Cyberstalking Victimization: An Empirical Assessment of RAT Among Female Iranian College Students

Abstract

The role of routine activity theory (RAT) as a guiding theoretical approach to understand online victimization has been well documented. However, the recent emphasis in criminology on its applicability to online victimization has largely been based on evidence from Anglo-American studies. This study fills this gap by testing the predictive utility of RAT for cyberstalking victimization, using data from a sample of female Iranian students. Our structural equation model showed that online exposure to motivated offenders, target suitability, and ineffective online guardianship were positively and significantly associated with cyberstalking victimization. Our results provide strong support for RAT, indicating its generalizability to a different sociopolitical context.

Keywords

routine activity theory cyberstalking victimization technology-enabled crime Iran

Introduction

Cyberspace offers an arena for new forms of crime, widening “the range of offenders and victims and the possible convergences between them” (Felson & Eckert, 2019, p. 146). Growing evidence indicates that the utility of criminological theories for explaining cybercrime and online victimization (Choi et al., in-press; Holt & Bossler, 2014; Reyns et al., 2018). Criminologists have taken an increasing interest in the applicability of the routine activity theory (RAT) to online victimization (Cohen & Felson, 1979; Felson & Eckert, 2019). RAT holds that three components are necessary for a crime event to take place: (1) motivated offenders, (2) suitable targets, and (3) the absence of capable guardianship (McNeeley, 2015; Pratt & Turanovic, 2016).

RAT has successfully explained different types of online victimization, such as cyberstalking (Leukfeldt & Yar, 2016; Marcum et al., 2010; Reyns et al., 2011, 2016), cyber-interpersonal violence (Choi, Cho, et al., 2019; Vakhitova et al., 2016), cyber dating abuse (Melander & Hughes, 2018; van Ouytsel et al., 2018), identity theft (Leukfeldt & Yar, 2016; Reyns & Randa, 2020), cyberbullying (Choi, Earl, et al., 2019; Näsi et al., 2017; Navarro & Jasinski, 2012), Internet fraud (Leukfeldt, 2014; Leukfeldt & Yar, 2016; Pratt et al., 2010; Reyns, 2015), and computer crime (Bossler & Holt, 2009; Choi, 2008; Holt et al., 2020; Leukfeldt & Yar, 2016). This research shows much promise for RAT to explain a range of online victimization. Nonetheless, several important questions remain unclear regarding the utility of RAT in cyberstalking victimization.

First, empirical evidence on cyberstalking victimization is predominantly based on Anglo-American studies; specifically, the majority of research has been conducted in the United States (e.g., Ngo & Piquero, 2020; Nobles et al., 2014; Reyns et al., 2011), along with some works in European countries (e.g., Dreßing et al., 2014; Leukfeldt & Yar, 2016). Although there is a growing empirical literature supporting the validity of RAT in accounting for online interpersonal violence in Asia (e.g., Choi, Cho, et al., 2019), the majority of this research is conducted using data from countries in a particular region of Asia (i.e., East Asia). Because Asia features a wide array of cultures, religions, political styles, and economies (Suzuki et al., 2018), cyberstalking phenomena that occur in East Asia and those observed in Islamic culture, such as the Islamic Republic of Iran, may differ starkly. Individuals who live in Islamic cultures often belong to a religious community and follow a set of rituals and traditions associated with Islam (Mirsepassi, 2010), while those from countries in East Asia view religious pluralism as the norm (Yun & Lee, 2017). The differences surrounding religious beliefs and behaviors can result in distinctive differences in patterns of cyberstalking perpetration and victimization between East Asia and the Middle East in terms of their effects on individuals’ routine activities in these societies (Felson & Eckert, 2019). These cultural and religious differences make Iran a valuable setting for examining cyberstalking victimization.

Living in a religious society influences one’s cognitive domain (e.g., self-control and self-regulation) in various ways (Mohammadkhani et al., 2015). School and family, the main agents of socialization, produce internalized religious norms and dictate individuals’ choices of lifestyle and behavior. A religious society also creates the conditions under which deviant members experience greater shame and lose social dignity. Additionally, religion can affect one’s moral behaviors by presenting a religious frame of reference. For example, one of the most important teachings of Islam is respecting other’s privacy and protecting their dignity and reputation. A religious society can also promote self-monitoring. This can occur through a range of mechanisms, such as interacting with individuals in a religious community, using strong religious norms to evaluate behavior, participating in self-regulatory religious practices, participating in a religious community, and performing religious practices, such as prayer (Mohammadkhani et al., 2015). Individuals who grow up in religious families and under the influence of religious teachings may be less likely to engage in behaviors that may invade others’ privacy or defame them. Simultaneously, the internalization of moral and religious codes leads to the formation of a self-regulatory approach in the individual and determines his/her lifestyle.

The application of RAT to cyberstalking victimization in different sociopolitical contexts, such as the Middle East, remains unknown because virtually no work has been done to investigate it. This is a critical omission not only because of the rapid growth of electronic exposure, which is not limited to Western societies, but also because RAT is proposed to apply generally, beyond any particular social setting (Felson & Eckert, 2019).

Although a group of authors has asserted the utility of RAT for cyberstalking victimization (Reyns et al., 2011, 2016), no independent replication of their findings has appeared. This lack of replication is detrimental to the systematic accumulation of knowledge because we cannot be certain that the original findings are not attributable to error or chance (Pridemore et al., 2018). Although low replication rates have been seen in criminological journals (McNeeley & Warner, 2015), research on online victimization involving new communication technologies is more susceptible to this problem.

The majority of empirical research exploring the relationship between online routines and online victimization has relied on traditional regression (e.g., ordinary least squares or logistic regressions), with few exceptions (refer to Choi, 2008). However, traditional regression is limited in the possible measurement errors it can identify because it assumes that all variables have been measured without error (Kline, 2016). Thus, the elements of RAT are generally measured with sole reliance on observed variables without considering error terms (Bossler & Holt, 2009; Navarro & Jasinski, 2012; Reyns et al., 2011).

This study addresses these three gaps in the literature by testing the predictive utility of RAT in cyberstalking victimization using data from a sample of 387 female college students in Iran. We built on prior research to operationalize the three key theoretical constructs and test them against cyberstalking victimization. We also employed structural equation modeling to test the direct effects of exposure on motivated offenders, target suitability, and ineffective online guardianship while incorporating the second-order latent constructs of these concepts.

Literature Review

Cyberstalking Victimization of College Students

Cyberstalking involves the use of electronic or Internet-capable devices, such as mobile phones or laptop computers to repeatedly pursue or harass an individual (Nobles et al., 2014). Although reported estimates regarding the prevalence of cyberstalking victimization vary depending on the methodology, such as the sample or reference period (Reyns et al., 2012), a large proportion of individuals (about 10% to 40%) appear to have experienced cyberstalking victimization. Research has found victimization by cyberstalking to be associated with adverse psychological and behavioral consequences, such as feelings of inner unrest, mistrust toward others, anger/aggression, depression, and social withdrawal (Dreßing et al., 2014; Parsons-Pollard & Moriarty, 2009). These psychosomatic, psychological, and social outcomes seem to be similar to those found in victims of traditional stalking (Tokunaga & Aune, 2017).

Scholars have investigated the methods used by cyberstalkers. Data from the 2006 Supplemental Victimization Survey supplement to the National Crime Victimization Survey indicate that cyberstalkers in the United States typically relied on e-mail (82.5%). However, they were also using instant messaging (35.1%), as well as blogs, and online message boards (12.3%) to send repeated unwanted messages (Baum et al., 2009). Other researchers have noted that cyberstalkers use more sophisticated methods as well, including computer spyware, listening devices and bugs, and digital video cameras (Marcum et al., 2016).

Another important aspect of cyberstalking victimization involves the relationship between cyberstalkers and their victims. Some researchers have focused exclusively on cyberstalking committed by romantic partners, who may use technology to monitor and track their partners, in what is called cyber dating abuse (Burke et al., 2011; Marcum et al., 2017; Marcum et al., 2016). A study using data from a sample of German social network users revealed that non-strangers—such as ex-partners (29.3%), or acquaintances and friends (28.5%)—are often the cyberstalkers; however, a large share of cyberstalking does involve perpetrators who are unknown to the victim (30.1%; Dreßing et al., 2014).

Empirical studies have reported that young females are the most vulnerable targets of cyberstalking victimization (Dreßing et al., 2014; Kraft & Wang, 2010). In this line of inquiry, it can be seen that cyberstalking is prevalent among students, especially among female students (Reyns et al., 2012; Spitzberg & Cupach, 2007). A recent survey by the Pew Research Center hints at why young adults may be more at risk of cyberstalking by demonstrating that young adults between the ages of 18 and 29 years use online platforms or messaging apps such as Snapchat or Instagram more frequently than older individuals (Perrin & Anderson, 2019). Due to the popularity of mobile messaging apps among young adults and their frequent involvement in online discussion forums (Duggan, 2015), it is likely that high levels of electronic exposure form a condition for the heightened risks of cyberstalking victimization among this population.

Theoretical Framework

Routine Activity Theory

Cohen and Felson (1979) presented RAT as a means of understanding trends in crime rates in the post-World War II United States. Specifically, they sought to explain the reason for the significant increase seen in direct-contact predatory crime in the United States in that period. They contended that trends in human activity patterns could account for changes in crime trends, and they tested the relationship between macro-level measures of routine activities (with an index of labor force participation) and the rates of five different crimes in the United States between 1947 and 1974. Their time-series analysis showed that the measures of routine activities were significantly associated with the five crime rates. Specifically, routine activities that led people to be away from those they trust and the property they value and that brought together people of different backgrounds increased the convergence of likely offenders and targets. Since its first formulation, RAT has remained as one of the most commonly used criminological theories, tested against a variety of offenses (Felson & Eckert, 2019; McNeeley, 2015; Spano & Freilich, 2009).

A current presentation of RAT can be found in the most recent edition of the book Crime and Everyday Life (Felson & Eckert, 2019). Here, we present the idea of RAT. RAT holds that criminal acts have three elements: (1) likely offenders, (2) suitable targets, and (3) the absence of capable guardians against the offense. Likely offenders may not be individuals with significant criminal tendencies. Instead, RAT proposes that anyone with many opportunities for crime can become a potential offender. In other words, RAT posits that any individual may be a motivated offender, so the presence of likely offenders is constant at any place or time. For this reason, Cohen and Felson (1978) devoted little discussion to this element in their original paper. Suitable targets are “any person or thing that draws the offender toward a crime” (Felson & Eckert, 2019, p. 28). Finally, capable guardians are those who can prevent criminal violations. They may not be police or security guards; this category includes family, friends, and other members of the public. Most importantly, potential targets can themselves be guardians. We also note that capable guardians can be objects as well, depending on the situation. For example, security cameras, home alarm systems, and guard dogs can be capable guardians under certain circumstances (McNeeley, 2015).

Notably, RAT focuses on crime events, not criminality, the focus of numerous criminological theories (Felson & Eckert, 2019). Although RAT was put forth as a macro-level explanation for crime, scholars have drawn on it as a micro-level explanation of victimization and supported its empirical validity (Felson & Eckert, 2019; McNeeley, 2015; Spano & Freilich, 2009). Individual-level measures of routine activities have been constructed using a series of everyday, routine behaviors that can potentially increase the exposure to motivated offenders (McNeeley, 2015; Pratt & Turanovic, 2016). For example, the amount of time spent outside of the house or times where one is out for the evening has been measured to capture individuals’ routine activities associated with interpersonal victimization (e.g., Averdijk, 2011; Felson & Eckert, 2019). A growing body of empirical literature has documented the utility of RAT for describing online victimization (Choi, Cho, et al., 2019; Holt et al., 2020; Ngo et al., 2020; Pratt et al., 2010; Reyns, 2015; Reyns et al., 2016). These researchers have measured routine online activities at the individual level. For example, the amount of time spent online or whether individuals have made a purchase online can be tracked and be used to represent routine online activities (Ngo et al., 2020; Pratt et al., 2010). Supporting the major tenet of RAT—that opportunity is the deciding factor—scholars have linked online victimization to online exposure to motivated offenders, target suitability, and ineffective online guardianship.

Several studies have been conducted to test the applicability of RAT to cyberstalking victimization—the specific focus of this article. In the first of these, Reyns et al. (2011) used a random sample of 974 college students to examine the effects of online exposure (e.g., the number of social network accounts), online guardianship (e.g., use of an online profile tracker), and online target attractiveness (e.g., gender) on cyberstalking victimization. Their study showed that not all variables derived from RAT were associated with cyberstalking victimization in an expected way. For instance, aggregate measures of target attractiveness were not significantly associated with any of cyberstalking victimization variables. Also, the use of a profile tracker, one of the indicators mirroring online guardianship, was significantly associated with increased odds of victimization. In a second study, Leukfeldt and Yar (2016) used data from Dutch citizens (n = 9,161) and performed multivariate analysis to test the generalizability of RAT to cyberstalking victimization. Their study showed that individuals performing online activities (e.g., using e-mail) and using the Internet Explorer browser were more likely to experience cyberstalking.

Lastly, Reyns et al. (2016) used data from another college sample (n = 850) to test whether offline guardianship (e.g., living with parents), online guardianship (e.g., deviant peers), and online target hardening (e.g., adding strangers as friends to social network accounts) are associated with cyberstalking victimization. Their investigation indicated that online guardianship (e.g., deviant peers) and target hardening (i.e., adding strangers as friends to social network accounts) were positively and significantly associated with cyberstalking, as predicted. However, offline guardianship, especially living with parents, increased the odds of cyberstalking victimization rather than decreasing them.

Although these studies are essential for documenting the applicability of RAT to cyberstalking victimization, some measures used were far from ideal for capturing the three necessary elements. For instance, when measuring guardianship, many studies of cyberstalking victimization have failed to consider key aspects of guardianship, such as user skill in safety and security regarding social media. Additionally, virtually no researchers have studied cyberstalking victimization using structural equation modeling, which permits the construction of latent variables to represent the three elements of a crime event. RAT theorists posit that individuals’ routines and lifestyles can be measured in a way that reflects the three necessary elements for a crime event to occur: likely offenders, suitable targets, and the absence of capable guardians (Felson & Eckert, 2019; Pratt & Turanovic, 2016). In other words, individuals’ routine behaviors should be understood with reference to these concepts. If these arguments hold true, these three components should be considered latent constructs that subsume the relevant indicators of routine activities. Structural equation modeling is a meaningful methodological approach for addressing these considerations because it enables researchers to create latent variables and test the empirical validity of the variables. Finally, additional replication studies using diverse samples are needed to ensure the validity of previous findings (McNeeley & Warner, 2015; Pridemore et al., 2018).

Although cyberstalking has not been empirically examined in an Iranian context before this study, some studies suggest that online interpersonal violence is becoming more widespread in Iran and is deserving of empirical scrutiny (Jaghoory et al., 2015; Kabiri et al., 2020; Shadmanfaat et al., 2020). For instance, Jaghoory et al. (2015) compared the prevalence of cyberbullying between Iranian adolescents and Finnish adolescents using the same instrument. Iranian adolescents were found to have higher scores for cyberbullying than Finnish adolescents on all measures (e.g., “another sent nasty SMS messages to you” or “another put up nasty pictures of you on the Internet”), both as perpetrators and victims. Other scholars who have studied technology-enabled crime using Iranian samples have also supported the finding that online interpersonal violence is prevalent in Iran and have called for more research on technology-enabled crime with Iranian samples (Kabiri et al., 2020; Shadmanfaat et al., 2020).

Current Study

This study extends prior research in three important ways. First, we used an Iranian sample to test the applicability of RAT to cyberstalking victimization. Because most previous research has relied on Anglo-American samples, it is important to examine the external validity of different samples and populations to replicate relationships. Second, we utilized original data obtained from a questionnaire that we created based on previous research. Collecting original data permits the inclusion of reliable and valid measures that have been used in previous research. Third, we applied a structural equation model to create latent constructs to represent exposure to motivated offenders, target suitability, and ineffective online guardianship. This analysis allowed us to consider both observed variables and error terms, which are often neglected in traditional regression (Kline, 2016).

Following previous research, we hypothesized that RAT is useful for understanding cyberstalking victimization among female Iranian students. The study hypotheses are as follows:

H₁: Exposure to motivated offenders is positively and significantly associated with cyberstalking victimization.

H₂: Target suitability is positively and significantly associated with cyberstalking victimization.

H₃: Ineffective online guardianship is positively and significantly associated with cyberstalking victimization.

Methods

The current study was intended to examine cyberstalking victimization based on RAT. Our initial goal was to use random sampling of the entire female student body of a large urban university in Iran, following previous findings that have shown that a much higher percentage of females tend to be cyberstalking victims relative to males (Dreßing et al., 2014; Reyns et al., 2012). However, we learned that the list of those who were sophomores or above maintained by the university was incomplete because many mobile phone numbers and e-mail addresses were missing. Thus, we relied on a sampling frame of only first-year female college students because we could obtain a complete list of them and their accurate contact information. The target population of this study was 5,665 first-year female college students. We randomly selected 410 of them from the university-provided list, and these students were invited to the university conference hall by messages sent via phone and e-mail. We collected original data consisting of valid responses from 387 female students who were enrolled in the fall 2018 semester by administering the survey at the end of that semester, yielding a response rate of 94.39%. Following approved institutional review board requirements, we explained the purpose of the study. Once the participating students provided consent to participate in the study, self-report questionnaires were distributed. Students were offered to choose either a gift card worth 50,000 Iranian tomans (approximately $1.20 in US currency) or a 2 GB flash drive in return for their participation in the survey.

The questionnaire was distributed in Persian. However, because we derived many survey items from previous research written in English, we used a professional translation agency to translate these items into Persian. The validity of the translated questionnaire was also reviewed and approved by three English department professors at the university. Table 1 presents descriptive statistics relating to age and marital status among participants, showing that 25.3% (98 respondents) of the respondents were 18 years old, 34.6% (134 respondents) were 19 years old, 16.8% (65 respondents) were 20 years old, 11.9% (46 respondents) were 21 years old, 6.5% (25 respondents) were 22 years old, and 4.9% (19 respondents) were 23 years old. Finally, 89.1% of the participants lived with both of their parents, and 10.9% of participants lived with one of their parents (single-parent family).

Table 1.

Sample Demographics.

Variable		Frequency	%
Age
	18 years old	98	25.3
	19 years old	134	34.6
	20 years old	65	16.8
	21 years old	46	11.9
	22 years old	25	6.5
	23 years old	19	4.9
Marital status
	Single	306	79.1
	Married	81	20.9

Dependent Variable

Cyberstalking victimization. To measure cyberstalking victimization, we used a modified version of the scales developed by Reyns et al. (2012) and Pereira and Matos (2016). Our scale included six items, and respondents were asked the following questions with regard to the previous month: (1) How many times have different people contacted you or attempted to contact you online on more than one occasion after you asked them to stop? (2) How many times have different people made unwanted sexual advances online toward you on more than one occasion after you asked them to stop? (3) How many times have different people harassed or annoyed you online on more than one occasion after you asked them to stop? (4) How many times have different people spoken to you online in a violent manner or threatened to physically harm you on more than one occasion after you asked them to stop? (5) How many times have different people sent exaggerated messages of affection to you on more than one occasion after you asked them to stop? (6) How many times have different people sent excessively needy, disclosive, or demanding messages to you on more than one occasion after you asked to stop? Responses ranged from 0 (nobody) to 4 (10 or more).

Independent Variables (RAT Constructs)

Exposure to motivated offenders. Exposure to motivated offenders was a second-order latent construct created from five measures: (1) risky online sexual behavior, (2) communication with strangers, (3) risky vocational activities, (4) online harassment, and (5) disclosure of personal information.

Risky sexual behavior online was measured using a scale adapted from Baumgartner et al. (2010). Participants were asked how many times they had participated in each of the following activities over the previous month: (1) searching for someone on the Internet to talk to about sex, (2) searching for someone on the Internet to have sex with, and (3) sending a photo or video using the Internet in which they were partly naked to someone they knew only online. The options ranged from 1 (never) to 5 (daily or almost daily).

Communication with strangers was assessed with a five-item self-report scale derived from Notten and Nikken (2016). Female college students’ tendency to communicate with strangers on social networking sites was measured by asking respondents how many times in the previous month they (1) had looked for new friends on the Internet, (2) had sent personal information (e.g., their full names, addresses, or phone numbers) to someone whom they had never met face to face, (3) had added people to their friend lists or address books whom they had never met face to face, (4) had sent photos or videos of themselves to someone whom they had never met face to face, and (5) had pretended to be different kinds of people than they really were. The responses ranged from 1 (never) to 5 (daily or almost daily).

Risky vocational activities were measured using a 3-item scale influenced by an instrument developed by Choi (2008). The respondents were asked to answer how many times in the previous month they (1) had opened any file or attachment from an unknown sender, (2) had clicked on a pop-up message that interested them (click or open links), and (3) had visited pages that were new to them. Each item featured a 5-point Likert-type response set, ranging from 1 (never) to 5 (daily or almost daily).

We borrowed the items for online harassment from Ybarra, Espelage, et al. (2007). Our scale consisted of three items asking how many times in the previous month respondents (1) had made a rude or nasty comment, (2) had harassed or embarrassed someone, and (3) had spread rumors about someone, whether they were true or not. Response categories to these three questions ranged from 1 (never) to 5 (daily or almost daily).

Female college students’ disclosure of personal information was measured using five items designed by Ybarra, Mitchell, et al. (2007) to assess the privacy control of Internet users. This measure consisted of items asking how many times in the previous month they had posted the following personal information in a public place online: their (1) real name, (2) telephone number, (3) school name, (4) age or birth year, or (5) picture. All items were reported on a 5-point Likert-type scale anchored at 1 (never) to 5 (daily or almost daily).

Target suitability. Target suitability was another second-order latent construct that was assessed with two measures: (1) users’ social media engagement and (2) time spent online on social networks.

Social media engagement was measured with a 4-item self-report scale developed by Navarro and Jasinski (2012). Participants were asked to indicate their social media engagement on a 5-point Likert-type scale ranging from 1 (never) to 5 (daily or almost daily) using questions asking how many times in the previous month they had done the following: (1) posted comments to something that a friend posted, (2) tagged people in posts, photos, or videos, (3) liked posts, photos, or videos of others, and (4) posted photos or videos.

Time spent online on social networks was measured with a single item adapted from Navarro and Jasinski (2012). Students were asked to indicate how often they went online during the previous month. The responses ranged from 1 (every few weeks) to 5 (several times a day).

Ineffective online guardianship. Ineffective online guardianship was the last second-order latent construct, created from the following three first-order latent constructs: poor social, personal, and parental guardianship.

Prior studies have shown that ineffective social guardianship increases the likelihood of online victimization (Bossler & Holt, 2009; Reyns et al., 2011). We used four items to capture poor social guardianship. Specifically, respondents were asked how likely it was that their friends would use the information that they have posted on social media networks to (1) persistently harass or annoy them, (2) speak forcefully to them, (3) threaten them, or (4) make unwanted sexual advances toward them. All items were rated on a 5-point Likert-type scale ranging from 1 (very unlikely) to 5 (very likely).

Poor personal guardianship was measured using a scale developed by Musharraf et al. (2018) that consists of eight items that are responded to on a 5-point Likert scale ranging from 1 (completely agree) to 5 (completely disagree) to capture users’ safety and security on social media. Students were asked to rate their agreement with the following items: (1) I can easily report a fake account pretending to be me, (2) I can easily hide any post that someone shared/tagged on my profile on the social networking sites that I mostly use, (3) I can easily block or restrict anyone on the social networking sites that I mostly use, (4) I can easily unfriend anyone on the social networking sites that I mostly use, (5) I can easily report any ID, post, image, or video as abusive/spam content on the social networking sites that I mostly use, (6) I can easily control the privacy settings of the social networking sites that I mostly use, (7) I can easily handle spam that is posted on my wall on the social networking sites that I mostly use, and (8) I can easily block anyone on the social networking sites that I mostly use.

To measure inadequate parental behavioral control of students’ social media activity, we revised and used the scales developed by Khurana et al. (2015) and Doty et al. (2018). Poor parental guardianship was designed to evaluate ineffective parental monitoring to protect students against potential cyber risks. Participants were asked how often in the previous month their parents had done the following: (1) restricted the amount of time they spent online, (2) monitored or tracked what they were doing online, such as tracking their Instagram pages or checking their search histories, (3) forbidden certain websites/pages that they might use, and (4) talked with them about right and wrong behaviors using technology. The response options ranged from 1 (almost always) to 5 (never).

Control variables. Several background variables were also included in the analyses: age and family structure. Age was a continuous variable. Living with both parents was coded as 0 and living with one of the parents (single-parent family) was coded as 1.

Analytic Strategy

The analyses proceeded along with a series of steps. First, we conducted various diagnostic tests to rule out the presence of multicollinearity issues; we also checked the validity and reliability of our measurement instruments. Second, we performed correlation analysis to investigate whether cyberstalking victimization is correlated with the elements of RAT. Third, we conducted ordinary least squares regression and structural equation modeling to assess whether exposure to motivated offenders, target suitability, and online guardianship predict cyberstalking victimization.

Results

Validity and Reliability of Measurement Instruments

Concerning the validity and reliability of measures, Table 2 shows that all scales exhibited high levels of internal consistency (α > .70; critical ratio > .70). We also tested the discriminant validity of our instruments by estimating the average variance (AVE) shared between a construct and its measures. AVE indices were higher than the .50 cutoff recommended by Fornell and Larcker (1981). First-order confirmatory factor analysis (CFA) for these scales indicated that the factor loadings for all items were greater than .50, and several fit indices confirmed that the model fit the data well (Kline, 2016). The normality of data was evaluated using skewness, kurtosis, and Mardia’s coefficient multivariate kurtosis. All variables were within the acceptable range of Mardia’s coefficient multivariate kurtosis, skewness, and kurtosis (skewness > 2; kurtosis > 7) for the sample size greater than 300 (Kim, 2013; Kline, 2016).

Table 2.

Validity and Reliability of Research’s Measurement Instruments.

	Factor Loadings	AVE (MSV)	CR	α	Range of Data	Skewness	Kurtosis	Multivariate Kurtosis^a
	Min–Max	AVE (MSV)	CR	α	Min–Max	Skewness	Kurtosis	Value (Critical Ratio)
Cyberstalking victimization	.631–.888	.501 (.194)	.855	.849	0–15	.246	–1.124
Exposure to motivated offenders				.891
Risky sexual online behavior	.762–.828	.637 (.321)	.840	.839	3–14	.214	–.131
Communication with strangers	.474–.872	.565 (.245)	.862	.853	5–20	.426	–.785
Risky vocational activities	.761–.866	.652 (.296)	.849	.838	3–14	.427	–.344
Online harassment	.750–.891	.671 (.296)	.859	.855	3–15	.280	–1.052
Disclosure of personal information	.781–.835	.640 (.321)	.842	.837	3–14	.071	–.720
Target suitability				.820
Social media engagement	.732–.803	.601 (.128)	.858	.857	4–18	.164	–.552
Time spent online on social networks	–	–	–	–	1–5	.147	–.862
Ineffective online guardianship				.907
Poor social guardianship	.654–.885	.658 (.148)	.884	.875	4–20	.246	–.634
Poor personal guardianship	.771–.866	.701 (.179)	.903	.936	8–39	.234	–.723
Poor parental guardianship	.782–.897	.651 (.179)	.937	.902	4–20	.324	–.809

Note. ^aMardia’s coefficient multivariate kurtosis.

Model Fit Summary: CMIN = 1081.969, df = 815, CMIN/df = 1.327, CFI = .972, IFI = .972, TLI = .969, SRMR = .048, RMSEA = .029, PClose = 1.000.

Table 3 reports the zero-order correlations between the independent and dependent variables. The results indicate moderately strong correlations (r > .27) between RAT constructs and cyberstalking victimization (p < .01).

Table 3.

The Zero-order Correlations Between Independent and Dependent Variables (N = 387).

Cyberstalking victimization

5.27

3.96

–

Risky sexual online behavior

6.90

2.27

.41**

–

Communication with strangers

11.08

3.76

.36**

.45**

–

Risky vocational activities

6.47

2.47

.40**

.46**

.34**

–

Online harassment

7.63

3.29

.36**

.42**

.25**

.40**

–

Disclosure of personal information

8.15

2.73

.37**

.48**

.34**

.35**

.47**

–

Social media engagement

8.70

2.89

.31**

.11*

.09*

.22**

.10*

.08*

–

Time spent online on social networks

2.83

1.23

.32**

.22**

.13**

.20**

.13**

.16**

.45**

–

Poor social guardianship

10.44

3.75

.27**

.09*

.14**

.15**

.14**

.13**

.16**

.06

–

Poor personal guardianship

20.78

7.91

.28**

.19**

.16**

.18**

.17**

.13**

.11*

.10*

.26**

–

Poor parental guardianship

10.60

3.93

.30**

.15**

.14**

.24**

.14**

.12**

.14**

.16**

.29**

.37**

Note. *p < .05. **p < .01.

Table 4 shows the results involving the direct effects of exposure to motivated offenders, target suitability, and online guardianship on cyberstalking victimization. Exposure to motivated offenders (Model 1), target suitability (Model 2), online guardianship (Model 3), and control variables (Model 4) were added to the model, respectively.

Regarding motivated offenders, Model 1 shows that risky sexual online behavior (β = .14; p < .05), communication with strangers (β = .16; p < .01), risky vocational activities (β = .19; p < .01), online harassment (β = .13; p < .05), and disclosure of personal information (β = .12; p < .05) were significantly and positively associated with cyberstalking victimization. Model 2 reveals the significant effects of target suitability variables on cyberstalking victimization; cyberstalking victimization was significantly and positively related to social media engagement (β = .17; p < .01) and time spent online on social networks (β = .13; p < .01). With regard to ineffective online guardianship, Model 3 shows that poor social guardianship (β = .12; p < .01), poor personal guardianship (β = .09, p < .05), and poor parental guardianship (β = .10; p < .05) directly affect cyberstalking victimization. Model 4 indicates that respondents’ age and family structure have no significant effect on cyberstalking victimization (p > .05).

Table 4.

Ordinary Least Squares Regression Predicting Cyberstalking Victimization (N = 387).

	Exposure to Motivated Offenders	Target Suitability	Online Guardianship	Control Variables	R-squared
	Model 1	Model 2	Model 3	Model 4
	b (β)	b (β)	b (β)	b (β)
Exposure to motivated offenders
Risky, sexual online behavior	.24* (.14)	.21* (.12)	.20* (.12)	.20* (.12)
Communication with strangers	17** (.16)	16** (.15)	14** (.13)	14** (.13)
Risky vocational activities	31** (.19)	23** (.14)	18* (.11)	18* (.11)
Online harassment	15* (.13)	15* (.12)	12* (.10)	12* (.10)
Disclosure of personal information	18* (.12)	17* (.12)	16* (.11)	16* (.11)
R-squared of model					.28
Target suitability
Social media activities		.23** (.17)	.20** (.14)	.20** (.14)
Time spent online on social networks		.42** (.13)	.40** (.12)	.40** (.12)
R-squared of model					.34
Ineffective online guardianship
Poor social guardianship			.12** (.12)	.12** (.12)
Poor personal guardianship			.05* (.09)	.05* (.09)
Poor parental guardianship			.10* (.10)	.10* (.10)
R-squared of model					.38
Control variables
Age				–.01 (.01)
Family structure (1 = Single-parent family)				–.03 (.01)
R-squared of model					.38

Note. *p < .05. **p < .01.

Table 5 (Figure 1) provides the findings from the structural model predicting cyberstalking victimization based on RAT. The model accounted for 48% of the variance in cyberstalking victimization. For the fitted model, the summary statistics of cyberstalking victimization (i.e., CMIN/df (1.205), CFI (.991), IFI (.991), TLI (.987), SRMR (.034), and RMSEA (.023) are better than critical values, and they represent the goodness-of-fit for the proposed model. The best predictors of cyberstalking victimization are exposure to motivated offenders (β = .40; p < .01), ineffective online guardianship (β = .25; p < .01), and target suitability (β = .24; p < .01) respectively.

Table 5.

Structural Equation Model of Cyberstalking Victimization (N = 387).

	b	β	R-squared
Exposure to motivated offenders ↓ Cyberstalking victimization	.99 **	.40
Target suitability ↓ Cyberstalking victimization	1.145 **	.24
Ineffective online guardianship ↓ Cyberstalking victimization	.22**	.25
R-squared of model			.479

Note. **p < .01.

Figure 1.

Note. Model Fit Summary: CMIN = 45.803, df = 38, CMIN/df = 1.205, CFI = .991, IFI = .991, TLI = .987, SRMR = .034, RMSEA = .023, PClose = .985. **p < .01.

Discussion

RAT maintains that the convergence of motivated offenders, suitable targets, and the absence of capable guardianship increase the likelihood of cyber victimization, proposing that opportunity is a key factor (Cohen & Felson, 1979; Felson & Eckert, 2019). The advent of new information technology provides new opportunities for motivated offenders to find potential victims. A growing body of empirical literature has documented the utility of RAT in accounting for online victimization (Choi, Cho, et al., 2019; Holt et al., 2020; Ngo et al., 2020; Pratt et al., 2010; Reyns et al., 2018). However, little research has been done on the extent of cyberstalking victimization in different countries. Similarly, it remains unclear whether the generalizability of RAT replicated in the United States could be supported by empirical research using international samples. Considering that the use of new communication technologies and victimization spans across the entire world, this lack of comparative research on cyberstalking victimization is problematic. This study sought to address these gaps in the literature using a sample of Iranian students. Specifically, we used original data consisting of 387 female college students in Iran, which is an understudied population, to examine risk factors for cyberstalking victimization based on RAT.

The results provided strong support for the external validity of RAT in the Iranian context, showing that the presence of motivated offenders, target suitability, and ineffective online guardianship can account for 48% of the variance in female college students’ cyberstalking victimization. Our empirical findings supported all three theoretical hypotheses that were tested. First, we hypothesized that greater online exposure to motivated offenders would result in increased risks for cyberstalking victimization. This hypothesis was supported. All measures representing exposure to motivated offenders were positively and significantly associated with cyberstalking victimization. Specifically, increased levels of online, risky sexual behavior, online harassment, disclosure of personal information, risky vocational activity, and communication with strangers online corresponded with increased risks for cyberstalking victimization. These findings are consistent with the findings from previous research on cyberstalking victimization (Reyns et al., 2011, 2016). Notably, our structural equation model revealed that the latent variable, representing exposure to motivated offenders and comprised of five, first-order latent constructs (i.e., online risky sexual behavior, online harassment, disclosure of personal information, risky vocational activity, and communication), was significantly associated with the likelihood of cyberstalking victimization in the expected direction.

Second, we hypothesized that target suitability is positively and significantly associated with cyberstalking victimization. Again, our second hypothesis was confirmed, showing that female college students with greater online target suitability had a heightened likelihood of cyberstalking victimization. Specifically, increased social media engagements and increased time spent online on social networks were related to higher risks of cyberstalking victimization. Female college students with more social activities—such as those who produced more comments, posts, and tags—were more likely to experience cyberstalking victimization. Also, our structural equation model revealed that the latent variable, representing target suitability and comprised of two first-order latent constructs (i.e., social media engagement and time spent online on social networks), was predictive of cyberstalking victimization.

Finally, ineffective online guardianship was expected to be positively associated with cyberstalking victimization. Relying on the proposition from RAT, we hypothesized that female college students with poor personal, parental, and social guardianship were more likely to experience cyberstalking victimization. Our study found strong support for the third hypothesis. The results demonstrated that the lack of safety and security skills regarding online social networks, such as blocking suspicious users, controlling privacy settings, handling spam, or reporting fake accounts, increased the likelihood of cyberstalking victimization. Additionally, ineffective parental and social guardianship increased the likelihood of cyberstalking victimization. The works of Reyns et al. (2011) and Bossler and Holt (2009) have shown that decreased levels of social guardianship (e.g., having cyberstalker friends) were positively and significantly associated with cyberstalking victimization, and our results support this link. Previous studies have also suggested that the lack of parental control over children’s social media use can increase the likelihood of children’s online victimization (Choi, 2008; Doty et al., 2018; Khurana et al., 2015). Our findings indicate that the risk of cyberstalking victimization increases when students use social networks without their parents’ effective control or guidance. Similarly, the second-order latent construct, representing ineffective online guardianship and consisting of three, first-order latent constructs (i.e., poor personal, parental, and social guardianship), was positively and significantly associated with cyberstalking victimization.

One key contribution of this study was its use of a non-U.S. sample to replicate the findings from previous research exploring the applicability of RAT to online victimization (Bossler & Holt, 2009; Choi, Cho, et al., 2019; Holt et al., 2020; Reyns et al., 2011; Reyns & Randa, 2020). Several scholars have noted that the field of criminology has often failed to produce meaningful replications, questioning the robustness of the findings from the research (McNeeley & Warner, 2015; Pridemore et al., 2018). We sought to assess the applicability of RAT to cyberstalking victimization, as seen in several studies, using different data and analyses and altering some elements to test the generalizability of the original results. We believe that our efforts in this study may contribute to the systematic accumulation of knowledge base regarding cyberstalking victimization by ensuring original findings. Additionally, our investigation of cyberstalking using the understudied population of Iranian college women should be welcomed because this line of work has tended to be confined to Anglo-American samples.

Although this study makes an important contribution to the criminological literature, our findings must be interpreted with several caveats in mind. First, we relied on data from female college students only, which could limit the generalizability of our findings. Nonetheless, it should be noted that we collected data only from female college students because prior research has shown that cyberstalking victims are predominantly young women (Dreßing et al., 2014; Kraft & Wang, 2010; Reyns et al., 2012). Second, the survey was distributed at the end of the fall 2018 semester, which means that the sample had only experienced a few months of life in college. This limited college experience may make them unrepresentative of older college students who have spent longer periods of time in college and may have experienced differential levels of cyberstalking. Third, we obtained data from a single source, raising concerns regarding common method variance. Fourth, measuring sensitive issues, such as cyberstalking victimization, via self-report methods can lead respondents to provide more socially acceptable answers, resulting in a social desirability bias (Remler & van Ryzin, 2015). Finally, although previous research has shown that RAT has its utility in explaining online victimization (Choi, Cho, et al., 2019; Pratt et al., 2010; Reyns, 2015), some factors that were not included in the current study, such as low self-control, could also increase the risk of being victimized online. The absence of established confounds and correlates of cybervictimization (e.g., self-control) could imply that the effects of RAT are inflated. For example, Reyns et al. (2018) revealed that those who have less self-control tend to place themselves in risky situations that heighten the possibility of cyberstalking victimization is high. Therefore, future researchers can consider personal traits such as risk-taking and sensation-seeking in their research on cyberstalking victimization so that we can better estimate the effects of the variables derived from RAT and understand the processes by which individuals with low self-control place themselves in close proximity to cyberstalkers.

Considering that exposure to motivated offenders can increase the likelihood of cyberstalking victimization, policymakers should develop programs to increase users’ awareness of the potential risks of reckless use of online social networks. To foster user awareness of social media security, He (2012) argued that it is critical to update and communicate social media policies regularly because social networking technology evolves quickly. Educational institutions may regularly provide students with educational and training programs to review the issues and policies regarding social media security. These programs may provide proper precautions to mitigate the threats and risks involving cyberstalking victimization and the reporting of such an incident. Our findings also suggest that online guardianship is a key dimension to consider reducing cyberstalking victimization. The majority of Iranian women live with their parents before they are married due to the patriarchal culture and the resulting difficulties in obtaining financial independence from (Sadeghi & Mirhosseini, 2011). Relatedly, many female students at the university commute to the college from their homes. This is common practice in Iran because many Iranian parents prefer their daughters to continue their education in the city or province where they live. Due to the patriarchal norms and gender stereotypes present in Iranian society, females are more strictly monitored by their parents than males. Fear of family dishonor leads Iranian families to be more attentive to their daughters’ behaviors than to their sons’. College-aged daughters are not an exception to this situation; instead, they are often strictly supervised because they are considered to be at the optimal age to marry. In this context, parents closely monitor college-aged daughters’ behaviors, whether online or offline, and this parental monitoring is considered to be acceptable, even if not necessarily desirable. As a result, physical proximity between parents and college-aged daughters may translate to effective guardianship, at least in an Iranian context.

Thus, promoting effective parental monitoring techniques regarding children’s use of social media may prevent cyberstalking victimization. Baldry et al. (2019) proposed that effective parenting strategies to lower the risks of their children’s cybervictimization includes sharing concerns with children, being aware of what is going on, providing relevant knowledge, and setting rules and restrictions.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

ORCID iDs

Jaeyong Choi

Julak Lee

Author Biographies

Saeed Kabiri, PhD, earned his master’s degree in sociology at the University of Guillan (2012) and also earned his PhD degree in social problems of Iran at the University of Mazandaran (2017). He has published several papers about the sociology and criminology of sports. His current research interests involve sports criminology.

Jaeyong Choi, PhD, is an assistant professor in the Department of Security Studies and Criminal Justice at Angelo State University. His research interests include criminological theory, police legitimacy, media, and criminal justice, and fear of crime.

Seyyedeh Masoomeh (Shamila) Shadmanfaat, MA, earned her master’s degree in sociology at the University of Guillan (2016). She has published several papers about sociology and criminology of sport with a focus on gender differences. Her current research interests involve gender sports criminology.

Julak Lee, PhD, is a professor in the Department of Industrial Security at Chung-Ang University. His research interests lie in the area of security management, comparative criminal justice, and policing.

References

Averdijk

(2011). Reciprocal effects of victimization and routine activities. Journal of Quantitative Criminology , 27(2), 125–149.

Baldry

A. C.

, Sorrentino

, & Farrington

D. P.

(2019). Cyberbullying and cybervictimization versus parental supervision, monitoring and control of adolescents’ online activities. Children and Youth Services Review , 96, 302–307.

Baum

, Catalano

, Rand

, & Rose

(2009). Stalking victimization in the United States . US Department of Justice.

Baumgartner

S. E.

, Valkenburg

P. M.

, & Peter

(2010). Assessing causality in the relationship between adolescents’ risky sexual online behavior and their perceptions of this behavior. Journal of Youth and Adolescence , 39(10), 1226–1239.

Bossler

A. M.

, & Holt

T. J.

(2009). On-line activities, guardianship, and malware infection: An examination of routine activities theory. International Journal of Cyber Criminology , 3(1), 400–420.

Burke

S. C.

, Wallen

, Vail-Smith

, & Knox

(2011). Using technology to control intimate partners: An exploratory study of college undergraduates. Computers in Human Behavior , 27(3), 1162–1167.

Choi

K.-S.

(2008). Computer crime victimization and integrated theory: An empirical assessment. International Journal of Cyber Criminology , 2(1), 308–333.

Choi

K.-S.

, Cho

, & Lee

J. R.

(2019). Impacts of online risky behaviors and cybersecurity management on cyberbullying and traditional bullying victimization among Korean youth: Application of cyber-routine activities theory with latent class analysis. Computers in Human Behavior , 100, 1–10.

Choi

K.-S.

, Earl

, Lee

J. R.

, & Cho

(2019). Diagnosis of cyber and non-physical bullying victimization: A lifestyles and routine activities theory approach to constructing effective preventative measures. Computers in Human Behavior , 92, 11–19.

10.

Choi

, Kruis

N. E.

, & Choo

(in press). Explaining fear of identity theft victimization using a routine activity approach. Journal of Contemporary Criminal Justice .

11.

Cohen

L. E.

, & Felson

(1979). Social change and crime rate trends: A routine activity approach. American Sociological Review , 44(4), 588–608. https://doi.org/10.2307/2094589

12.

Doty

J. L.

, Gower

A. L.

, Sieving

R. E.

, Plowman

S. L.

, & McMorris

B. J.

(2018). Cyberbullying victimization and perpetration, connectedness, and monitoring of online activities: Protection from parental figures. Social Sciences , 7(12), 265.

13.

Dreßing

, Bailer

, Anders

, Wagner

, & Gallas

(2014). Cyberstalking in a large sample of social network users: Prevalence, characteristics, and impact upon victims. Cyberpsychology, Behavior, and Social Networking , 17(2), 61–67.

14.

Duggan

(2015). Mobile messaging and social media 2015 . https://www.pewresearch.org/internet/2015/08/19/mobile-messaging-and-social-media-2015/

15.

Felson

, & Eckert

M. A.

(2019). Crime and everyday life (6th ed.). Sage Publications.

16.

Fornell

, & Larcker

D. F.

(1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research , 18(1), 39–50.

17.

(2012). A review of social media security risks and mitigation techniques. Journal of Systems and Information Technology , 14(2), 171–180.

18.

Holt

T. J.

, & Bossler

A. M.

(2014). An assessment of the current state of cybercrime scholarship. Deviant Behavior , 35(1), 20–40.

19.

Holt

T. J.

, Wilsem

J. van.

, van. de Weijer

, & Leukfeldt

(2020). Testing an integrated self-control and routine activities framework to examine malware infection victimization. Social Science Computer Review , 38(2), 187–206.

20.

Jaghoory

, Björkqvist

, & Österman

(2015). Cyberbullying among adolescents: A comparison between Iran and Finland. Journal of Child and Adolescent Behavior , 3(6), 1–7.

21.

Kabiri

, Shadmanfaat

S. M.

, Choi

, & Yun

(2020). The impact of life domains on cyberbullying perpetration in Iran: A partial test of Agnew’s general theory of crime. Journal of Criminal Justice , 66, 101633.

22.

Khurana

, Bleakley

, Jordan

A. B.

, & Romer

(2015). The protective effects of parental monitoring and internet restriction on adolescents’ risk of online harassment. Journal of Youth and Adolescence , 44(5), 1039–1047.

23.

Kim

H.-Y.

(2013). Statistical notes for clinical researchers: Assessing normal distribution (2) using skewness and kurtosis. Restorative dentistry & endodontics , 38(1), 52–54.

24.

Kline

R. B.

(2016). Principles and practice of structural equation modeling (4th ed.). Guilford Publications.

25.

Kraft

, & Wang

(2010). An exploratory study of the cyberbullying and cyberstalking experiences and factors related to victimization of students at a public liberal arts college. International Journal of Technoethics (IJT) , 1(4), 74–91.

26.

Leukfeldt

E. R.

(2014). Phishing for suitable targets in the Netherlands: Routine activity theory and phishing victimization. Cyberpsychology, Behavior, and Social Networking , 17(8), 551–555.

27.

Leukfeldt

E. R.

, & Yar

(2016). Applying routine activity theory to cybercrime: A theoretical and empirical analysis. Deviant Behavior , 37(3), 263–280.

28.

Marcum

C. D.

, Higgins

G. E.

, & Nicholson

(2017). I’m watching you: Cyberstalking behaviors of university students in romantic relationships. American Journal of Criminal Justice , 42(2), 373–388.

29.

Marcum

C. D.

, Higgins

G. E.

, & Poff

(2016). Exploratory investigation on theoretical predictors of the electronic leash. Computers in Human Behavior , 61, 213–218.

30.

Marcum

C. D.

, Ricketts

M. L.

, & Higgins

G. E.

(2010). Assessing sex experiences of online victimization: An examination of adolescent online behaviors using routine activity theory. Criminal Justice Review , 35(4), 412–437.

31.

McNeeley

(2015). Lifestyle-routine activities and crime events. Journal of Contemporary Criminal Justice , 31(1), 30–52.

32.

McNeeley

, & Warner

J. J.

(2015). Replication in criminology: A necessary practice. European Journal of Criminology , 12(5), 581–597.

33.

Melander

, & Hughes

(2018). College partner violence in the digital age: Explaining cyber aggression using routine activities theory. Partner Abuse , 9(2), 158–180.

34.

Mirsepassi

(2010). Democracy in modern Iran: Islam, culture, and political change . New York University Press.

35.

Mohammadkhani

, Yeganeh

, & Karimpour

(2015). Role of religious orientation and self-control in prediction of drug addiction potential. Journal of Health Care , 17(3), 248–259.

36.

Musharraf

, Bauman

, Anis-ul-Haque

, & Malik

J. A.

(2018). Development and validation of ICT self-efficacy scale: Exploring the relationship with cyberbullying and victimization. International Journal of Environmental Research and Public Health , 15(12), 2867.

37.

Näsi

, Räsänen

, Kaakinen

, Keipi

, & Oksanen

(2017). Do routine activities help predict young adults’ online harassment: A multi-nation study. Criminology & Criminal Justice , 17(4), 418–432.

38.

Navarro

J. N.

, & Jasinski

J. L.

(2012). Going cyber: Using routine activities theory to predict cyberbullying experiences. Sociological Spectrum , 32(1), 81–94.

39.

Ngo

F. T.

, & Piquero

A. R.

(2020). Low self-control, harassment perpetration, and stalking victimization among Asian college students. Victims & Offenders , 15(1), 78–102.

40.

Ngo

F. T.

, Piquero

A. R.

, LaPrade

, & Duong

(2020). Victimization in cyberspace: Is it how long we spend online, what we do online, or what we post online? Criminal Justice Review , 45(4), 430–451.

41.

Nobles

M. R.

, Reyns

B. W.

, Fox

K. A.

, & Fisher

B. S.

(2014). Protection against pursuit: A conceptual and empirical comparison of cyberstalking and stalking victimization among a national sample. Justice Quarterly , 31(6), 986–1014.

42.

Notten

, & Nikken

(2016). Boys and girls taking risks online: A gendered perspective on social context and adolescents’ risky online behavior. New Media & Society , 18(6), 966–988.

43.

Ouytsel

J. van.

, Ponnet

, & Walrave

(2018). Cyber dating abuse victimization among secondary school students from a lifestyle-routine activities theory perspective. Journal of Interpersonal Violence , 33(17), 2767–2776.

44.

Parsons-Pollard

, & Moriarty

L. J.

(2009). Cyberstalking: Utilizing what we do know. Victims and Offenders , 4(4), 435–441.

45.

Pereira

, & Matos

M. J.

(2016). Cyber-stalking victimization: What predicts fear among Portuguese adolescents? European Journal on Criminal Policy and Research , 22(2), 253–270.

46.

Perrin

, & Anderson

(2019). Share of U.S. adults using social media, including Facebook, is mostly unchanged since 2018 . https://www.pewresearch.org/fact-tank/2019/04/10/share-of-u-s-adults-using-social-media-including-facebook-is-mostly-unchanged-since-2018/

47.

Pratt

T. C.

, Holtfreter

, & Reisig

M. D.

(2010). Routine online activity and internet fraud targeting: Extending the generality of routine activity theory. Journal of Research in Crime and Delinquency , 47(3), 267–296. https://doi.org/10.1177/0022427810365903

48.

Pratt

T. C.

, & Turanovic

J. J.

(2016). Lifestyle and routine activity theories revisited: The importance of “risk” to the study of victimization. Victims & Offenders , 11(3), 335–354.

49.

Pridemore

W. A.

, Makel

M. C.

, & Plucker

J. A.

(2018). Replication in criminology and the social sciences. Annual Review of Criminology , 1, 19–38.

50.

Remler

D. K.

, & Ryzin

G. G. van

. (2015). Research methods in practice: Strategies for description and causation (2nd ed.). Sage Publications.

51.

Reyns

B. W.

(2015). A routine activity perspective on online victimisation. Journal of Financial Crime , 22(4), 396–411.

52.

Reyns

B. W.

, Fisher

B. S.

, & Randa

(2018). Explaining cyberstalking victimization against college women using a multitheoretical approach: Self-control, opportunity, and control balance. Crime & Delinquency , 64(13), 1742–1764.

53.

Reyns

B. W.

, Henson

, & Fisher

B. S.

(2011). Being pursued online: Applying cyberlifestyle–routine activities theory to cyberstalking victimization. Criminal Justice and Behavior , 38(11), 1149–1169.

54.

Reyns

B. W.

, Henson

, & Fisher

B. S.

(2012). Stalking in the twilight zone: Extent of cyberstalking victimization and offending among college students. Deviant Behavior , 33(1), 1–25.

55.

Reyns

B. W.

, Henson

, & Fisher

B. S.

(2016). Guardians of the cyber galaxy: An empirical and theoretical analysis of the guardianship concept from routine activity theory as it applies to online forms of victimization. Journal of Contemporary Criminal Justice , 32(2), 148–168.

56.

Reyns

B. W.

, & Randa

(2020). No honor among thieves: Personal and peer deviance as explanations of online identity fraud victimization. Security Journal , 33, 228–243.

57.

Sadeghi

, & Mirhosseini

(2011). A Study on Factors Affecting Victimization of Woman Living in Tehran. Women in Development and Politics , 9(1), 35–62.

58.

Shadmanfaat

S. M.

, Choi

, Kabiri

, & Lee

(2020). Impact of social concern on cyberbullying perpetration in Iran: Direct, indirect, mediating, and conditioning effects in Agnew’s social concern theory. Deviant Behavior . https://doi.org/10.1080/01639625.2020.1753152

59.

Spano

, & Freilich

J. D.

(2009). An assessment of the empirical validity and conceptualization of individual level multivariate studies of lifestyle/routine activities theory published from 1995 to 2005. Journal of Criminal Justice , 37(3), 305–314.

60.

Spitzberg

B. H.

, & Cupach

W. R.

(2007). The state of the art of stalking: Taking stock of the emerging literature. Aggression and Violent Behavior , 12(1), 64–86.

61.

Suzuki

, Pai

C.-F.

, & Islam

M. J.

(2018). Systematic quantitative literature review on criminological theories in Asia. Asian Journal of Criminology , 13(2), 129–151.

62.

Tokunaga

R. S.

, & Aune

K. S.

(2017). Cyber-defense: A taxonomy of tactics for managing cyberstalking. Journal of Interpersonal Violence , 32(10), 1451–1475.

63.

Vakhitova

Z. I.

, Reynald

D. M.

, & Townsley

(2016). Toward the adaptation of routine activity and lifestyle exposure theories to account for cyber abuse victimization. Journal of Contemporary Criminal Justice , 32(2), 169–188.

64.

Ybarra

M. L.

, Espelage

D. L.

, & Mitchell

K. J.

(2007). The co-occurrence of Internet harassment and unwanted sexual solicitation victimization and perpetration: Associations with psychosocial indicators. Journal of Adolescent Health , 41(6), S31–S41.

65.

Ybarra

M. L.

, Mitchell

K. J.

, Finkelhor

, & Wolak

(2007). Internet prevention messages: Targeting the right online behaviors. Archives of Pediatrics & Adolescent Medicine , 161(2), 138–145.

66.

Yun

, & Lee

(2017). The relationship between religiosity and deviance among adolescents in a religiously pluralistic society. International Journal of Offender Therapy and Comparative Criminology , 61(15), 1739–1759.