Assessing risk factors and impact of cyberbullying victimization among university students in Myanmar: A cross-sectional study

Background Cyberbullying is a global public health concern with tremendous negative impacts, not only on the physical and mental health of students but also on their well-being and academic performance. However, there are very few studies on cyberbullying among university students, especially in Myanmar. This study aims to determine the percentage of university students who suffered cyberbullying victimization in the last 12 months, and the association between students’ socio-demographic characteristics, adverse events following cyberbullying and cyberbullying victimization. Methods A cross-sectional study was conducted among university students aged 18 years and older at one medical university in Magway, Myanmar. A total of 412 students (277 males and 135 females) participated in the study. Data were collected from August to September, 2018 using a self-administered questionnaire. Multiple logistic regression analyses (models I and II) were performed to estimate the unadjusted (UOR) and adjusted odds ratios (AOR), and 95% confidence intervals (CI). Results In total, 40.8% of males and 51.1% of females in the study had suffered cyberbullying victimization in the past 12 months. In model I, students who had been studying at the university for 3 years or less (AOR = 1.81; 95% CI 1.14–2.85), and who had witnessed psychological, physical or sexual violence, or cyberbullying in their neighborhoods, (AOR = 2.95; 95% CI 1.48–5.91) were more likely to have suffered cyberbullying victimization in the past 12 months. In model II, being a victim of cyberbullying was associated with difficulties in concentrating and understanding lectures (AOR = 3.96; 95% CI 1.72–9.11), and substance abuse (AOR = 2.37; 95% CI 1.02–5.49). Non-resident students were at a higher risk of being cyberbullying victims than their resident peers (AOR = 1.86; 95% CI 1.04–3.34). Conclusion Two out of five students had suffered cyberbullying victimization in the past 12 months, and only half of the victims discussed their experience(s) with someone else. Students who suffered cyberbullying victimization faced academic difficulties and started or increased smoking, betel chewing or alcohol drinking. Counter measures to prevent and mitigate the adverse events related to cyberbullying victimization are urgently needed among university students in Myanmar. Periodic screening for cyberbullying, counseling services, cyber-safety educational programs, and awareness raising campaigns should be implemented.

adjusting the title so as to emphasize this sample age difference.
The structure of the paper is satisfactory, but some of the elements or contents should be moved to different sections of the paper (e.g. the research procedure description, sample characteristics), so as to better suit the study presentation.
The study design is valid, the sample size and characteristics are satisfactory, while the chosen title and research area of the paper present a scientifically important subject, due to the scarcity of research findings on this topic.

THE CONS (recommendations for improvement)
A) Page 4 -..."To surf the worldwide information technology tide...."this section feels disconnected from the rest of the section content, so it would be recommended inserting latest data findings from a meta-analysis by Kowalski, R.M., Giumetti, G.W., Schroeder, A.N., Lattanner, M.R., 2014. Bullying in the digital age: a critical review and metaanalysis of cyberbullying research among youth. Psychol. Bull. 140, 1073-1137. This would help the authors better explain the importance of some of the studied socioeconomic characteristics. B) Page 5 -Methods, Study area and participants -The authors should explicitly state the age status of the sample, because the sample used is very diversified age-wise, while age has been found to be a very significant factor when studying cyberbullying among the adolescent population.
C) The instrument's methodological properties or metric characteristics need more thorough clarification. Was the instrument used in previous similar studies, and what are its methodological properties (validity, reliability)? How many questions or scales and what kind of questions were used, and how was the intensity or frequency measured (Likert scale, semantic differential, scale type...)? This is recommended, because the authors state that "data were collected with a pre-tested, self-administered F) It would be recommended the authors use the term "Socioeconomic characteristics / status" or "Socio-demographic characteristics / traits" instead of using the term "Background characteristics", as it does not describe the studied variables correctly. G) Page 8 -instead of using "Magway" as a point of reference or variable name, I would suggest using the "urban / rural" variable distinction when presenting and discussing results (e.g., Table 1., Background data). Table 1 should be revised so as to clearly show significant differences between the proposed variables (within the groups), as the significance is not visible in the table. For these kind of analyses, a t-test for independent samples (between 2 groups; e.g. marital status) or ANOVA (3 groups or more; e.g. age group) would be recommended, as chisquare tests are usually implemented with smaller samples containing less than 30 participants.

H)
I) The authors should decide on the style of data presentation, as it is uncommon to describe the results via text, and then provide the same data visually / graphically in a table (e.g. Table 2, page 9). It should be mutually exclusive, so the authors should decide what style of data presentation to use and implement it uniformly throughout the paper.
J) The notes presented under Table 2 (page 10) related to the types of cyberbullying should be presented in the Introduction / Theoretical background of the paper, where such definitions of cyberbullying and related behavior should be referenced and clarified. Table 3 is clear and presented well, but Table 4 needs revision, as such a style of presenting regression analysis results is unusual. Table 4 is confusing and complex, as the data is not validly and clearly presented. Regression analysis should be presented using APA standards, with percentages of variance explained, and clear indicators of R 2 , F for change in R 2 , and R Square Change, as well as B, SE(B) and β coefficients as indicators of significant predictors. Maybe some examples from similar papers using the same type of statistical analysis would be helpful to the authors. L) Discussion, page 14 -the authors reference studies on the relation of bullying and poor academic achievement, but it remains unclear whether the referenced studies report findings on traditional bullying or cyberbullying? ... "Many studies have demonstrated that among students, bullying is strongly correlated with poor academic performance or outcomes [24,[31][32][33]. Students who suffered peer bullying received lower grades and, faced academic difficulties and/or worsened academic performance compared with their non-bullied peers [24,25,[31][32][33]." M) Considering the number of authors and their personal contribution to this paper, the Discussion section of the paper seems to be somewhat lacking, as it aims to describe only a segment of the results presented in the paper, while omitting study findings on various socio-economic characteristics that were established and presented in the authors' proposed study (risk behavior, suicide, use of substances, etc.). The authors should revise and further explain all their results, as socioeconomic variables (aside from gender or age) were found in recent studies to be significant in explaining cyberbullying perpetration and victimization.