Explaining Variables of Adolescents’ Cyberstalking Behavior:
Internet Addiction, Gender and Educational Level
N. van Ewijk
11082844
Master’s thesis: Forensic Child and Youth Care Sciences Graduate School of Child Development and Education University of Amsterdam
Advisor: Dr. Inge B. Wissink Second advisor: Dr. Esther A. Rutten Amsterdam, 07-02-2018
2
Abstract
This study attempts to give insight in cyberstalking behavior among 927 high-school students by examining three possible explaining variables: internet addiction, gender and educational
level. Participants filled in an online questionnaire including the PIUQ-SF-6 to measure internet addiction (Demetrovics et al., 2016), and a translated (in Dutch) and updated version
of the ECIPQ (Brighi et al., 2012) with one additional question to measure cyberstalking. Hierarchical regression analysis was used to analyze the data. Results indicated that the three
variables together significantly explained around 10% of cyberstalking behavior. Internet addiction proved to be the strongest explaining variable with almost 8% explained variance,
followed by gender and, subsequently, educational level. Moreover, this study showed that males, students with lower educational levels and with more characteristics of internet addiction had the highest scores on cyberstalking behavior. Results of this study can help
with identifying high-risk groups for cyberstalking and, therefore, prevention and intervention methods can be better targeted and adjusted. In turn, this may help to reduce
cyberstalking behavior. Implications for future research are discussed as well.
Keywords: Cyberstalking, Internet Addiction, Gender-Differences, Educational Level,
3
Explaining Variables of Adolescents’ Cyberstalking Behavior
Internet has become a key element in the life of adolescents. Throughout their daily lives, adolescents use smartphones, tablets, laptops and other electronic devices to access the internet. The online platform offers them an environment where they can socialize with friends and strangers, listen to music, enjoy gaming, do online gambling, online shopping and it provides an environment of unlimited information. Though internet has enriched their lives in many ways, misuse of internet can have tremendous effects. Cyberstalking is one of many ways to misuse the internet. It is defined as the repeated and unwanted pursuit of an
individual through electronic devices (Reyns, Henson, & Fisher, 2010). Cyberstalking includes harassment, intimidating and threatening behavior, as well as unwanted sexual advances(Dreβing, Bailer, Anders, Wagner, & Gallas, 2014; Finn, 2004; Navarro, Marcum, Higgins, & Ricketts, 2016; Paullet, Rota, & Swan, 2009; Reyns et al., 2010; Reyns, Henson, & Fisher, 2011). As these behaviors happen online, it is very hard for parents, school and others involved to identify and tackle them. Therefore, this study aims to provide more insight in cyberstalking1 and it’s risk factors. First, I will attempt to do this by examining a
possible association between Internet Addiction (further: IA) and cyberstalking. Though a high level of use of the internet is a condition for cyberstalking, in few studies a possible association between excessive use of the internet and cyberstalking has been examined so far (Navarro et al., 2016). Therefore, in the current study, the association between the level of IA and cyberstalking will be examined. Furthermore, analyses of gender differences and
differences in educational level regarding both IA and cyberstalking and regarding the association between IA and cyberstalking will be performed. Cyberstalking reflects a criminal activity and, consequently, perpetrators of cyberstalking probably operate in secret while covering up their online activities. Gender and educational level are factors that could be identified with ease compared to cyberstalking. If this study could relate gender and educational level to cyberstalking, it could help to classify groups that may be at risk of engaging in cyberstalking behavior. All taken together, results of this study could provide schools (and parents) with valuable information in order to be able to identify persons ‘at risk’ and to prevent cyberstalking from happening. In turn, this might help to reduce both the number of individuals that are in trouble because of cyberstalking perpetration and the individuals that are victimized due to cyberstalking.
1 This study focuses on perpetrators of cyberstalking. If cyberstalking is mentioned, this refers to perpetrators
4
To illustrate the seriousness of cyberstalking, next, a real example is presented. In 2012, Adam Savader hacked into a woman’s AOL account and found some nude pictures she had saved for private use. Savader knew this woman from high school and contacted her through text messages using a Google Voice phone number. He demanded her to send him more photos, otherwise he would spread her nude pictures. On top of that he asked several personal questions regarding her sex life. To demonstrate the seriousness of his demands, he sent her a Facebook picture of her mother. Police found out Savader had contacted 14 other women, all of them receiving online threats and had received unwanted messages up to 80 times a day. He was finally arrested in April 2013 and faced federal charges of cyberstalking and internet extortion (“Adam Savader Plea Agreement”, 2013).
Studies examining the prevalence rates of cyberstalking perpetration show a
prevalence of around 5.0% (Navarro et al., 2016; Reyns, Henson, & Fisher, 2012). This rate could be an underestimation as there is no adequate measuring instrument for cyberstalking and perpetrators might intentionally or unknowingly withhold from answering truthfully (Paulhus, 2003). Another issue regarding the establishment of a prevalence rate concerning cyberstalking is it’s overlap with cyberbullying. Building on the concept of (offline) bullying, cyberbullying is described as the intentional and repeated behavior to harm another person through use of electronic devices, which creates an imbalance of power between the bully and the victim (Herrera-López, Casas, Romera, Ortega-Ruiz, & Del Rey, 2017; Olweus, 2012; Smith, 2015; Tokunaga, 2010; Veenstra, 2011). When comparing cyberstalking and cyberbullying there seem to be quite some similarities. According to Article 285b of the Criminal Code in the Netherlands (2017), behavior can be classified as stalking (or more precise: harassment) if it is: (1) intentional; (2) repeated; (3) unlawfully; (4) unwanted and (5) with the intention to force someone to do, or withhold them from doing something or to cause fear. According to above mentioned description, both cyberstalking and cyberbullying are intentional, repeated and unwanted. (Cyber)Bullying itself is not punishable by law, but can be placed under other punishable behaviors like insulting (article 266 of the Criminal Code in the Netherlands, 2017), slander (article 261 of the Criminal Code in the Netherlands, 2017) or stalking/harassment (article 285b of the Criminal Code in the Netherlands, 2017). Following this, cyberbullying, or at least the more extreme forms, can be classified as unlawful,
similarly as with cyberstalking. Combee (2008) compared cyberbullying and cyberstalking in a descriptive manner and argued that the intention behind the two behaviors is the most discriminating factor. She mentioned that a bully does not automatically intends to cause fear or force or withhold someone from doing something, while a cyberstalker does. Nevertheless,
5
it can be argued if the imbalance of power between the bully and the victim isn’t self-evident for intending some form of fear. In this light, cyberbullying does meet the fifth element of cyberstalking (i.e. harassment) as stated above. To recap, all five characterizing elements of cyberstalking as classified in the Criminal Code of the Netherlands seem to be important defining elements of cyberbullying as well. It is important to find consensus in the
similarities and differences between the two concepts because it can affect prevalence rates. Furthermore, research has focused more on cyberbullying resulting in more information about cyberbullying. For instance, there are adequate questionnaires measuring cyberbullying and prevention methods as well and these might also be helpful in fighting cyberstalking. On the other hand, it is not sure whether these questionnaires and prevention methods are equally effective in measuring and combating cyberstalking. It is also not clear whether the variables that explain cyberbullying (perpetration) are similar to the variables that explain
cyberstalking. Therefore, more information, specifically focusing on cyberstalking, is needed. The current study will focus on the explanation of ‘cyberstalking’, as it is currently formulated in the literature (Reyns et al., 2010). Following Reyns and colleagues (2010) we define cyberstalking as the repeated and unwanted pursuit of an individual through electronic devices, as already described. And as previously mentioned as well, internet is a condition for cyberstalking, for cyberstalking cannot be classified as such if it is not happening online. To better understand the possible existence of an association between IA and cyberstalking, it is important to first clarify the concept of IA. In the late 90’s the phenomenon of the internet took over lives, which led Young (1998) to propose a new disorder: Internet Addiction (IA). In the literature the term IA is controversial, but in this study it is synonymous to compulsive, pathological or problematic internet use (Demetrovics, Szeredi, & Rózsa, 2008; Laconi, Rodgers, & Chabrol, 2014; Meerkerk, Van den Eijnden, Vermulst, & Garretsen, 2009). IA can be defined as the maladaptive and pathological pattern of internet use, which is time consuming and leads to clinically significant impairments in health or interpersonal
relationships. It is characterized by a loss of self-control which results in an inability to stop using the internet. Consequently, social lives, including home, work and school, suffer (Laconi, Rodgers, & Chabrol, 2014; Wu, Lee, Liao, & Chang, 2015). As it appears that adolescents are at an increased risk of becoming addicted to the internet (because they are still developing their cognitive control and boundary setting skills; Casey, Tottenham, Liston, & Durston, 2005; Leung, 2008; Liu & Potenza, 2007), the current study will focus on high-school students. Kuss, Van Rooij, Shorter, Griffiths and Van de Mheen (2013) examined the prevalence of IA within a sample of Dutch high-school students using a self-report
6
questionnaire that included the Compulsive Internet Use Scale and the Quick Big Five Scale. They found a prevalence of 3.7%, which corresponds with other studies investigating IA among adolescents in other Western-countries (Kaltiala-Heino, Lintonen, & Rimpelä, 2004; Kormas, Critselis, Janikian, Kafetzis, & Tsitsika, 2011; Liu, Desai, Krishnan-Sarin, Cavallo, & Potenza, 2011). This prevalence rate might appear low, but considering a classroom holds around 25 students, it means that every classroom holds at least one student that has IA. Moreover, a more recent study showed that around 12.8% of adolescents are at risk for IA behavior (Tsitsika et al., 2016), meaning one out of eight adolescents shows premature signs (such as compulsive tendencies, emotional changes and behavioral problems) that could lead up to IA. To summarize, relatively many adolescents are addicted to the internet or show at-risk behavior for IA. With the constant growth of the internet, it’s possibilities and topics broaden as well, which will attract even more individuals. Consequently, the group of
adolescents with (a risk for developing) IA will grow, which in itself is unwelcome, but could be even more undesired if an association between IA and cyberstalking exists.
Navarro and colleagues (2016) are among the few who have examined the possibility of an association between IA and cyberstalking. They suspected an association based on four areas of overlap between characteristics of persons with IA and of cyberstalkers. First of all, cyberstalkers are usually referred to as loners, who are emotionally disturbed and seek for attention and companionship online. Acier and Kern (2011) reported that clients with IA were also predominantly introvert and withdrawn individuals, who experienced difficulties
expressing themselves. Secondly, both populations of cyberstalkers and of persons with IA displayed a heightened tendency towards addiction (Griffiths, Rogers, & Sparrow, 1998; Griffiths, 2001). Third, due to characteristics of the internet such as anonymity and
depersonalization, there is a heightened risk of excessive and compulsive behavior (Griffiths, 2001). As previously described, IA is characterized by a loss of self-control which results in an inability to stop using the internet (Laconi et al., 2014; Wu, et al., 2015), while
cyberstalkers often become obsessed with their victim while disregarding warnings to stop (“http://cyberangels.org” as cited in Deirmenjian, 1999; Stephens, 1995). Finally, both persons with IA and cyberstalkers predominantly exist of men (D’ovidio & Doyle, 2003). Navarro et al. (2016) examined the association between IA and cyberstalking in 1.617 high school students. IA was measured as a dichotomous explaining variable, with cyberstalking perpetration as the dichotomous dependent variable. The results of their logistic regression analyses indeed showed that individuals who possessed more characteristics indicating IA, had an increased risk of being engaged in cyberstalking perpetration. However, it should be
7
noted that Navarro et al. (2016) examined cyberstalking by only asking one question. Though this question embedded the key characteristic of cyberstalking, namely repeatedly seeking contact after someone has asked you not to, the reliability of the results is somewhat
questionable. Furthermore, the sample of high-school students was obtained in a rural country in western North Carolina and therefore, results from this study are hard to generalize to other geographical areas, like for instance the Netherlands. Despite these limitations, the study conducted by Navarro et al. (2016) was the first to indicate a significant association between IA and cyberstalking. If the present study could replicate the results of Navarro et al. (2016), this would strengthen the support for an association between IA and cyberstalking behavior. If that will be the case, youngsters who demonstrate internet addictive behaviors could then be regarded as at-risk-groups for engaging in cyberstalking behavior. This could help in the development of prevention methods that are more specifically aimed at this particular group, and that are therefore more efficient.
To further improve prevention methods, it is also important to look at other possible risk factors for cyberstalking, besides IA, such as gender. Based on gender theories of
violence, men would be at a higher risk of engaging in (offline) stalking behavior, as they are believed to be more prone to use controlling behavior to show toughness, authority and dominance (Langhinrichsen-Rohling, Selwyn, & Rohling, 2012). Additionally, D’ovidio and Doyle (2003) already suggested that cyberstalking perpetrators are predominantly male (80%). This enormous gender gap is, to say the least, remarkable, and could be due to the fact that D’ovidio and Doyle (2003) only examined cases concerning aggravated assault. As the concept of cyberstalking consists of more than just aggravated assault, we can expect smaller yet still significant gender differences in the general level of cyberstalking. Reyns et al. (2012) referred to WHOA’s (i.e. Working to Halt Online Abuse, 2009) statistics which stated that 42% of stalkers were male (versus 31% female stalkers).2 These figures show
there is still an apparent gender gap. Burke, Wallen, Vail-Smith, and Knox (2011) examined gender differences in specific cyber pursuit acts among undergraduates. Results showed that women were more likely to use monitoring behaviors like excessive mail and phone contact, and monitoring Facebook pages. In contrast, men were more likely to use surveillance methods like a hidden camera or GPS and were more likely to post inappropriate photos. To recapitulate, gender seems to play a part in various cyberstalking behaviors. Building on this, in the current study it will be examined whether gender works as an explanatory variable for
8
cyberstalking, besides IA. Moreover, the current study will examine whether gender affects the association between IA and cyberstalking as well, meaning that gender affects the strength of the association between IA and cyberstalking. Such a so-called interaction effect could be expected as girls that are addicted, tend to use the internet in very different ways compared to boys that are addicted to the internet. Hetzel-Riggin and Pritchard (2011), for instance, reported that girls who are depressed (an important risk factor for IA; Ha & Hwang, 2014; Ko, Yen, Chen, Yeh, & Yen, 2009), seem to search for online communication in blogging, chatting and social networking to decrease their depression. In contrast, depressed boys were looking for ways to express hostility and violence, therefore spending their time with online gaming (Ko et al., 2009; Yen, Ko, Yen, Wu, & Yang, 2007). So, there seems to be gender differences in the activities online and the function of these activities. In this regard, it could be expected that gender does not only play a role in explaining the mean level of cyberstalking but also in explaining the strength of the relation between IA and
cyberstalking, although firm evidence for the direction of effects is lacking. If gender does play a significant role, prevention and treatment efforts should be adjusted accordingly.
Another interesting factor that might affect adolescents’ level of cyberstalking is educational level. As previously mentioned, Navarro et al. (2016) already reported that perpetrators of cyberstalking were more likely to have higher educational levels. In contrary, Combee (2008) only found cyberstalkers in low educational levels (6.9%; although her sample only consisted of two respondents that were eventually classified as cyberstalkers). Though cyberstalking has gained more attention in the past few years, researchers have primarily focused on college students and victims of cyberstalking. Unfortunately, this means that both younger individuals and individuals with lower educational levels, as well as
perpetrators of cyberstalking, have often been excluded. Hence, aside from above mentioned studies, results regarding the explanative role of educational level in cyberstalking are scarce. Fortunately, more information is available regarding the relation between educational level and cyberbullying. For instance, a study by Veenstra (2011) reported that individuals with a low educational level were significantly more involved in threatening and excluding someone online (i.e. cyberbullying). Moreover, results indicated that, overall, there were fewer
cyberbullies within the higher educational level group compared to within the lower educational level group. As cyberbullying and cyberstalking overlap, this may suggest a lower prevalence rate of cyberstalkers among students with higher educational levels (which will be in accordance with the findings of Combee, 2008, but not with the findings of
9
cyberstalking might need to focus more on adolescents in the lower educational levels. On the other hand, other researchers indicated that a high educational level could pose as a risk factor for IA (Bakken, Wenzel, Götestam, Johansson, & Øren,2009). In their study, they examined the prevalence and factors associated with IA among 3.339 Norwegian citizens and found that individuals with a lower educational level spent less time on the internet. These results correspond with results of another study, which even stated that high educational level was a strong predictor for being at risk for IA (Macur, KirÁly, Maraz, Nagygyörgy, &
Demetrovics, 2016). To summarize, above mentioned studies show that high educational level could increase the chances of becoming addicted to the internet. Furthermore, the prevalence of cyberbullies within the lower educational levels seems to be higher compared to the prevalence within the higher educational levels. In light of these seemingly contrasting findings, it is important to further study these variables within one study using a broad normative sample of adolescents who differ in educational level (and gender) in order to be able to gain more information regarding these associations and possible moderators. As such, this study’s outcomes will give insight in the role that educational level plays in explaining cyberstalking, and whether it affects the possible association between IA and cyberstalking. In case educational level does significantly affect cyberstalking in either or both ways, treatment and preventive efforts should pay attention to these findings accordingly.
All taken together, this study aims to replicate the results found by Navarro et al. (2016) by examining the association between IA and cyberstalking. Moreover, limitations from above mentioned study will be minimized by adding more questions regarding
cyberstalking, and by retracting a normative sample from multiple high schools and different geographical areas in the Netherlands. Besides examining the association between IA and cyberstalking, this study aims to investigate gender differences and educational level differences in both IA and cyberstalking as well, and in the association between these two concepts. More information about possible gender differences and educational level differences in these deviant online behaviors and in the associations between them can be used to further develop adequate prevention and intervention methods. If this study yields significant results concerning the variables under study in relation to cyberstalking, specific implications for prevention and intervention methods will be discussed in a latter section. With the increase and severity of the consequences of cyberstalking attacks (for both the victims and the perpetrators), the importance of identifying adolescents who are at risk for engaging in these behaviors cannot be stressed enough.
10
Method Sample
A total of 38 schools were approached to participate in a broader research project (focusing on adolescents and their online behavior), of which seven schools (18%) responded in a positive manner and participated. In total 1.071 students participated in the current study. A total of 63 students did not give consent (13), were inconclusive in giving consent (15), or did not complete the questionnaire (35). A total of 72 students did not meet the inclusion criteria regarding age (57) or educational level (15). That is, participants had to be of a minimal age of 12 and a maximum age of 18 years and had to follow secondary education (i.e. either low vocational level ‘VMBO’, higher educational ‘HAVO’ or pre-university level ‘VWO’). All these 135 students were therefore excluded resulting in a final sample of N = 927 (87% of the original 1.071 students). In average, the participants were 14.48 years old (SD = 1.17, range 12-18) and included both boys (N = 469; 51%) and girls (N = 459; 49%). The high-school students varied in educational level, from low (VMBO = 1; 35%) to high (HAVO; VWO = 2; 65%) educational level. An overview of demographic characteristics, as well as descriptive statistics, is presented in Table 1.
Procedure
The current study was part of a larger research project examining online behavior of adolescents and associated risk factors. The board of Ethics of the Department of Child Development and Education of the University of Amsterdam provided ethical approval for
this project (2017-CDE-7732). Schools were either approached through mail, phone or
face-to-face via a known contact person of one of the researchers. Each school was provided with a letter of information regarding the purpose of the study and the criteria to participate. All students obtained an extensive information letter about the study and afterwards they were given an option to decline participation. All parents of the students were informed via the schools as well. If parents refused their child to participate they could indicate this by using a consent form that was added to the information letter.
The participants received a link to an online questionnaire to measure, among other variables, cyberstalking and IA. The questionnaire was filled out during class that the high-school students attended. Teachers had been informed about the questionnaire before the classes took place. A researcher or an instructed teacher (and researcher available by phone) was present to answer any questions of the students.
11
Table 1
Demographic Characteristics and Descriptive Statistics (N= 927)
Variables Percentage (%) M S.D. Cronbach’s Alpha
Age 14.48 1.17 Gender 50.6* Educational level 65.3** Dutch Nationality 96.5*** PIUQ-SF-6 2.03 .71 .718 ECIPQ 1.20 .33 .811 Note:
*Males versus females (49.4%)
**Higher educational level versus lower educational level (34.7%)
***Other countries of origin included: Poland (.8%), Afghanistan (.3%), China (.3%), Belgium (.2%), France (.2%), Somalia (.2%), Burundi (.1%), Colombia (.1%), Eritrea (.1%), Greece (.1%), Ireland (.1%), Latvia (.1%), Morocco (.1%), Portugal (.1%), Turkey (.1%), United States of America (.1%), South-Africa (.1%)
Instruments
Demographic characteristics. Questions about gender (1 = male, 2 = female) and
educational level (1= ‘VMBO’ versus 2 = ‘HAVO’ or ‘VWO’) were asked in the first
segments of the questionnaire.
Internet addiction was measured with the ‘Problematic Internet Usage
Questionnaire-Short-Form’ (PIUQ-SF-6) provided by Demetrovics et al. (2016). The PIUQ-SF-6 uses six
questions to determine the level of IA (see Table 2). The questions were translated into Dutch and adjusted to fit the current period of time (Wissink & Rutten, 2017a), and scored on a 5-point Likert-type scale (1 = never at all and 5 = all the time). Demetrovics et al. (2016) examined the psychometric characteristics of the PIUQ-SF-6 and found a Cronbach’s alpha
of α =.77 for the total scale. The current study’s data showed a Cronbach’s alpha of α = .72
for the total scale (see Table 1).
Cyberstalking was measured with items of a translated and adjusted version of the
European Cyberbullying Intervention Project Questionnaire (ECIPQ; Brighi et al., 2012) and
one additional item (adjusted and translated version: ECIPQ-REV Dutch; Wissink & Rutten, 2017b). The ECIPQ consists of 22 questions concerning cyberbullying victimization and perpetration (Brighi et al., 2012). The ECIPQ is useful for the present study as it focuses on repeated online behavior and (partially) on perpetrators of cyberbullying. More similarities between cyberbullying and cyberstalking have previously been discussed. As this study focuses on perpetration of cyberstalking, only the questions concerning cyberbullying perpetration were included in the analysis (see Table 3.). The final questionnaire was
comparable with questions used in previous studies examining cyberstalking (DreBing et al., 2014; Finn, 2004; Reyns et al., 2011; Reyns et al., 2012). The questions were scored on a
12
Table 2
Questions based on the Problematic Internet Use Questionnaire Short-Form (PIUQ-SF-6 of Demetrovics et al., 2016; -translated and updated by Wissink & Rutten, 2017a)
English version Dutch version
1. How often do you spend time on the internet (including
Facebook/WhatsApp/Snapchat/games) while you’d rather sleep?
1. Hoe vaak besteed je tijd op ‘internet’ (dus ook:
Facebook/Whatsapp/Snapchat/games), terwijl je liever zou slapen?
2. How often do you feel tense, annoyed or stressed as a result of not spending the desired amount of time on the internet?
2. Hoe vaak voel je je gespannen,
geïrriteerd, of gestrest als je ‘internet’ niet zo lang kan gebruiken als je zelf wil? 3. How often have you tried to reduce the
amount of time you spend on the internet, but did not succeed?
3. Hoe vaak komt het voor dat je je
voorneemt om minder tijd ‘online’ door te brengen maar dat je dat niet wilt lukken?
4. How often have you tried to
conceal/mask the time you spend on the internet?
4. Hoe vaak probeer je te verbergen hoeveel tijd je ‘online’ doorbrengt?
5. How often do people complain about the
time you spend online? 5. Hoe vaak klagen mensen in jouw omgeving over dat je teveel tijd ‘online’ besteedt?
6. How often do you feel depressed, nervous or moody if you’re not spending time on the internet, and this feeling disappears as soon as you are able to go online again?
6. Hoe vaak komt het voor dat je je
depressief, chagrijnig of gespannen voelt wanneer je niet op het ‘internet’ bent en dat deze gevoelens stoppen zodra je weer online bent?
5-point Likert-type scale (1 = never, 2 = once every two months, 3 = once a month, 4 = once a week, 5 = multiple times a week). Del Rey et al. (2015) examined the ECIPQ resulting in an overall Cronbach’s alpha of α = .96 and a Cronbach’s alpha of α =.93 regarding the perpetration factor. A Dutch translation of the perpetration subscale of the ECIPQ was available and was used as a starting point for the Dutch version that was used in the current study (Brighi et al., 2012; Wissink & Rutten, 2017b). Small adjustments were made to make the perpetration subscale suitable for the current period of time. Finally, one additional item was included to fully capture the concept of cyberstalking, namely: ‘I have repeatedly contacted someone online or through social media even after that person asked me not too’. This item was retrieved from the study by Navarro et al. (2016). The importance of the added item is well reflected as it is used in multiple other studies examining cyberstalking (DreBing et al., 2014; Finn, 2004; Reyns et al., 2011; Reyns et al., 2012). Aforementioned item was translated into Dutch by the author and approved by the supervisor of this study. It was added to the adjusted questions from the ECIPQ and scored accordingly. With this addition the total
13
scale of 12 items were believed to capture the full spectrum of cyberstalking (see Table 3).
Analysis from the present study showed a Cronbach’s alpha of α =.81 for the total scale (see
Table 1).
Table 3
Questions regarding the measurement of Cyberstalking
English version Dutch version
1. I have said mean things to someone through social media or online.* 2. I have said mean things about
someone through social media or online.*
3. I have threatened someone through social media or online.*
4. I have hacked into someone’s account and stole personal information (e.g. through email or social networking accounts).*
5. I have hacked into someone’s account and pretended to be him/her (e.g. through WhatsApp or social networking accounts).*
6. I have made a fake account and pretended to be someone else (e.g. on Facebook or WhatsApp).*
7. I have put personal information about someone online.*
8. I have put embarrassing movies or pictures about someone online.* 9. I have edited pictures or movies of
someone that were put online.* 10. I have excluded someone or ignored
someone in a group app or on other social media.*
11. I have spread rumors about someone on the internet.*
12. I have repeatedly contacted someone online or through social media even after that person asked me not to.**
1. Ik heb gemene dingen tegen iemand gezegd via social media of online.* 2. Ik heb gemene dingen over iemand
gezegd tegen anderen via social media of online. *
3. Ik heb iemand bedreigd via social media of online.*
4. Ik heb het account van iemand gehackt en persoonlijke gegevens gestolen (bijv. via email of accounts van sociale netwerken).*
5. Ik heb het account van iemand gehackt en gedaan alsof ik hem/haar was (bijv. via WhatsApp of accounts van sociale netwerken).*
6. Ik heb een nep-account aangemaakt en gedaan alsof ik iemand anders was (bijv. op Facebook of Whatsapp).* 7. Ik heb persoonlijke gegevens over
iemand online gezet.*
8. Ik heb gênante filmpjes of foto’s van iemand online gezet. *
9. Ik heb foto’s of filmpjes van iemand die online waren gezet bewerkt.* 10. Ik heb iemand uitgesloten of
genegeerd in een groepsapp of op andere sociale media.*
11. Ik heb roddels over iemand verspreid op internet.*
12. Ik heb online of via social media contact gezocht met iemand, nadat diegene mij had gevraagd dit niet te doen.**
Note:
*European Cyberbullying Intervention Project Questionnaire (ECIPQ; Brighi et al., 2012). ** Navarro, Macrum, Higgins and Ricketts (2016)
Plan of analysis
14
of respondents. Next, multiple independent t-tests were used to determine possible gender and educational level differences in cyberstalking. Finally, regression analysis was used to determine the relationship between the independent variables (gender, educational level and IA) and cyberstalking. Besides, interaction terms were included to examine if there were gender differences or educational differences in the association between IA and
cyberstalking. This was done by performing a hierarchical regression analysis with at the first step gender and educational level, at the second step IA, and at the third step the interactions between the variables at step 1 and 2 (as independent variables) and cyberstalking as the dependent variable. Analysis of normality showed a non-normal distribution of the
cyberstalking data. However, in very large samples, non-normality should not be considered a major problem while analyzing and interpreting data (Field, 2009). Nevertheless, log transformation was used on the non-normal distributed cyberstalking data.
Results
In Table 1 descriptive statistics regarding the study variables are shown. As both scales showed good internal consistency, a mean score for each respondent was computed for both scales. Overall, a mean cyberstalking score of 1.20 out of 5.00 was obtained from the current sample (see Table 1).
Multiple independent t-tests were conducted to determine if there were differences regarding gender and educational level in cyberstalking. Results confirmed boys scored significantly higher on cyberstalking than girls, t(782) = 3.596, p <.001, while no significant educational level difference was found in cyberstalking (see Table 6). Analysis of data showed a non-equal variance distribution between boys and girls.Therefore, an additional Mann-Whitney test was performed with gender as independent, and cyberstalking as
dependent variable. This test confirmed a significant difference between both gender groups,
U = 97669.00, p <. 05. Furthermore, this study’s sample was skewed towards a high
educational level. As higher education forms in the Netherlands take one or two more years to complete, the possibility of age as a confounding variable was examined. A scatterplot with age at the X-axis and cyberstalking at the Y-axis was examined and showed a virtually horizontal line, indicating that cyberstalking scores remained similar throughout the studied ages (12-18 years). It therefore seems that age did not influence the results concerning cyberstalking. Consequently, age was not included as a controlling variable.
Table 4
Answers regarding Internet Addiction (N= 927)
15
Next, two separate independent t-tests with gender and educational level as an independent variable, and IA as dependent variable were performed. The results showed no significant differences between boys or girls, and also no differences between low or high educational level in IA (see Table 6). Moreover, a weak but significant linear correlation was found between IA and cyberstalking, r = .281, p < .000 (two-tailed).
Finally, a three stage hierarchical regression analysis was conducted with
cyberstalking as the dependent variable. Gender and educational level were entered at the first step3, IA at the second step and interaction effects between the three independent
variables at the third step. The hierarchical multiple regression results showed that the third step with the interaction effects did not contribute significantly to the explanation of
cyberstalking, F change = 1.610, p > .05. Therefore, the results of the first two steps are presented here. At step one, only gender contributed significantly to the regression model (β
= -.119, p <.001; see Table 7). Moreover, Model 1 accounted for 1.7% of the variation in
cyberstalking and this change in R2 was significant, F(2, 875) = 7.661, p < .05. Adding IA to
the regression model explained an additional 7.9% of the variation in cyberstalking and this change in R2 was significant as well, F(1, 874) = 76.470, p <.001.
Table 5
Answers regarding Cyberstalking (N= 927)
Questions Never Sometimes - Always
1. I have said mean things to someone through social media or online.
73.0% 25.7%
3 The variable cyberstalking showed serious violations of the assumption of normality of residuals. Therefore,
log transformation was used on the non-normal cyberstalking data. Comparison showed no to little improvement resulting in usage of non-transformed data in this study.
- Always 1. How often do you spend time on the internet (including
Facebook/WhatsApp/Snapchat/games) while you’d rather sleep? 19.1% 75.6% 2. How often do you feel tense, annoyed or stressed as a result of
not spending the desired amount of time on the internet? 42.7% 52.0% 3. How often have you tried to reduce the amount of time on the
internet? 38.8% 55.8%
4. How often have you tried to conceal/mask the time you spend on
the internet? 61.4% 33.0%
5. How often do people complain about the time you spend online? 40.7% 53.6% 6. How often do you feel depressed, nervous or moody if you’re not
spending time on the internet, and this feeling disappears as soon as you are able to go online again?
71.5% 22.9%
Note: Not all questions were answered by all 927 participants. For that reason, above mentioned percentages don’t add up to 100%
16 2. I have said mean things about
someone through social media or online.
3. I have threatened someone through social media or online.
4. I have hacked into someone’s account and stole personal
information (e.g. through email or social networking accounts). 5. I have hacked into someone’s
account and pretended to be
him/her (e.g. through WhatsApp or social networking accounts).
6. I have made a fake account and pretended to be someone else (e.g. on Facebook or WhatsApp). 7. I have put personal information
about someone online.
8. I have put embarrassing movies or pictures about someone online. 9. I have edited pictures or movies of
someone that were put online. 10. I have excluded someone or
ignored someone in a group app or on other social media.
11. I have spread rumors about someone on the internet. 12. I have repeatedly contacted
someone online or through social media even after that person asked me not to. 73.5% 94.5% 95.5% 96.1% 94.2% 95.8% 82.3% 93.6% 84.3% 91.6% 94.8% 25.3% 4.5% 3.4% 2.9% 4.8% 3.2% 17.1% 5.5% 15.1% 7.3% 4.1%
Note: Not all questions were answered by all 927 participants. For that reason, above mentioned percentages don’t add up to 100%
All three4 included variables in Model 2 showed significant value for the explanation
of cyberstalking, with IA as the most important variable (β = .282, p <.001). The positive beta shows that when adolescents reported a higher level of IA, they also reported a higher level of cyberstalking. Moreover, the beta’s of gender and educational level show boys and participants with lower educational levels to have significantly higher scores on
cyberstalking.
4 As educational level was only significant in model two, the possibility of educational level as moderator
variable has been explored. An additional simple linear regression was performed with an interaction effect between educational level and IA as independent variable, and cyberstalking as dependent variable. Results showed a significant effect , F(1, 891) = 32.735, p <.000, implying that educational level poses as a moderator variable on the relationship between IA and cyberstalking.
17
Discussion
The results of the current study provide us information regarding multiple possible explanatory variables for cyberstalking. Internet addiction (further: IA) turned out to have the highest significant value for the explanation of cyberstalking in itself, but also when controlled for the other predictor variables gender and educational level. When looking at the results concerning gender, the results showed that boys scored significantly higher on
cyberstalking than girls. Results also showed gender to have unique explaining value for cyberstalking (i.e. when controlled for IA and educational level). The same applies for educational level, because educational level had explaining value for cyberstalking, even
Table 6
Results regarding Multiple Independent T-tests
Dependent var. Independent var. M SE t Df Sig (2-tailed)
Cyberstalking Gender 3.596 782 .000* Boys 1.23 .387 Girls 1.16 .158 Educational level 1.546 587 .123 High 1.22 .310 Low 1.18 .351 Internet addiction Gender .587 878 .557 Boys 2.04 .728 Girls 2.02 .696 Educational level 1.216 878 .224 High 2.05 .701 Low 1.99 .732 Note: *Significant at p < .001 Table 7
Summary of Hierarchical Regression Analysis for Variables explaining Cyberstalking
Model Variables included R square
change Change F change Sig. F Beta Sig.
1 .017 7.661 .001** Gender -.119 .000*** Educational level -.058 .085 2 .079 76.470 .000*** Gender -.113 .000*** Educational level -.070 .031* IA .282 .000*** 3 .005 1.610 .185
18
when controlled for IA and gender. Moreover, the results of the analyses indicate that adolescents with lower educational levels reported higher levels of cyberstalking behavior. Overall, it seems that boys in the lower educational levels and with more characteristics of IA have the greatest chances of exhibiting cyberstalking behavior.
Navarro et al. (2016) first examined the possibility of an association between IA and cyberstalking. The current findings are in line with results found in their study. In other words, IA positively correlates with cyberstalking and explains around 8% of cyberstalking behavior. This means that someone who reported higher levels of IA, also reported higher levels of cyberstalking behavior. As the internet is still a growing phenomenon which increasingly attracts individuals from all areas (Armstrong, Phillips, & Sailing, 2000), it can be expected that more and more individuals will get addicted to the internet, which would mean that cyberstalking behavior could increase as well. The fact that adolescents are still developing their cognitive control and boundary setting skills (Casey et al., 2005; Leung, 2008; Liu & Potenza, 2007), can further increase their chances of becoming addicted to the internet and also their chances of engagement in cyberstalking behavior. Furthermore, Hitchcock (2006) stated that cyberstalking can turn into offline stalking if obsession for a victim exceeds their need fulfillment online. Adolescents are, on average, more impulsive, less able to see the consequences of their actions and more prone to engage in riskier
behavior compared to adults (Gardner & Steinberg, 2005; Scott, 1992; Steinberg et al., 2008). Therefore, their chances of showing cyberstalking behavior and perhaps also of cyberstalking behavior turning into physical stalking might be higher. Further research should contemplate on this. To summarize, adolescents possibly have a higher risk of becoming addicted to the internet and, following results from this study, this could mean an increase in cyberstalking behavior as well.
Results of the current study further indicate that, compared to girls, boys are
predominantly involved in cyberstalking behavior. Gender seems to have significant value for the explanation of cyberstalking, meaning the chances of being male and getting involved in cyberstalking are higher compared to the chances of being female and exhibiting
cyberstalking behavior. These results are comparable to other studies investigating gender and cyberstalking. For example, Reyns et al. (2011) summarized from WHOA’s statistics (2008) and concluded perpetrators of cyberstalking to be primarily male, who had been in a prior relationship with the victim and often lived relatively far away from their victim. Moreover, Thorp (2004) and Ha and Hwang (2014) stated that females leant more towards online relational harassment, while males tended to sexually harass their victims online.
19
These findings, in combination with the results found in the current study, could perhaps be used in the establishment of some kind of a ‘cyberstalker profile’, which will be helpful in setting up prevention and intervention methods.
This study also shows individuals with lower educational level to exhibit more deviant online behavior characteristic of cyberstalking, like hacking into someone else’s account or contacting someone even after he/she asked this individual to stop. These findings are similar to the results of a study in which perpetrators of cyberstalking were only found in low educational levels (Combee, 2008). However, this finding contradicts with what other researchers have claimed. For example, Navarro et al. (2016) reported that cyberstalkers are usually well-educated and could be ‘the best performing students in the classroom’. This distinction in results might be due to different participation groups. While the current study’s researcher tried to reach participants from all over the Netherlands, it’s sample might not correspond with the participants used in the study by Navarro and colleagues (2016) who were originated in a rural country in western North-Carolina. Moreover, Navarro et al (2016) included students that were between 9th graders and 12th graders with various Grade Point
Average (GPA) scores. Consequently, this study’s sample has included younger students (roughly students that were between 7th graders and 12th graders) which could have yielded
different results. However, the current study showed that age did not seem to affect the results concerning cyberstalking, it is expected that the contradicting results from the current study and that of Navarro et al. (2016) are not due to sample differences regarding age. However, geographical dissimilarities might have caused this deviation in results. Still, it is recommendable to take into consideration that individuals with lower educational levels could form a significant proportion of perpetrators of cyberstalking. To further elaborate on this, Gregorie (2001) stated that most of the prosecuted cyberstalking cases ‘did not involve technically complex forms of stalking’. In fact, cyberstalkers tended to ‘simply’ use
anonymous re-directed e-mails combined with some software to cover their tracks. Navarro and colleagues (2016) might overestimate perpetrators of cyberstalking by stating they are more sophisticated and therefore are able to outwit computer systems and exhibit
cyberstalking behavior through multiple technological pathways. The current study seems to point more in the direction of Gregorie (2001) describing perpetrators of cyberstalking to have lower educational levels who might have no necessity for high computer literacy to engage in cyberstalking behavior.
Though findings from this study provide more insight in the explanatory variables of cyberstalking, future research is essential in capturing more aspects of both IA and
20
cyberstalking. For instance, this study did not discriminate between different types of
cyberstalkers. McFarlane and Bocij (2003) aimed to identify and describe different categories of cyberstalking perpetrators. Their examination of 24 participants, aged between 14 and 67 years, resulted in four major categories: the vindictive cyberstalker, the composed
cyberstalker, the intimate cyberstalker, and the collective cyberstalker. Future studies should examine how IA is related to the above mentioned categories of cyberstalkers, while
additionally paying attention to possible differences in gender and educational background between the different ‘types’ of cyberstalkers. On a side note, McFarlane and Bocij (2003) had a predominantly female sample (22 females against two males). Also in light of the current study’s findings that males seem more prone to report cyberstalking behavior, forthcoming studies should examine if the categories McFarlane and Bocij (2003) proposed are similarly applicable to women and men and/or girls and boys. Likewise, more research should take place to see if different types of (extreme) internet usage, like for example gaming or social networking, yield different results on the association with cyberstalking. Previous studies indicated that females tend to use the internet more for communication platforms as chat rooms (74%, Shariff & Gouin, 2006), while males tend to use the internet more for gaming (Ha & Hwang, 2014). It should be examined if certain types of internet addicts, for instance gamer addicts versus social media addicts, are more prone to
cyberstalking behavior than others.
Furthermore, as previously discussed, cyberstalking has quite some overlap with cyberbullying. The current literature provides no clear distinction between the two concepts. If cyberbullying and cyberstalking remain understood as two separate concepts, a clear distinction between the two should be established and future studies should be focused on the further examination of the differences (in explaining factors, in consequences, etc.). This will also help researchers to develop an adequate cyberstalking questionnaire and to investigate cyberstalking more accurately. A more precise demarcation of both cyberstalking and
cyberbullying (and the possible associated legal consequences) is also important from a legal perspective. Therefore, it will be of high interest to find consensus on this topic.
Finally, acknowledgment of shortcomings of this study is in place. First, to the best of the author’s knowledge, there was no preset-questionnaire available regarding cyberstalking. Previous studies examining cyberstalking mostly used one or two questions retracted from relevant publications (DreBing et al., 2014; Finn, 2004; Navarro et al., 2016; Reyns et al., 2011; 2012). Those researchers seemed to agree upon the importance of including one particular question reflecting cyberstalking (namely: repeatedly seeking contact after
21
someone has asked you not to). In the present study, this question was added to an existing questionnaire regarding cyberbullying. It showed good inter-item-correlations and good internal consistency, meaning all questions had high probability of measuring the same concept. It can be discussed, however, that it is unclear which concept the questionnaire exactly measured though,as the questions used in the current study also resemble less serious forms of cyberstalking which can be placed under cyberbullying as well. Following this, the validity of the measurement of cyberstalking in the current study could be lower than desired.
Second, the sample of the current study showed a non-normal distribution of the cyberstalking data. Although expected, student’s cyberstalking scores turned out to be skewed to the right, meaning most students scored relatively low on cyberstalking (see Table 1). Field (2009) argued that sample sizes above 40 respondents can counteract non-normal distribution of data. Even though this study far exceeded this referred sample size, and transformation of data did not result in different conclusions, it is necessary to interpret results in light of this limitation.
Third, this study was part of a larger study. Though participating researchers tried to minimalize the number of questions included, it cannot be argued that students were asked to put in quite an effort of concentration. Cape and Phillips (2015) stated that the ideal length of a questionnaire does not exceed beyond 23 minutes, otherwise fatigue can affect the quality of the data gained from the questionnaire. The current study’s questionnaire completion time was estimated between 30 and 45 minutes, which does not reflect this ideal length.
Furthermore, the questions for this study were included in the latter part of the overall questionnaire. Cape and Phillips (2015) stated that a respondent spends more time on
questions positioned earlier in a questionnaire. Both the length of the questionnaire as well as the position of the questions in the questionnaire could have affected the data, and therefore the results found. Nevertheless, both scales showed adequate to good reliability which provides confidence in the results found in the current study.
Finally, the possibility of students giving socially desirable answers should not be neglected. Krumpal (2011) argued that sensitive questions which refer to behaviors that are taboo, illegal or socially sanctioned are more often affected by the social desirability bias. This implies that there is a reasonable chance of underestimation of the true prevalence and frequency of the examined behavior. As cyberstalking can be classified as undesirable and illegal behavior, odds are reasonable that respondents answered in a socially desirable
manner. The researchers involved in the current study tried to minimize the social desirability bias by reducing the presence of the interviewer (i.e. individually answering questions on the
22
computer) and ensuring the anonymity of the respondent (i.e. not asking for the respondent’s name). Krumpal (2011) described these measures as effectively reducing deliberate
misreporting on sensitive topics. As this may be, it is important to keep in mind that results from this study could be an underestimation of true cyberstalking behavior among
adolescents and that the association between IA and cyberstalking could be inflated, because of the social desirability bias (i.e. adolescents who are more honest in reporting –high- on IA, will also be more honest in reporting –high- on cyberstalking, so that the association between the two will be higher because of this artifact).
Despite these limitations, results from the current study provide first glances of a much needed insight in cyberstalking behavior among adolescents. With the increase of internet used in our daily lives, the relatively ‘fragile’ or impulsive state of mind most teenagers are in, and the ease to commit online negative behavior (fairly) anonymous, it is important to keep up with the newest ‘trends’ among adolescents. Explaining abnormal behavior like cyberstalking is necessary and this study provides handles to do so by giving insight in the variables that provide an explanation for cyberstalking. Future research should focus on further specifying the profile of perpetrators of cyberstalking as this may provide more specific risk factors and risk factors could be more easily recognized than cyberstalking behavior itself. This is possible, for instance, by examining the different categories of
cyberstalkers mentioned by McFarlane and Bocij (2003) and the explaining variables for each category. Furthermore, as the current study’s findings support the assumption of a correlation between IA and cyberstalking, future research could also focus on different types of IA and cyberstalking behavior.
With the determination of an at-risk-group of cyberstalking, parents, teachers and other guardians can prematurely interfere when they recognize certain risk factors. Hence, prevention methods can be used to target individuals at risk of engaging in cyberstalking behavior, and consequently minimize chances of behaviors escalating into cyberstalking. For instance, redirecting individuals at risk for cyberstalking towards other stimulating and positive activities like sports, music or other extracurricular activities at school could perhaps deflect from excessive use of the internet and cyberstalking behavior. Moreover, school psychologists could take an active role in educating students (and teachers) on cyberstalking and intervening with students at-risk for cyberstalking behavior, as well as developing school guidelines regarding the manners in which cyberstalking behavior should be addressed. Setting up these guidelines will prove to be difficult as the border between ‘normal adolescent behavior’ and punishable behavior is unclear and hard to establish. The phrase
23
‘children should be children’ should not be taking lightly as adolescents still have to figure out social boundaries and appropriate behaviors, and in that light should be giving the opportunity to do so without too much interference from adults. In other words, adolescents should be able to explore life, make mistakes and learn from these experiences. So, where do we draw the line? What behaviors should we prohibit? Cyberstalking can have a major impact on its victims so perhaps we should follow the Criminal Code of The Netherlands regarding cyberstalking (i.e. harassment). Accordingly, guidelines could state that it is mainly necessary to act against online behavior if it is intentional, repeated, unwanted and with the intention to force someone to do, or withhold them from doing something or to cause fear. School employees could then intervene by first finding out who perpetrates these behaviors, address them and (individually) explore ways to minimize chances of recidivating. Moreover, intervention methods can be better applied if underlying behaviors, like IA, can be treated as well or at least will be taken into account when treating cyberstalking behavior.
The current study shows the importance of examining cyberstalking behavior. Limited research concerning the subject is available, but this study’s results underline the necessity of exploring and discussing cyberstalking further. If future research focuses on specifying the profile of perpetrators of cyberstalking and the correlates, this behavior might be better prevented or at least better minimized in the future.
24
References
Acier, D., & Kern, L. (2011). Problematic internet use: Perceptions of addiction counsellors.
Computers & Education, 56, 983-989.
Adam Savader Plea Agreement (2013). Retrieved from
https://www.scribd.com/document/185709216/Adam-Savader-Plea-Agreement#from_embed
Armstrong, L., Phillips, J. G., & Sailing, L. L. (2000). Potential determinants of heavier Internet usage. International Journal of Human-Computer Studies, 53,537-550.
Bakken, I. J., Wenzel, H. G., Götestam, K. G., Johansson, A., & Øren, A. (2009). Internet addiction among Norwegian adults: a stratified probability sample study. Scandinavian
Journal of Psychology, 50, 121-127.
Burke, S. C., Wallen, M., Vail-Smith, K., & Knox, D. (2011). Using technology to control intimate partners: An exploratory study of college undergraduates. Computers in
Human Behavior, 27, 1162-1167.
Brighi, A., Ortega, R., Scheitauer, H., Smith, P.K., Tsormpatzoudis, C., Barkoukis, V., Del Rey, R., et al. (2012). European Cyberbullying Intervention Project Questionnaire (ECIPQ). University of Bologna. Unpublished Manuscript.
Cape, P., & Phillips, K. (2015). Questionnaire length and fatigue effects: the latest thinking and practical solutions. White paper. Retrieved from
www.surveysampling.com/site/assets/files/1586/questionnaire-length-and-fatiigue-effects-the-latest-thinking-and-practicalsolutions.pdf.3
Casey, B.J., Tottenham, N., Liston, C., & Durston, S. (2005). Imaging the developing brain: what have we learned about cognitive development? Trends in Cognitive Sciences, 9, 104-110.
Combee, K. H. (2008). Cyberpesten of Cyberstalken? Een onderzoek naar cyberpesten en cyberstalking onder middelbare scholieren in Nederland. Retrieved from https://vubis.politieacademie.nl/webopac/FullBB.csp?WebAction=ShowFullBB&Enco dedRequest=ny*D3*3EZ*09f*AC*B7*AE*D1r*85*96*83*B0&Profile=Default&Opa cLanguage=dut&NumberToRetrieve=10&StartValue=9&WebPageNr=1&SearchTerm 1=2008.2.71383&SearchT1=&Index1=Index1&SearchMethod=Find_1&ItemNr=9 Criminal Code of the Netherlands. (2017). Retrieved from
http://wetten.overheid.nl/BWBR0001854/2017-09-01
Deirmenjian, J. (1999). Stalking in Cyberspace, Journal of the American Academy of
25
Del Rey, R., Casas, J. A., Ortega-Ruiz, R., Schultze-Krumbholz, A., Scheithauer, H., Smith, P., & Guarini, A. (2015). Structural validation and cross-cultural robustness of the European Cyberbullying Intervention Project Questionnaire. Computers in Human
Behavior, 50, 141-147.
Demotrovics, Z., Szeredi, B., & Rózsa, S. (2008). The three-factor model of Internet addiction: The development of the Problematic Internet Use Questionnaire. Behavior
Research Methods, 40, 563-574.
Demetrovics, Z., Király, O., Koronczai, B., Griffiths, M. D., Nagygyörgy, K., Elekes, Z., & Urbán, R. (2016). Psychometric properties of the problematic internet use questionnaire short-form (PIUQ-SF-6) in a nationally representative sample of adolescents. PloS one, 11.
D’Ovidio, R., & Doyle, J. (2003). A study on cyberstalking: Understanding Investigative Hurdles. FBI Law Enforcement Bulletin, 73, 10-17.
Dreβing, H. M.D., Bailer, J., Anders, A., Wagner, H., & Gallas, C. (2014). Cyberstalking in a large sample of social network users: Prevalence, characteristics, and impact upon victims. Cyberpsychology, Behavior, and Social Networking, 17, 61-67.
Field, A. (2009). Discovering statistics using SPSS. Sage publications.
Finn, J. (2004). A survey of online harassment at a university campus. Journal of
Interpersonal Violence, 19, 468-483.
Gardner, M., & Steinberg, L. (2005). Peer influence on risk taking, risk preference, and risky decision making in adolescence and adulthood: An experimental study. Developmental
Psychology, 41, 625–635.
Gregorie, T. (2001). “Cyberstalking: Dangers of the information Superhighway”. National Center for Victims of Crime. Retrieved from
http://www.victimsofcrime.org/docs/src/cyberstalking—dangers-on-the-information-superhighway.pdf.
Griffiths, M., Rogers. L., & Sparrow, P. (1998). Crime and IT: Stalking the Net. Probation
Journal, 45, 138-141.
Griffiths, M. (2001). Sex on the Internet: Observations and implications for Internet sex addiction. The Journal of Sex Research, 38, 333-342.
Ha, Y.-M., & Hwang, W. J. (2014) Gender differences in Internet addiction associated with psychological health indicators among adolescents using a national web-based survey.
26
Herrera-López, M., Casas, J. A., Romera, E. M., Ortega-Ruiz, R., & Del Rey, R. (2017). Validation of the European Cyberbullying Intervention Project Questionnaire for Colombian Adolescents. Cyberpsychology, Behavior, and Social Networking, 20, 117-125.
Hetzel-Riggin, M., & Pritchard, J. R. (2011). Predicting problematic internet use in men and women: The contributions of psychological distress, coping style, and body esteem. Cyberpsychology, Behavior and Social Networking, 14, 519–523.
Hitchcock, J.A. (2006). Net Crimes and Misdemeanors: Outmaneuvering Web Spammers, Stalkers and Con Artists, New Library World, 108, 89-90.
Kaltiala-Heino, R., Lintonen, T., & Rimpelä, A. (2004). Internet addiction? Potentially problematic use of the Internet in a population of 12–18 year-old adolescents. Addiction
Research & Theory, 12, 89-96.
Ko, C. H., Yen, J. Y., Chen, C. S., Yeh, Y. C., & Yen, C. F. (2009). Predictive values of psychiatric symptoms for Internet addiction in adolescents: A 2-year prospective study.
Archives of Pediatrics and Adolescent Medicine, 163, 937–943.
Kormas, G., Critselis, E., Janikian, M., Kafetzis, D., & Tsitsika, A. (2011). Risk factors and psychosocial characteristics of potential problematic and problematic internet use among adolescents: a cross-sectional study. BMC public health, 11, 595.
Krumpal, I. (2011). Determinants of social desirability bias in sensitive surveys: a literature review. doi: 10.1007/s11135-011-9640-9
Kuss, D. J., Van Rooij, A. J., Shorter, G. W., Griffiths, M. D., & Van de Mheen, D. (2013). Internet addiction in adolescents: Prevalence and risk factors. Computers in Human
Behavior, 29, 1987-1996.
Laconi, S., Rodgers, R. F., & Chabrol, H. (2014). The measurement of Internet addiction: A critical review of existing scales and their psychometric properties. Computers in
Human Behavior, 41, 190-202.
Langhinrichsen-Rohling, J., Selwyn, C., & Rohling, M. L. (2012). Rates of bidirectional versus unidirectional intimate partner violence across samples, sexual orientations, and race/ethnicities: A comprehensive review. Partner Abuse, 3, 199-230.
Leung, L. (2008). Leisure boredom, sensation seeking, self-esteem, and addiction. Mediated
interpersonal communication, 359-381.
Liu, T. C., & Potenza, M. N. (2007). Problematic Internet use: clinical implications. CNS
27
Liu, T. C., Desai, R. A., Krishnan-Sarin, S., Cavallo, D. A., & Potenza, M. N. (2011). Problematic Internet use and health in adolescents: data from a high school survey in Connecticut. The Journal of Clinical Psychiatry, 72, 836.
Macur, M., Király, O., Maraz, A., Nagygyörgy, K., & Demetrovics, Z. (2016). Prevalence of problematic internet use in Slovenia. Slovenian Journal of Public Health, 55, 202-211. McFarlane, L., & Bocij, P. (2003). An exploration of predatory behavior of cyberspace:
Towards a typology of cyberstalkers. First Monday, 9, 1-13.
Meerkerk, G.-J., Van Den Eijnden, R. J. J. M., Vermulst, A. A., & Garretsen, H. F. L. (2009). The Compulsive Internet Use Scale (CIUS): Some psychometric properties.
Cyberpsychology & Behavior, 12, 1-6.
Navarro, J. N., Marcum, C. D., Higgins, G.E., & Ricketts, M. L. (2016). Addicted to the thrill of the virtual hunt: Examining the effects of Internet addiction on the cyberstalking behaviors of juveniles. Deviant Behavior, 37, 893-903.
Olweus D. (2012) Cyberbullying: an overrated phenomenon? European Journal of
Developmental Psychology, 9, 520–538.
Paulhus, D. L. (2003). Self-presentation measurement. In: Fernandez-Ballesteros, R. (ed.).
Encyclopedia of Psychological Assessment, 858-860.
Paullet, K. L., Rota, D. R., & Swan, T. T. (2009). Cyberstalking: An exploratory study of students at a Mid-Atlantic University. Issues in Information Systems, 10, 640-649. Reyns, B. W., Henson, B., & Fisher, B. S. (2010). A situational crime prevention approach to
cyberstalking victimization: Preventive tactics for Internet users and online place managers. Crime Prevention and Community Safety, 12, 99-118.
Reyns, B. W., Henson, B., & Fisher, B. S. (2011). Applying Cyberlifestyle-Routine Activities Theory to cyberstalking victimization. Criminal Justice and Behavior, 38, 1149-1169. Reyns, B. W., Henson, B., & Fisher, B. S. (2012) Stalking in the twilight zone: Extent of
cyberstalking victimization and offending among college students. Deviant Behavior,
33, 1-25.
Scott, E. S. (1992). Judgment and reasoning in adolescent decisionmaking. Villanova Law
Review, 37, 1607–1669.
Shariff, S., & Gouin, R. (2006). Cyber-dilemmas: Gendered hierarchies, new technologies and cyber-safety in schools. Atlantis: Critical Studies in Gender, Culture & Social
Justice, 31, 27-37.
Smith, P. K. (2015). The nature of cyberbullying and what we can do about it. Journal of
28
Steinberg, L., Albert, D., Cauffman, E., Banich, M., Graham, S., & Woolard, J. (2008). Age differences in sensation seeking and impulsivity as indexed by behavior and self-report: Evidence for a dual systems model. Developmental Psychology, 44, 1764–1778.
Stephens, G. (1995). Crime in Cyberspace. Futurist, 29, 24-29.
Thorp, D. (2004) Cyberbullies on The Prowl in Schoolyard. Retrieved from http://australianit.news.com.au/articles/0,7204,9980900^15322
^^nbv^15306,00.html
Tokunaga, R. S. (2010). Following you home from school: A critical review and synthesis of research on cyberbullying victimization. Computers in Human Behavior, 26, 277–287. http://dx.doi.org/10.1016/j.chb.2009.11.014
Tsitsika, A. K., Andrie, E. K., Psaltopoulou, T., Tzavara, C. K., Sergentanis, T. N., Ntanasis-Stathopoulos, I., Richarson, B. C., Chrousos, G. P. & Tsolia, M. (2016). Association between problematic internet use, socio-demographic variables and obesity among European adolescents. The European Journal of Public Health, 26, 617-622.
Veenstra, S. (2011). Cyberbullying: an explanatory analysis. (Unpublished doctoral thesis). Retrieved from
https://www.politieacademie.nl/kennisenonderzoek/kennis/mediatheek/PDF/82738.pdf Working to Halt Online Abuse (2009). 2008 Cyberstalking Statistics. Retrieved from
http://www.haltabuse.org/resources/stats/2008Statistics.pdf.
Wissink, I. B., & Rutten, E. A. (2017a). Translation of the Problematic Internet Usage
Questionnaire. Internal Document. Amsterdam: University of Amsterdam.
Wissink, I. B., & Rutten, E. A. (2017b). Translation of the European Cyberbullying
Intervention Project Questionnaire. Internal Document. Amsterdam: University of
Amsterdam.
Wu, C.-Y., Lee, M.-B., Liao, S.-C., & Chang, L.-R. (2015). Risk factors of Internet addiction among Internet users: An online questionnaire survey. PLoS ONE, 10, e0137506. doi:10.1371/journal.pone.0137506.
Yen, J., Ko, C., Yen, C. F., Wu, H. Y., & Yang, M. J. (2007). The comorbid psychiatric symptoms of Internet addiction: Attention deficit and hyperactivity disorder (ADHD), depression, social phobia, and hostility. Journal of Adolescent Health, 41, 93–98.
Young, K. (1998). Internet addiction: The emergence of a new clinical disorder.
