Sexual violence among male sexual minority college students in Guangdong, China: a cross-sectional study

This study aims to provide an estimate of lifetime sexual violence (SV) among male sexual minority college students with differing sexual orientations and contribute to the evidence base on the association between SV experience, stress, and depressive symptoms. The data stemmed from the 2019 Internet Survey on HIV/AIDS and Sexual Health in Colleges and Universities in Guangdong province, China. Lifetime SV victimisation (including non-sexual harassment, physical contact sexual harassment, and rape), level of stress, and depressive symptoms were examined among 1441 male sexual minority students from 37 colleges and universities in Guangdong province. Linear regression analyses were adopted to examine associations between independent variables and depressive symptoms. 23.4% (337/1441) of the participants reported lifetime non-contact harassment, while 8.8% (127/1441) and 2.4% (35/1441) of the participants reported physical contact sexual harassment and rape, respectively. Homosexual students reported the most physical contact sexual harassment (48/127, 37.8%) and bisexual students reported the most non-contact sexual harassment (102/337, 30.3%) and rape (10/35, 28.6%) among sexual minority participants; 85.6% participants reported high levels of stress. The experience of SV and the level of stress were significantly associated with depressive symptoms, and stress completely mediated the path from SV experiences to depressive symptoms. This study specifically showcases SV as a pervasive problem affecting Chinese male sexual minority college students and highlighting the severity of stress and depressive symptoms among this population. These findings emphasise the urgency of implementing effective practices and policies to address SV and depressive symptoms among male sexual minority college students in China.


Introduction
Research suggests that sexual violence (SV) occurs at alarmingly high rates on college campuses, with an incidence ranging from 9% to 97% worldwide (Campbell et al., 2021;Dzau & Johnson, 2018;Klein & Martin, 2019;Yoon et al., 2010). According to the World Health Organization (WHO), SV is 'any sexual act or attempt to obtain a sexual act by violence or coercion, acts to traffic a person or acts directed against a person's sexuality, regardless of the relationship to the victim' (Krug et al., 2002). Forms of SV vary and can include sexual harassment, unwanted sexual behaviour, sexual coercion, rape, or sexual assault (Krug et al., 2002;Martin-Storey et al., 2018). In the United States, for example, one in five women and one in 71 men be raped at some point in their lives (Black et al., 2011). The majority of literature on the prevalence of SV on college campuses covers populations in the United States. There is a previous Dannuo Wei and Jinghua Li are Co-first author meta-analysis in China showed that the pooled incidence of SV against college students was 2.66% (95% confidence interval [CI]: 1.07%-6.44%) (Li et al., 2020b). Studies found that risk factors associated with SV victimisation among college students included non-heterosexual identity, binge drinking, Greek life (e.g., membership in fraternities or sororities), participation in casual sexual encounters, and experiences of childhood sexual abuse or SV before college (Mellins et al., 2017;Ray et al., 2018;Soto-Marquez, 2016). Meanwhile, SV perpetration was associated with engagement in risky behaviours, prior mistreatment by teachers or bullying by others, and having parents with indulgent-responsive but not demanding parenting style (Wang et al., 2015).
Central discussions surrounding SV on college campuses have focused primarily on heterosexual female who has experienced SV rather than sexual minorities including gay, lesbian, or bisexual of either gender. Empirical studies have shown that sexual minorities are more vulnerable to SV (Coulter & Rankin, 2020;Edwards et al., 2015;Klein & Martin, 2019;Martin-Storey et al., 2018). Gay, lesbian, and bisexual students and students who are uncertain of their identities report more SV compared with their heterosexual peers (Coulter et al., 2017;Murchison et al., 2017;Porter & McQuiller Williams, 2013;Soto-Marquez, 2016). Dekeseredy et al. found that 40% of sexual minority college students experienced one or more types of sexual assault, compared with 24% of heterosexual participants in a South Atlantic part of the United States (DeKeseredy et al., 2017). Ford and Soto Marquez also found a 20% higher rate of stalking victimisation among sexual minority students compared with heterosexual students, and homosexual and bisexual men reported sexual assault at frequencies similar to those reported by heterosexual women in the United States (Soto-Marquez, 2016).
Experiences of SV can cause psychological and physical adverse outcomes. In the United States, 81% of female and 35% of male among general population reported significant short-term or long-term impacts of SV, such as post-traumatic stress disorder (PTSD), depression, anxiety, and suicide (Black et al., 2011;Espelage et al., 2018). A previous review showed that students who were sexually harassed were more likely to use or misuse alcohol, have physical illnesses, and have general clinical symptomology after the incident (Klein & Martin, 2019). SV experiences were also found to be associated with depression, experiences of psychological distress, PTSD, and low levels of life satisfaction among college students (Buchanan et al., 2009;Wolff et al., 2017). A cross-sectional survey of 83 universities in the United States found that survivors of SV were more likely to experience anorexia, anxiety, bulimia, depression, insomnia, stress and suicidal ideation in comparison with students who had not experienced SV, with adjusted odds ratios (ORa) ranging from 1.92 to 3.46 (Dilip & Bates, 2021). These studies mostly focused on the heterosexual population, especially heterosexual females, and paid less attention to the sexual minority population, especially the male minority group.
Upon entering university, many students in China begin living independently for the first time and may lose support networks of family and friends. Sexual minority college students are even more vulnerable in terms of their mental health due to social discrimination or the internalisation of stigma due to their sexual orientation. Studies have shown that sexual minority populations face a disproportionate burden of poor mental health compared to heterosexual students (King et al., 2008). Przedworski et al. found that sexual minority students in Minnesota were significantly more likely to report frequent mental distress compared with heterosexual students (OR range, 1.6-2.7) (Przedworski et al., 2015). This disparity has been partially attributed to 'minority stress' (Lehavot & Simoni, 2011;Murchison et al., 2017). According to the minority stress theory, which has been commonly used to understand the occurrence of SV (Martin-Storey et al., 2018) and to explain how social stigma affects mental health among sexual minorities, higher vulnerability to violent victimisation results from the stigmatisation associated with minority identities (Meyer, 2003). These finding are largely based on studies in western country, however, the experiences of SV, stress, mental health outcomes among sexual minority students and the association between them remain under-researched in China.
At present, research on college students' SV experience in China is insufficient, and studies on the experiences of SV in male minority groups are even more rare. The most recent studies on university students in China studies have shown that the prevalence of SV among Chinese college students ranges from 6.6% to 38.5% (Li & Zhu, 2015;Wang et al., 2015Wang et al., , 2016Wang et al., , 2019. A study in Hong Kong found a victimoffender overlap in sexual offending among young adults (Chan, 2020). None of these studies, however, have noted the sexual orientation of the participants or specifically mentioned the SV experience among male sexual minority college students. Reports have confirmed that experiences of SV in male sexual minority populations are associated with human immunodeficiency virus (HIV) and risky sexual behaviour, making this an issue deserving of more attention (Aho et al., 2014). Some researchers not only measured the SV prevalence among male students but also compared differences in SV experiences among populations with different gender and sexual identities (Martin-Storey et al., 2018;Soto-Marquez, 2016). Studies have stressed the importance of disaggregating data on the sexual identity of the individual experiencing SV (K. M. Edwards et al., 2015). In contrast, little is known about the prevalence of sexual victimisation among Chinese male sexual minority college students, or whether the types of the victimisation differ by sexual orientation.
While our society is well aware of female victims, male victims of SV are often forgotten or neglected. Male sexual minorities who have been sexually offended may have the same feelings as other survivors of SV, but they may face additional challenges due to societal attitudes, stereotypes about masculinity, or stigma related to sexual orientation, which may cause more harm to mental health. Although violence perpetrators are mainly male, it is not clear whether gay men who only date men experience more SV than bisexual men who date both men and women, or vice versa, within the male sexual minority group. Accurate prevalence estimates are crucial for drawing attention to the health burden of SV on male sexual minority college students, and the possible mental health consequences of SV victimisation need to be confirmed. Moreover, previous studies of students at universities in China have noted that stress is a significant factor affecting college students' mental health, with high stress levels positively associated with depression (Li, 2020;Ye et al., 2018). Based on the minority stress theory, stress is worthy of further study as an important factor associated with both SV and mental health among sexual minorities.
Accordingly, the current study sought to explore 1) the prevalence of non-contact sexual harassment, physical contact sexual harassment, and rape and differences in the experiences of SV among individuals with differing sexual orientations in a group of male sexual minority college students; 2) the prevalence of depression and the stress levels among this population; 3) the associations between different SV experiences, stress and depression, and the potential mediating effect of stress. We hypothesised that 1) the experience of sexual violence differs among male sexual minority college students with different sexual orientation; 2) the experience of sexual violence is positively associated with depression, with a mediating effect of stress.

Sampling Procedure
For this study, we collected cross-sectional data as part of the 2019 Internet Survey on HIV/AIDS and Sexual Health in Colleges and Universities in Guangdong province, which was promoted by the Guangdong Provincial Department of Education. We conducted the survey from 2 to 15 December 2019 relied on multi-stage stratified cluster random sampling. First, we took universities and colleges in Guangdong province as sampling units and stratified them according to administrative region, school level, and school type.
We randomly selected 37 (25%, 37/154) schools in total. Second, based on the proportion of students at the selected universities to the total number of students in the province, we calculated the sample size for each selected university/ college. Next, with consideration of practicality and feasibility, we selected the most commonly declared majors of the sampled universities/colleges and determined the sample size for each major based upon the proportion of students in each major. Finally, we took the class as the sampling unit, stratified it according to grade level, and randomly selected a class from each grade, with all classes in equal proportion.
We included participants who met the following criteria: 1) at least 18 years old and 2) currently enrolled in undergraduate or tertiary education in Guangdong province. At the beginning of the online questionnaires, we asked students to click a button to confirm their consent to participate in the survey anonymously. Finally, we collected 45,392 valid questionnaires through the online platform, including responses from 20,556 (45.3%, 20,556/45392) males. Of the latter, 1441 (7.0%, 1441/20556) were sexual minorities (i.e., self-reported their sexual orientation as non-heterosexual) and were included in the final data analysis.

Socio-Demographic Information
We collected the socio-demographic characteristics, including age, grade, major, sexual orientation, birthplace (including first-tier cities, second and third-tier cities, county and rural areas, the cities were ranked according to the report by China Business News New First-Tier Cities Research Institute, which ranked cities based on brand business data, user behavior data of Internet companies and city big data from data institutions (Bill's website, 2020); current intimate relationship status and place of residence (e.g., on campus, off campus). In the measurement of sexual orientation, considering the diversity of sexual identity, the options we set included 'Homosexual', 'Bisexual', 'Asexual', and 'Uncertain'. We also set an option of 'Other' as an umbrella term for students who did not want to disclose information.

Sexual Violence Experience
Students were asked if they had experienced any of the following types of SV: (1) verbal sexual harassment (including inappropriate sexual advances, off-colour jokes, sexually explicit/abusive taunting, or conversations that escalated in uncomfortable and/or aggressive ways); (2) exposure to others gazing at sensitive parts of the body, sexually suggestive gestures or body movements; (3) exposure to or receipt of pictures, photographs, or words about sex; (4) inappropriate sexual advances made through online games 1 3 or chat; (5) exposure to others' genitalia; (6) uncomfortable physical touching, hugging, or kissing with sexual intent; (7) attempted forced sexual contact (vaginal, anal, or oral sex); or (8) forced sexual contact (vaginal, anal, or oral sex). The above items refer to a previous survey in Norway, in which the researchers divided the items into verbal harassment, non-verbal harassment, and physical contact sexual harassment (Sivertsen et al., 2019). In this study, we made a distinction according to the severity of violence: students who chose any of (1) to (5) were rated as having experienced non-contact sexual harassment; students who chose (6) or (7) were rated as having experienced physical contact sexual harassment; and students who chose (8) were rated as having experienced rape. In addition, we added up the eight items of SV experiences to get the variable 'number of types of SV experienced'.

Stress
We measured the level of stress using the College Stress Scale, developed by Li et al. (Li & Mei, 2002). This scale measured general stress among college students through 30 items and can be divided into three subscales: academic stress (i.e., Unsatisfactory academic performance; Examination pressure; Competition among peers), personal stress (i.e., Poor appearance; Tension between peers; Poor physical health), and negative life events (i.e., Failed an exam; Being accused in public; Make a fool of oneself in public). The scale uses a 4-point score range (0 = no stress; 1 = mild stress; 2 = medium stress; 3 = severe stress), and the scores of all items are summed to reach an overall score, with higher scores indicating greater stress. A score above 45 was considered to indicate a high level of stress (Li & Mei, 2002). This scale has been used widely to determine stress levels among college students in China (Shen & Ye, 2018;Wu & Sun, 2019). The reliability of this scale in this study was good, with a Cronbach's alpha value of 0.97.

Depressive Symptoms
We used a 10-item version of the Centre of Epidemiological Studies Depression Scale (CES-D) to measure the level of depressive symptoms among participants (Radloff, 1977). The past week experience of the participants was measured using items like "I felt depressed", "I felt afraid", "I can't sleep well". The scale uses a 4-point score range (0 = less than one day per week; 1 = one to two days per week; 2 = three to four day per week; 3 = five to seven days per week). We added all items to obtain the total scores, with a higher score indicating more severe depressive symptoms. The Cronbach's alpha value for this scale was 0.85 in this study.

Statistical Analysis
First, we conducted a descriptive analysis of all variables. Second, we compared the socio-demographic characteristics, levels of stress, and severity of depressive symptoms of the participants with different experiences of violence. We included variables with significant statistical differences (p < 0.05) in the regression as background variables. Third, we compared the experiences of SV between different levels of stress and different sexual orientations. We conducted a single factor one-way analysis of variance for continuous variables and chi-square tests for categorical variables. Depressive symptoms used as the continuous variable; thus, we adopted linear regression analyses to examine associations between independent variables and depressive symptoms. We then fit multiple linear regression models to investigate associations between three SV variables and the dependent variables by adjusting the background variables. In addition to the p values, we obtained the unadjusted regression coefficient beta (β u ) from univariate regression models and the adjusted regression coefficient beta (β a ) from multiple regression models. We performed data analyses using SPSS 21.0 (SPSS Inc., Chicago, IL, USA), and considered a p value <0.05 as statistically significant.
Finally, we used AMOS 22.0 to perform structural equation modelling (SEM) via the maximum likelihood method to test the hypothesised mediation effect of stress on the association between the number of types of SV experienced and depressive symptoms. The total, direct, and indirect effects of the mediation model were estimated using bootstrapping, which is more powerful than the commonly used Sobel test or causal steps approach (Preacher & Hayes, 2008). A bias-corrected 95% confidence interval (CI) for each direct and indirect path was reported based on 5000 bootstrap samples. To evaluate the overall model fit, we used indices, including the χ 2 /df ratio, comparative fit index (CFI), incremental fit index (IFI), and root-mean-square error of approximation (RMSEA). The level of statistical significance was set at p < 0.05 (two sided).

Results
Details of the background characteristics and SV experiences are presented in Table 1. The average age of the 1441 students was 20.3 (SD = 1.4). Approximately 20.6% (297/1441) of the students self-reported as homosexual, 25.0% (360/1441) as bisexual, 11.5% (166/1441) as asexual, and 12.4% (185/1441) chose not to disclose their orientation. More than half of the students were born in cities (25.2% in first-tier cities and 27.1% in second and third-tier cities), and 30.2% (435/1441) came from rural areas. Most of the students were single (69.3%, 999/1441); only 18.3% of students reported they were in ongoing relationships. Most of the students resided on campus (85.5%, 1237/1441). The average score on the stress scale was 68.9 (SD = 22.6), with approximately 85.6% (1234/1441) of the students reporting high levels of stress. Students also reported a high level of depressive symptoms, with a mean CES-D score of 13.8 (SD = 6.8), 72.4% (1043/1441) students scored 10 or higher on the CES-D scale.

Sexual Violence Experience
Overall, 24.3% (350/1441) of students reported any SV experience; 23.4% (337/1441) of them had experienced noncontact sexual harassment, 8.8% (127/1441) had experienced physical contact sexual harassment, and 2.4% (35/1441) had been raped during their lifetimes. The most common form of non-contact harassment experienced by participants was being shown photos or words about sex (15.3%), followed by verbal harassment (13.7%). Background variables that differed across experiences of SV include major, birthplace, sexual orientation, current intimate relationship status, and place of residence. Levels of stress and depressive symptoms also differed significantly among participants with different experiences of SV (Table 2). Overall, bisexual males most frequently reported non-contact sexual harassment (30.3%) and rape (28.6%), while homosexual males most frequently reported physical contact sexual harassment (37.8%, Table 3).

Associations between Sexual Violence Experiences, Stress, and Depressive Symptoms
After adjusting significant background variables, the lifetime experiences of non-contact sexual harassment, physical contact sexual harassment, and rape were all positively associated with depressive symptoms (β a range: 0.055-0.058, p < 0.05). Personal stress, academic stress, and negative life events were also significantly associated with depressive symptoms, with β a of 0.678, 0.642, and 0.571, respectively ( Table 4). Figure 1 shows three model scenarios in which background variables were all controlled. Model 1 did not consider the mediating effect, Model 2 added stress as a mediator, and Model 3 added the direct path between SV experiences and depressive symptoms based on Model 2. The results show that there was little difference in model fit between Model 2 and Model 3, but both of these models yielded a significantly better fit than Model 1. In Model 3, the indirect effect value of stress 0.33 (95% CI: 0.16-0.51) was significant (p = 0.001), while the direct path from SV experience to depressive symptoms was not significant, indicating that stress completely mediated the association between SV experience and depressive symptoms ( Table 5). The path coefficients of the three models are also presented in Fig. 1. We also further analysed the mediating effects of different sources of stress. The results showed that all three sources of stress have a mediating effect between SV and depressive symptoms, with personal stress having the largest effect (see Supplement document).

Discussion
In this study, we investigated the prevalence of three types of SV victimisation, in addition to the level of stress and severity of depressive symptoms among a group of male sexual minority college students in Guangdong province, China. We compared the differences in SV experiences by individuals with differing sexual orientations and explored the associations between SV experience, stress, and depressive symptoms. We found that the prevalence of non-contact sexual harassment was highest, with over 20% of male sexual minority students having experienced sexual harassment in different ways, the most common being the showing of sex-related images or words and verbal harassment. Physical contact sexual harassment, including sexual touching and attempted penetration, was relatively rare, and rape was least frequently experienced. The lifetime prevalence of SV victimisation among male sexual minority college students in our study was slightly lower than the prevalence reported by general female students (4.7%-20.5%) but significantly higher than the prevalence among general male students (0.4%-3.9%) in a Norwegian study among a mix of heterosexual, bisexual, asexual, and other college students conducted by Sivertsen et al. that used the same survey items (Sivertsen et al., 2019). In both studies, the prevalence of verbal harassment was high (13.7% in this study vs. 15.4% in the Norwegian study). Indecent exposure and attempted rape were more frequently reported in this study (8.8% and 3.7%) than by general female students in Norway (2.3% and 2.9%) (Sivertsen et al., 2019). Based on the results of this study, male sexual minority college students experience a great deal of SV, which merits public attention.
To compare this research with similar studies on college students in China, we identified a study in 2015 that explored the experiences of SV among 2060 college students in Guangzhou; the results showed that 13.4% of students reported at least one type of SV victimisation during the past 12 months, among which 1.6% of males and 1.5% of females experienced rape (Wang et al., 2015). Foreign studies mostly have measured the incidence of SV in the past year or since matriculation (Fedina et al., 2016;Klein & Martin, 2019). Moreover, few separate reports have discussed specifically the prevalence of SV among male sexual minority college students. Ford and Soto-Marquez found that homosexual and bisexual male students reported sexual assault (including physical force, attempted physical force, and incapacitation) after enrolment at similar percentages (24.3% and 17.7%, respectively) to those reported by straight women (Soto-Marquez, 2016), and at rates higher than those found in our study. Variations in the recall time, study sample, violence definition, and study location reduce the comparability of published studies to the current study. However, these studies collectively show that SV victimisation is a pervasive problem among both male and female college students and extend the literature by documenting the high prevalence of SV victimisation, especially attempted rape and rape, specifically among male sexual minority students (Edwards et al., 2015;Fedina et al., 2016;Johnson et al., 2016;Klein & Martin, 2019;Martin-Storey et al., 2018;Soto-Marquez, 2016). Edwards et al. noted that the findings may have been influenced by the inclusion of all sexual minorities in a single group and called for studies that further disaggregate data by sexual orientation (Edwards et al., 2015). In the current study, we compared the experiences of SV among male students with differing sexual orientations. We found that homosexual and bisexual male students reported higher rates of victimisation compared with students who were asexual, uncertain of their orientation, and/or identified themselves as "other" orientations, suggesting that there is a relationship between sexual orientation and SV. Past studies have also found gender to be a vital contextual variable for understanding SV against sexual and gender minorities. Transgender/nonbinary students, for example, report the most forms of SV, compared with their cisgender peers (Coulter et al., 2017;Martin-Storey et al., 2018). Future research on SV in universities should pay special attention to gender and sexual identification among sexual minority students.  Minority stress theory can help to explain the vulnerability of sexual minority students to SV (Meyer, 2003). Previous studies have marked minority stress as an important risk factor for SV and note that sexual/gender-minority stressors may lead to behaviours (e.g., higher alcohol use, internalised homophobia, or violence perpetration) that make them more likely to experience SV (Edwards & Sylaska, 2013;Johnson et al., 2016;Murchison et al., 2017). Sexual minority students may be targeted specifically due to their minority identification, as a reflection of the perpetrators' general prejudice against sexual minority populations. Relevant departments and institutions can use this knowledge as a starting point to develop more targeted prevention and intervention strategies for SV against sexual minorities. Minority stress theory was also used to understand the mental health disparities among sexual minorities, which propose that experiences of external prejudice (i.e., distal stressors) and internal stress processes, like internalised homophobia and concealment (i.e., proximal stressors), may contribute to the elevated risk of psychological distress among sexual minorities (Sun et al., 2020).
Based on this theory, we measured students' levels of general stress, which can be affected by minority stress and depressive symptoms, to explore the association between SV victimisation, stress, and depressive symptoms. Overall, the students in our study reported a much higher level of depressive symptoms than the average college student in China (Gao et al., 2020;Li et al., 2020a, b), providing further evidence of the specific vulnerability of sexual minorities to mental illness stemming from SV. Meanwhile, the experiences of non-contact sexual harassment, physical contact sexual harassment, and rape are all significantly associated with depression, although the effect size was small, which is consistent with previous findings (Fedina et al., 2016;Klein & Martin, 2019). SV found that stress was positively associated with depression and completely mediated the effect of SV experiences on depression. The findings suggest that reducing students' levels of stress, especially personal stress and academic stress, may effectively prevent the occurrence of depressive symptoms among male sexual minority students. Continued study of SV prevalence, adverse effects, and theoretical explanations of college campuses are necessary.
At present, China lacks a clear legal definition for sexual harassment and other forms of SV. The laws and regulations that protect the legitimate rights and interests of individuals who has experienced SV are not yet complete or inclusive. Existing laws addressing SV include the Anti-Domestic Violence Law, the Marriage Law, and the Public Security Administration Punishment Law; however, no clauses explicitly protect sexual minorities. Once violence occurs, it is difficult for the individuals who has experienced SV to prove the case; their legitimate rights and interests cannot be fully protected under the current law (Ling, 2012). Besides a lack of legislation, there is a lack of relative legal and moral education in the country. Individuals, schools, families and the general society do not pay enough attention to SV. Considering that violence is embedded in social context and culture, SV, it is necessary for multiple institutions to work together in a coordinated manner to reduce the incidence of SV and the negative impact of experiencing violence among college students. First, government agencies should pay more attention to this issue and formulate or improve relevant laws and regulations to deter potential violence perpetrators. The legal rights and interests of sexual minorities should also be considered when formulating laws. Legislation to provide formal social support for individuals who has experienced SV can help to promote their helpseeking behavior. Public media organisations should publicise relevant knowledge about SV to raise awareness among college students.
Moreover, campus SV resources or services are necessary. According to the theoretical framework proposed earlier by Liang et al., there are three relevant processes of seeking help in the violence context: defining the problem, deciding to seek help, and selecting a source of support (Liang et al., 2005); the college or university should be active in all three stages. First, the campus should set up relevant student health centers, in which the professionals should use a unified definition of SV when identifying individuals who has experienced SV (Halstead et al., 2017); School-wide SV-related education activities are needed to raise awareness about SV and help define the problem. According to previous research, college students have an overall lack of knowledge of the existence of campus SV services, and specific knowledge about how to utilize them was lacking as well (Garcia et al., 2012;Hayes-Smith & Levett, 2010). To facilitate individuals who has experienced SV to seek help and to select a source of support, increasing visibility and accessibility of the resources or services offered to students was vital. It is also important that university faculty and staff be taught how to respond appropriately when disclosed to. At the same time, more understanding should be given to sexual minority students, and interventions should be implemented to improve the acceptance of and respect for sexual minorities on campus and relieve their minority stress. These measures may help lower the high incidence of SV experienced by this population, as well as the level of mental health burden that individuals who has experienced SV may suffer.
Being one of the first studies on the prevalence of SV among male sexual minority college students in China, the current study provides additional evidence that SV is a substantial problem experienced by this population and highlights the severity of stress and depression among sexual minority students. This study used specific behavioural measures, as recommended by previous studies, to better detect the occurrence of SV (Fedina et al., 2016). As there may be regional differences in the study samples, more representative domestic studies are still needed. In addition, future research should consider developing more targeted measurement tools based on sexual minorities to assess the prevalence of specific types of SV, as existing scales were developed based on the experiences of heterosexual individuals in western country. Considering the cultural differences across countries, a standardised cross-culture scale would be particularly useful for comparing the results of studies in different regions. Future studies should continue to include more information on the occurrence of SV to allow for better targeted interventions, appropriate treatments, and legal services. Longitudinal studies with a wide range of predictors are crucial to determining the causal pathways and long-term interventions.
This study has several limitations. First, although the results show SV was associated with depression, the size of the effect was small; and we treat stress and depression as continuous variables in regression models which, compare to treat as dichotomous variables, lacked clinical and epidemiological implications. The impact of SV on the mental health of male sexual minority college students and other factors', such as minority stress, association with depression need to be further confirmed. More research is needed to validate the applicability of the Minority stress theory among sexual minority SV victims in China. Second, the items used to assess the victimisation of SV may not be consistent with those used in other Chinese studies, partially due to the nonuniform definition of SV. This may have resulted in a lack of comparability between the study results. Third, we did not measure the frequency of violence, ask participants whether their experiences of SV occurred after enrolment, or solicit information on the perpetrators. Measuring sexual victimisation "since entering college" is necessary to accurately estimate the extent of campus sexual assault and distinguish these experiences from lifetime, childhood, and adolescent sexual victimisation (Fedina et al., 2016). Moreover, sexual identity was used to measure sexual orientation; we may have found different results if we used measures of sexual behaviour or attraction (Matthews et al., 2014). Finally, participants were recruited from 37 colleges and universities in Guangdong province; although this population is somewhat representative, the results may not be generalisable to the entire country.

Conclusion
Sexual violence is common among male sexual minority students and can have an adverse effect on their psychological health. Sexual violence experiences differ among students with different sexual orientations. Meanwhile, male minority college students suffer from high levels of stress and depression, and stress mediates the association between sexual violence victimisation and depression in this population. Coordinated public action should be taken to address these issues based on the incidence of sexual violence against male minority college students. These findings can inform practices and policies in response to sexual violence and depression among sexual minority college students in China.
Acknowledgments The manuscript has been read and approved by all named authors and all of them have contributed to the development of this research. The authors would like to thank all participants who have taken part in this study and all field workers for their support on data collection.
Author Contributions Jinghua Li, Fengsu Hou and Jing Gu contributed to the study conception and design. Material preparation, data collection and analysis were performed by Dannuo Wei, Huifang Xu, Lirui Fan, Yaoyao Du and Xiaoting Chen. The first draft of the manuscript was written by Dannuo Wei and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding This research was supported by the Research on the status of AIDS prevention education in colleges and universities in China (51000-71020384), the Science and Technology Innovation Committee of Shenzhen Municipality (JCYJ20190809162411393), the National Natural Science Foundation of China (81803334, 71774178).

Data Availability
The data sets used and analyzed in the current study are available from the corresponding author on reasonable request.

Financial Interests
The authors declare they have no financial interests.

Conflict of Interest
The authors have no competing interests to declare that are relevant to the content of this article.

Consent to Publish
The manuscript has been read and approved by all named authors.
Ethics Approval Ethic approval for the study was obtained from the Ethics Committee of Sun Yat-Sen University (2019[065]).
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.