Cyber Victimization on College Campuses

Abstract

Although much is known about the cross-sectional associations between cyber victimization and the negative socioemotional outcomes associated with this experience, not much is known about the longitudinal associations among college students. The purpose of the present study was to examine longitudinal, bidirectional associations between cyber victimization, suicidal ideation, depression, and anxiety among college students, using cross-lagged models. These relationships were examined over 4 years. Participants were 1,483 college students (M _age = 24.67; 60% female; 35% White, 15% Black/African American, 10% Latino/Latina, 6% Asian, and 4% biracial) from Southeastern universities in the United States. They completed self-reports of face-to-face and cyber victimization and questionnaires on suicidal ideation, depression, and anxiety at four time points over 4 years. Findings revealed that cyber victimization contributed to suicidal ideation, depression, and anxiety over time, and that suicidal ideation, depression, and anxiety each contributed to cyber victimization over time as well. Such findings suggest bidirectional relationships between these variables, although there were differences in the size of the bivariate relationships. In particular, the magnitudes of the associations were stronger when cyber victimization predicted suicidal ideation, depression, and anxiety. Recommendations are provided to help reduce or eliminate cyber victimization among students on college campuses.

Keywords

cyber victimization suicidal ideation depression anxiety college student

Introduction

Not only have many children and adolescents grown up in an increasingly digitally connected world, but so have many college students. Many college students go online at least once a day, enjoying the opportunities afforded by the digital age, including the decreasing costs of digital technologies, having access to a vast array of information at their fingertips, and being able to communicate with anyone in just a matter of minutes (A. Smith, Rainie, & Zickuhr, 2011).

Despite the opportunities offered by digital technologies, there are also some risks associated with college students’ technology usage, including Internet addiction, identity theft, and cyber victimization (Copes, Kerby, Huff, & Kane, 2010; Finn, 2004; Weigman & van Schie, 1998; Wright & Li, 2012). High levels of technology use increase the risks associated with cyber victimization (Alvarez-Garcia, Nunez Perez, Gonzalez, & Perez, 2015; Holt, Fitzgerald, Bossler, Chee, & Nq, 2016; Mishna, Khoury-Kassabri, Gadalla, & Daciuk, 2012; Perrin & Duggan, 2015). Therefore, college students’ immersion in digital technology increases their online risks, particularly their exposure to cyber victimization.

Regardless of the connection between digital technology use and cyber victimization, little attention has been given to college students’ exposure to this behavior. Most of the research on cyber victimization has focused on children and adolescents, leading to the mistaken perception that these behaviors are only a concern for younger segments of the population. Not recognizing college students’ vulnerability to experiencing cyber victimization might increase their risk of negative socioemotional outcomes. Although there is still more research needed in this area, some studies have been conducted on cyber victimization among college students typically utilizing concurrent research designs, making it difficult to understand the long-term predictors and consequences related to this experience. The present study utilized cross-lagged models to simultaneously examine reciprocal associations between cyber victimization, suicidal ideation, depression, and anxiety, while also controlling for previous levels of these variables and face-to-face victimization. These relationships were examined among college students over 4 years, with an assessment each year, yielding a total of four waves of data.

Cyber Victimization Among College Students

Defined as being purposefully embarrassed or intimated by others in a repetitive and hostile way through digital technologies, cyber victimization is typically conceptualized as an extension of face-to-face traditional victimization (Ferdon & Hertz, 2007; Kowalski & Limber, 2007; Topcu, Erdur-Baker, & Aydin, 2008; Wolak, Mitchell, & Finkelhor, 2007; Ybarra, Diener-West, & Leaf, 2007). Cyber victimization includes an imbalance of power between the perpetrator and the victim (Grigg, 2010). These experiences are typically malicious, and they can include receiving abusive e-mails, being impersonated, and receiving harassing messages via instant messenger, social networking sites, and text messages (Walker, Sockman, & Koehn, 2011; Wolak et al., 2007; Wright & Li, 2012; Ybarra & Mitchell, 2004). Other behaviors can include being the target of harassment, insults, physical threats, social exclusion, and humiliation. It can also include having explicit videos distributed of oneself and being the victim of happy slapping and flaming (Rideout, Roberts, & Foehr, 2005; P. K. Smith et al., 2008). Happy slapping involves a group of people who randomly insult another person, film the incident on a mobile phone, and then post the images or videos online. Flaming involves someone posting a provocative or offensive message in a public forum with the intention of provoking an angry response or agreement from members of the forum.

The earliest studies on cyber victimization included investigations of the prevalence rates, with most of this research conducted on children and adolescents. In one of the earliest studies to examine cyber victimization among college students, Finn (2004) found that 10–15% of her sample experienced e-mail and/or instant messenger victimization. Digital technologies have changed a lot since 2004, and to follow-up this research, Walker, Sockman, and Koehn (2011) found that 56% of college students in their sample were targeted via Facebook. Furthermore, Wright and Li (2012) found that the majority of their college-aged participants reported experiencing cyber victimization every 2–4 months. Ultimately, prevalence rates of cyber victimization among college students varies ranging from 8.6% to 43.3% (Crosslin & Crosslin, 2014; Finn, 2004; Hinduja & Patchin, 2010; Schenk & Fremouw, 2012). Variations in prevalence rates might reflect differences in the samples used, measurement of cyber victimization, and definitions of cyber victimization. Moving beyond prevalence rates, some researchers have focused on the correlates of cyber victimization. In these studies, high levels of Internet use and cyberbullying involvement as adolescents, and low self-control and self-esteem are linked to college students’ experience of cyber victimization (Balakrishnan, 2015; Bossler, & Holt, 2010; Hemphill & Heerde, 2014; Ngo & Paternoster, 2011). Despite evidence that cyber victimization is a concern among college-aged individuals, research is slow to develop on these experiences among this age-group.

Consequences of Cyber Victimization for College Students

Because most of the research on cyber victimization has been conducted among children and adolescents, it is not surprising that only a handful of studies have been conducted on the consequences associated with these behaviors among college students. Exposure to cyber victimization increases college students’ risk for eating disorders, suicide, and depression (Cowie, 2013). College students who are victims of cyberbullying are also at an increased risk for not only depression but also anxiety, phobias, and paranoia (Cowie, 2013; Schenk & Fremouw, 2012). Victims’ isolation might further increase, as they are not likely to form new friendships following cyber victimization (Crosslin & Crosslin, 2014). Kokkinos, Antoniadou, and Markos (2014) also found evidence that depression and anxiety increased cyber victimization exposure among college students. Considering the literature reviewed in this section, it is clear that there are associations among cyber victimization and negative socioemotional outcomes.

The Present Study

Considering that there is a lack of longitudinal studies conducted on cyber victimization and the negative socioemotional outcomes associated with this experience among college students, this study investigated the reciprocal relationships between cyber victimization, suicidal ideation, depression, and anxiety, while controlling for prior levels of these variables and face-to-face victimization. Autoregressive cross-lagged panel analyses were used to examine these potential reciprocal relationships over 4 years, equating to four time periods (Bollen & Curran, 2006). Such models help to control for stability effects and the concurrent relationships among the variables examined in this study, which provide a more controlled examination of the changes in associations over time.

Based on the concurrent research evidence among college students and the longitudinal study results from the literature on cyber victimization among children and adolescents, bidirectional relationships were hypothesized between cyber victimization, suicidal ideation, depression, and anxiety over 4 years ( Cowie, 2013; Crosslin & Crosslin, 2014; Schenk & Fremouw, 2012; Wright, 2015; Ybarra et al., 2007). These bidirectional or reciprocal relationships were expected because research suggests that cyber victimization is a predictor of these variables, and that these variables are also predictors of cyber victimization (Cappadocia, Craig, & Pepler, 2013; Ybarra & Mitchell, 2004). Given the lack of longitudinal designs, particularly those utilizing cross-lagged models, it is not clear whether these associations might be stronger for a particular ordering of the variables; for example, the magnitude might be higher when cyber victimization predicted suicidal ideation when compared to suicidal ideation predicting cyber victimization. Another important aim of this study was to also control for face-to-face victimization when examining the associations among the variables in this study and to specify time parameters for college students’ self-reports of victimization. Many studies focusing on college students do not specify time parameters (e.g., within the past 60 days and within the past year) when assessing cyber victimization. This is an important consideration because it is not clear whether college students are reporting lifetime prevalence rates or current rates of cyber victimization. In addition, specifying time parameters can strengthen the argument that cyber victimization is a concern for college campuses.

Method

Participants

The sample consisted of 1,483 college students enrolled in one of three 4-year universities located in the Southeastern United States, with total enrollment numbers ranging from 29,636 to 59,770. Participants were between the ages of 21 and 29 (M _age = 24.67 years; 60% female). The majority of the participants self-identified as White (65%), followed by Black/African American (15%), Latino/Latina (10%), Asian (6%), and biracial (4%). Most of the students were college seniors, with 10% classified as juniors. Students reported family incomes ranging from US$23,000 to US$461,000. Most students came from middle-class families and about 85% received financial assistance, either through loans, scholarships, or work–study programs. These demographic variables are similar to those for the overall student population at each of the universities.

Procedures

Participants were recruited during their first year in college. They were initially recruited through the psychology subject pool in which they received course credit and/or extra credit for completing research related to psychology in their Introduction to Psychology courses. The opportunity for participation in this study was posted on the psychology subject pool website, with a short description describing the nature of the study and the length of students’ expected participation. The study was advertised as occurring over 4 years, and that students would be expected to complete surveys once a year for a total of 4 times. The first wave of data collection occurred in the fall of the students’ first year in college. They came to a research laboratory for this initial wave of data collection. Before completing any documents, students were screened to determine their eligibility. Students had to be first-year college students to qualify for the study. Thirty-five students were not eligible for the study, although they still received credit for attending the session. After determining eligibility, students completed an informed consent document, which further explained the study, the length of the study, what they would be expected to do over the 4 years, and how they would be compensated (US$10 per wave for a total of US$40 over the course of the study). Participants were informed that they could still participate in the study, even if they were unsure of whether they could commit for the 4-year duration. They were also informed that participation was voluntary and that if they decided not to participate or to stop participating during the survey, then they would still receive credit through the psychology subject pool. Thirty-five students declined to participate and they were thanked for their time and given credit. After reading the informed consent, 2,227 first-year college students agreed to participate in the study. Before completing the questionnaires, an interview was conducted with college students to gather demographic information. Students confirmed again that they were first-year college students, provided their gender, income levels of their parents (348 participants were unsure), and information about their financial assistance. Next, students were asked for their permanent home address, a school-affiliated e-mail address, a personal e-mail address, and their phone number. The purpose of this information was explained in the interview as providing a way for the research personnel to contact them over the next 4 years. This personal information was linked to the demographic information and data through an ID. The linkage of this information was important, as the nature of the questionnaires included sensitive information in which students might pose a risk to themselves. Therefore, it was necessary to be able to link back to participant information in the event that a student needed to be referred for psychological counseling on their schools’ campus. The study was approved by all three universities’ institutional review boards, and APA standards were followed throughout the completion of this study. At all waves of data collection, participants completed measures on self-reported cyber victimization, self-reported face-to-face victimization, suicidal ideation, depression, and anxiety.

Data were also collected from participants 2 years, 3 years, and 4 years later. Participants were contacted via the phone before sending them information about the online questionnaires. The purpose of this initial phone interview was to reverify their permanent home address, school-affiliated e-mail address, and personal e-mail address. This information was updated as needed over the remaining 3 years. There were 52 participants who did not have a phone number that was working during Wave 2. Research personnel were able to get in contact again with 10 of these participants, and a phone meeting was scheduled to verify their personal information. After making contact with participants, they received an e-mail with a website directing them to the online questionnaires. Before completing questionnaires, participants read an informed consent document. If they agreed to participate, they were taken to the questionnaires. If they did not agree to participate, they were not taken the questionnaires and thanked for their time. The same procedure was used for Wave 3 and Wave 4. After completing the online questionnaires, participants were asked for an e-mail address that they checked often. An Amazon gift card of US$10 was e-mailed to students who wished to receive one after each wave of data collection (depending on the wave around 60–73% requested the gift card). At Wave 2, 42 participants were unreachable and 112 declined to participate (Wave 2 attrition = 154). The reason for declining participation was not asked. At Wave 3, 68 participants were unreachable by phone, but 8 were able to be contacted via e-mail. There was attrition of 60 participants who were unreachable and 78 who declined to participate (Wave 3 attrition = 138). At Wave 4, 65 participants were unreachable by phone, but 2 were able to be contacted again through e-mail. Therefore, attrition at Wave 4 was 63 participants who were unreachable and 167 who declined to participate (Wave 4 attrition = 230). Attrition rates were between 9% and 15%, such that 522 of the 2,227 college students from the first wave did not participate by the end of the fourth wave. This attrition rate is not alarming as the longer the follow-up period, the greater the chance for dropout, and many studies typically reported 30–70% dropout, which is higher than 15% for the current study (Fischer, Dornelas, & Goethe, 2001).¹ The total participation rate at each school was 521 for School A, 514 for School B, and 501 for School C.

Measures

Self-reported cyber victimization

Participants were asked to report how often they experienced a variety of negative behaviors online or through text messages. Some of these behaviors included having rumors spread about oneself, being teased, having mean or hurtful comments posted about oneself, and having someone threaten to hurt them (Wright & Li, 2012). They rated 10 items on a scale of 1 (never) to 9 (daily). The instructions of the questionnaire asked participants to respond based on what has happened within the last 60 days. All behaviors were combined to form final scores on self-reported cyber victimization, with Cronbach’s αs of .88 (Wave 1), .87 (Wave 2), .91 (Wave 3), and .90 (Wave 4).

Face-to-face victimization

Similar to the previous questionnaire, participants reported how often they were victimized by negative behaviors in the offline world (Wright, Li, & Shi, 2014). The items were similar to the “Self-Reported Cyber Victimization” questionnaire, except the “online or through text messages” portion of the item was removed, and they were also rated on a scale of 1 (never) to 9 (daily). Participants were asked to respond to the questions by thinking about their experiences within the last 60 days. All items were combined to form final scores on self-reported victimization. Cronbach’s αs were acceptable for each wave (.90 for Wave 1, .86 for Wave 2, .86 for Wave 3, and .88 for Wave 4).

Suicidal ideation

Suicidal ideation was assessed by asking participants two questions (i.e., during the past 12 months, how often have you seriously considered attempting suicide, and during the past 12 months, did you make a plan about how you would attempt suicide; Center for Disease Control and Prevention [CDC], 2011). The original items from the CDC (2011) included yes or no responses. This response set was updated to a 5-point Likert-type scale or this study, with participants rating the 2 items on a scale of 1 (never) to 5 (very often). Both items were combined to form four scores for suicidal ideation. Cronbach’s αs were .73 (Wave 1), .73 (Wave 2), .71 (Wave 3), and .70 (Wave 4).

Depression

The Beck Depression Inventory was used to assess depression among the college students (Beck, Steer, & Brown, 1996). There were 21 items with four statements per item. Participants were asked to pick the statement that best fit how they have been feeling within the past 2 weeks. For example, participants had to pick on a scale of 0 (I do not feel sad) to 3 (I am sad and unhappy that I can’t stand it) just how sad they have felt. Items were combined to form four scores on depression, with Cronbach’s αs of .86 (Wave 1), .88 (Wave 2), .88 (Wave 3), and .86 (Wave 4).

Anxiety

The Beck Anxiety Inventory was used to examine participants’ symptoms of anxiety over the past month, including the day they completed the questionnaire (Beck & Steer, 1993). Participants read 21 items representing a symptom of anxiety (e.g., numbness or tingling and feeling hot), and they rated them on a scale of 0 (not at all) to 3 (severely—it bothered me a lot). All 21 items were combined to form a final score of anxiety symptoms. Cronbach’s αs were acceptable (.83 for Wave 1, .81 for Wave 2, .81 for Wave 3, and .84 for Wave 4).

Data Analysis

To test the study’s hypotheses, various autoregressive cross-lagged path models were constructed within Mplus 7.31 (Bollen & Curran, 2006) to examine the longitudinal bivariate associations among cyber victimization, suicidal ideation, depression, and anxiety from Wave 1 to Wave 4. Using this framework, all variables were regressed on proceeding variables, making it possible to investigate bivariate effects between variables while also controlling for previous levels of these variables over time. Face-to-face victimization was allowed to predict all variables in the models in an effort to control for the high correlation among these variables and to examine the associations beyond the impact of face-to-face victimization (Wright & Li, 2012). The model provides information about comparative fit index (CFI), root mean square error of approximation (RMSEA), and standardized root mean square (SRMR) to determine model fit. Before performing the analyses, the missing completely at random (MCAR) test was performed to determine whether data were missing completely at random (Little & Rubin, 1989; Rubin & Coplan, 2010). Results from this analysis suggested that all data were missing completely at random, χ² = 76.79, df = 99, p = .95. Consequently, only complete cases, with participants who had information at all four waves, were included in the analyses. There were 53 instances of missing data from the cases that were not used for the analyses. The final sample size was 1,483. Several multigroup path analyses were conducted to examine ethnicity and gender differences between cyber victimization, suicidal ideation, depression, and anxiety across the four waves. These were compared to freely estimated and constrained models, and then the χ² differences were examined (Hancock & Mueller, 2006). Findings revealed that the analyses did not vary based on ethnicity and gender, and consequently, ethnicity and gender were not included in the final model to retain more parsimonious models.

Results

Before conducting the analyses for the hypotheses, descriptive statistics were computed for each of the study’s variables (see Table 1). Correlational analyses revealed that cyber victimization at Wave 1, Wave 2, Wave 3, and Wave 4 was related positively with anxiety, depression, and suicidal ideation, assessed at all four waves as well.

Table 1.

Correlations among all waves of cyber vicitmization, suicidal ideation, depression, and anxiety.

Variable	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
1. W1 CV	—
2. W2 CV	.67***	—
3. W3 CV	.66***	.70***	—
4. W4 CV	.62***	.67***	.74***	—
5. W1 SI	.39***	.37***	.42***	.48***	—
6. W2 SI	.36***	.36***	.43***	.48***	.70***	—
7. W3 SI	.34***	.33***	.45***	.46***	.71***	.69***	—
8. W4 SI	.35***	.41***	.42***	.41***	.71***	.72***	.70***	—
9. W1 DEP	.51***	.53***	.49***	.50***	.74***	.75***	.68***	.73***	—
10. W2 DEP	.48***	.49***	.46***	.50***	.68***	.67***	.69***	.71***	.68***	—
11. W3 DEP	.46***	.51***	.50***	.51***	.69***	.69***	.63***	.67***	.67***	.69***	—
12. W4 DEP	.46***	.56***	.46***	.46***	.70***	.69***	.65***	.70***	.67***	.63***	.72***	—
13. W1 ANX	.26***	.29***	.30***	.29***	.22**	.21**	.23**	.19*	.43***	.44***	.41***	.42***	—
14. W2 ANX	.21**	.31***	.32***	.29***	.19*	.20*	.26***	.20*	.41***	.41***	.40***	.41***	.71***	—
15. W3 ANX	.25***	.22**	.28***	.27***	.19*	.18*	.23**	.23*	.35***	.41***	.36***	.39***	.70***	.67***	—
16. W4 ANX	.18*	.17*	.24**	.25**	.23**	.18*	.21*	.25**	.34***	.36***	.42***	.34***	.70***	.66***	.63***	—
M (SD)	5.10 (.67)	5.06 (.66)	5.19 (.71)	5.23 (.72)	2.43 (1.01	2.51 (1.00)	2.48 (.99)	2.50 (.99)	1.87 (.67)	1.91 (.68)	1.90 (.63)	1.91 (.66)	2.01 (.76)	2.08 (.73)	2.08 (.77)	2.13 (.78)

Note. W1 = Wave 1; W2 = Wave 2; W3 = Wave 3; W4 = Wave 4; CV = cyber victimization; SI = suicidal ideation; DEP = depression; ANX = anxiety.

*p < .05. **p < .01. ***p < .001.

Relationship Between Cyber Victimization and Suicidal Ideation

The reciprocal relationship between cyber victimization and suicidal ideation was examined through an autoregressive cross-lagged model, with paths specified between all waves of cyber victimization and suicidal ideation from Waves 1 to 4. The model had excellent fit, χ² = 96.13, df = 12, CFI = .99, RMSEA = .05, and SRMR = .04. The coefficients for the paths revealed stability for both cyber victimization and suicidal ideation over time (see Figure 1). Wave 1, Wave 2, Wave 3, and Wave 4 cyber victimization predicted all waves of suicidal ideation, after controlling for prior suicidal ideation, cyber victimization, and face-to-face victimization, and all waves of suicidal ideation predicted cyber victimization, although the magnitudes of these associations were weaker.

Figure 1.

Autoregressive cross-lagged model of cyber victimization and suicidal ideation. Standard coefficients are displayed. *p < .05.

Relationship Between Cyber Victimization and Depression

To examine the reciprocal relationship between cyber victimization and depression, an autoregressive cross-lagged model was created. Paths were added across all waves of cyber victimization and depression from Waves 1 to 4. The model had excellent fit, χ² = 119.23, df = 12, CFI = .97, RMSEA = .06, and SRMR = .05. Similar to cyber victimization and suicidal ideation, the results from the model with cyber victimization and depression indicated that there is stability in these constructs over time as well (see Figure 2). In addition, all waves of cyber victimization related to each wave of depression while accounting for prior levels of these constructs and face-to-face victimization, and each wave of depression predicted cyber victimization, although the strength of these associations was lower.

Figure 2.

Autoregressive cross-lagged model of cyber victimization and depression. Standard coefficients are displayed. *p < .05.

Relationship Between Cyber Victimization and Anxiety

The reciprocal relationship between cyber victimization and anxiety was investigated using an autoregressive cross-lagged model. Like the previous models, paths were specified from all waves of cyber victimization to all waves of anxiety. The model demonstrated excellent fit, χ² = 123.03, df = 12, CFI = .96, RMSEA = .06, and SRMR = .05. All waves of cyber victimization predicted anxiety over time, after accounting for previous waves of these variables and face-to-face victimization, and the findings from the model also revealed that these constructs were stable over time (see Figure 3). Furthermore, anxiety also predicted cyber victimization over time. However, the strength of the magnitudes was weaker when compared to the prediction of anxiety from cyber victimization.

Figure 3.

Autoregressive cross-lagged model of cyber victimization and anxiety. Standard coefficients are displayed. *p < .05.

Discussion

The available research-linking cyber victimization to suicidal ideation, depression, and anxiety among college students have relied on cross-sectional studies (Kokkinos, Antoniadou, & Markos, 2014; Schenk & Fremouw, 2012; Tennant, Demaray, Coyle, & Malecki, 2015; Wright & Li, 2012, 2013b). Literature on this topic among samples of children and adolescents revealed that cyber victimization predicted increases in suicidal ideation, depression, and anxiety (Bannink, Broeren, van de Looij-Jasen, de Waart, & Raat, 2014; Hay & Meldrum, 2010; Shpiegel, Klomek, & Apter, 2015). Examining suicidal ideation, depression, and anxiety as antecedents and consequences of cyber victimization among college students helps to understand how this topic, typically conceptualized as a problem in primary and secondary schools, affects colleges and universities. The aim of this study was to examine the reciprocal relationship of cyber victimization to suicidal ideation, depression, and anxiety across 4 years among college students, after controlling for face-to-face victimization.

The results of the present study replicated those found in the cross-sectional studies on the linkages between cyber victimization and socioemotional difficulties among college students (Chen & Huyang, 2014; Feinstein, Bhatia, & Davila, 2014; Kokkinos et al., 2014; Landoll, La Greca, & Lai, 2013; Schenk & Fremouw, 2012; Tennant et al., 2015). The findings of the present study extend these results by revealing longitudinal and bidirectional associations of cyber victimization to college students’ socioemotional difficulties (i.e., suicidal ideation, depression, and anxiety). These associations were highly stable over time, such that prior cyber victimization predicted later cyber victimization and socioemotional difficulties over 4 years. Being frequently and consistently subjected to cyber victimization might diminish college students’ abilities to deal effectively with their victimization, potentially leading them to suicidal ideation, depression, and anxiety. The chronic nature of cyber victimization that was experienced by some college students in this study indicates that their declining ability to deal with this experience might increase their risk for further victimization and socioemotional difficulties (Ybarra, Alexander, & Mitchell, 2005).

The magnitudes of the relationships between cyber victimization and depression as well as between cyber victimization and anxiety were stronger than the reverse association (i.e., depression predicting cyber victimization and anxiety predicting cyber victimization). Cyber victimization tends to be “a more persistent and pervasive form of aggression,” which might increasingly worsen its connection to socioemotional difficulties over time among college students (Tennant et al., 2015). In addition, cyber victimization constantly bombards the victim with negative messages, potentially heightening victims’ experiences of these behaviors, which makes it difficult to perceive positive interactions. Consequently, this experience might reduce victims’ self-concept, contributing to greater levels of depression and anxiety (Lipton, 2011). Research also revealed that cyber victimization might be just as severe as traditional face-to-face forms of victimization as many times victims are not able to receive sanctuary in their homes, as cyber victimization can follow them just about anywhere that they have access to digital technologies (Kowalski & Limber, 2012).

Recommendations and Limitations

This study was one of the first to examine the longitudinal and reciprocal relationships between cyber victimization and various socioemotional problems, including suicidal ideation, depression, and anxiety, among college students. The four-wave longitudinal design was a strength of this research. Additional research should account for college students’ offending behavior, especially considering that cyberbullying perpetration is highly correlated with cyber victimization (Wright & Li, 2013a). This study relied on self-report for face-to-face and cyber victimization, suicidal ideation, depression, and anxiety, and follow-up research should include multiinformants such as friends and parents. Furthermore, it might be possible to examine clinical forms of suicidality, depression, and anxiety to better understanding their longitudinal linkages to cyber victimization. The time parameters for the variables examined in this study were different. For example, cyber victimization was assessed over 60 days, while suicidal ideation was examined over 1 year. Follow-up research should aim to consider these variables within the same time frame to better understanding the associations among the variables examined in this study. Another important consideration is that the participants in this study were from a convenience sample of college students taking psychology courses during their freshman year. Follow-up research should aim to understand cyber victimization among other adults as well.

Suggestions for Colleges and Universities

It is imperative for college and university administrators, faculty, staff, and students to join together to discourage cyber victimization on their campus. Colleges and universities need to have a clear policy regarding cyber victimization. The policy should make it clear that cyber victimization is discouraged and discuss the sanctions at the school level against perpetrators. Furthermore, materials should be distributed during freshmen orientations about the dangers of cyber victimization, how to tell if someone has been victimized online, and include information about resources on campus that students can turn to if they feel like they have been cyber victimized. Another resource should include the counseling center, which should have trained and knowledgeable staff that are able to recognize the dangers of cyber victimization among college students and the negative socioemotional, academic, and behavioral consequences.

The establishment of theoretically based and empirically proven prevention and intervention programs are needed, specifically designed for college students. Many programs have focused on stopping or reducing cyber victimization, but these programs are typically designed for children and adolescents between the ages of 11 and 14 (e.g., Lee, Zi-Pei, Sanstrom, & Dalal, 2013; Williford et al., 2014), with some programs aimed at adolescents until the age of 17 (e.g., Wolfer, Schultze, Zagorscak, Gobel, & Scheithauer, 2014) or among students from 14 to 20 years (e.g., Menesini, Nocentini, & Palladino, 2012). Programs are especially needed at the college level. Such programs should focus on the perpetration of cyberbullying, cyber bystander behavior, and cyber victimization. To date, only one theoretically based and empirically tested prevention program has been published on reducing cyberbullying among college students (Doane, Kelley, & Pearson, 2016).

Conclusions

This study adds to a growing body of literature on cyber victimization and the associated socioemotional difficulties among college students. Consistent with the cross-sectional studies on the socioemotional outcomes related to cyber victimization, this study found reciprocal relationships between cyber victimization, suicidal ideation, depression, and anxiety over 4 years. Further replications of this research should be undertaken, particularly to understand the magnitude differences found in this study. Too long has the mentality been that cyber victimization is a problem associated with childhood and adolescence. This mentality stunts the growth of research in the area of cyber victimization on college campuses, which hinders our ability to effectively deal with these behaviors. More attention should be given to addressing cyber victimization on college campuses in an effort to raise awareness of these behaviors. Awareness is instrumental to combating cyber victimization on these campuses.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Note

References

Alvarez-Garcia

Nunez Perez

J. C.

Gonzalez

A. D.

Perez

C. R.

(2015). Risk factors associated with cyber victimization in adolescence. International Journal of Clinical and Health Psychology, 15, 226–235.

Balakrishnan

(2015). Cyberbullying among young adults in Malaysia: The roles of gender, age and Internet frequency. Computers in Human Behavior, 46, 149–157.

Bannink

Broeren

van de Looij-Jansen

P. M.

de Waart

F. G.

Raat

(2014). Cyber and traditional bullying victimization as a risk factor for mental health problems in adolescents. PLoS One, 9, e94026.

Beck

A. T.

Steer

R. A.

(1993). Beck anxiety inventory manual. San Antonio, TX: Harcourt Brace.

Beck

A. T.

Steer

R. A.

Brown

G. K.

(1996). Manual for the beck inventory-II. San Antonio, TX: Psychological Corporation.

Bollen

K. A.

Curran

P. J.

(2006). Latent curve models: A structural equation perspective. New York, NY: Wiley.

Bossler

A. M.

Holt

T. J.

(2010). The effect of self-control on victimization in the cyberworld. Journal of Crime Justice, 28, 227–236.

Cappadocia

M. C.

Craig

W. M.

Pepler

(2013). Cyberbullying: Prevalence, stability and risk factors during adolescence. Canadian Journal of School Psychology, 28, 171–192.

Center for Disease Control and Prevention [CDC]. (2011). Self-directed violence surveillance: Uniform definitions and recommended data elements. Retrieved from: http://www.cdc.gov/violenceprevention/pdf/Self-Directed-Violence-a.pdf

10.

Copes

Kerley

Huff

Kane

(2010). Differentiating identity theft: An exploratory study of victims using a national victimization survey. Journal of Criminal Justice, 38, 1045–1052.

11.

Cowie

(2013). Cyberbullying and its impact on young people’s emotional health and well-being. Psychiatrist, 37, 167–170.

12.

Crosslin

K. L.

Crosslin

M. B.

(2014). Cyberbullying at a Texas university—A mixed-methods approach to examining online aggression. Texas Public Health Journal, 66, 26–31.

13.

Doane

A. N.

Kelley

M. L.

Pearson

M. R.

(2016). Reducing cyberbullying: A theory of reasoned action based video prevention program for college students. Aggressive Behavior, 42(2), 136–146. doi:10.1002/ab.21610

14.

Feinstein

B. A.

Bhatia

Davila

(2014). Rumination mediates the association between cyber-victimization and depressive symptoms. Journal of Interpersonal Violence, 29, 1732–1746.

15.

Ferdon

C. D.

Hertz

M. F.

(2007). Electronic media, violence, and adolescents. An emerging public health problem. Journal of Adolescent Health, 41, 1–5.

16.

Finn

(2004). A survey of online harassment at a university campus. Journal of Interpersonal Violence, 19, 468–483.

17.

Fischer

E. H.

Dornelas

E. A.

Goethe

J. W.

(2001). Characteristics of people lost to attrition in psychiatric follow-up studies. Journal of Nervous & Mental Disease, 189, 49–55.

18.

Grigg

D. W.

(2010). Cyber-aggression: Definition and concept of cyberbullying. Australian Journal of Guidance and Counseling, 20, 143–156.

19.

Hancock

G. R.

Mueller

R. O.

(Eds.) (2006). Structural equation modeling: A second course. Charlotte, NC: Information Age Publishing.

20.

Hay

Meldrum

(2010). Bullying victimization and adolescent self-harm: Testing hypotheses from General Strain Theory. Journal of Youth & Adolescence, 39, 446–459.

21.

Hemphill

S. A.

Heerde

J. A.

(2014). Adolescent predictors of young adult cyberbullying perpetration and victimization among Australian youth. Journal of Adolescent Health, 55, 580–587.

22.

Hinduja

Patchin

J. W.

(2010). Bullying, cyberbullying, and suicide. Archives of Suicide Research, 14, 206–221.

23.

Holt

T. J.

Fitzgerald

Bossler

A. M.

Chee

(2016). Assessing the risk factors of cyber and mobile phone bullying victimization in a nationally representative sample of Singapore youth. International Journal of Offender Therapy and Comparative Criminology, 60(5), 598–615. doi:10.1177/0306624X14554852

24.

Kokkinos

C. M.

Antoniadou

Markos

(2014). Cyber-bullying: An investigation of the psychological profile of university student participants. Journal of Applied Developmental Psychology, 35, 204–214.

25.

Kowalski

R. M.

Limber

S. P.

(2007). Electronic bullying among middle school students. Journal of Adolescent Health, 41, 22–30.

26.

Kowalski

R. M.

Limber

S. P.

(2012). Psychological, physical, and academic correlates of cyberbullying and traditional bullying. Journal of Adolescent Health, 53, 513–520.

27.

Landoll

R. R.

La Greca

A. M.

Lai

B. S.

(2013). Aversive peer experiences on social networking sites: Development of the Social Networking-Peer Experiences Questionnaires (SN-PEQ). Journal of Research on Adolescence, 23, 695–705.

28.

Lee

Zi-Pei

Svanstrom

Dalal

(2013). Cyber bullying prevention: Intervention in Taiwan. PLoS One, 8, e64031.

29.

Lipton

J. D.

(2011). Combating cyber-victimization. Berkeley Technology Law Journal, 26, 1104–1126.

30.

Little

R. J. A.

Rubin

D. B.

(1989). The analysis of social science data with missing values. Sociological Methods and Research, 18, 292–326.

31.

Menesini

Nocentini

Palladino

B. E.

(2012). Empowering students against bullying and cyberbullying: Evaluation of an Italian peer-led model. International Journal of Conflict and Violence, 6, 314–321.

32.

Mishna

Khoury-Kassabri

Gadalla

Daciuk

(2012). Risk factors for involvement in cyber bullying: Victims, bullies and bully-victims. Children and Youth Services Review, 34, 63–70.

33.

Ngo

F. T.

Paternoster

(2011). Cybercrime victimization: An examination of individual and situational level factors. International Journal of Cyber Criminology, 5, 773–793.

34.

Perrin

Duggan

(2015). Americans’ internet access: 2000-2015. Retrieved from http://www.pewinternet.org/2015/06/26/americans-internet-access-2000-2015/

35.

Rideout

V. J.

Roberts

D. F.

Foehr

U. G.

(2005). Generation M: Media in the lives of 8-18-year-olds: Executive summary. Menlo Park, CA: Henry J. Kaiser Family Foundation.

36.

Rubin

K. H.

Coplan

(2010). The development of shyness and social withdrawal. New York, NY: Guilford Press.

37.

Schenk

A. M.

Fremouw

W. J.

(2012). Prevalence, psychological impact, and coping of cyberbully victims among college students. Journal of School Violence, 11, 21–37.

38.

Shpiegel

Klomek

A. B.

Apter

(2015). Bullying, cyberbullying, depression and suicide ideation among youth. Comparing online to paper-and-pencil questionnaires. International Journal of Child and Adolescent Health, 8, 161–167.

39.

Smith

Rainie

Zickuhr

(2011). College students and technology. Retrieved from http://www.pewinternet.org/2011/07/19/college-students-and-technology/

40.

Smith

P. K.

Mahdavi

Carvalho

Fisher

Russell

Tippett

(2008). Cyberbullying: Its nature and impact in secondary school pupils. Journal of Child Psychology and Psychiatry, 49, 376–385.

41.

Tennant

J. E.

Demaray

M. K.

Coyle

Malecki

C. K.

(2015). The dangers of the webs: Cybervictimization, depression, and social support in college students. Computers in Human Behavior, 50, 348–357.

42.

Topcu

Erdur-Baker

Capa

A. Y.

(2008). Examination of cyber-bullying experiences among Turkish students from different school types. CyberPsychology & Behavior, 11, 644–648.

43.

Walker

Sockman

Koehn

(2011). An exploratory study of cyberbullying with undergraduate university students. Tech Trends, 55, 31–38.

44.

Weigman

Schie

(1998). Video game playing and its relations with aggressive and prosocial behaviors. British Journal of Social Psychology, 37, 367–378.

45.

Williford

Elledge

L. C.

Boulton

A. J.

DePaolis

K. J.

Little

T. D.

Salmivalli

(2014). Effects of the KiVa antibullying program on cyberbullying and cybervictimization frequency among Finnish youth. Journal of Clinical Child & Adolescent Psychology, 42, 820–833.

46.

Wolak

Mitchell

Finkelhor

(2007). Does online harassment constitute bullying? An exploration of online harassment by known peers and online-only contacts. Journal of Adolescent Health, 41, 51–58.

47.

Wolfer

Schultze

Zagorscak

Gobel

Scheithauer

(2014). Prevention 2.0: Targeting cyberbullying@School. Prevention Science, 15, 879–887.

48.

Wright

M. F.

(2015). Cyber victimization and perceived stress: Linkages to late adolescents’ cyber aggression and psychological functioning. Youth & Society, 47(6), 789–810. doi:10.1177/0044118X14537088

49.

Wright

M. F.

(2012). Kicking the digital dog: A longitudinal investigation of young adults’ victimization and cyber-displaced aggression. CyberPsychology, Behavior, & Social Networking, 15, 44–454.

50.

Wright

M. F.

(2013a). The association between cyber victimization and subsequent cyber aggression: The moderating effect of peer rejection. Journal of Youth & Adolescence, 42, 662–674.

51.

Wright

M. F.

(2013b). Normative beliefs about aggression and cyber aggression among young adults: A longitudinal investigation. Aggressive Behavior, 39, 161–170.

52.

Wright

M. F.

Shi

(2014). Chinese adolescents’ social status goals: Associations with behaviors and attributions for relational aggression. Youth & Society, 46, 566–588.

53.

Ybarra

M. L.

Alexander

Mitchell

K. J.

(2005). Depressive symptomology, youth internet use, and online interactions: A national survey. Journal of Adolescent Health, 36, 9–18.

54.

Ybarra

M. L.

Diener-West

Leaf

P. J.

(2007). Examining the overlap in internet harassment and school bullying: Implications for school intervention. Journal of Adolescent Health, 41, S42–S50.

55.

Ybarra

M. L.

Mitchell

K. J.

(2004). Online aggressor/targets, aggressors, and targets: A comparison of associated youth characteristics. Journal of Child Psychology and Psychiatry, 45, 1308–1316.