self efficacy toward service to willingness to report. One path was through campus connectedness and trust in the college suppo rt system and the other was through campus connectedness alone. The effect of campus
55 connectedness on willingness to report remained mostly direct while the effect of delinquency on willingness to report entirely was direct. As in the previous two models, no association was identified between fear of negative evaluation and willingness to report.
56 Table 3 1. Descriptive statistics, internal consistency estimates, and interfactor correlations between study variables Items Min Max M SD 1 2 3 4 5 6 1. Willingness to report 20 20 120 83.12 24 97 .97 1 .00 2. Trust in the college support system 6 6 23 1 7.38 2.28 .81 .18 ** 1 .00 3. Campus connectedness 14 14 84 60.02 13.02 .92 .28 ** .24 ** 1 .00 4. Fear of negati ve evaluation 11 11 44 25.53 7.92 .94 .04 .13 ** .29 ** 1 .00 5. Delinquency 4 4 8 4.62 .92 .8 0 .23 ** .2 2 ** .09 .08 1 .00 6. Self efficacy toward service 5 5 25 20.96 3.44 .9 0 .05 .23 ** .35 ** .25 ** .10* 1 .00 7. Social desirability 1 0 10 20 13.81 1.51 .35 Note. p < .05*, p < .01** Table 3 2. Goodness of fit test and index values for tested models df CFI NNFI RMSEA Acceptable fit val u es .95 .95 .09 M Model 1742.88* 176 .95 .97 .10 Model 1 1815.38 176 .93 .97 .11 Model 2 1782.88 176 .95 .97 10 Model 3 1679.59 159 .95 .98 .09 Note M Model = measurement model ; CFI = comparative f it index; NNFI = non normed fit index; RMSEA = root mean square error approximation; p < .01
57 T able 3 3 F actor loadings and factor variances by construct for the measurement model Willingness to report Campus connectedness Fear of negative e valuation Trust in the college support system Self efficacy toward service Delinquency Item S U S U S U S U S U S U 1 .90 1.00 .71 1.00 .83 1.00 .60 1.00 .80 1.00 .81 1.00 2 .81 .90 .72 1.02 .74 .89 .65 1.09 .78 .97 .92 1.14 3 .85 .95 71 1.01 .78 .95 .70 1.22 .89 1.12 .65 .80 4 .82 .90 .63 .91 .81 .98 .60 1.01 .90 1.14 .59 .72 5 .90 .99 .86 1.23 .87 1.06 .68 1.14 .85 1.05 6 .93 1.03 .88 1.25 .88 1.07 .50 .83 7 .85 .95 .71 1.00 .84 1.01 8 .97 1.09 .78 1 .10 .84 1.01 9 .79 .88 .72 1.01 .72 .88 10 .94 1.05 .78 1.09 .84 1.01 11 .71 .78 .82 1.16 .79 .96 12 .88 .98 .71 1.00 13 .84 .93 .81 1.14 14 .90 1.00 .77 1.09 15 .77 .85 16 .98 1.10 17 .70 .78 18 .92 1.03 19 .87 .97 20 .66 .73 Factor variance .81 .50 .68 .36 .64 .65 Note U = unstandardized factor loading; S = standardized factor loading; All factor loadings and factor variances are significant, p < .01.
58 Table 3 4 Effects decomposition for study variables Endogenous variables Willingness to report Trust in the college support system Campus connectedness Total Direct Indirect Total Direct Total Direct effect effect effect effect effect effect Effect Exogenous variables r r r r r r r Mode l 1 Delinquency .27 .23 04* 24 .2 4 --Self efficacy toward service .06* 02 04* .2 3 .2 3 --Campus connectedness .28 .28 -----Fear of negative evaluation .02 .02 -----Trust i n the college support system 1 8 .1 8 -----Model 2 Delinquency .25* 23* .02 19 19 --Self efficacy toward service .05* .02 .0 3* 22* .2 2 --Campus connectedness .29 .27 .03 .21 .21 --Fear of negative evaluation .0 2 0 1 .02 .1 2 .1 2 --Trust in the college support system .14 .14 -----Model 3 Delinquency 25 23 .02 .1 9* 1 9* --Self efficacy toward service .1 4 .0 1 13 21* .2 1 .38 .38 Campus connectedness .32 .29 .03 .22 .22 --Fear of negative evaluation .03 01 .02 .12 .12 --Trust in the college support system .14 .14 -----Note. The symbol -means the ef fect is not in the model; p < .05
59 Figure 3 1. Model 1: The hypothesized structural equation model Note: FNE = fear of negative evaluation; DEL = delinquency; SETS = self efficacy toward service; CC = campus connectedness; TICSS = trust in the college support system; REP = willingness to report; significant effects in bold (p <.05). FNE DEL SETS CC TICSS .2 4 .23 REP .02 .28 .1 8 .2 3 .02
60 Figure 3 2. Model 2: Allowing direct effects of campus connectedness and fear of negative evaluation on trust in the college support system Note: FNE = fear of negative evaluation; DEL = delinquency; SETS = self efficacy toward service; CC = campus connectedness; TICSS = trust in the college support system; REP = willingness to report; significant effects in bol d (p <.05). Figure 3 3. Model 3: Allowing a direct effect from self efficacy toward service to campus connectedness. Note: FNE = fear of negative evaluation; DEL = delinquency; SETS = self efficacy toward service; CC = campus connectedness; TICSS = trust in the college support system; REP = willingness to report; significant effects in bold (p <.05). FNE DEL SETS CC REP TICSS .29 .2 2 .21 .01 .2 3 .12 .19 .01 .1 4 .38 FNE DEL SETS CC REP TICSS .27 .21 .2 2 .02 .23 .1 2 19 .01 .1 4
61 CHAPTER 4 DISCUSSION threats of vi olence in campus communities. The primary aim of this study was to test report threats of violence. Specifically, the effects of trust in the college support syste m, campus connectedness, and self efficacy toward service were expected to be positively related to willingness to report threats. On the other hand, delinquency and fear of negative evaluation were expected to be negatively related to willingness to repor t. Additionally, trust in the college support system was expected to mediate relationships between delinquency and self efficacy Structural equation modeling with weighted least squares was used to test the aforementioned relationships. Three simultaneous equation models were estimated in addition to a comprehensive measurement model for all latent factors. Following the establishment of an adequate measuremen t model, a structural equation model (model 1) was specified to test associations among latent variables. All exogenous variables were expected to have direct effects efficacy toward service and delin quency were expected to be partially mediated by the influence of trust in the college support system. However, model 1 demonstrated marginal fit to the data. Thus, a second simultaneous equation model (model 2) was specified, which allowed for the testing of putative relationships among study variables. As compared to model 1, model 2 evidenced better fit and it allowed campus connectedness and fear of negative evaluation to have direct effects on trust in the college support system. This model then was re specified to test the effect of self efficacy toward service
62 on campus connectedness (model 3). Overall, model 3 demonstrated the best fit and it provided the most comprehensive view of the relationships among study variables. Trust in the College Suppor t System willingness to report threats. Therefore, students who trust in college administrators, faculty members, and police officers are more likely to report threats of violence than students are who do not trust these figures. In general, students who trust in members of the college support system feel comfortable sharing safety concerns with school officials, feel supported in general by the support system, and believe that campus officials would respond quickly and efficiently to crises. Findings obtained from studying peers of prominent and prospective school shooters indicate that adolescents who report threatening peers generally have positive relations with adults, believe their reports will be taken seriously, and think that authority figures will address threats appropriately (Gaughan et al., 2001; Greenberg, 2007; Pollack et al ., 2008; Stueve et al., 2006). Moreover, adolescents who are socially bonded with adults display an increased willingness to report threats if they think that their reports will be believed and that the information they provide will be protected (Brank et al., 2007). On the other hand, adolescents who do not report threatening peers tend to believe that they may experience negative consequences from authority figures for sharing information about potential threats. In the extreme, some of these adolescents may fear being interrogated, harshly questioned, or even suspended from school for reporting a threatening friend or peer (Pollack et al., 2008). trust in the college support system also may influence t heir crime reporting behaviors in general.
63 For example, one study found that 24% of students did not report being a victim of a campus crime (e.g., theft, assault) due to a lack of trust in campus police to respon d effectively (Sloan III et al., 1997). Documented reasons for not reporting in the aforementioned study include believing that police would not be able to recover lost property, find the offender, and solve the crime. Furthermore, a small number of studen ts (1%) did not report their victimizations out of a fear of being harassed or threatened by police officers. S elf E fficacy Toward Service Quite unexpectedly, no direct association was found between self efficacy toward service to report threats of violence. Instead, this relationship was completely mediated by other variables. For example, students who trust in the campus support system, feel connected to the campus community, and believe they can have an impact on campus displa y a greater willingness to report threats. However, students who feel efficacious in their ability to report threats yet do not trust members of the college support system or feel weakly bonded to the campus community still may not report threats. Thus, th e belief that one can have a positive impact in the campus community does not facilitate threat reporting by itself. In its place, it is important for students to feel connected to the campus community and to trust in members of the college support system to facilitate threat reporting. This finding is supported by results from a study that found college students unwilling to trust authority figures if they were unable to express their opinions or concerns with them openly (Lind, Kanfer, & Earley, 1990). Furthermore, students who believe they are discriminated against tend to have low levels of connectedness to the campus community even if they have a strong sense of ethnic pride and self esteem (Lee, 2005). Therefore, students who feel disenfranchised, di sconnected, or discriminated against in the campus community may be less
64 willing to trust members of the campus support system and be less willing to report threats even if they are confident in their ability to communicate them. Campus Connectedness Camp directly and indirectly. Thus, the association between campus connectedness and willingness to report is consistent with results suggesting that school climate and student t eacher relationships intentionally harm others (Brinkley & Saarnio, 2006; Syvertsen et al., 2009). Results from the current study and the latter two suggest that stud ents who feel bonded with others, develop strong interpersonal relationships with peers and educators, and feel comfortable on campus, may take protective actions to preserve the safety of the campus such as reporting the presence of threats of violence. In general, students who feel strongly connected to their campus community are likely to have many interpersonal relationships, value the welfare of these individuals, and even develop a sense of pride related to belonging to a specific college or universi ty (Summers, Svinicki, Gorin, & Sullivan, 2002). However, on the other hand, students who are poorly bonded with members of the campus community tend to have fewer interpersonal relationships, are less aware of social issues at their school, and may even h ave negative perceptions of specific individuals on campus or their college as a whole (Pascarella, Edison, Nora, Hagedorn, & Terenzini, 1996 ; Summers et al., 2002 ) Further, adolescents who do not share prior knowledge of school attacks generally report f eeling loosely connected to the school community and uncomfortable communicating with others at school, especially authority figures (Pollack et al., 2008). Results of this study indicate that students who feel connected to the campus community and trust in the college support system display a greater willingness to report threats of violence.
65 On the other hand, students who feel disconnected from the campus and display low trust in members of the college support system are less likely to report threats of violence. Similarly, One study found that the majority (85%) of students do not know personally or have any contact with campus police officers and almost no students (less than 1%) know a campus police officer very well (e.g., on a first name basis) (Revels, 1999). Furthermore, when questioned about the accessibility of campus law enforcement officers, almost half of students (49%) reported that the officers s (Revels, 1999). Thus, weak student police relationships may be a barrier to threat reporting and port threats to the campus community. Delinquency general, students who report a history of delinquent behavior display a low willingness to report threats of viole nce. Similarly, delinquent adolescents appear to be less likely to report weapon carrying students (Brank et al., 2007) and threats of violence than non delinquent adolescents are (Brinkley & Saarnio, 2006). One study even found 74% of adolescents who endo rsed having carried a weapon to school to state that they would not report another weapon carrying student, even if the student threatened to harm other students (Brinkley & Saarnio, 2006). Although delinquent students may be in a favorable position to re port potential threats of violence due to their increased likelihood of associating with delinquent peers (Thornberry et al., 1994) and the association between weapon carrying behaviors and other delinquent behaviors (Miller et al., 2002; Presley et al., 1 997), some barriers appear to prevent them from doing so. Delinquent students may belong to social networks that condone delinquency and reinforce
66 antisocial behavior. For example, snitching (i.e., informing police or other authorities of illegal activity) can lead to social ostracism and place informants at risk for retaliation from peers in safety may prevent some individuals from reporting if they think tha t they may be identified as an informant or be perceived negatively by peers (Haynie, 2001) Additionally, cooperating with authority figures may cause some delinquent individuals to feel ineffective and inferior for having to appeal to an external agency to address a problem that they were unable to do so personally (Rosenfeld et al., 2003). Fear of Negative Evaluation Although college students generally are concerned with peer approval ( Bro wn et al., 1997; Perkins, 2002) and often engage in maladaptive b ehaviors (e.g., binge drinking) to conform to social norms (Borsari & Carey, 2001; Scholly, Katz, Gascoigne, Holck, & Scholly, 2005), no threats of violence. Perh aps tragic attacks at Virginia Tech, Northern Illinois University, and other educational institutions may have caused students to take threats of violence seriously and overcome fears that may be associated with incurring negative peer evaluations. Alterna tively, reporting threats of violence may be socially acceptable or even condoned in non delinquent student groups. In this vein, students may receive peer, family, institutional, or community support for attempting to mitigate threats of violence to the c ampus community. Further, external willingness to report. For example, empathy motivation to avoid negative social evaluation s when they are presented with an opportunity to help a depressed student (Fultz, Batson, Fortenbach McCarthy, & Varney, 1986)
67 Other Findings Associations between demographic characteristics and variables pertinent to research questions also were explored in this study. Specifically, associations betwe race/ethnicity, grade (i.e., sophomore, junior, senior), place of residence (i.e., on/off campus) and willingness to report threats were explored None of these associations were significant. These results differ from those obtained with adolescents. One study found a negative less willing to report threats than were eighth graders ( Brinkley & Saarnio, 2006 ). Grade level peer pressure. For example, the influence of peer perceptions on student behavior tends to decline in college, especially for risk taking behavior (Gardner & Steinberg, 2005). Additionally, seriously than younger students would. Compared to their male counterparts, female adolescent s exhibit a greater willingness to report threats (Brinkley & Saarnio, 2006) and the presence of weapons at school (Brank et al., 2007). Their tendency to engage in lower rates of maladaptive behaviors, delinquency, and use more mental health services due to their greater openness toward others (Davies et al., 2000) could imply that females would be more willing to report threats. In this study, males may have been more likely to report threats of violence than males in the general population. This finding students who participated to meet a class requirement were enrolled in educational psychology classes. Male psychology students have been found to display more advanced critical thinking skills and higher maturity levels than male nursing and business students do (Walsh & Hardy, 1999). Therefore, the male students in this study also may have been elevated in these qualities,
68 thus leading them express a greater wi llingness to report threats than would be expected from males in different programs of study. or trust in the college support system. One study found Hispanic adol escents to be slightly less likely to report weapons carriers as compared to their non Hispanic peers (Brank et al., 2007). However, in the same study, White/Caucasian, African American, and students who identified fer in reporting. Additional research is needed to understand the influence of race/ethnicity on the aforementioned variables (i.e., willingness to report, trust in the college support system) as well as to determine if findings from this study generalize to educational institutions that have a different demographic makeup. Compared to males, females in this study displayed greater trust in the college support system. This result is consistent with findings suggesting that females tend to be more trusting in situations in which they feel obligated to perform pro social actions (e.g., donating money) and avoid personal risks (Chaudhuri & Gangadharan, 2003 ). Additionally, compared to males, females tend to be more trusting of high trust targets (e.g., physici ans, police officers) (Rotter, 1980). Thus, they also may be more likely to trust figures of authority in campus communities. Limitations Several limitations of this study deserve mention. First, a low response rate (17%) was obtained from the general student population. Whether participants differ in meaningful ways from students who were solicited and did not provide data is not known. Efforts were made to control for a potential sampling bias through the inclusion of a sample of students who partici pated to meet a class requirement. However, the use of a multi method sampling approach (i.e., contacting participants through different mediums [e.g., email, mail, phone contact]) may have increased the overall response rate for this study (Dillman, Smyth & Christian, 2008), as
69 well as protected against a potential sampling bias. Second, all participants were enrolled in classes at one large university located in the Southern U.S. Weapons carrying rates on college campuses vary by geographic location (Mei lman et al., 1998 willingness to report threats of violence. Thus, the results from this study may not generalize to students attending markedly different educational institutions. Third, approximately 70% of participants in this study were female. Females were over sampled as they comprise 54% of the general student body of the institution in which data were collected (Peterson Education, 2010). Forth, all participants either chose to participate voluntarily or were enrol led in an educational psychology class in which participation was required. Therefore, these students, especially male students, may differ from the general student body in important qualities (e.g., critical thinking) (Walsh & Hardy, 1999) and this may ha ve influenced their responses to study questions. As a result, the generalizability of these findings for males may be limited. Fifth, not all possible causes of endogenous variables (e.g., willingness to report, trust in the campus support system) were in cluded in this study. Additional causes of these variables warrant future research. Sixth, several different structural equation models were estimated to protect against the possibility of an alternative model demonstrating superior fit. However, not all p ossible models were tested. Alternative models may exist that demonstrate equivalent fit yet describe different relations among study variables. Seventh, this study relied only on self report measures. Future studies may benefit from using different method s (e.g., observations, interviews, records) to assess variables of interest. Additionally, this study used a cross sectional design. Efforts are needed to examine these and other variables over time to determine the reliability of these results. Lastly, th
70 data may be misleading and thus lead to inaccurate conclusions despite efforts that were made to control for social desirability. Conclusions Despite the se limitations, results from this investigation support the expectation that trust in the college support system, campus connectedness, self efficacy toward service, and campus connectedness to the campus community may improve their willingness to report threats of violence to authorities who may be able to mitigate them. On the other hand, students with a history of delinquent behavior appear to be less willing to report threats of violence in general. Therefore, additional research is needed to explore qualities that may increase delinquent hreats. Perhaps anonymous reporting options such as those that have been recently implemented at several large universities may encourage delinquent students to report threats without incurring possible negative consequences for doing so. In conclusion, t he task of stopping a violent attack on a college campus is analogous to capturing a bolt of lightning on film: the task is impossible without being in the right place at the right time. Although the results of this study are unlikely to inform the efforts of those who are in the right place at the right time to mitigate threats of violence, they may encourage the formation of policies and practices designed to prevent future acts of violence. Instead of being unduly influenced by rare albeit tragic acts of violence in campus communities, it is important to recognize that college campuses generally are safe (Bromley, 1995) and that students who feel bonded with members of the campus community are likely to take protective actions to preserve campus safety. T hus, in lieu of controversial increases in security technology, concealed weapons, harsh disciplinary policies, and criminal profiling techniques that may produce an
71 illusion of safety, better solutions to reducing threats of violence may result from facil itating greater trust in members of the college support system, as well as from engendering a greater
72 APPENDIX A: INFORMED CONSENT FORM CONSENT TO ACT AS A RESEARCH PARTICIPANT Protocol Tit le: Contributing factors to college students threat reporting Note: Please read this consent document carefully before you decide to participate in this study. Purpose of the research study: The purpose of this study is to assess factors that may contrib threat to campus or public safety. What you will be asked to do in the study: Should you choose to participate, you will be asked to answer various questions that asse ss your subjective levels of self efficacy toward service (e.g., your beliefs about how responsible you are for causing positive social outcomes), campus connectedness (e.g., how much you feel a part of the campus milieu), delinquency (unlawful or deviant acts you may have committed), fear of negative evaluation (e.g., how much you dread Time required: 30 45 minutes Potential Risks : You may experience some distress from filling out some personal questions. If you experience significant discomfort, you are encouraged to contact the University of Florida Counseling Center (P301 Peabody Hall, 352 392 1575). Confidentiality: Your identity will remain anonymous, as you are not to provide any personally identifiable information (e.g., name, student ID #) anywhere on the questionnaire. All of your responses will be kept confidential. When the study is completed and the data have been analyzed, your answers will be destroyed. You r individual answers will not be used in any report, scientific meetings, institutional policies, or published materials that may result from this research. Voluntary participation: You can only participate if you are 18 years of age or older. Your part icipation in this study is completely voluntary. There is no penalty for not participating and you have the right to withdraw from the study at anytime without consequence. Whom to contact if you have questions about the study: Michael Sulkowski, M.Ed., School Psychology Program, firstname.lastname@example.org phone (352) 392 0724. Thomas Oakland, Ph.D., Professor, School Psychology Program, email@example.com phone (352) 273 4283 Whom to contact about your rights as a research participant in the study: IRB02 Office, Box 112250, University of Florida, Gainesville, FL 32611 2250; phone 392 0433.
73 Agreement: I have read the information provided for the study as described herein I will express this agreement by completing the following questions. Note: Students were presented with the following information after consenting to participate in the study. Participants of this study may win any one of the following: A free Apple ipo d Nano 8GB digital music player valued at $200, a gift card for the Outback Steakhouse valued at $50.00 (quantity = 1), a gift card for Sports Authority valued at $40 (quantity = 1), a gift card for Barnes & Noble valued at $25 (quantity = 1), a gi valued at $25 (quantity = 2), a gift card for the Regal Entertainment Group valued at $10 (quantity = 5), and gift ca rds for Starbucks valued at $5.00 (quantity = 12). Incentives will be provided to participants who respond in the following order: The 1 st and 2 nd respondents will each receive one gift card for TGI Fridays Restaurant valued at $25. The 100 th 112 t h responders will each receive one gift card for Starbucks valued at $5.00. The 200 th 205 th respondents will each receive one gift card for the Regal Entertainment Group valued at $10. The 500 th respondent will receive one gift card for Barnes & No ble valued at $25. The 600 th respondent will receive one gift card for Sports Authority valued at $40. The 700 th respondent will receive one gift card for the Outback Steakhouse valued at $50.00 The 800 th respondent will receive one gift card fo Grill valued at $25. The 1000 th respondent will receive one free Apple ipod Nano 8GB digital music player valued at $200.
74 APPENDIX B : P ARTICIPANT MEASURES DEMOGRAPHIC QUESTIONNAIRE Please select the options t hat best correspond to you 1. Please select your gender ____ Male ____ Female 2. Which racial or ethnic group best describes you? ___ __ Caucasian/Non Hispanic White ___ ___Hispanic/ Latino _____ Black/African Ameri can ______Asian ___ __ Mixed race 3. I am presently a: ____ sophomore ____ junior ____senior 4. What is your academic major? ______________________________________________ 5. Do you . (Check one) _______ Live on campus _______ Live off campus 6. I am . (Check all that apply) _______ A student athlete ______ A member of a social fraternity/sorority ___ ___ A member of an ext ra curricular school club ____ __ A member of an honor society
75 CAMPUS CONNECTEDNESS SCALE (Lee & Robbins, 1995) Rate the degree to which you AGREE or DISAGREE with each statement using the following scale: (1 = Strongly Disagree and 6 = Strongl y Agree). There is no right or wrong answer. Do not spend too much time with any one statement and do not leave any unanswered. 1 2 3 4 5 6 Strongly Disagree Disagree Mildly Disagree Mildly Agree Agree Strongly Agree 1. There are people on campus wit h whom I feel a close bond 1 2 3 4 5 6 2. I really don't feel that I belong around the people that I know on campus 1 2 3 4 5 6 3. I feel that I can share personal concerns with other students 1 2 3 4 5 6 4. I am able to make connections with a d iverse group of people 1 2 3 4 5 6 5. I feel distant from the other students 1 2 3 4 5 6 6. I have no sense of togetherness with my peers 1 2 3 4 5 6 7. I can relate to my fellow classmates 1 2 3 4 5 6 8. I catch myself losing all sense of conn ectedness with college life 1 2 3 4 5 6 9. I feel that I fit right in on campus 1 2 3 4 5 6 10. There is no sense of brother/sisterhood with my college friends 1 2 3 4 5 6 11. I don't feel related to anyone on campus 1 2 3 4 5 6 12. Other stude nts make me feel at home on campus 1 2 3 4 5 6 13. I feel disconnected from campus life 1 2 3 4 5 6 14. 1 2 3 4 5 6
76 SELF EFFICACY TOWARDS SERVICE SCALE (Weber, Weber, Sleeper, & Schneider, 200 4) Please use the following scale to respond to each item. Select the number theta corresponds to the answer that best describes the extent to which you agree or disagree with the statement. 0 1 2 3 4 Strongly Disagree Disagree Neither Agree/Disagre e Agree Strongly Agree 1. I can have a positive impact on social problems. 0 1 2 3 4 2. I can help people with disabilities. 0 1 2 3 4 3. I have confidence in my ability to help others. 0 1 2 3 4 4. I can make a difference in my community. 0 1 2 3 4 5. Each of us can make a difference in the lives of the less fortunate. 0 1 2 3 4
77 BRIEF FEAR OF NEGATIVE EVALUATION SCALE REVISED (Carleton, McCreary, Norton, & Asmundson, 2006) Please carefully read and answer e ach question below. Indicate HOW TRUE each statement is for you by selecting the number next to each question. 0 1 2 3 Not at all characteristic of me Slightly characteristic of me Moderately characteristic of me Extremely characteristic of me 1. I worry about what other people will think of me 0 1 2 3 2. It bothers me when people form an unfavorable impression of me. 0 1 2 3 3. I am frequently afraid of other people noticing my s hortcomings. 0 1 2 3 4. I worry about what kind of impression I make on people. 0 1 2 3 5. I am afraid that others will not approve of me. 0 1 2 3 6. 0 1 2 3 7. When I am talking to someone, I worry about what they may be thinking about me. 0 1 2 3 8. I am usually worried about what kind of impression I make. 0 1 2 3 9. If I know someone is judging me, it tends to bother me. 0 1 2 3 10. Sometimes I think I am too concerned with what ot her people think of me. 0 1 2 3 11. I often worry that I will say or do wrong things. 0 1 2 3
78 SELF REPORT DELINQUENCY SCALE (Piquero, MacIntosh, & Hickman, 2002) Please carefully read and answer each question below. Indicate number of t imes that you engaged in each of the following behaviors. Answer each question by selecting the number next to each question. OVER THE PAST YEAR, have you? No Yes 1. Sold illegal drugs. 0 1 2. Used force to get money or things from others. 0 1 3. Stolen (or tried to steal) things worth between $5 and $50. 0 1 4 Been involved in a physical fight. 0 1
79 MARLOWE CROWNE SOCIAL DESIRABILITY SCALE SHORT FORM X2 (Strahan & Gerbasi, 1972) Listed below are a number of state ments concerning personal attitudes and traits. Read each item and decide whether the statement is true (circle the T ) or false (circle the F ) as it pertains to you. 1. I never hesitate to go out of my way to help someone in trouble. T F 2. I have nev er intensely disliked anyone. T F 3. T F 4. There have been times when I felt like rebelling against people in authority even though I knew they were right. T F 5. out of something. T F 6. T F 7. I am always courteous, even to people who are disagreeable. T F 8. I would never think of letting someone else be punished for my wrongdoings. T F 9. There ha ve been times when I was quite jealous of the good fortune of others. T F 10. I am sometimes irritated by people who ask favors of me. T F
80 TRUST IN COLLEGE SUPPORT SYSTEM (Sulkowski, unpublished) Please carefully read and answer each question below. Indicate how much you either AREE or DISAGREE with each statement is for by circling the number next to each question. 1 2 3 4 Strongly disagree Disagree Agree Strongly agree 1. If a crisis happened on campus, my college would hand le it well. 1 2 3 4 2. 1 2 3 4 3. The college responds too slowly in difficult situations. 1 2 3 4 4. It is difficult to share safety concerns with school officials. 1 2 3 4 5. My school is po orly prepared to handle crisis situations. 1 2 3 4 6. There is a good support system on campus for students going through difficult times. 1 2 3 4
81 THREAT REPORTING PASSAGES (Sulkowski, unpublished) Please read the passages below and ans wer the questions that follow Passage 1. Following the sudden loss of his older brother last month, Chris has gone from being normally quiet and reserved around other people to being blunt and confrontational. He blames his ts who were in a car that struck and instantly killed his brother. The other students attributed the accident on the bad weather and slick streets, but Chris is convinced that they were intoxicated and that a cover up took place. He believes that the cam pus police gave them special treatment since the students were well connected. Currently, Chris is extremely upset and frustrated. While he is in the process of grieving, his mood swings back and forth from sadness to rage when he thinks about the tragic loss of his brother. Chris believes that must do something. Pretend that you overheard Chris say the following: 1. Based on the former information and his statement, how willing would you be to report Chris? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Chris say the following: shots an own hands. I bet there are a lot of people out there who would sympathize with my cause. If we
82 2. Based on th e former information and his statement, how willing would you be to report Chris? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Chris say the following: 3. Based on the former information and his statement, how willing would you be to report Chris? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Chris say the following: their door, knock, and when the door opens 4. Based on the former information and his statement, how willing would you be to report Chris? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1
83 Passage 2. Thomas always tries to be the center of attention. Whether in class or out in public, he always wants to get a rise out of others. Sometimes his peers find his antics amusing, but other times, he goes too fa r and gets himself into trouble. Last Friday, he was arrested for being drunk and disorderly at a campus sponsored event and now he is facing legal charges and in jeopardy of loosing his scholarship. Thomas does not know what to do and for the first time he feels like he cannot talk his way out of the situation. He recently started an online group where students can get together and protest the oppressive campus atmosphere. He has appointed himself the leader and spokesperson of the group. Pretend that you overheard Thomas say the following: th go up the night before so nobody will see me. Then, when the sun rises, I will start sniping people with my rifle. 5. Based on the former information and his statement, how willing would you be to report Thomas? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Defini tely report 6 5 4 3 2 1 Pretend that you overheard Thomas say the following: 6. Based on the former information and his statement, how willing would you be to report Thomas?
84 Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Thomas say the following: bang 7. Based on the former information and his statement, how wil ling would you be to report Thomas? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Thomas say the following: 8. Based on the former information and his statement, how willing would you be to report Thomas? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1
85 Passage 3. Jessica is enraged about being rejected by a social sorority. She feels l ike an outcast, and since she is a new student on campus, she does not know many people. Throughout her adolescence, Jessica experienced significant periods of depression and she continues to feel hopeless about the future and is desperate for things to c hange. Jessica has recently been stalking a male student named Jeff. She will often call him to hear his voice and hang up without speaking, drive by his apartment, and send him cryptic messages from an anonymous email address. Last Friday, Jessica saw Jeff and another female student walking together while holding hands. Afterward, she ran back to her dorm room and started furiously drawing pictures of Jeff and the other girl being killed. Pretend that you overheard Jessica say the following: nted to do so, I could shoot her 9. Based on the former information and his statement, how willing would you be to report Jessica? Definitely would report Probably would report Somewhat willing to report Not r eally willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Jessica say the following: when he 10. Based on the former information and his statement, how willing would you be to report Jessica?
86 Definitely would report Probably would report Somewhat willing to report Not really willing to re port Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Jessica say the following: 11. Based on the former information and his statement, how willing would you be to report Jessica? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Jessica say the following: 12. Based on the former information and his statement, how willing would you be to report Jessica? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1
87 Passage 4. Ever since childhood, Ray had wanted to be a doctor. After c ompleting his undergraduate education, he had planned to attend medical school. However, despite applying to many different programs, Ray was not accepted into a single medical school. Ray now feels like burned his class notes and gave away many of his personal possessions to other students. Pretend that you overheard Ray say the following: 13. Based on the former information and his statement, how willing would you be to report Ray? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Ray say the following: t 14. Based on the former information and his statement, how willing would you be to report Ray? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Ray say the following:
88 15. Based on the former information and his statement, how willing would you be to report Ray? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Ray say the following: 16. Based on the form er information and his statement, how willing would you be to report Ray? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 P assage 5. Alex and his ex girlfriend had often disagreed with each other, Alex was shocked by the breakup. He was even more shocked when he found out that h is ex girlfriend had been ex girlfriend told him that he was stifling her and that they had nothing in common. Upon hearing this, Alex reacted violently. He s creamed at his girlfriend to get out of his apartment, smashed a picture frame on the ground, and then punched the wall with his fist. Since Alex had
89 previously spent most of his free time with his ex girlfriend, he now feels isolated, alone, and betrayed by her actions. Pretend that you overheard Alex say the following: 17. Based on the former information and his statement, how willing would you be to report Alex? Definitely would report Pr obably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Alex say the following: 18. Based on the former information and his statement, how willing would you be to report Alex? Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you o verheard Alex say the following: 19. Based on the former information and his statement, how willing would you be to report Alex?
90 Definitely would report Probably would report Somewhat willing to report Not really willing to report Probably wouldn't report Definitely report 6 5 4 3 2 1 Pretend that you overheard Alex say the following: 20. Based on the former information and his statement, how willing would you be to report Alex? Definitely would report Probably would report Somewhat willing to report Not really will ing to report Probably wouldn't report Definitely report 6 5 4 3 2 1
91 LIST OF REFRENCES American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text revision). Washington, DC: Author. Ame rican Psychological Association. (2002). Ethical principles of psychologists and code of conduct. American Psychologist, 57, 1060 1073. Balk, D. E. (2001). College student bereavement, scholarship, and the university: A call for university engagement. Deat h Studies, 25 67 84. Balk, D. E. (1997). Death, bereavement and college students: A descriptive analysis. Mortality, 2 207 220. Band, S. R., & Harpold, J. A. (1999). School violence: Lessons learned. FBI: Law Enforcement Bulletin, 68 9 16. Baron, R. M., & Kenny, D. A. (1986). The moderator mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173 1182. Barrios, L. C., Everett, S. A., Sim on, T. R., & Brener, N. D. (2000). Suicidal ideation among U. S. college students: Associations with other injury risk behaviors. Journal of American College Health, 48 229 233. Beck, K. H., & Treiman, K. A. (1996). The relationship of social context of drinking, perceived social norms, and parental influence to various drinking patterns of adolescents. Addictive Behaviors, 21 633 644. Bohn, K. (2008, February 15). Company: Gunman, V irginia Tec h shooter used same Web dealer. CNN Retrieved on February 15 2008. http://www.cnn.com/2008/US/02/15/ university.shooting/index.html Borsari, B., & Carey, K. B. (2001). Peer influences on college drinking: A review of the research. Journal of Substance Abuse, 13 391 424. Boynton A. (2003). Securing college campuse s in the face of terrorism. Campus Law Enforcement Journal, 33 15 17. Brank, E. M., Woolard, J. L., Brown, V. E., Fondacaro, M., Luescher, J. L., Chinn, R. G., & Miller, S. A. (2007). Will they tell? Weapons reporting by middle school youth. Youth Violen ce and Juvenile Justice, 5 125 146. Beretvas, S. N., Meyers, J. L., & Leite, W. L. (2002). A reliability generalization study of the Marlowe Crowne Social Desirability Scale. Educational and Psychological Measurement, 62, 570 589.
92 Brinkley, C. J., & Saar nio, D. A. (2006). Involving students in school violence prevention: Are they willing to help? Journal of School Violence, 5 93 116. Bromley, M. L. (1995). Comparing campus and city crime rates: A descriptive study. American Journal of Police, 14 131 14 5 Brown, B. B., Dolcini, M. M., & Leventhal, A. (1997). Transformations in peer relationships at adolescence: Implications for health related behavior. In Schulenberg, J., Maggs, J. L., & Hurrelmann (Eds.), Health risks and developmental transitions duri ng adolescence (pp. 161 189). Boston, MS: Cambridge University Press. Callaway, R. L., Gehring, D. D., & Douthett, T. J. (2000). Two year college compliance with the notice requirement of the Campus Security Act. Community College Journal of Research and Practice, 24 181 91. Carleton, R. N., McCreary, D. R., Norton, R. J., & Asmundson, G. J. (2006). Brief Fear of Negative Evaluation Scale Revised. Depression and Anxiety 23 297 303. Carleton, R. N., Collimore, K. C., & Asmundson, G. J. (2007). Social anx iety and fear of negative evaluation: Construct validity of the BFNE II Journal of Anxiety Disorders, 21, 131 141. Chapin, J. (2001). Optimistic bias regarding campus violence. Current Research in Social Psychology, 6 1 11. Chaudhuri, A., & Gangadharan, L. (2003). Gender differences in trust and reciprocity. Working paper, economics. Wellesley College. Clark, D. M., & Wells, A. (1995). A cognitive model of social phobia. In Heimberg, R., Liebowitz, M., Hope, D. A., & & Schneier, F. R. (Eds.), Social phob ia: diagnosis, assessment and treatment (pp. 69 93). New York: Guilford Press. Cornell, D., Sheras, P., Kaplan, S., Levy Elkon, A., McConville, D., McKnight, L., & Posey, J. ( 2004b ). Guidelines for responding to student threats of violence: Field test of a threat assessment approach. In M. J. Furlong, P. M. Kingery, & M. P. Bates (Eds.), Appraisal and prediction of school violence: Context, issues, and methods Binghamton, NY: Haworth Press. Cornell, D. G., Benedek, E. P., & Benedek, D. M. (1987). Characte ristics of adolescents charged with homicide: Review of 72 cases. Behavioral Sciences & the Law, 5 11 23. Cornell, D. G., Sheras, P. L., Kaplan, S., McConville, D., Douglass, J., Elkon, A., McKnight, L., Branson, C., & Cole, J. (2004a). Guidelines for st udent threat assessment: Field test findings. School Psychology Review, 33 527 546. Cornell, D. G. (1990). Prior adjustment of violent juvenile offenders. Law and Human Behavior, 14, 569 577.
93 Crowne, D. P., & Marlowe, D. (1960). A new scale of social des irability independent of psychopathology. Journal of Consulting Psychology, 24, 349 354. Davies, J., McCrae, B. P., Frank, J., Dochnahl, A., Pickering, T., Harrison, B., et al. (2000). Identifying male college students' perceived health needs, barriers to seeking help, and recommendations to help men adopt healthier lifestyles. Journal of American College Health, 48, 259 268. Dillman, D. A., Smyth, J. D., & Christian, L. M. (2008). Internet, mail, and mixed mode surveys: An Arbor, MI: Wiley & Sons. Dodge, M. (2006). Juvenile police i nformants: Friend ship, persuasion, and pretense. Youth Violence and Juvenile Justice, 4 234 246 Ellen, J. M., Boyer C. B., Tschann, J. M., & Shafer, M. A. (1996). isk for STDs and HIV i nfection. Journal of Adolescent Health, 18 177 1 81. Elliott, D. S., Ageton, S. S., & Huizinga, D. (1985). Explaining delinquency and drug use Beverly Hills, CA: Sage. Ephraim, T. A. (1997). Adolescent coping strategies after a suic ide or other loss by death: A retrospective study. Dissertation Abstracts International, 58 4083. Farrington, D. P., Loeber, R., Stouthamer Loeber, M., Van Kammen, W. B., & Schmidt, L. (1996). Self reported delinquency and a combined delinquency seriousn ess scale based on boys, mothers, and teachers: Concurrent and predictive validity for African Americans and Caucasians Criminology, 34 493 518. Fein, R. A., Vossekuil, B., Pollack, W. S., Borum, R., Modzeleski, W., & Reddy, M. (2002). Threat assessment in schools: A guide to managing threatening situations and to creating safe school climates. Washington, DC: U.S. Secret Service & U.S. Department of Education. Ferguson, C. J. (2008). The school shooting/violent video game link: Causal relationship or mo ral panic? Journal of Investigative Psychology and Offender Profiling, 5, 25 37. Flora, D., & Curran, P. (2004). An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data. Psychological Methods, 9, 466 491. Fox, J. A., & Savage, J. (2009). Mass murder goes to college: An examination of changes on college campuses following Virginia Tech. American Behavioral Scientist, 52, 1465 1485. Friedman, E. (2008, February 15). Who w a s the Illinois school shooter?. ABC News. Fultz, J., Batson, C. D., Fortenbach, V. A., McCarthy, P. M., & Varney, L. L. (1986). Social evaluation and the empathy altruism hypothesis. Journal of Personality and Social Psychology, 50, 761 769.
94 Garcia, C. A. (2003). School safety technology in America: Current use and perceived effectiveness. Criminal Justice Policy Review, 14 30 54 Gardner, M., & Steinberg, L. (2005). Peer influence on risk taking, risk p reference, and risky decision making in adolescence and adulthood: An experimental s tudy Developmental Psychology, 41, 625 635. Gaughan, E., Cerio, J. D., Myers, R. A. (2001). Lethal violence in schools Alfred, NY: Alfred University Press. Gehring, D. D., & Callaway, R. L. (1997). Compliance with the notice requirement of the Campus S ecurity Act. College and University, 73 13 18. Greenberg, S. F. (2007). Active shooters on college campuses: Conflicting advice, roles of the individual and first responder, and the need to maintain perspective. Disaster Medicine and Public Health Prepar edness, 1 57 61 Hagerty, B. M., & Williams, A. (1999). The effects of sense of belonging, social support, conflict, and loneliness on depression. Nursing Research, 48 215 219 Hansen, W. B., Raynor, A. E., & Wolkenstein, B. H. (1991). Perceived person al immunity to the consequences of drinking alcohol: The relationship between behavior and perception. Journal of Behavioral Medicine, 14 205 240. Harshman, E., S. Puro, and L. A. Wolff. (2001). The Clery Act: Freedom of information at what costs to stud ents? About Campus, 3 18. Harnisch, T. L. (2008, November). Concealed weapons on state college campuses: In pursuit of individual liberty and collective security A Higher Education Policy Brief. American Association of State Colleges and Universities. Haynie, D. L. (2001). Delinquent peers revisited: Does network structure matter? American Journal of Sociology, 106, 1013 1057. Heinzmann D., & St. Clair Chicago Tribune Retrieved on January 8, 2 009 Helweg Larsen, M. (1999). Optimistic biases in response to the 1994 Northridge Earthquake: The role of personal experience. Basic and Applied Social Psychology, 21 119 129. Hsu, M. H., Ju, T. L., Yen, C. H., Chang, C. M. (2007). Knowledge sharing b ehavior in virtual communities: The relationship between trust, self efficacy, and outcome e xpectati ons. International Journal of Human Computer Studies, 65 153 1 69. Hu, L. T., & Bentler, P. M. (1999). Cutoff Criteria for Fit Indexes in Covariance Struct ure Analysis: Conventional Criteria Versus. Structural Equation Modeling, 6, 1 55.
95 Hurst, Y. G., & Frank, J. (2000). How kids view cops the nature of juvenile attitudes toward the police. Journal of Criminal Justice, 28 189 202. Hurtado, S., & Carter, D. F. (1997). Effects of college transition and perceptions of the campus racial climate on Latino college students' sense of belonging. Sociology of Education, 70 324 345. Janosik, S. M., & Gregory, D. E. (2003). The Clery Act and its influence on campus law enforcement practices. NASPA Journal, 41 1 18. Janosik, S. M. (2001). The impact of the Campus Crime Awareness Act of 1998 on student decision making. NASPA Journal, 38 348 360. Jones, W. H., Sansone, C., & Helm, B. (1983). Loneliness and interper sonal judgments. Personality and Social Psychology Bulletin, 9 437 441 Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Data analysis in social psychology. In D. T. Gilbert, S. T Fiske, & G. Lindzey (Eds.), The handbook of social psychology. New York: O xford University Press. Kline, R. B. (2005). Principles and practice of structural equation modeling, second edition. New York, NY: The Guilford Press. Langman, P. (2009). Rampage school shooters: A typology. Aggression and Violent Behavior, 14, 79 86. Lau S., & G. E., Gruen. (1992). The social stigma of loneliness: Effect of target person's and perceiver's sex. Personality and Social Psychology Bulletin, 18 182 189. Lee, R. (2005). Resilience against discrimination: Ethnic identity and other group orien tation as protective factors for Korean Americans. Journal of Counseling Psychology, 52, 36 44. Lee, R. M., & Robbins, S. B. (1995). Measuring belongingness: The social connectedness and the social assurance scales. Journal of Counseling Psychology, 42 23 2 241. Lee, R. M., Sexton, J. D., & Keough, K. A. (2002). Social connectedness, social appraisal, and perceived stress in college women and men. Journal of Counseling & Development, 80 355 361. Leite, W. L., & Beretvas, S. N. (2005). Validation of score s on the Marlowe Crowne Social Desirability Scale and the Balanced Inventory of Desirable Responding. Educational and Psychological Measurement, 65 140 154 Lind, E. A., Kanfer, R., & Earley, P. C. (1990). Voice, control, and procedural justice: Instrumen tal and noninstrumental concerns in fairness judgments. Journal of Personality and Social Psychology, 59, 952 959.
96 Lipka, S. (2009, January 30). In campus Chronicle of Higher Education, 55, A2 A16. MacNab, B. R., & Worthley, R. (2008). Self efficacy as an intrapersonal predictor for internal whistleblowing: A U.S. and Canada examination. Journal of Business Ethics, 79 407 421. Marx, G. T. (1998). Ethics for the new surveillance. The Information Society, 1 4 171 185. McGee, J. P., & DeBernardo, C. R. (1999). The Classroom Avenger. The Forensic Examiner, 8 16 28. Meilman, P. W., Leichliter, J. S., & Presley, C. A. (1998). Analysis of weapon carrying among college students, by region and institution type. Journal of American College Health, 46 291 200. Mercer, J. (1996). The legacy o f Danny Rollings Retrieved on January 10, 2009, from About Gainesville: http://www.afn.org/~riffer/projects/hogtown/5.html Miller, M., Hemenway, D., & Wechsler, H. (2002). Guns and gun threats at college. Journal of American College Health, 51 57 65. Morse, J. (2000, April 24). Looking for trouble: More and more schools are trying to spot the potential killers in their midst. But what abou t the innocents? Time pp. 50 58 Muthn, B., & Asparouhov, T. (2002). Latent variable analysis with categorical outcomes: Multiple group and growth modeling in Mplus. Mplus Web Notes, 4, 1 22. Muthn, B., du Toit, S. H. C., & Spisic, D. (in press). Robust inference using weighted least sq uares and quadratic estimating equations in latent variable modeling with categorical and continuous outcomes. Psychometrika. Muthen, L. K., & Muthen, B. O. (1998). Los Angeles: Muthen & Muthen. Muthn, B. O., & Satorra, A. (1995). Tech nical aspects of Muthn's LISCOMP approach to estimation of latent variable relations with a comprehensive measurement model. Psychometrika, 60, 489 503. North, C. S., Smith, E. M., & Spitznagel, E. L. (1994). Posttraumatic stress disorder in survivors of a mass shooting. American Journal of Psychiatry, 151 82 88. Norris, F. H. (2007). Impact of mass shootings on survivors, families, and communities. PTSD Research Quarterly, 18, 1 8. O'Toole, M. (2000). The school shooter: A threat assessment perspective Quantico, VA: National Center for the Analysis of Violent Crime, Federal Bureau of Investigation.
97 Pascarella, E. T., Edison, M., Nora, A., Hagedorn, L. S., & Terenzini, P. T. (1996). Influences on students' openness to diversity and challenge in the first year of college. Journal of Higher Education, 67, 143 156. Patrick, K., Covin, J. R., Fulop, M., C alfas, K., Lovato C., & Patrick, K. (1997). Health risk behaviors among California college students. Journal of American College Health, 45 265 72. Perkin s, H. W. (2002). Social norms and the prevention of alcohol misuse in collegiate contexts. Journal of Studies on Alcohol, 14 164 172. Pershing, J. L. (2002). Whom to betray? Self regulation of occupational misconduct at the United States Naval Academy. D eviant Behavior, 23 149 175. Peterson Education: University of Florida student body. (2010). Retrieved from http://www.petersons.com/colleg eprofiles/Profile.aspx?inunid=9188&reprjid=12&sponsor =1&tabid=10005 Pinel, E. C. (1999). Stigma consciousness: The psychological legacy of social stereotypes. Journal of Personality and Social Psychology, 76 114 128. Pi q uero, A. R., MacIntosh, R., & Hic kman, M. (2002). The validity of a Self reported Delinquency Scale: Comparisons across gender, age, race, and place of r esidence. Sociological Methods & Research, 30, 492 529 of working undercover. Journal of Criminal Justice, 21 383 395. Pollack, W. S ., Modzeleski W., & Rooney, G. (2008). Prior knowledge of p otential school based violence: Information students learn may prevent a targeted attack. Washington, DC: U.S. Secret Service & U.S. Department of Education. Presley, C. A., Meilman, P. W., & Cashin, J. R. (1997). Weapon carrying and substance abuse among college students. Journal of American College Health, 46 3 9. Quiroga, A. M. (1992). Studies of the polychoric corr elation and other correlation measures for ordinal variables. Unpublished doctoral dissertation, Acta Universitatis Upsaliensis. Rapee, R. M., & Heimberg, R. G. (1997). A cognitive behavioral model of anxiety in social phobia. Behaviour Research and Therap y, 35 741 756. Reddy, M., Borum, R., Berglund, J.,Vossekuil, B., Fein, R., & Modzeleski, W. (2001). Evaluating risk for targeted violence in schools: Comparing risk assessment, threat assessment, and their approaches. Psychology in the Schools, 38, 157 1 72. Revels, D. W. (1999). Attitudes of campus students and campus police personnel toward campus law enforcement at UW Stout). Retrieved from: http://www.uwstout.edu/lib/thesis/1999/1999revels.pdf
98 Rosenfeld, R., Jacobs, B. A., & Wright, R. (2003). Snitching and the code of the street. British Journal of Criminology, 43 291 309. Rotenberg, K. J. (1994). Loneliness and interpersonal trust. Journal of Social and Clinical Psychology, 13 152 173. Rotter, J. B. (1971). Generalized expectancies for interpersonal trust. American Psychologist, 26 443 452. Rotter, J. B. (1980). Interpersonal trust, trustworthiness, and gullibility. Ameri can Psychologist, 35, 1 7. Scholly, K., Katz, A. R., Gascoigne, J., Holck, P. S., & Scholly, K. (2005). Using s ocial norms theory to explain perceptions and sexual health behaviors of undergraduate college students: An e xploratory study. Journal of Americ an College Health, 53 159 1 66. Schuh, J. H. (1999). Guiding principles for evaluating student and academic affairs partnerships. New Directions for Student Services, 87 85 92. Schumacker, R. E., & Lomax, R. G. (1996). quation modeling. Mahwah, NJ: Erlbaum. Schwarz, E. D., Fapa, F., & Kowalski, J. M. (1991). Malignant memories: PTSD in children and adults after a school shooting. Journal of the American Academy of Child & Adolescent Psychiatry, 30 936 944 Schwarzer. R (1993). Measurement of perceived self efficacy : Psychometric scales for cross cultural research. Berlin, Germany: Freie Universitat Berlin. Sewell, K. W., & Mendelsohn, M. (2000). Profiling potentially violent youth: Statistical and conceptual problems. Social Policy, Research, and Practice, 3 147 169. Skiba, R. J., & Knesting, K. (2002). Zero tolerance, zero evidence: An analysis of school disciplinary practice. In R.J. Skiba & G.G. Noam (Eds.). New directions for youth development (pp. 17 43). San Fr ancisco: Jossey Bass. Skiba, R., & Peterson, R. (1999). The dark side of zero tolerance: Can punishment lead to safe schools? Phi Delta Kappan, 80 372 394. Skiba, R., Reynolds, C. R., Graham, S., Sheras, P., Conoly, J. C., & Garcia Vazquez, E. (2006). Ar e zero tolerance policies effective in the schools? An evidentiary review and recommendations. Zero Tolerance Task Force Report. The American Psychological Association. Skiba, R. J., Michael, R. S., Nardo, A. C., & Peterson R. L. (2002). The color of di scipline: Sources of racial and gender disproportionality in school punishment. Urban Review, 34 317 342.
99 Sloan III, J. J., Fisher, B. S., & Cullen, F. T. (1997). Assessing the student right to know and Campus Security Act of 1990: An analysis of the vi ctim reporting practices of college and university students. Crime & Delinquency, 43 148 168 Smith, G. E., Gerrard, M. & Gibbons, F. X. (1997). Self esteem and the relation between risk behavior and perceptions of vulnerability to unplanned pregnancy in college women. Health Psychology, 16 137 146. Strahan, R., & Gerbasi, K. C. (1972). Short, homogeneous versions of the Marlowe Crowne Social Desirability Scale. Journal of Clinical Psychology, 28 191 19 3. Stueve, A., Tehranifa r P., Wilson Simmons R., Slaby, R., Link, B. G. (2006). Rethinking the bystander role in school violence prevention. Health Promotion Practice, 7, 117 124. Summers, J., Beretvas, S., Svinicki, M. & Gorin, J. (2005). Evaluating collaborative learning an d community. Journal of Experimental Education, 73 165 188. Summers, J. J., Svinicki, M. D., Gorin, J. S., & Sullivan, T. A. (2002). Student feelings of connection to the campus and openness to diversity and challenge at a large research university: Evide nce of progress? Innovative Higher Education, 27, 53 64. Syvertsen, A. K., Flanagan, C. A., & Stout, M. D Journal of Educational Psyc hology, 101, 219 232. Tinto, V. (1993). Leaving college: Rethinking the causes and cures of student attrition (2nd ed.). Chicago: University of Chicago Press. Tobin, T., Sugai, G., & Colvin, G. (1996). Patterns in middle school discipline records. Journa l of Emotional and Behavioral Disorders, 4, 82 94. Toomey, K. A. (2004). Practice pointer: The snitch rule. Utah State Bar, 17 24 26. Trappler, B., & Friedman, S. (1996). Posttraumatic stress disorder in survivors of the Brooklyn Bridge Shooting. American Journal of Psychiatry, 153 705 707 Tschannen Moran, M., & Hoy, W. (1998). Trust in schools: A conceptual and empirical analysis. Journal of Educational Administration, 36 334 352. Unnever, J. D., & Cornell, D. G. (2004). Middle school victims of bull ying: Who reports being bullied? Aggressive Behavior, 30, 373 388. U.S. Department of Education Office of Postsecondary Education (n.d.). Campus security data analysis cutting tool. Retrieved November 22, 2009, from http://ope.ed.gov/security/
100 Utterback, J., & Caldwell, J. (1989). Proactive and reactive approaches to PTSD in the aftermath of campus violence: Forming a traumatic stress react team. Journal of Traumatic Stress, 2 171 183. Verlinden, S., Hersen, M., Thomas, J. (2000). Risk factors in school shootings. Clinical Psychology Review, 20 3 56. Vossekuil, B., Fein, R. A., Reddy, M., Borum, R., & Modzeleski, W. (2000). The final report and findings of the Safe School Initiative: Implications for the pre vention of school attacks in the United States. Washington, DC: U.S. Secret Service and U.S. Department of Education. Walsh, C. M., & Hardy, R. C. (1999). Dispositional differences in critical thinking related to gender and academic major. Journal of Nurs ing Education, 38, 149 155. Wasko M. L., & Faraj, S. (2005). Why should I share? Examining social capital and knowledge contribution in electronic networks of practice. MIS Quarterly, 29 35 58. Watson, D., & Friend, R. (1969). Measurement of social eval uative anxiety. Journal of Consulting and Clinical Psychology, 33 448 457. Weber, P. S., Weber, J. E., Sleeper, B. R., Schneider, K. L., & Weber, P. S. (2004). Self efficacy toward service, civic participation and the business student: Scale development and validation. Journal of Business Ethics, 49 359 3 69. Westmarland, L. (2005). Police ethics and integrity: Breaking the blue code of silence. Policing and Society, 15 145 165. Yu, C. Y. (2002). Evaluating cutoff criteria of model fit indices for late nt variable models with binary and continuous outcomes. Doctoral dissertation, University of California, Los Angeles. Zaheer, A., McEvily, B., & Perrone. B. (1998). Does trust matter? Exploring the effects of interorganizational and interpersonal trust on performance. Organization Science, 9 141 159. Zdziarski, E. L., Dunkel, R. W., & Rollo, J. M. (2007). Campus crisis management: A comprehensive guide to planning, prevention, response, and recovery. New York, NY: Wiley, John and Sons.
101 BIOGRAPHICAL SKE TCH Michael Lee Sulkowski was born in Getzville, N ew York in 1984. He spent his formative years in Amherst, NY and graduated from Williamsville North High School in 2002. Michael then matriculated at Canisius College and earned Bachelors of Arts degrees in history and p sychology in 2006. He entered the School Psychology Program at the University of Florida in 2006 and earned a Master or Education degree in 2007. Michael plans to earn his Doctor of Philosophy in school p sychology in 2011 from the University of Florida after the completion of his professional internship in psychology. He will then be a postdoctoral fellow at the Rothman Center for Pediatric Neuropsychiatry in the Department of Pediatrics at the University of South Florida. During his gradua te career, he has received the following awards and honors: The University of Florida College of Education Outstanding Graduate Research Award (2011), t he Everett L. Holden & Marian G. Holden Memorial Scholarship (2010), the Melissa Institute Belfer Aptman Dissertation Research Award (2009), the American Psychological Association D ivision 55 Patrick H. DeLeon Award (2009), the A merican Academy of School Psychology Irwin Hyman/Nadine Lambert Memorial Scholarship (2009), the Florida Association of School Psyc h ology Doctoral Graduate Studies Award (2009), Numerous Res earch Travel Awards (2008 2010 ), the Joseph Lillian Damon Scholarship (2007), and the Norman F. Nelson Fellowship (2007). Michael also has published the following articles and book chapters: Sul kowski, M. L., & Lazarus, P. J. (in press). Contemporary responses to violent attacks on college campuses. Journal of School Violence. Saklofske, D., Joyce, D. J., Sulkowski, M. L., & Climie, E. (in press). Models of personality assessment for children an d adolescents. In C. R. Reynolds (Ed.), Principles and Models of Personality Assessment Oxford University Press.
102 Sulkowski, M. L., Storch, E. A., & Lewin, A. B. (in press). D Cycloserine augmentation of fear extinction and exposure based anxiety treatme nt In J. E. Murray (Ed.), Exposure therapy: New developments (pp. XX XX). Hauppauge, NY: Nova Publishers. Long, S., Sulkowski, M. L., Dempsey, A. G. (in press). Cyber victimization and psychosocial adjustment in youth. In S. Mortella (Ed.), Psychology of Victimization (pp. XX XX). Hauppauge, NY: Nova Publishers. Sulkowski, M. L., Dempsey, J., & Dempsey, A. G. (in press). Effects of stress and coping on binge eating in female college students. Eating Behavior Sulkowski, M. L., West, J., & Lazarus, P. J. (in press and dazed but not out. Communiqu. Sulkowski, M. L., Mariaskin, A., & Storch, E. A. (2011). Obsessive compulsive spectrum disorder symptoms in college students. Journal of American College Health, 59 342 348. doi: 10.1080/07448481.2010.511365 Sulkowski, M. L., Wingfield, R. J., Jones, D., & Coulter, W. A. (2011). Response to intervention and interdisciplinary collaboration: Joining hands to support children and families. Journal of Applied School Ps ychology, 27, 1 16. doi: 10.1080/15377903.2011.565264 Sulkowski, M. L., Mancil, T. L., Jordan, C., Reid, A., Chakoff, E., & Storch, E. A. (2011). Validation of a classification system of obsessive compulsive spectrum disorder symptoms in a non clinical sa mple Psychiatry Research. doi: 10.1016/j.psychres.2011.01.015 in campus communities. Psychology of Violence, 1, 53 65. doi: 10.1037/a0021592 Dempsey, A. G ., Sulkowski, M. L., Dempsey, J., & Storch, E. A. (2010). Has cyber technology produced a new group of peer aggressors? CyberPsychology, Behavior, and Social Networking. doi: 10.1089/cyber.2010.0108 Pence, S. L., Jr., Aldea, A., Sulkowski, M. L., & Storch E. A. (2010). Cognitive behavioral therapy in adults with obsessive compulsive disorder and borderline intellectual functioning: A case series of three patients. Journal of Developmental and Physical Disabilities. doi: 10.1007/s10882 010 9200 6 Lazarus, P. J., & Sulkowski, M. L. (2010). Oil in the water, fire in the sky: Responding to technological/environmental disasters. Communiqu, 39, 1 & 16 17. Pence, S. L., Jr., Sulkowski, M. L., Jordan, C., & Storch, E. A. (2010). When exposures go wrong: Trouble shooting guidelines for managing difficult scenarios that arise in exposure based treatment for obsessive compulsive disorder. American Journal of Psychotherapy, 64, 39 53.
103 Sulkowski, M. L., Jordan, C., Reid, A., Graziano, P. A., Shalev, I., & Storch, E. A. (2009). Relations between impulsivity, anxiety, and obsessive compulsive symptoms in a non clinical sample. Personality and Individual Differences 47, 620 625. Jordan, L. M., Jordan, C., Sulkowski, M. L., Reid, A., Geffken, G. R., & Storch, E. A. (20 09). treatment. Psychology, Health & Medicine, 14, 680 688. Dempsey, A. G., Sulkowski, M. L., Nichols, B., & Storch, E. A. (2009). Differences between peer vic timization in cyber and physical settings and associated adjustment in early adolescence. Psychology in the Schools, 46, 962 972. Sulkowski, M. L., Jordan, C., & Nguyen, M. L. (2009). Current practices and future directions in psychopharmacological train ing and collaboration in school psychology. Journal of Canadian School Psychology, 24, 237 244. Shalev, I., Sulkowski, M. L., Geffken, G. R., Rickets, E., & Storch, E. A. (2009). Letter to the Editor: Long term durability of cognitive behavioral therapy treatment gains for pediatric obsessive compulsive disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 48, 766 768. Salloum, A., Sulkowski, M. L., Sirrine, E., & Storch, E. A. (2009). Overcoming barriers to using empirically supp orted therapies to treat childhood anxiety disorders in social work practice. Child and Adolescent Social Work Journal, 26, 259 273 Shalev, I., & Sulkowski, M. L. (2009). Relations between distinct aspects of self regulation to symptoms of impulsivity an d compulsivity. Personality and Individual Differences, 47, 84 88. Sulkowski, M. L., Mariaskin, A., Jordan, C., & Storch, E. A. (2009). Obsessive compulsive disorder and the obsessive compulsive disorder spectrum: A review of research. In G. H. Lassiter ( Ed.), Impulsivity: Causes, control and disorders (pp. 31 58). Hauppauge, NY: Nova Publishers. Sulkowski, M. L., Storch, E. A., Geffken, G. R., Ricketts, E., Murphy, T. K., & Goodman, W. K. (2008). Concurrent validity of the Yale Brown Obsessive Compulsiv e Scale Symptom checklist. Journal of Clinical Psychology, 64, 1338 1351 Oakland, T. O., Mpofu, E., & Sulkowski, M. L. (2007). Temperament styles of Zimbabwe and U.S. children. Journal of Canadian School Psychology, 21, 139 153.