Gender Differences in Cyberstalking: The Roles of Risk, Control, and Opportunity Factors in Social Media

1. Introduction

Stalking refers to the act or crime of intentionally and repeatedly following or harassing another person in circumstances that would cause a reasonable person to fear injury or death due to the threat expressed or implied. Cyberstalking is this same act or crime performed through information and communication networks, and there is no significant difference between the two definitions other than the location of the stalking (Bocij, 2006). Cyberstalking began to emerge as one of the main issues studied by criminologists, who dully described the damage of cyberstalking in the early days (Bennett et al., 2011; Reyns et al., 2012). However, scholars have shifted their attention to the causality of perpetration with the help of traditional sociological theories, such as strain theory, social learning theory, and self-control theory (Curry & Zavala, 2020; Fissel et al., 2021; Marcum et al., 2017; Wilson et al., 2023); or psychological factors, such as hostility, jealousy, and attachment problems (Deans & Bhogal, 2019; Toplu-Demirtas et al., 2020). Even so, empirical research on cyberstalking is still lacking, and more evidence is needed.

This study explores explanatory factors for cyberstalking on SNS (Social Network Service), especially focusing on gender differences among college students. According to the 2020 Cyber Violence survey conducted by the National Information Society Agency (NIA)1. in South Korea, only 0.8% of youths surveyed had offending experiences of cyberstalking, while this figure was 19.1% for adults. In this regard, this study intends to examine the cyberstalking of adults, especially university students, who have relatively higher offense experiences of cyberstalking. In addition, this study examined whether the explanatory factors for cyberstalking are differentiated by gender.

Previous studies suggest that men are often the perpetrators of stalking, while women are the victims. Furthermore, there are many cases in which the victim is an ex-girlfriend or lover when the perpetrator is male, whereas for female perpetrators, the victim is a same-gender friend (Meloy & Boyd, 2003; Sheridan et al., 2014). It is also argued that the motive for a male stalker is a high level of obsession or retaliation for the opposite gender who informed him of break-up, while that for female stalkers is an extension of bullying by same-gender peers (Purcell et al., 2010). In addition, stalking by men is somewhat more violent than that by women, and includes physical violence (James & Farnham, 2003; Rosenfeld & Lewis, 2005). Based on these characteristics, it can be presumed that the causes of stalking between men and women are somewhat different. This study analyzes whether the motives and causes of cyberstalking differ between men and women.

Cyberstalking can be viewed as an extension of traditional stalking (or offline stalking, general stalking), but it has different characteristics (Bocij, 2006). Unlike traditional offline stalking, perpetrators and victims in cyberstalking often do not know each other, and the stereotypical role assignment of male perpetrators and female victims is not necessarily the case (or applicable) anymore in cyberstalking (Borrajo et al., 2015a; Deans & Bhogal, 2019). Nevertheless, cyberstalking studies (Dreßing et al., 2014) show that, although the rate of female perpetrators has increased compared to conventional stalking, it is still a norm that males are perpetrators and females are victims, and the target is often former girlfriends or lovers. Of course, some studies (Reyns et al., 2012) have shown that the perpetrators and victims tend to know each other rather than an unknown relationship. Is cyberstalking similar to traditional stalking? However, because cyberstalking today occurs on SNS or through instant mobile messengers rather than blogs and cafes in conventional Internet networks, it is highly likely that it would occur between people who know each other, so it may have characteristics similar to those of conventional stalking in reality.

In this study, we used various explanatory factors for SNS stalking and attempted to verify it. The existing factors can be divided into risk, control, and opportunity factors, and the interactive mechanism of these three factors is examined. The risk factor refers to the motive and causative factor that induces cyberstalking, the control factor is the factor that prevents cyberstalking, on the contrary, and finally the opportunity factor is the situational opportunity factor for the occurrence of stalking. This study examines the main risk effect on cyberstalking and the interaction effect between risk and control factors and between risk and opportunity factors.

The most notable aspect of the study is the application of such interaction effects to men and women separately and examination of the gender differences in the results. Previous studies have shown gender differences in the causes of conventional stalking (Purcell et al., 2010; Sheridan et al., 2014); the research question is whether gender differences exist in the main effects and interaction effects of three factors – risk, control, and opportunity – in cyberstalking. We take a look at the heterogeneous characteristics between men and women in the main effects and interaction effects of three factors of risk, control, and opportunity, and try to provide answers for the difference.

As such, this study attempts to deal with cyberstalking on SNS, such as Facebook and Instagram, due to the recent increases in the use of SNS as well as the rate of cyberstalking. Through these attempts, we will examine whether men and women differ in their risk, control, and opportunity factors for cyberstalking. Specifically, we examine the main effects and interactive effects of the risk, control, and opportunity factors in cases of cyberstalking, and examine the implications and alternatives from the results.

2. Literature Review

2.1. Risk Factors in Cyberstalking and Gender Differences

Stalking can be classified into three types: simple obsession, affectionate obsession, and delusion (Zonna et al., 1998). Simple obsessional is derived from a form of attachment or restoration of a relationship with a former spouse or lover, while affectionate obsessional arises from an expression of affection for someone they love. The delusional type of stalking arises from the illusion (false belief) that the victim loves the perpetrator. In the case of cyberstalking, apart from obsession and delusion, retaliation against the victim is also proposed as a major of offense (Mishra & Mishra, 2008). On the other hand, McFarlane and Bocij (2003) classified cyberstalking into four types of obsessions, retaliation, harassment and grouping. It can be said that obsession and affection are not the only motives; retaliation, punishment, harassment, and collectively in a majority of cases.

Here, in presenting the main causal factors of cyberstalking, we distinguish the explanatory factors of male and female stalking according to the above classification, as well as to the claim (Purcell et al., 2010) that males are mainly obsessed with former lovers, whereas females engage in same-sex bullying. Thus, we discuss the causal factors of cyberstalking with emphasis on the following two factors.

First, as we see in the cases of simple obsession and affection obsession, the main forms of stalking begin with the so-called obsession. Obsession is a major factor in men’s stalking behavior toward their former lovers and is known to be caused by a lack of attachment. It is argued that stalking arises out of confusion in attachment between the actor and his/her parents during childhood. Previous studies on attachment problems show that attachment with parents in childhood is very important, and lack of affection or violence from parents is more likely to manifest in the form of preoccupied, anxious, and insecure attachments rather than stable attachments. These characteristics of abnormal attachment are prevalent among stalkers suffering from relational problems (Davis et al., 2000; Dutton & Winstead, 2006; Patton et al., 2010). Individuals with attachment disorders want to control others in the relationship even in coercive ways because they want to get attention and reassurance from them, which leads to an increased outbreak of stalking. Moreover, fear of rejection, suspicion of others, and probable breakup of intimate relationships tend to produce anger and subsequent violence in extreme cases.

Second, cyberstalking is not just a matter of affection and obsession but is rather motivated by violent relationships such as bullying, retaliation, and punishment against others (Strawhun et al., 2013). Cyberstalking is very similar to cyberbullying, as it is a form of cyberbullying (or violence) that has traits of both, obsession and bullying. Furthermore, offenders blame and punish targets in the form of stalking. Just as the main cause of cyberbullying is retaliation and punishment against someone who has done wrong or harmed the perpetrator by spreading false rumors about them, stalking is also aimed at punishing and harassing the target, because cyber stalkers think victims deserve punishment and harassment. Offenders justify their actions through these processes, which also serve as a major explanatory factor (Fissel et al., 2024; Lanares et al., 2021; Santos et al., 2024).

In particular, heterosexual men with insecure attachment disorders tend to be obsessive with their partners and try to dominate them, leading to stalking. Although few stalking studies have applied attachment discussion to men and women separately, dating violence studies show that, unlike women, men often try to control their partner in their relationship with ex-partners and often use violence to vent their resentment against separation. Therefore, if male violence is somehow related to attachment problems (Follingstad et al., 2002), and if cyberstalking is an extension of offline stalking, then attachment is a more suitable predictor for male stalking as compared to female stalking.

Bullying and retribution in stalking tend to occur more often in women than in men. Similar to the statement that bullying is more prevalent among women than men, the forms of warning and punishment against those who stand out could be more likely to be suitable in cases of female stalking. Just as women’s stalking tends to justify their actions by blaming the person who provides the cause (Langhinrichsen-Rohling, 2012), it takes the form of bullying against female peers to punish and deny out of hate, which is quite different from male stalking. In other words, the motive for punishment, such as denying and blaming the other, can explain female stalking better in the form of bullying. Borrajo and colleagues (2015b) suggested that the justification belief of violence in explaining cyber dating abuse was more significant in women and acted as a form of self-defense.

2.2. Gender Differences in Control and Opportunity Factors

2.2.1. Control and Opportunity Factors

While the risk factor has a positive (+) relationship with cyberstalking, the control factor has a negative (-) relationship. Here, rather than the motive and risk of cyberstalking, we focus on the factors that can control and prevent it. The most representative theory on the control factor is the social control theory (Hirschi, 1969). This theory presents several social bonding factors as an answer to the question of why people do not commit crimes rather than why they commit crimes. In the context of cybercrime, social bond assuming individual moral attitude and ethical sense as moral belief is also important. In cyberspace, cyberstalking is more permissible because of the characteristics of anonymity and noncontact, so a lack of control over former traits makes people commit cyberstalking more. In that sense, possessing a moral attitude toward cyberstalking can be a major factor in controlling and preventing stalking. In a stalking study on college students, Fox et al. (2011) suggested that the attitude toward stalking was the main explanatory factor for both the perpetration and victimization of stalking. In their cyberstalking analysis, Lee et al. (2020a) also suggested that a moral attitude toward stalking was the main factor of cyberstalking. Therefore, individual morality is a major control factor in cases of stalking.

Self-control is another major control factor found to be significant in previous studies. The general theory of crime by Gottfredson and Hirschi (1990) noted that most crimes occur immediately and impulsively so that one’s criminal behavior is differentiated by internal disposition, implying an individual’s capacity to manage and control immediate gratification and impulsivity. Self-control would account for the outbreak of any kind of crime, and even cybercrimes including cyberstalking can be explained by a main factor of self-control. In stalking research (Davis et al., 2012), self-control is evaluated as one of the most important factors in the application of similar self-regulation theory (Baumeister et al., 1994). In other words, it is easy to act accidentally in cyberspace, and a person with low self-control can easily commit cyberstalking, but if the person with high self-control can refrain from and control the behavior, it can be seen that the possibility of violation is low.

On the other hand, in other discussions on crime, its opportunity factor has been emphasized. For example, the lifestyle theory (Hindelang et al., 1978) and routine activity theory (Cohen & Felson, 1979; Meier & Miethe, 1993) emphasize that there should be an opportunity for crime to occur anyway. Unlike the control factor that has a negative (-) relationship with crime, opportunity is a factor that promotes crime and thus is positively (+) related to it. These discussions can also be applied to cybercrime, suggesting that the more opportunities and conveniences there are for crime to occur, the more cybercrime would occur unaffectedly (Holt & Bossler, 2008; Marcum et al., 2010). In cyberspace, it is easy to commit a crime with a single click, and there are many accidental opportunities, such as the absence of surveillance, which can be seen as a major cause of crime.

Opportunity factors, such as non-face-to-face (or untact) and anonymous situations, can act as major factors in cybercrime (Barlett, 2015; Lowry et al., 2016; Wright, 2013). Here, attention is paid to the anonymity of the many opportunity factors. In the case of cyberstalking, anonymity is a major explanatory factor (Philips & Morrissey, 2004). Bocij (2006) also finds that, in addition to disinhibition emerging from anonymity, the ease of opportunity and technical ability are other major factors of cyberstalking. As such, opportunity is another major factor that facilitates cyberstalking.

2.2.2. Gender Differences in Their Moderating Effects

Apart from the risk factors, cyberstalking would be more easily explained when control factors are withheld or opportunity factors are disclosed. As explained above, the control factor acts (works) negatively (-) on cyberbullying, whereas the opportunity factor acts (works) positively (+). Such control and opportunity factors can work as a buffer or catalyst integrally in the action of risk factors.

This argument is based on existing integrated theory. Agnew’s (1992) general strain theory posits that while strain is a major risk, crime is more likely to occur when social or internal control is low, covering another factor building up an integrated theory. To some extent, this type of integrated model has been supported in subsequent studies (Mazerolle & Maahs, 2000; Morash & Moon, 2007). In addition, according to the coercion and social support theory by Colvin et al. (2002), both coercion and supporting factors need to be considered integrally when we divide explanatory factors into two parts: coercion and support. Referencing a study where self-control and morality were utilized as the main factors of control (Barton-Crosby & Hirtenlehner, 2021), Lee and colleagues (2021) explained cyberbullying with the main effects and the interaction effects between risk factors and control factors, where the former utilized cyberbullying victimization and cyberbullying peers, while the latter used morality and self-control.

In previous studies, morality and self-control were found to be significant in both genders, showing no difference between men and women in inhibiting crime (Burton et al., 1998; Ishoy & Blackwell, 2019; Shekarkhar & Gibson, 2011; Weerman et al., 2016). However, when we consider the role of gender difference in the buffering effect of control factors, they appeared to be more significant among women, even if there were not many supporting materials. Jang and Johnson (2005) compared the gender effect of strain on crime and suggested that religiosity had a buffering effect on crime in women; this finding was supported by Jang (2007). This implies that women are less likely to commit crimes if they have the morality and self-control to inhibit criminal activity. In addition, Mears et al. (1998) examined the gender-differentiated buffering effect of morality in the relationship between delinquent peers and crime and suggested that morality did not buffer the influence of delinquent peers on crime for men, while it played a buffering role for women. This implies that the probability of a criminal act declines with a high level of morality, even if youth have many delinquent peers. Furthermore, the moderating effect of self-control as a controlling factor in the relationship between delinquent peers and crime was greater for women than for men. Cheung and Cheung (2010) showed that the interaction effect of self-control on the influence of strain on delinquency was greater for women than for men. In other words, these outcomes imply that men are unable to control their morality or self-control in the action of a certain risk, but women are more likely to be able to control their morality or self-control in the process.

On the other hand, it is also likely that the risk factors of cyberstalking would work together with the opportunity factors. In fact, scholars who emphasize opportunity factors have argued that motivated offenders commit crimes when there are opportunities for crime, such as suitable targets and the absence of guardians working together (Cohen & Felson, 1979; Meier & Miethe, 1993). As the opportunity factor used to be treated as a major factor interacting with various risk factors in previous cybercrime studies (Lee et al., 2020b), it is likely that criminal opportunity factors would interact with risk factors in explaining cyberstalking in an integrated way. In other words, in the risk factors mechanism, there should be an opportunity for risk to be activated more, and such opportunity factors would promote the impact of risk factors on deviant behaviors.

Some studies on opportunity factors by gender show that opportunity operates more effectively for men than for women. Augustyn and McGloin (2013) argued that women do not easily engage in deviant behavior, even by chance, due to parental control or internal restraints, whereas men have a high plausibility of involvement in delinquency if opportunities arise due to the manifestation of masculinity in certain circumstances. In addition, Marcum et al. (2014), studying the impacts of various factors on cyberbullying by gender, also suggested that opportunity factors appear to be more important to men. Some research results have shown that factors such as anonymity—a representative variable known as one of the opportunity factors—have a stronger influence on men than on women (Wang et al., 2021). Although few previous studies have dealt with opportunity factors by gender as a moderating variable, we can propose that the effect of opportunity factors as moderators precipitating the impact of risk factors on criminal behavior such as anonymity would be greater for men than for women.

3. Current Study and Research Questions

To explain cyberstalking, this study proposes a model that focuses on risk, control, and opportunity factors (separately and simultaneously), and empirically tests the effects of these factors on cyberstalking. First, it investigates the main effects of risk factors for cyberstalking. Second, it investigates the interaction effects of risk factors with both control and opportunity factors, respectively, in explaining cyberstalking. Specifically, it is predicted that two types of interaction effects will be tested, such that there will be an interaction effect between the risk factor and the control factor, as well as an interaction effect between the risk factor and the opportunity factor. The former interaction is called suppressing interaction, whereas the latter is termed precipitating interaction.

We used lack of attachment and denial of victim risk factors, as the main types and motives of the risk factors are obsession and retribution. According to the discussion on attachment as a risk factor, it is likely that a person with attachment confusion will commit stalking because of problems in personal relationships such as anger or frustration in the relationship. From the standpoint of previous stalking research (Follingstad et al., 2002), the lack of attachment seems to occur more likely in men who try to control their relationship with the opposite sex or partner. Denial of the victim refers to denying and blaming the victim for providing reasons for stalking, as well as chastising the victim seems to be more evident for female stalking than for male stalking (Borrajo et al., 2015b; Langhinrichsen-Rohling, 2012).

Research question 1: Does lack of attachment and denial of the victim affect cyberstalking? Are there gender differences in the impact of these factors?

Lack of attachment and denial of victim are the main risk factors for cyberstalking, but we posit that stalking would be more likely to occur when control factors such as morality and self-control are low, and less likely to occur when the control factors are high. In the same manner, we posit that the effects of risk factor on stalking would contribute more to cyberstalking when the opportunity factors are high. In other words, lack of attachment and denial of the victim as risk factors will interact negatively (-) with morality and self-control as control factors, while interacting positively (+) with anonymity as an opportunity factor. Thus, based on previous studies (Jang, 2007; Marcum et al., 2014; Mears et al., 1998; Wang et al., 2021), it can be assumed that there will be differences between men and women in the moderating effect of control and opportunity factors in such a way that the former will be greater for women, while the latter will be greater for men.

Research question 2: Do risk factors (lack of attachment and denial of the victim) and control factors (morality and self-control) have negative (-) interaction effects with cyberstalking? Are there gender differences in the impact of these factors?

Research question 3: Do risk factors (lack of attachment and denial of the victim) and opportunity factors (anonymity) have positive (+) interaction effects on cyberstalking? Are there gender differences in the impact of these factors?

4. Data and Methods

4.1. Sample and Procedure

In this study, we examined SNS cyberstalking among adult college students. For the survey, we selected 60 students from each of the five universities in Seoul, South Korea. Ine university was selected from each of five regions in Seoul: east, west, north, south, and center. A quota sampling was applied within each university with three demographic variables: gender, academic major and academic level. The Survey was conducted over a two-week period in July 2018, with approval from the Institutional Review Board of the first author’s institution (IRB#). Participants were informed that their participation was completely voluntary, and their responses would be treated as confidential and protected by the law. A total of 304 male and female students were surveyed. Among them, 270 SNS users were finally selected and analyzed.

We conducted OLS regression analysis using SPSS version 24. The baseline model examined the effect of two risk factors on cyberstalking by gender, controlling sociodemographic variables. In addition, we incorporated interaction terms with each moderating variables. To mitigate multicollinearity, the interaction terms were mean-centered.

4.2. Measurements

We use the actual perpetration of cyberstalking as the dependent variable in the current study. While previous studies have used inconsistent measurements for this variable (Curry and Zavala, 2020; Marcum et al., 2014, 2017), we specifically define cyberstalking as more than just accessing or monitoring a partner’s social media account without consent. In our context, it refers to the act of stalking, which includes sending persistent messages to a partner against his/her wishes or causing anxiety and fear. To operationalize this definition, participants were asked to respond to the following questions for the past year, “I have continuously sent messages on SNS even though the victim rejected it,” “I have consistently sent a friend request on SNS even though the victim rejected it,” “I have uploaded a post that could cause fear or anxiety in a victim on SNS,” “I have ever sent a message or made a comment that could cause fear or anxiety against the victim on SNS”. Four items have four-response scale: “never (score=0),” “once or two (score=1),” “three to eight times(score=2),” and “nine or more times(score=3).” Each score (0-3) of four items was summed and the natural log-transformation of the summed score was applied to reduce positive skewness in the variable.

Lack of attachment as a risk factor corresponding to attachment avoidance and attachment anxiety was constructed with 12 items among the items on the attachment scale developed by Brennan et al. (1998), such as “I feel uncomfortable with being too close to others” and scored on a 5-point Likert scale, ranging from “not at all true (1)” to “very true (5).” The responses were then summed, resulting in a composite score (alpha=.922).

Denial of the victim utilized items from the discussion of techniques of neutralization (Sykes & Matza, 1957) in stalking situations. Three items were selected from the study by Borrajo et al. (2015b), such as “Anyone stalked is responsible for the victimization to a degree” and scored on a 5-point Likert scale (alpha=.890).

The morality of cyberstalking, as a control factor to control the impact of the risk factor, consists of one question regarding how bad the cyberstalking behavior is on a 5-point Likert scale from “not bad at all (1)” to “very bad (5)”.

Self-control, based on a previous study (Grasmick et al., 1993), consisted of two questions for each of the following six traits: impulsivity, risk-seeking, simple task-seeking, activity, selfishness, and temperament. Thus, 12 questions in total including “I often act impulsively” were asked to get answers on a 5-point scale, ranging from “not at all true (1)” to “very true (5).” The responses were then summed, resulting in a composite score and finally inverse-coded (alpha=.823).

Anonymity, as an opportunity factor, was asked to check whether they used SNS mainly in anonymous situations. Four questions were asked: “Actually I do not use my real name on SNS,” and responses were made on a 5-point scale (alpha=.906).

As a sociodemographic control variable, age was recorded by year of birth. In addition, the family’s subjective economic level was classified into 5 categories from “low” to “lower-middle,” “middle,” “upper-middle,” and “high.”

5. Results

Table 1 presents descriptive statistics of the variables used in this study. The study sample consisted of 145 males (53.7%) and 125 females (46.3%). The average age of the respondents was 21.37 ranging from to 18–28 years, with an average of 21.9 years old for males and 20.9 for females. The average value of the family’s economic level was 3.18 ranging 1-5 with 3.24 for males and 3.13 for females.

The average lack of attachment as an independent variable was 32.452 ranging between 12-60 with males scoring 34.064, a little higher than females (31.062). The average denial of victims was 5.024 in the range of 3–15 with males scoring 6.408, which was significantly higher than that of females (4.166). Morality, as a control factor, had an average of 4.390 ranging from to 1-5 with males scoring 3.944, which was lower than that of females (4.778). The average self-control score was 37.478 in the range of 12-60, and males scored 36.960, which was lower than that of females (37.747). As an opportunity factor, the average of anonymity in the range of 4–20 was 10.048, and males scored 10.912, which was higher than that of females (9.303). Finally, the average logged score of cyberstalking experience, the dependent variable, has an average of .512 ranging between 0–2.56 with males (.814) scoring much higher than females (.252).

Prior to the analysis of the major research question, Table 2 present the effects of the risk factors on cyberstalking. In the case of males, as shown in Table 2, lack of attachment was significant with the largest effect size (β = .524), followed by denial of the victim (β = .347), both at the p<.001 level. In the case of females in Table 2, denial of victim had the greatest effect (β = .315) at the p<.001 level. However, lack of attachment was not significant for females, which is an important difference between males and females, which supports the predictions of this study to some extent.

Table 3-1 and 3-2 present the results of the interaction effect between risk, control, and opportunity factors. Analysis 1 presents the interaction effect between the risk factors and the control factor, morality. The direct and independent influence of morality was significant for men (β = -.805) at the p<.001 level, but not for women (β = -.157). Next, we investigated the interaction effects between risk factors (lack of attack and denial of victim) and morality. It showed that the interaction effect between lack of attachment and morality was not significant for both men and women, but the interaction effect between denial of victim and morality was significant at the p<.05 level only for women (β = -.303) even if it was still not significant for men. This means that the moderating effect as a control factor for morality works more in the case of women.

Analysis 2 presents the interaction effect between the risk factors and the control factor, self-control. The direct and independent influence of self-control was not significant for both men and women. Next, we investigated the interaction effects between risk factors (lack of attack and denial of victim) and self-control, which is a control factor. The interaction effects between risk factors and control factors were not significant for men, but only one interaction effect was significant for women, that is, between denial of victim and self-control (β = -.211) at the p<.05 level. This means that the moderating effect of self-control as a control factor is more important for women.

Finally, in Analysis 3, the interaction effect between the risk factors and the opportunity factor for anonymity is provided. Unlike the previous steps, the direct and independent influence of anonymity on cyberstalking was significant for both men (β = 0.263) and women (β = 0.319) at the p<.001 level. Next, the interaction effects between risk factors (lack of attack and denial of victim) and anonymity (an opportunity factor) were investigated. Result showed that the interaction effect between lack of attachment and anonymity was significant for men (β =0.251), but it was not significant for women. It showed that the interaction effect between denial of victim and anonymity was not significant for both men and women. This implies that the moderating effect of anonymity as an opportunity factor works more in the case of men.

6. Discussion

To explain cyberstalking in the use of SNS, this study investigated the influence of risk factors for cyberstalking and also tried to test the interaction effects focusing on risk, control, and opportunity factors simultaneously. The proposed hypothesis posits that there will be a negative (−) interaction (risk factor × control factor) and a positive (+) interaction (risk factor × opportunity factor). Even if the risk of committing cyberstalking commission is high, its probability is low if the condition of the control factor is high, or that of the opportunity factor is low. In contrast, even if the risk of a cyberstalking commission is low, the probability of a real commission increases if the condition of the control factor is low or that of the opportunity factor is high. In this study, we used lack of attachment and denial of victim as risk factors, morality and self-control as control factors, and anonymity as opportunity factors.

Furthermore, this study predicted that the interaction between risk and moderating factors (control and opportunity) would differ by gender. It posits that the impact of the risk factors - lack of attachment and denial of victim – would be differentiated by gender such that lack of attachment would be important for men and denial of victim for women. In the action of these risk factors, the interaction effect is also differentiated by gender in such a way that the moderating effect of control factors will be greater for women than for men, and that of opportunity factors such as anonymity will be greater for men than for women. Thus, the interaction between risk factors (lack of attachment and denial of victim) and control factors (morality and self-control) would be greater in women, while that between risk factors (lack of attachment and denial of victim) and opportunity factors (anonymity) would be greater in men.

The research outcomes support the general hypotheses of this study. In the case of men, stalking originates from the obsession with a former lover of the opposite sex. For men, denial of victim was also significant, but the effect of lack of attachment was stronger. On the other hand, unlike men, women’s lack of attachment was not significant, and the effect of denial of victim was significant, suggesting that stalking is an extension of cyberbullying and is committed as a form of punishment to deny or blame the other person. Moreover, the effect of the control factor is greater for women, with the interaction effect of denial of victim x morality and the interaction effect of denial of victim x self-control being more significant for women. This suggests that the two control factors reduced the impact of denial of victim, a key risk factor for women.

On the other hand, the facilitating role of the opportunity factor was predicted to be greater for men than for women. However, the role of opportunity factors in interaction is different for men and women such that while the interaction effect of lack of attachment × anonymity appeared more significant in men, even though denial of victim × anonymity was not significant in both men and women. What it suggests is that, as predicted, the opportunity factors in the interaction effects facilitate and precipitate the action of risk factors on cyberstalking for men.

Out of these study outcomes, we found that the cause of cyberstalking behavior is somewhat different between men and women and that the moderating effects of control factors and opportunity factors on cyberstalking through risk factors could also be differentiated by gender. This implies that the action of risk factors in cyberstalking needs to be identified separately for each gender. For men, addressing the problems of fixation and parental nurturing is critical to preventing cyberstalking. This finding is consistent with previous research that emphasized the importance of fixation or attachment style (Follingstad et al., 2002), even though it did not examine gender differences. For both women and men, retaliation, punishing, and blaming others are important causes of cyberstalking and the change of the justification belief is necessary. These results imply that gender-specific solutions are required, such as counseling and healing for male stalkers to reduce their obsession with their former partner or lover, and for women to get along and foster relationships with their same-sex friends. Furthermore, a gender-differentiated policy needs to be pursued in search of countermeasures that contribute to controlling the action of the risk. For example, moral and ethics education may be more effective and suitable for women, whereas the removal of opportunities for cyberstalking would be more effective and suitable for men. In addition,

Now that empirical research on cyberstalking is scarce, this study could be considered to be meaningful and contributable for developing the subject in this field in bits and pieces. In addition, it is also meaningful to clarify the process by which the moderating action of control and opportunity factors with risk factors on cyberstalking can be differentiated by gender. However, a limitation of this study is that the study sample was collected from college students. In addition, this study assumes that men reported stalking women and women reported stalking their same-sex friends, but in the future, it is necessary to study more in-depth through stalking surveys that identify the target-gender of each stalker including intimate partner stalking (Gordon& Dardis, 2023; Reiss et al., 2022; Smoker & March, 2017). While this study provides an example , future research will need to build a more robust integrative model that includes risk, control, and opportunity factors, and may require the construction of gender-integrated or gender-specific models. Future research should continue to expand the sample by targeting not only college students but teenagers and adults as well.

7. Conclusions

This study found that the effects of lack of attachment and denial of victims as risk factors for cyberstalking differed by gender. As a risk factor, the influence of lack of attachment was greater in men, while denial of the victim was greater in women Furthermore, in the context of risk factors, we predicted that the moderating effect of the control factor would be greater for women, and that of the opportunity factor such as anonymity would be greater for men. The results show that the moderating effects of the control factors were greater for women in such a way that the interaction between denial of victim and morality and that between denial of victim and self-control were significant for women, while the moderating effect of the opportunity factor was greater for men in such a way that the interaction between lack of attachment and anonymity was more significant for men. These findings suggest that the causal processes of cyberstalking and the respective moderating actions can be differentiated across gender, which implies that a gender-differentiated policy needs to controlling the action against cyberstalking.