Research of Attitudes toward Online Violence—Significance of Online Media and Social Marketing in the Function of Violence Prevention and Behavior Evaluation

Abstract

The main aim of this paper is to analyze attitudes toward online violence in order to identify the current state of this problem through psychosocial analysis, as well as the possibility of preventing this kind of violence through the prism of online media and social marketing. The authors developed a conceptual model and conducted an online quantitative research study. The survey was conducted among 277 respondents in Montenegro, which is a developing country and as such is experiencing transitions in many segments. The data were analyzed using standard statistical tests, an analysis of variance (ANOVA) test, logistic regression and the Relative Importance Index. The results of the research showed that specific vulnerable groups such as women are more exposed to online violence. The results of the logistic regression revealed that female respondents aged 18 to 24 were 76.3% more exposed to online violence than respondents over 55. Regarding the aspect of marital status, the results showed that divorced women were most exposed to online violence, 74% more than widows, while single respondents had a 54.2% greater chance of being a victim of online violence compared to widows. In terms of education, the results of the research highlighted that female respondents with a higher level of education were significantly less likely to be victims of online violence than those with a lower level of education. Respondents who had completed doctoral studies had a 48.6% lower chance of being victims of online violence compared to the respondents who had only completed high school. In terms of online media being used as a tool for spreading online violence, 36.89% of respondents identified Facebook, while 21.61% identified Instagram as the main medium of this type of violence. Thus, the research identified the necessity of continuous work focused on informing the population about online violence, as well as raising awareness about measures that should be carried out in order to reduce online violence, because of the significant influence of online media on forming attitudes toward online violence, which was confirmed by 85.19% of respondents. Furthermore, the results showed that online media—in particular, social marketing—may have a key role in the prevention of online violence, as its preventive role was confirmed by 96.76% of respondents. The findings of the research may be extremely useful for different policymakers in order to help them to better understand and identify cases of online violence, especially against women, as well as to create adequate strategies for its prevention through the prism of online media and social marketing.

1. Introduction

The processes of globalization, internationalization and the implementation of information and communication technologies (ICT) have significantly changed people’s lives, and the world has become one huge “global village” [1]. ICT technologies have many benefits from many different points of view, especially considering their informative and educational function. The number of ICT users is constantly growing, communication is becoming much faster and easier than ever before and new channels of communication are being introduced [2].

Although there are many benefits from the implementation of ICT technologies, there are also a large number of challenges. One of the most important of these is violence, which represents an old–new phenomenon that has been allowed to exist not only in the real but also in the virtual world. The characteristics and consequences of traditional and online violence are almost the same, and they are mostly realized simultaneously and coexist [3].

However, according to some authors, online violence is even more dangerous because of the fact that it is much more difficult to escape because of the impossibility of feeling safe in any place, even in the places in which victims should feel safest, such as at home [4,5]. Online violence is not limited by place or time [6], while the general public, which often witnesses online violence, makes victims feel more helpless and unprotected [3,7]. Additionally, online media communication is characterized by greater anonymity, which encourages unconstrained aggressive forms of commenting and the posting of offensive messages, which boosts violence even more online compared to the real environment [8,9,10]. Content is made available on the Internet very quickly, and there is a large number of people who may be witnesses of violence [4,5], as well as passive bullies. As the research of Barlińska et al. suggests [11], the online environment increases the likelihood of negative bystander behavior through the sharing of offensive posts and photos due to the absence of direct contact with the victim, which reduces the sense of empathy and guilt. The targets of online violence are both children and adults, and in addition, a bully very often does not have a picture of the extent to which they are hurting the victim and causing damage [12]. At the same time, the bully is very often anonymous, and it is not easy to identify them, which makes them more confident because there is no fear of being sanctioned [13].

Online violence may be defined as any kind of communication, including the exchange of textual, audio or video content, via the Internet that endangers the safety of an individual or makes them feeling frightened and upset [14]. It may take many forms, such as sending aggressive or threatening emails and messages, sharing personal photos and videos without approval, Internet trolling, hacking and Internet monitoring. The most common types of online violence are violence in social media, cyberbullying, cyberstalking, harassment, gang violence, cyber-suicide, etc. [15]. In the last few years, a new phenomenon called Internet banging has appeared, in which social media sites such as Facebook, Twitter and YouTube are used to trade insults or make threats, which may end in homicide or victimization [15]. The term cyberbullying represents a type of bullying in which different technologies are used, such as “cell phones, computers and tablets, as well as communication tools including social media sites, text messages, chat and websites, willful and repeated harm inflicted through the use of computers and other electronic devices” [16].

Some authors [17] distinguish between direct and indirect online violence. Direct online violence includes sending disturbing messages, insults and threats on social networks directly to a victim; publishing private data, photos or videos without approval; impersonation; etc. In contrast, indirect violence is realized through a third party (the account or profile of another person, etc.). With the continuous and rapid progress of ICT technologies, new types of violence against women are also appearing. As the research of Lewis et al. [18] indicated, women engaged in feminism are especially exposed to different forms of online violence, even more than in the offline world. Although their exposure to online violence depends on the time spent online, more than 80% of feminists that are highly active online media users experienced flaming, trolling and harassment.

Since the beginning of the 21st century, a large number of research papers and publications regarding online violence have appeared, and most of them refer to adolescents [19]. In recent years, the focus has been put on women as well. The number of studies on Montenegro is quite small, and they predominantly refer to online violence toward women as well as children.

One of the most important studies is the research of the European Union Agency for Fundamental Rights from 2015 and Organization for Security and Co-operation in Europe (OSCE) [20], which is about violence. However, this research predominantly refers to traditional violence, while online violence is included to a lesser extent. In 2017, the European Institute for Gender Equity published a survey that refers exclusively to online violence toward women and girls [21]. In 2019, NGO (Non-governmental organization) Safe Women’s House carried out the project “Stop the connection to online violence”, which was supported by the European Union and carried out by The United Nations Development Program (UNDP) and included research among 700 women and girls in the capital city of Podgorica regarding online violence toward women and girls [22]. Although this research provided some significant results regarding this issue, its main disadvantage was the fact that it was carried out only in the capital city, while the other towns in Montenegro were not included, as well as the fact that the respondents were only women and girls which, in the opinion of authors of this research, significantly diminishes its importance and relevance.

In 2016, UNICEF and the Government of Montenegro presented the publication “Children online—possibilities, risks and safety” which presents the results of the research regarding online violence toward children [23]. The research was used as an introduction for UNICEF global campaign “Stop the violence online” which was realized in Montenegro as well.

Online violence is, according to the mentioned studies, a common and present phenomenon in Montenegro; on the basis of the results of the research carried out in Montenegro, it was concluded that there is neither a significant amount of research on online violence nor scientific papers regarding this issue and that the research mostly focuses on the problem of online violence toward women, girls and children. In addition, recent researchers only examined the opinions, attitudes and perceptions of females and children, while the male population is completely excluded from research. There are also no scientific data about online violence toward specific marginalized groups, which could also be a very interesting part of the research. On the other side, studies regarding children were focused on violence in general, while online violence was only considered as a segment and was not given appropriate importance. However, the research on the topic of online media and online violence impacting children is of great importance, considering the effects that the virtual world has on the creation of their attitudes and behavior. As the studies by Castro and Osório and by Gentile et al. suggest, the social and online media pressures have a strong influence on adolescents’ social norms and the adopted patterns of behavior, including negative ones [24,25].

Online media may be considered a very powerful tool that influences opinions, perceptions and attitudes in different areas, so it is expected that their influence regarding online violence is significant as well. Specifically, in recent decades, the influence of online media has been rapidly growing, and they have become an integral part of everyday routine in the modern world [26]. In fact, they represent a real revolution in the communication process [27,28] and one of the most crucial elements of modern human society, which significantly influences not only its social norms and values but its culture as well [29,30]. In these circumstances, the concept of social marketing is gaining importance with the emergence of new paradigms, discourses, approaches and perspectives [31], and it has significantly broadened and deepened its scope [32] especially in the last decade. Many authors state that online media may shape attitudes and opinions not only in a positive way, but also in a negative way, and they bring many negative effects to their consumers and society as a whole [33,34,35,36,37]. The study by Barter et al. [38] indicated that online media tools (such as social networks) can even increase the number of ways in which bullies can stalk and harm their victims in real life. Additionally, the research of Anderson et al. [39] revealed that the impact of online media on the adoption of aggressive forms of behavior is common for all societies, regardless of their cultural norms and beliefs. The previously mentioned studies carried out in Montenegro predominantly focused on these negative effects of online media, and none of them treated the positive effects of online media, especially social marketing and its possible use for the prevention of online violence.

Taking into consideration the recent research studies carried out in Montenegro, in order to fill the literature gap, the authors considered the female population, which represents the largest part of the sample, as the focus of the research, but they also considered it very useful to include the male population, which may also be the victim of online violence, in the research in order to better identify and understand the existing situation regarding online violence in Montenegro. Contrary to some of the mentioned studies, which targeted only the capital city, the authors included all Montenegrin municipalities in the research. It is also important to mention that none of the studies considered the possibility of the introduction of prevention of online violence in the curricula at different educational levels in the Montenegrin education system and its importance, and, bearing in mind the importance of raising awareness of the local population about this issue, the authors included this segment in the research as well. As was mentioned before, previous research studies carried out in Montenegro mostly focused on online media as a tool for online violence, but the authors of this research also wanted to identify the respondents’ perception of online media, especially social marketing, not only in raising awareness of online violence but more importantly in its prevention.

The results of the research may be used as a basis for future research in this field, as well as the guidelines for creators of public policies. In that sense, the authors believe that, in addition to researchers and scientists, these results may be useful for governmental institutions, organizations, bodies, NGOs, etc., which deal with online violence, as well as for many other stakeholders who may find this issue interesting.

The paper itself is divided into five sections. The first section provides an overview of previous research studies that have been carried out. In the second section, the authors briefly describe the hypotheses, materials and methods, including a detailed description of the research methodology, while in the third part, results of the research are presented. In the fourth section, the authors discuss the results, while in the final part they draw conclusions, paying special attention to the implications of the research and the main research limitations while providing brief and useful recommendations for future research in this field.

2. Hypotheses Development, Materials and Methods

Based on the relevant literature and using data obtained from empirical research in Montenegro as well as in other European countries, three hypotheses were defined to investigate the relationship between the analyzed variables and respondents’ attitudes toward online violence.

It is interesting to point out the resource theory that is based on the assumption that the powerful always dominate over the less powerful [40]. In this sense, the powerful ones are men because of the fact that they have greater income, social standing and knowledge, skills and competences thanks to their employment [41]. At the same time, official statistics from Canada from 2004 show that women who were separated reported 7 times and divorced ones twice the prevalence of violence compared to married women in the same year [42]. Some studies also showed that women with disabilities are 40% more exposed to violence than others, and what is also worrying is the fact that they are mostly exposed to severe violence [43]. Based on the results of the mentioned studies regarding violence, the authors wanted to examine if some specific groups are more exposed to online violence and, if so, to identify them. Bearing in mind all of these data, the first hypothesis was defined:

Hypothesis 1 (H1).

Specific vulnerable groups such as women are more exposed to online violence than other ones.

Some research studies showed that there are many negative emotional, social and other outcomes of experiencing violence from many different points of view and that most victims significantly change their routines and behavior after an act of violence [44]. The changes in behavior are more obvious when the violence took place in public, especially for women and young girls who then start to avoid certain places, change the way they dress and look in public and very often begin to carry some tools in order to protect and defend themselves in public. These reactions may cause different fears and frustrations, especially among women and girls, and they may lead to numerous changes in behavior, which may further influence their engagement and integration in society in the future [45,46].

The fact is that there is a difference in behavior between people who experienced violence and those who did not, but the authors wanted to go one step further and try to identify if there is a difference between opinions and attitudes regarding some general issues about online violence depending on whether respondents experienced it or not, as well as depending on their age and the level of education. In accordance with this, the second hypothesis was defined:

Hypothesis 2 (H2).

Attitudes toward online violence depends on respondents’ online violence experience, age, marital status and level of education.

Online media are considered strong marketing media, and their popularity has continuously grown in the last few years [47,48]. This is especially true for social networks, which play a crucial role in the modern world [49]. Despite their benefits from many different points of view, there are still many risks and challenges that should be exposed, and one of them is online violence. Taking into consideration the fact that this violence is strongly connected with online media, the authors tried to identify if it is possible to use online media, as well as social marketing, in the opposite direction—as a powerful tool for the prevention of online violence. This led to the third hypothesis:

Hypothesis 3 (H3).

Online media and social marketing have a significant role in the formation of perceptions and attitudes towards online violence and its prevention.

The conceptual model is presented in Figure 1 below.

The given conceptual model is based on the causality between respondents’ attitudes towards online violence and three other constructs of the research: the belonging to a specific vulnerable group, the demographic characteristics of respondents and the role that online media have as a medium of online violence, as well as an instrument for its prevention at the same time, especially through the implementation of social marketing. Specifically, although any user of online media can be a potential victim of online violence, this is especially true for members of specific vulnerable groups, such as women. Additionally, the respondents’ attitudes towards online violence also depend on whether they already had an experience with online violence, as well as on their demographic characteristics, such as age, marital status and level of education. The respondents of distinctive demographic characteristics are faced with different levels of risk of online violence, but they also have a different level of knowledge about forms that online violence can take and the ways in which they can protect themselves. Finally, the respondents are also affected by online media, as this is the main medium of online violence, but online media can also serve as a very important instrument for educating people regarding the ways in which this type of violence can be recognized and prevented. The understanding of the relationships between the respondents’ attitudes towards online violence and the given three constructs is a prerequisite for the evaluation of the behavior of potential and real victims, which serves as a basis for the implementation of different measures aimed at online violence prevention and the usage of online media and social marketing in the appropriate way.

Bearing in mind the research goals, formulated hypotheses, methods and results of previous studies in this field, as well as their previous research experience in online violence, the authors developed a questionnaire (which is given in Appendix A). The questionnaire was sent to 500 respondents in Montenegro via numerous mailing lists and Viber groups in October 2019. In order to test the validity of the questionnaire, a pilot survey was conducted among 20 respondents. According to their feedback, the content of the questionnaire was improved, and the final version was adopted. The poll lasted for 15 days, and 277 participants completed the survey. It showed a high answer rate of 55.4%, which confirms the attractiveness of the topic itself.

The final form of the questionnaire consisted of 17 questions divided into four groups. The first group encompassed multiple-choice questions related to the demographic characteristics of respondents—their gender, age, level of education and marital status. The second group consisted of questions related to the respondents’ attitudes towards online violence in general, as well as its presence in Montenegro. The attitudes were measured using a multiple-choice form of questions, as well as 5-point Likert scale, where 1 represented a strong disagreement and 5 a strong agreement with the given statements. The third group encompassed multiple-choice questions related to the online violence experienced by respondents or persons close to them. The aim of these questions was to enable the identification of the forms that online violence usually takes, the media it is mostly based on and the ways in which the victims usually react to it. Finally, the fourth group of multiple-choice questions was related to the respondents’ attitudes regarding the efforts made by public institutions and society aimed at the prevention of online violence, and the contribution that social marketing potentially can make in that process.

Although the topic is rather specific, it should be noted that the survey was organized in complete accordance with ethical standards. The research did not examine patients and medical documentation, but only people’s attitudes and perceptions of online violence. The survey was anonymous and did not include any personal information of respondents. Respondents voluntarily participated in the survey, and all of them agreed to the publication of the research results. As all participants gave their informal consent for inclusion before they participated in the study, the conduction of the research did not require any additional permission. Hence, considering all previously mentioned, the Ethics Committee did not require any additional permission for conducting the research.

The authors used Cronbach’s alpha to test the reliability of the questionnaire. The calculated value of Cronbach’s alpha is 0.796, and this is a satisfactory level in social sciences, which led to the conclusion that the provided data are appropriate for further analysis. In order to identify the underlying relationships between specific groups of variables derived from the questionnaire, as well as their impact on respondents’ attitudes toward online violence, exploratory factor analysis was applied. The analysis showed that the measured variables were grouped into three main factors, and their relationship is defined with the given conceptual model of the research.

The SPSS program (Statistics 20, released by IBM Corp. in Armonk, NY, USA) was used for the analysis of data, as well as descriptive statistics, the ANOVA test, logistic regression and Relative Importance Index. The ANOVA test is also well known as the analysis of variance, and it is used for testing the significance of differences [50,51]. This model is focused not only on the variables but also on their interaction's, which represents one of its main advantages. Regarding the Likert scale, the data were measured on an interval scale by placing them in the place of neither neutral nor moderate [52]. In the case of interval scale data, the ANOVA is applicable. For the Likert scale data defined from 1–5 with equal intervals, the midpoint is measured with the value 3 [53,54]. Likert scale labeling options do not create any difference in the data distance, because the code is the same. Therefore, the usage of parametric tests provides the best results [55]. Additionally, in order to further investigate the impact of online violence on women as a particularly vulnerable group, logistic regression was applied. Relative Importance Index (RII) was used in order to determine which factors of the analysis are of the greatest importance for the attitudes of respondents about online violence. The results of the research are presented in the next section.

3. Results

Although one of the main goals of this research was to point out the problem and characteristics of online violence against women in Montenegro, the survey included both women and men. We decided at the very beginning that it is important to present the kind of experiences the population in Montenegro has with online violence, regardless of gender. In the remaining part of the research, we focused on a more detailed analysis of online violence against women as it has been shown that they face this problem more often than men. In order to determine and explain different aspects of the analysis of online violence, the authors started from the analysis of the characteristics of the respondents using the descriptive statistics method, which is described in

Table 1. Characteristics of respondents.

Gender
Female		196		70.8
Male		81		29.2
Age			Level of Education
	N	Weighted%		N	Weighted%
18–24	85	30.7	Primary school	1	0.4
25–29	45	16.2	High school	40	14.4
30–34	43	15.5	College	5	1.8
35–40	53	19.1	Faculty	123	44.4
41–45	27	9.7	Specialist	40	14.4
45–55	19	6.9	Master’s	52	18.8
More than 55	5	1.8	PhD	16	5.8

.

Descriptive statistics obtained by analyzing the answers from the questionnaire indicates the basic characteristics of the respondents. Respondents were mostly female (70.8%), while the participation of males was 29.2%. Most of the respondents were from the younger population: a total of 85 respondents belonged to the age group of 18–24 years, which represents 30.7% of the sample. The next group, according to the number of respondents, represents people between 35 and 40 years old: in total, 53 respondents or 19.1% of the sample. The next was the group of respondents who were between 25 and 30, which represents 16.2% of the sample, and a group 30–34 years old, which makes up 15.5% of the sample. The smallest groups of the respondents belong to the older population: 19 respondents from the age group 45–55 participated in the survey, which is 6.9% of the sample, and 5 respondents who were over 55, which represents 1.8% of the sample. In most cases, respondents had a university degree (124 respondents or 44.4% of the sample), 108 of them had specialist, master or Ph.D. diploma, while only one respondent, which represents 0.4%, completed only primary school.

The results of the survey showed that 52.71% were victims of online violence, of which 38.99% were female, while the proportion of male respondents in this category was 13.72%. As many as 32.49% of the total respondents were women who experienced online violence more than once. Victims of online violence were in most cases members of the younger population. Specifically, a total of 14.44% of the respondents belonged to the group of 18–24 years. The oldest population of respondents, i.e., over 55 years of age, were the least vulnerable to online violence. Respondents with a high school diploma degree were the most frequent victims of online violence. Survey results showed that 23.1% of respondents with a diploma degree experienced online violence several times. A detailed overview of the characteristics of respondents who were victims of online violence is given below in the

Table 2. Characteristics of respondents who were victims of online violence.

Variables		Victims of Online Violence
		Once	More than Once
Gender	Female	61.64%	12.33%
	Male	22.60%	3.42%
Age	18–24	27.40%	5.48%
	25–29	8.22%	3.42%
	30–34	15.75%	2.05%
	35–40	17.81%	3.42%
	41–45	8.90%	1.37%
	45–55	4.79%	0.00%
	More than 55	1.37%	0.00%
Level of education	Primary school	0.00%	0.00%
	High school	13.70%	2.05%
	College	1.37%	0.00%
	Faculty	43.84%	4.79%
	Specialist	8.22%	3.42%
	Master	13.70%	3.42%
	PhD	3.42%	2.05%

.

Based on the characteristics of respondents who were victims of online violence (Table 2), it may be concluded that it is women who are most often the victims of online violence. Of the total number of women who have been victims of online violence, 61.64% have experienced online violence once, while 12.33% have experienced online violence more than once. On the other hand, these percentages are much lower for the male population. Specifically, only 6.17% of male respondents stated that they were victims of online violence once, and the same percentage experienced it more than once. Hence, it is not surprising that the participation of men in this research was lower, owing to their lack of interest in the topic of online violence. The research confirms that women face online violence more often than men. This was also presented in research studies in this field [21,22,42] and is why we focus on them in the second part of the analysis. Furthermore, the results of the survey clearly confirm that victims of online violence were most often members of the younger population, which is also shown by other research studies that were carried out [20,21,22,23]. In fact, the research showed that 27.40% of the respondents who were victims of online violence were between 18 and 24 years old and experienced violence once. The results of the research confirmed our first hypothesis that specific vulnerable groups such as women are more exposed to online violence than the other ones, which was also stated in some previous studies [42,43].

The survey also tried to answer the question of whether the respondents had the same experience with online violence in a particular online space, regardless of gender, age or level of education (

Table 3. ANOVA test of differences in characteristics of respondents depending on their experience with online violence in a particular online space.

ANOVA
		Sum of Squares	df	Mean Square	F	Sig.
Gender	Between Groups	1.536	7	0.219	1.010	0.426
	Within Groups	41.519	191	0.217
	Total	43.055	198
Age	Between Groups	62.813	7	8.973	3.737	0.001
	Within Groups	458.624	191	2.401
	Total	521.437	198
Level of education	Between Groups	24.885	7	3.555	1.958	0.063
	Within Groups	346.783	191	1.816
	Total	371.668	198

).

In the ANOVA test, the initial hypothesis defined that respondents had equal experience with online violence in a particular online space, regardless of gender, age or educational level. The results of the ANOVA test indicate that the respondents had the same or similar experience of violence in a specific online space, whether female or male and regardless of their educational level, with an error risk of 5%. The ANOVA test also confirmed that there is a significant difference in experience with violence on a particular online platform depending on the age of the respondents. The descriptive statistics confirmed these results, showing that the younger generation is most likely to experience online violence on Instagram, while the older generation is primarily experiencing violence on Facebook.

In order to further investigate the consequences of the previous conclusion, according to which women are a particularly vulnerable group of online violence, in continuation of the research, we examined which categories of the female population, in terms of age, marital status and education, are particularly affected by this type of violence. In order to reach this, logistic regression and Relative Importance Index (RII) were used.

The aim of the application of the logistic regression was to determine whether women who were victims of online violence belonged to a specific age group and whether they had a particular marital status or a specific level of education. In other words, we tried to find out whether there was a group by age, marital status or education within the female population included in the research that was especially affected by online violence. Logistic regression is most often used to quantify the effects of the interaction between the independent variables as well as to rank their importance [56].

The dependent variable in logistic regression is the impact that is experienced with online violence, while the independent variables are respondents’ age, marital status and level of education. The main results are described below.

Before the analysis of the model, the hypothesis that there was no relation between the dependent and independent variables in logistic regression was tested. The table below summarizes all results (

Table 4. Model fitting test for logistic regression of experience with online violence of women in relation to their age, marital status and level of education.

Model Fitting Information
Model	Model Fitting Criteria	Likelihood Ratio Tests
	−2 Log Likelihood	Chi-Square	Df	Sig.
Intercept Only	109.010
Final	27.685	81.326	14	0.000

).

The model was tested by comparing the model without an independent variable (109.010) with the model with an independent variable (27.685). With 14 degrees of freedom, χ² is 81.326, which is significant at the level of 0%. In other words, the results of the model fitting test for logical regression confirmed that the model is appropriate and that the null hypothesis, which refers to the non-existence of the relation between the independent and dependent variables, can be rejected.

The results of the evaluation of the logistic regression model are presented in the following table (

Table 5. Logistic regression of experience with online violence of women in relation to their age, marital status and level of education.

Have You ever Been a Victim of Online Violence?		B	Std. Error	Wald	df	Sig.	Exp(B)
Yes	Intercept	−19.058	0.98	378.183	1	0.000
	Age = 18–24	0.567	0.089	40.587	1	0.045	1.763
	Age = 25–29	0.554	0.045	151.563	1	0.000	1.740
	Age = 30–34	0.342	0.032	114.223	1	0.000	1.408
	Age = 35–40	0.312	0.044	50.281	1	0.040	1.366
	Age = 41–45	0.225	0.021	114.796	1	0.000	1.252
	Age = 45–55	0.205	0.029	49.970	1	0.003	1.228
	Age = More than 55	0 c			0
	Marital status = Divorced	0.554	0.077	51.765	1	0.013	1.740
	Marital status = Single	0.433	0.063	47.238	1	0.044	1.542
	Marital status = Married	−0.251	0.041	37.478	1	0.037	0.778
	Marital status = Widow	0 c			0
	Level of education = PhD	−0.665	0.076	76.563	1	0.007	0.514
	Level of education = Faculty	−0.214	0.032	44.723	1	0.025	0.807
	Level of education = Master	−0.376	0.043	76.461	1	0.006	0.687
	Level of Education = Specialist	−0.112	0.012	87.111	1	0.000	0.894
	Level of education = High school	0.231	0.028	68.063	1	0.010	1.260
	Level of education = College	0 c			0

Legend: ^c the reference category.

).

The analysis of the results showed that certain groups of women were more affected by online violence than others. In particular, the logistic regression confirmed that young women were the most affected by online violence. One of the reasons for this is that the younger population is more present on online platforms compared to the older population, but also the fact that the younger population uses online platforms for work and entertainment every day, unlike the older population, which uses only certain types of platforms. The results showed that female respondents aged 18 to 24 were 76.3% more exposed to online violence than respondents over 55. Furthermore, the results indicate that respondents aged 25 to 30 are 74% more exposed to online violence, respondents aged 30 to 34 have a 40.8% higher chances of experiencing online violence, respondents aged 35 to 40 are 36.6% more likely to be victims of online violence, respondents aged 41 to 45 have a 25.2% higher chances of being victims of online violence and respondents who are between 45 and 55 years are 22.8% more exposed to online violence in comparison with the respondents of the oldest age group, that is, the group over 55 years old.

If we analyze the experience of women experiencing online violence from the aspect of marital status and take widowed women as the base or reference group, then the results of logistic regression show that divorced women are 74% more exposed to online violence in comparison to those who are widows. On the other hand, if the respondents are single, then the chance of being victims of online violence is 54.2% higher compared to widows. Married women are 22.2% less likely to be victims of online violence than widowed respondents. These results may be easily explained by taking into consideration the fact that the divorced ones most often use online platforms to get in touch and establish a connection with a person of the opposite sex and thus become an easy target for online violence. The same situation is true with respondents who are alone. On the other hand, it was shown that married respondents often visit online platforms related to running a household and family but have less free time to spend on online platforms, so they are less likely to be victims of online violence.

Finally, if we observe the level of education and take respondents who completed high school as the reference group, we come to some important conclusions regarding the relationship between the level of education and online violence. Specifically, respondents who have completed doctoral studies have a 48.6% lower chance of being victims of online violence; if they have a master’s degree, they are 31.3% less likely to be victims of online violence; those with a faculty degree have a 19.3% lower chance to be victims of online violence; and those with specialist studies are 10.6% less likely to experience online violence compared to respondents who completed high school only. It was considered that the more educated the respondents are, the better informed they are about what online violence is, how to recognize it and how to avoid it. Respondents who have completed only high school are 26% more often victims of online violence in comparison to those who have completed college.

In the continuation of the research, which concerns online violence and the female population, we wanted to analyze the most important attitudes and opinions of respondents about online violence in Montenegro. We performed the analysis using Relative Importance Index.

The research tried to answer the question about the attitude of women towards online violence and what the key factors are that determine it. Respondents rated the significance of the stated attitudes to online violence on a scale from 1 to 5. The survey tried to identify respondents’ opinions on whether they think that online violence is very frequent nowadays, whether its level is continuously increasing, whether its consequences are quite different from consequences of other types of violence, whether they consider women or members of marginalized groups to be more exposed to online violence, whether men commit online violence more often than women, whether online violence is a common form of peer violence, whether online abusers are usually violent in real life as well, whether all kinds of violence should be reported to relevant institutions, whether they believe that relevant institutions have adequate mechanisms to deal with online violence and whether they consider that it is necessary to work continuously on raising awareness about online violence and its consequences, as well as the measures and activities that may be carried out in order to sanction it. In order to examine which of these factors has the greatest significance in forming the attitudes of respondents to online violence, but also to rank all other factors by importance, Relative Importance Index (RII) [48,49,50] was used.

Relative Importance Index (RII) in this research was applied by equation [48,49,50]:

RII = (1 × N_A + 2 × N_B + 3 × N_C + 4 × N_D + 5 × N_E)/(5 × N)

(1)

in which 1, 2, 3, 4 and 5 represent Likert scale values used by respondents in order to estimate the value of each observed factor, and N_A, N_B, N_C, N_D and N_E represent the total number of respondents who estimated a specific value to the observed factor. The value of 5 is used as the highest value, while N stands for the total number of respondents.

Table 6. The ranking of factors of online violence using Relative Importance Index.

Variables	1	2	3	4	5	RII	Rank
Online violence is very frequent nowadays in Montenegro	7	9	40	69	71	0.791837	6
The level of online violence is continuously increasing	8	4	43	63	78	0.803061	5
Consequences of online violence are not significantly different from consequences of other types of violence	9	19	51	72	45	0.727551	8
Girls/Women are more imposed to online violence than men	6	20	49	51	70	0.762245	7
Marginalized groups are more exposed to online violence than unmarginalized ones	11	20	70	57	38	0.692857	11
Men commit online violence more often than women	7	29	52	60	48	0.715306	9
Online violence is a common form of peer violence	5	7	43	64	77	0.805102	4
Online abusers are usually violent in real life as well	7	25	67	53	44	0.704082	10
All kinds of violence should be reported to relevant institutions	6	4	22	47	117	0.870408	3
Relevant institutions in Montenegro have adequate mechanisms to deal with cases of online violence	49	54	72	17	4	0.470408	12
It is necessary to work continuously on raising awareness about online violence and its consequences	5	7	11	40	133	0.894898	2
It is necessary to work continuously on raising awareness about measures and activities that may be carried out in order to sanction online violence	6	5	12	37	136	0.897959	1

presents the ranking of 12 attitudes or factors that define the state of online violence in Montenegro.

Based on the results of RII analysis, we can draw the following conclusions. For the respondents, the fact that it is necessary to work continuously on raising awareness about what online violence is and its consequences, as well as on online violence sanctions, are of the greatest importance for forming the respondents’ attitudes and state of online violence in Montenegro. Therefore, the respondents believed that online violence is very frequent nowadays in Montenegro, but the population does not have enough knowledge to recognize and react to it or sanction it in an appropriate way. This is why these two attitudes take the first two places on the RII ranking list. The third most important is the attitude according to which online violence should be reported. This third factor indicates that respondents in Montenegro are aware of online violence, but they believe that this awareness of the possible serious and long-lasting consequences of this kind of violence is still not sufficient among the population. Therefore, their answers indicate that further steps should be taken to raise the awareness of local population about online violence and that people should be educated to report online violence and sanction it properly if they find themselves in the role of victims. Considering that the victims are most often women and other vulnerable groups, then these results should be kept in mind in order to improve the existing situation. What is worrying is the fact that, in terms of importance on this list, the attitude that the relevant institutions in Montenegro have adequate mechanisms to deal with cases of online violence is in the last place. Therefore, according to the first factors on this list, the respondents indicated that they are aware that online violence is a big and significant problem in society, but they are obviously not sure that the relevant institutions understand this problem and that they adequately solve it.

The results of the research also showed that the respondents, who were victims of online violence, in 72.5% of cases characterized the violence as harassment by calling and sending messages by the perpetrator of violence; in 12.2% of cases, the violence was committed by stealing identity or creating a false profile; and 5.3% of cases were sexual harassment. It is important to mention that 3.1% of respondents described the violence as threats/insults and publishing and distribution of photos and videos without the permission of the victim of violence, while 2.3% of cases of online violence were manifested in the form of blackmail and 1.5% of cases took the form of stalking and persecution.

In order to better understand what happened after the experience with online violence, we examined how the respondents reacted to the violence and whom they turned to for help. The results of our research showed that the respondents, after experiencing online violence, predominantly addressed people they trusted, i.e., family members and friends, and talked to them about this unpleasant experience. Forty percent of respondents reacted to online violence in this way. Further, 20% of respondents who experienced online violence decided to block the perpetrator of violence on social networks. Interestingly, as many as 9.6% of respondents with experience with online violence decided not to react to this type of violence. On the other hand, when it came to seeking help, respondents, after experiencing online violence, most often turned to friends for help (in 48.9% of cases). Furthermore, 38.5% of respondents who experienced online violence decided to turn to family members for help, and 9.6% of respondents thought that they should turn to the relevant institutions for help. The percentage of respondents who needed the help of a psychiatrist or psychologist was 2.2%, and 0.7% of respondents turned to friends from social networks.

In the last part of the analysis, the authors wanted to put emphasis on the importance of the crucial role that online media and other forms of communication may have, especially for the purpose of prevention of online violence. The results of the survey regarding online tools used for spreading online violence are presented in the graph below (Figure 2).

Respondents who have experienced online violence personally or had someone close to them experience it identified the main online medium that was the target for this kind of violence. The survey’s results indicate that in Montenegro, the social network Facebook is where online violence against women is most common. As many as 36.89% of the female respondents had experience with online violence on this social network. Another tool is Instagram, where 21.61% of female respondents have experienced online violence, and Viber was in third place. The least amount of space for online violence in Montenegro was reported on YouTube and Twitter.

The survey tried to determine the role of online media in the prevention of online violence. The authors of the research considered this set of questions very important, taking into account the fact that most of the research studies considered online media as a factor of online violence, but none of them, according to the authors’ findings, discussed the possibility of using them as a tool for the prevention of online violence. The results are shown in the following graph (Figure 3).

According to the results of the survey, it can be concluded that online media are extremely important in the prevention of online violence. The results also showed that more than half of the respondents believe that the role of online media is of great importance in the prevention of online violence. Respondents also believe that online violence may be better inhibited by using online media. Very few respondents believed that the media cannot be used in order to prevent online violence.

The survey also tried to identify if social marketing should be used in the prevention of online violence (Figure 4).

Research findings suggest that social marketing should be used to raise awareness of what online violence is and what its consequences are in order to prevent it in the future. As many as 76.9% of respondents believe that social marketing should be continually used to prevent online violence, while only 1.44% of them thought that social marketing could not prevent the occurrence of online violence. As it is already well known, online violence is a phenomenon that has appeared and whose influence is increasing together with the increase of importance of online media. The results of the research confirmed the third hypothesis that online media may, and we would add should, be used as a powerful tool for the prevention of online violence, and we would add that in the future their role may become even more important.

One important question that was included in the questionnaire and which, according to the findings of the authors, was not treated in any research before, is the inclusion of specific courses regarding the prevention of online violence into the curriculum of certain levels of the education system in Montenegro (Figure 5).

The survey found that 36.82% of respondents thought that this should be introduced only in primary schools, but 36.10% of them also believed that courses regarding the prevention of online violence should be introduced at all educational levels. Very few thought that this issue should not be included in the curriculum (only 1.44% of respondents), and even fewer (0.72%) thought that this topic should be included only in the curriculum of higher education. These results may be extremely important for policymakers in education in order to not only assure the quality of the education system in Montenegro but also to improve the quality of life by prevention of one of the largest problems, namely online violence, especially against women, regarding these particular target groups. Bearing in mind the fact that modern technology represents an integral part of young people’s lives, as well as the fact that online violence that is based on it cannot be stamped out with traditional measures and instruments, the courses regarding online violence should be an integral and obligatory part of the curriculum at all levels of the educational process. Special attention in these courses in primary and secondary education should be placed on different types of online violence, the importance of its timely detection, its consequences, different models of prevention and the role of different stakeholders in prevention and responding to online violence. At the same time, such courses should also be introduced at a higher education level with a more detailed overview of all the mentioned issues, as well as the introduction of the new ones such as theories of online violence; its biological, social and psychological contributors; national and international norms and standards regarding this issue; and media portrayal of online violence. Although the precise measurement of the effectiveness of such programs is quite challenging, the positive effects of formal education about any form of violence have already been confirmed in several projects conducted in the USA and Australia [56,57,58]. Hence, the experience obtained through formal education programs about online violence implemented in other countries can help policy decision-makers in Montenegro create and adopt similar courses that will be adjusted to the local conditions and the existing formal education system.

In this way, the awareness about online violence and the importance of adequately dealing with it would be raised from the beginning of the educational process, and not only would young people would develop the knowledge, skills and competencies that are necessary in order to meet the needs and demands of the modern market, but more importantly, they would be better prepared to become a member of society in which they live. In other words, the social dimension of the education process that was unjustly neglected for many years would be strengthened, which would lead to the improvement of the quality of the educational process in Montenegro.

4. Discussion

The obtained results of the research study clearly confirmed the first hypothesis that specific vulnerable groups such as women are more exposed to online violence than others. Thus, the results of the research clearly show that victims of online violence are in most cases members of the younger population: 27.40% of the respondents who were victims of online violence were between 18 and 24 years old. The results also confirmed that women are more exposed to online violence than men. Of the total number of women who have been victims of online violence, 61.64% have experienced online violence once, while 12.33% have experienced online violence more than once. On the other hand, these percentages are much lower for the male population: only 6.17% of male respondents stated that they were victims of online violence once, and the same percentage experienced it more than once. The emphasis in the analysis was therefore put on women. Furthermore, we examined which categories of the female population, in terms of age, marital status and education, are most exposed to online violence. The results showed that victims of online violence were most often the females of the younger population aged 18 to 24, while respondents of the older population over 55 years were the least vulnerable to online violence. These results are expected if we bear in mind the fact that the younger population is more present on online platforms compared to the older population, as well as the fact that the younger population uses online platforms for work and entertainment every day, unlike the older population, who use only certain types of platforms. While the younger generation is most likely to experience online violence on Instagram, the older one primarily experiences violence on Facebook, which is also expected bearing in mind the fact that Instagram is extremely popular among younger generations in Montenegro, while the older generation still prefers Facebook. It is interesting to mention that women who are divorced or single are more exposed to online violence, which may be explained by their more common use of online platforms in order to make different social connections, which makes them an easy target for online violence. This additionally confirms the hypothesis that specific vulnerable groups among the women themselves are more exposed to online violence than others. The results of the research regarding the relationship between online violence and education go in the same direction, so women who have a lower level of education are more exposed to online violence than those who are more educated. The more educated the respondents are, the better they are informed about what online violence is, how to recognize it and finally how to avoid it. These results are completely in accordance with some other studies carried out in different parts of the world [41,42,43].

The results of the survey indicate that in Montenegro the social networks Facebook (36.9%) and Instagram (21.8%) are the places where online violence is most common, which is expected considering that these networks are the most popular and most commonly used in Montenegro. The most common type of violence, in 72.5% of cases, is characterized by violence through harassment by the perpetrator of violence calling and sending messages; in 12.2% of cases, the violence was committed by stealing identity or creating a false profile; and in 5.3% of cases, it was characterized as sexual harassment, which is also expected considering the results of other studies carried out in Western Balkan countries including Montenegro [20,21,22,23].

However, as was previously mentioned, these research studies targeted only specific target groups such as women and adults, while some of them focused only on the capital city, excluding other municipalities in Montenegro. As was stated earlier, this was one of the main motives of the research, because the authors wanted to fill the obvious gap that exists in the literature and to try to analyze and discuss this issue by including all important categories.

The results of the research confirmed our second hypothesis that there is a difference between attitudes toward online violence depending on whether respondents experienced online violence or not, especially taking into consideration the age and the level of education. Hence, some general views regarding the presence of online violence, the existence of specific groups more exposed to it, the necessity of its reporting, the capacity of institutions to solve the problems regarding online violence, the necessity of raising awareness of the local population about online violence, its consequences and the measures that should be carried out, etc. are formed based on the personal attitude of the individuals and not on the basis of their specific experience with online violence, and they are not influenced by the age or level of education of respondents.

Based on the results, it may be concluded that the respondents believe that online violence is very frequent nowadays in Montenegro, but the population does not have enough knowledge to recognize and react to it in an appropriate way. It is encouraging that respondents are completely aware that online violence is a significant problem in Montenegrin society and that work should continuously be done on raising awareness about what online violence is and its consequences, as well as about the measures that should be carried out in order to prevent it. What is quite worrying is the fact that respondents strongly believe that all cases of online violence should be reported, but they are obviously not sure that the relevant institutions understand this problem or that they have the capacities to solve the cases of online violence and sanction them properly. The fact that the respondents of the survey do not have a satisfactory level of trust in relevant institutions explains the fact that only 9.6% of respondents, after experiences with online violence, turn to the relevant institutions in order to seek help, while most of them turn to friends (48.9%) and family members (38.5%) for help. Only 2.2% of respondents needed the help of a psychiatrist or psychologist, which is also quite worrying, and this should be taken into consideration, especially bearing in mind the possible consequences of online violence.

The survey tried to determine the role of the media, with a special emphasis on online media, in the prevention of online violence. The authors of the research considered this set of questions very important considering that most research studies treated online media as the factor of online violence, but none of them, according to the authors’ findings, discussed the possibility of using them as a tool for the prevention of online violence. This was another important motive of the research, and thus the authors wanted to fill the literature gap regarding this extremely important issue. Specifically, although online media, especially social networks, are the place where online violence takes place, according to some authors [59], they may also be used in order to prevent the existence of online violence and its possible consequences. The survey pointed out that most of the respondents consider online media to be of great importance in the prevention of online violence and online violence to be better inhibited by the use of online media. At the same time, as many as 76.9% of respondents believe that social marketing should be continually used to prevent online violence, while only 1.44% of them thought that social marketing cannot prevent the occurrence of online violence. In accordance with this, the results of the survey confirmed our third hypothesis that online media, as well as social marketing, are of great importance in the prevention of online violence and that they should and must be used more intensively for this purpose.

The results regarding the presence of online violence in Montenegro suggest that it is necessary to define some measures and carry out a set of different activities that may contribute to solving the existing problems regarding online violence. The authors believe that social marketing may be of huge importance in raising awareness about online violence and the necessity of taking certain actions and that it may be an extremely powerful tool for its prevention. On the other hand, the introduction of particular courses regarding the prevention of online violence at all levels of education may also be extremely useful and may significantly contribute to raising awareness about online violence and possible models of its prevention.

5. Conclusions and Implications

The implementation of ICT technologies have brought many challenges that we face, and online violence is one of them. Violence as one well-known phenomenon found its place in the virtual environment, and its consequences are even more far-reaching than ever before.

Although several studies have already examined respondents’ attitudes toward online violence in Montenegro, the authors of this research found a certain literary gap that should be filled. Specifically, some of the mentioned studies included only women and girls in their research, some of them are focused only on the capital city, and some of them treat violence in general without extracting online violence as a new phenomenon that should be discussed and analyzed itself. The authors of this research therefore decided to conduct research that would fill this obvious gap and enable a thorough analysis of this phenomenon from many points of view.

The research has shown that online violence is a present phenomenon in Montenegrin society. The authors of the research created and empirically tested a model that focused on the relationship between different factors and respondents’ perceptions and attitudes to online violence. The study has shown that some specific vulnerable groups such as women are more exposed to online violence than others. Considering these results, the analysis was focused on women to identify if there are some vulnerable groups within this category regarding age, marital status and level of education. The results clearly showed that younger female respondents were more likely to be victims of online violence than older ones, divorced and single women were more likely to be victims than married ones and those with a lower level of education have significantly higher chances to become victims of online violence than those with a higher level of education. The results also showed that, in addition to there being no significant differences regarding perceptions and attitudes toward online violence between respondents who experience this kind of violence and those who do not, some general views, opinions and attitudes about online violence are influenced by the age or level of education of respondents. Unlike previously mentioned research studies, which predominantly emphasized online media as a space where online violence happens, this research identifies the possibility of the use of online media, especially social marketing, as a powerful tool for its prevention, which, in the authors’ opinions, represents an added value of this research. The results clearly showed that online media, especially social marketing, may be of high importance for raising awareness not only about online violence and its consequences but also about activities and measures that should be carried out in order to sanction online violence and thus become an important and powerful tool for its prevention. The research also addresses the necessity of introducing specific courses regarding online violence at all levels of education in order to further promote the importance of this issue.

The research showed that online violence is presently quite common in Montenegro and that its influence is continuously increasing, but the respondents are completely aware of it and strongly believe that steps should be continuously taken to raise awareness about what online violence is and about its consequences, as well as about measures that should be carried out in order to prevent and sanction it. Although the respondents are convinced that all cases of online violence should be reported, there is an evident lack of confidence in relevant institutions, which is quite worrying and should be considered in the future, especially in the process of preparation and implementation of different online marketing strategies, which may significantly help regarding this issue.

Taking into account the fact that this research includes some issues that have not been discussed and analyzed in any research that has been carried out in Montenegro so far, its results may be extremely important for different policymakers and professionals in this field because they may help them to understand better the impact of online media on perceptions of online violence as well as to use it as a tool for its prevention.

The authors of the paper also strongly believe that this paper provides not only a practical but also an important theoretical contribution, especially considering that the concept of online violence in this paper is analyzed from different perspectives, thus filling the existing literature gap.

Considering the importance of the issue itself, the results may also be used by researchers as a base for future research in this field. This is extremely important bearing in mind the current COVID-19 pandemic, during which, according to some studies, online violence is increasing exponentially [60]. This is expected, considering the fact that more than half of the world’s total population was under lockdown conditions by April 2020, while different quarantine measures and self-isolation policies increased the use of the internet by 50–70% [61], which increased online violence [62]. In these circumstances, online media, especially social marketing, may have a crucial role in the prevention of online violence in the future, and different stakeholders such as governments, civil society organizations and internet intermediaries should put concerted efforts into creating and implementing adequate online communication and social marketing strategies of prevention of online violence. However, although the role of social marketing in education about online violence and its prevention is evident, further research in this field is needed in order to investigate what specificities the online social marketing strategies should contain, considering the fact that various target groups have different approaches to their usage of online media. Additionally, in order to improve the applied social marketing strategies, as well as other programs implemented in the formal education system, precise measurements of their effects are needed. Considering the difficulties identified in those evaluation processes, this issue should also be further investigated.