Factors Associated with Gender-Based Violence Perpetration by Male High School Students in Eastern Ethiopia

Abstract

Gender-based violence (GBV) perpetration is a severe problem among youths in schools and the community at large. As most studies focused on adults, this study aimed to investigate factors associated with GBV by male high school students in eastern Ethiopia. An institutional based cross-sectional survey was conducted with 1,109 male students aged 15–24 years in eastern Ethiopia in December 2018 using self-administered WHO Multi-Country Study on Women’s Health and Life Events and UN Multi-Country study on perpetration. Poisson regression model was used to determine the prevalence ratio. Multivariable analysis showed that having girlfriend, chewing khat, drinking alcohol, watching pornography and being sexually active increased the prevalence of perpetrating any type of GBV (emotional, physical, sexual, and overall GBV) by young male in the last 12 months. Compared to those who did not smoke tobacco, smokers had an prevalence of perpetrating emotional violence (adj. PrR = 1.22, 95% CI = 1.03–1.44) and any type of GBV (adj. PrR = 1.15, 95% CI = 1.00–1.31). Young males who had a mother who had completed Grade 9–12 (adj. PrR = 0.75, 95% CI = 0.59–0.96) had a 25% lower prevalence of perpetrating emotional violence compared to male students who had an illiterate mother. Young males who had used illicit drugs, smoked shisha (adj. PrR = 1.39, 95% CI = 1.10–1.77), and were married (adj. PrR = 1.89, 95% CI = 1.15–3.12) had a higher prevalence of perpetrating sexual violence compared to young males who hadn’t used smoked shisha and were not married, respectively. Young males who had discussed reproductive health with their family (adj. PrR = 1.13, 95% CI = 1.03–1.24) and who chewed khat (adj. PrR = 1.26, 95% CI = 1.02–1.55) had a higher prevalence of perpetrating any GBV compared to young men who hadn’t discussed reproductive health with their parents and who hadn’t chewed khat, respectively. Any type of GBV (emotional, sexual, or physical violence) was significantly associated with several individual and relationship factors. Effective prevention and intervention programs should be focused on education of the influence of pornography, prevention of substance use and interventions for those in romantic relationships could all assist in reducing GBV.

Keywords

Factors associated gender-based violence perpetration high school students male Ethiopia

Introduction

Gender-based violence (GBV) perpetration is a global severe problem among youths in schools and the community at large. GBV affects youths in a school setting because it hinders education, causes severe psychological and physical harm, reduces students’ willingness to go to school, (Fincham & Smith, 2010; UNGEI, 2014) and reduces students’ achievements at school and at home (Demise et al., 2002). According to the World Health Organization (WHO) Multi-Country Study, UN Multi Country Study, and previous studies, GBV perpetration is defined as at least one or more acts of physical and or sexual, and or emotional violence committed against a partner or non-partner (Abeya et al., 2011; Fulu et al., 2013; Jewkes et al., 2013; Philpart et al., 2009; WHO, 2005b).

GBV reflects the imbalance power between women and men. Gender inequality (i.e., male superiority), and social norms accepting of violence are the key determinants of GBV. In this context, the social attitude that support male superiority and acceptance of violence towards women may increase the likelihood that men commit GBV (Graaff & Heinecken, 2017; Jewkes & Abrahams, 2002). Like other African countries, Ethiopia is a patriarchal society and perpetration of violence is often considered as an acceptable behavior, which supports male dominance and sexual entitlement and the expectation of females to be submissive (Bekele et al., 2012; Berhane, 2004; Central Statistical Agency [CSA] Ethiopia & ICF Macro, 2016; Decker et al., 2013; Gossaye et al., 2003; Graaff & Heinecken, 2017; Jewkes & Abrahams, 2002; Philpart et al., 2009; Worku & Mulugeta, 2002).

Results of a UN Multi-Country study in Asia and the Pacific noted that GBV perpetration was associated with alcohol misuse, sexual practice and sexual domination of women, lack of income, previous history of victimization, norms of masculinity, low empathy, depression, low education, experiences of childhood, and involvement in gangs and related activities (Decker et al., 2013; Fulu et al., 2013; Jewkes et al., 2013).

Several researchers, particularly in developed countries, have identified several individual and relationship factors that were risk factors for GBV perpetration. For example, witnessing parental violence, substance use, watching pornography, and risky sexual behaviors were related to GBV perpetration (Abbey et al., 2003; Abbey et al., 2011; Abbey et al., 2004; Carr & VanDeusen, 2004; Wilkins et al., 2014). However, most of these studies have been conducted at community level.

There is scanty of research focused on GBV perpetration by male high school students for a number of reasons. First, researchers in Africa have examined the risk factors of GBV perpetration particularly focused on sexual violence perpetration among the general population and students in higher educational institutions (Abrahams et al., 2004; Bekele et al., 2012; Chirwa et al., 2018; Philpart et al., 2009; Russell et al., 2014). However, risk factors for psychological and physical violence perpetration have not been studied among youths in a high school setting.

Second, existing studies in Ethiopia have exclusively focused on risk factors for female GBV victimization (Arnold et al., 2008; Bekele et al., 2011; Dibaba, 2007; Mekuria et al., 2015; Worku & Mulugeta, 2002) with the exception of two studies of GBV perpetration by males (Bekele et al., 2012; Philpart et al., 2009). These two studies, conducted in a college and a high school, found that multiple sexual partners and frequent pornography and substance use were significantly associated with sexual violence perpetration by male high school students, (Bekele et al., 2012) and khat use, combined use of alcohol and khat, and witnessing parental violence were risk factors for perpetrating GBV by male college students (Philpart et al., 2009). However, they did not investigate the relationship between risk factors and the types of GBV perpetration.

Third, studying multiple risk factors for GBV perpetration by type of violence can assist in the targeting of interventions and prevention strategies in an emerging country like Ethiopia, where resources must be spent where they will have the most impact. Lastly, young men in eastern Ethiopia are especially vulnerable to risky behaviors such as using khat, smoking tobacco, smoking shisha, and having multiple sexual partners. These risky behaviors are common in this area compared to other regions in Ethiopia (Getinet et al., 2007).

The most common substances used in Ethiopia among young people is khat, alcohol, and shisha. Khat is a leafy green plant (Catha edulis) mostly planted in the eastern and south-western parts of Ethiopia. Young people mostly use this green leaf for recreational purposes, to change their mood and enhance stimulation (Alemayehu, 2012; Philpart et al., 2009). Shisha, traditionally known as “Gaya” in Ethiopia, is a sweetened and flavored substance (tobacco) smoked through a water pipe pot. The young people mostly used shisha as an accompaniment of khat for recreational purposes and social togetherness (Alemayehu, 2012; Philpart et al., 2009).

However, no research has been conducted on these risky behaviors in connection with different types of GBV perpetration in the Ethiopian school setting. Therefore, the present study aimed to investigate multiple factors associated with three forms of GBV perpetration (emotional, physical, and sexual) by male high school students in eastern Ethiopia.

Methods and Materials

Study Area and Sample

An institution-based cross-sectional survey was conducted in East Hararghe Zone (525 km east of Addis Ababa, the capital of Ethiopia), in December 2018. The desired sample size was calculated using a single population proportion formula based on the assumption that the previous research conducted on sexual violence perpetration by male was 70% (Bekele et al., 2012). The study used a multi-stage sampling technique. First, the number of schools and students was obtained from the Zone Education Office. In the second stage, five schools were selected using simple random sampling from the total number of schools in the zone. In the third stage, the classrooms were selected using simple random sampling from each grade level, and all male students from those classes were invited to participate. Students who were absent during data collection were excluded. The study sample was proportionally allocated to each selected school based on their size. A total of 1,109 male students were invited and 1,064 agreed to take part with a response rate of 96.50%.

Data Collection Methods and Tools

Data were collected using pretested self-administered questionnaires. The questionnaires were adapted from the WHO Multi-Country Study on Women’s Health and Life Events survey, the Ethiopian Demographic and Health Survey (EDHS), (CSA Ethiopia & ICF Macro, 2016; Fulu et al., 2013; Jewkes et al., 2013; Straus et al., 1996; WHO, 2005b) and previous relevant published articles (Abeya et al., 2011; Bekele et al., 2012; Philpart et al., 2009). The original questionnaires (female experiences) were changed into male questionnaires (GBV perpetration) for this study by the authors and English instructor. We have changed the way of asking the questions, so they are appropriate to male perpetration rather than female experiences of GBV—the acts of GBV remained the same as in the questions asked of females. These scales have been previously used in Ethiopia to determine GBV perpetration by male college students but the scale for emotional violence has not been validated for college students. Using this scale is valid and reliable because it clearly measures any type of GBV (Abeya et al., 2011; Bekele et al., 2012; CSA Ethiopia & ICF Macro, 2016; Philpart et al., 2009; WHO, 2005b). The survey was translated into Afaan Oromo (the regional language) and back to English by an independent translator to ensure consistency. The survey included socio-demographic factors, sexual history, the experience of substance use, and lastly, perpetration of GBV.

We recruited data collectors who were health professionals (e.g., nursing, public health, psychiatry). Data collectors were male, and aged 24–37 years, and were all spoke Afaan Oromo fluently (the regional language). Three days of intensive training were provided for data collectors and supervisors with emphasis on the study objectives, sampling methods, tools and data collection methods, ethical conduct, details in the information statement, sensitive questions, and the process for addressing any participants who may become distressed. A pre-test was done on 5% of the sample size in one of the non-selected schools in East Hararghe before the commencement of the study. The questionnaires were revised based on the skipping patterns and translations. The study questions were read aloud to the students by data collectors, while the students marked their answers, and questionnaires were then returned to data collectors (whether they were completed or uncompleted) in an envelope and anonymous. Ethical clearance was secured from the University of Newcastle Human Research Ethics Committee (H-2018-0031) and Haramaya University, College of Health and Medical Science (Ref. No. IHRERC /137/2018). This indicates that the study met the Australian National Health and Medical Research Council guidelines for the conduct of research with human participants. Official permission in writing was obtained from the Zonal Education Office Representatives, District office and the school’s principal. As the study involved a single survey, the ethics committee recommended that formal consent was not required. Participants' consent was implied by the completion of the survey. This process would enhance the assurances of anonymity given to the students. Students could decide not to take part by not attending the survey session, or not completing a survey within the survey session (by placing a blank or incomplete survey in the provided envelope). No reason for non-participation or withdrawal were required, and there were no consequences for participation or non-participation.

Measures

Physically violent, sexually violent, and emotionally violent was assessed using six, three and four items, respectively. A participant was considered to be physically, sexually, and emotionally violent if he answered ‘yes’ to any of these items (Abeya et al., 2011; Bekele et al., 2012; CSA Ethiopia & ICF Macro, 2016; Philpart et al., 2009; WHO, 2005b). The prevalence of any type of GBV was assessed as the perpetration of at least one act of emotional, and or physical, and or sexual violence committed against a partner or non-partner. All items were measured as dichotomous variables (no/yes) (Abeya et al., 2011; Bekele et al., 2012; CSA Ethiopia & ICF Macro, 2016; Philpart et al., 2009; WHO, 2005b) (Table 1).

Table 1.

Question Used in the Study to Assess Emotional Violence, Sexual Violence and Physical Violence, Eastern Ethiopia, December 2018.

Types of GBV	Questions	Yes	No
Emotionally violent	In the past 12 months have you ever: Insulted/made a partner/non-partner woman feel bad about herself? Belittled or humiliated a partner/non-partner woman in front of others? Scared/intimidated a partner/non-partner woman on purpose (e.g. by the way you looked at her, by yelling or threats)? Threatened a partner/non-partner woman /someone she cares about?
Physically violent	In the past 12 months, have you partner/non-partner ever: Slapped or thrown something at a partner/non-partner woman that could hurt her? Pushed or shoved a woman? Beaten a partner/non-partner woman with your fists or with something else that could hurt her? Kicked or dragged a woman? Been choked or burnt a partner/non-partner woman on purpose? Threatened a partner/non-partner woman verbally or actually by using a gun, knife, or other weapon?
Sexually violent	In the past 12 months have you ever: Physically forced a partner/non-partner woman to have sexual intercourse. Physically forced a partner/non-partner woman to perform any other sexual acts she did not want to? Forced a partner/non-partner woman with threats or in any other way to perform sexual acts she did not want to (degraded or humiliated)?

Information regarding socio-demographic variables were measured as follows: childhood residence (urban and rural); educational level (Grade 9, Grade 10, Grade 11, and Grade 12); academic performance (good and above, average, and poor), and marital status (yes married, yes girlfriend, and no). Parents and family living circumstances variables were measured as follows: father’s education (illiterate, grades 1–8 complete, grades 9–12 complete, above Grade 12, and do not know); mother’s education (illiterate, grades 1–8 complete, grades 9–12 complete, above grade 12, and do not know); family help (yes and no); receiving enough pocket money from parents (yes and no); family income status (better-off, average, and poor); witness of violence as a child (yes and no); family control (tight, average, and loose/free); mother’s occupation (farmer, government employee, merchant, housewife, no permanent employee, and other); and father’s occupation (farmer, government employee, merchant, no permanent employee, and other). Substance use and sexual experiences variables were measured as follows: ever chewed khat (yes and no); ever smoked a cigarette (yes and no); ever drunk alcohol (yes and no); peers who drank (yes and no); ever used drugs like “Shisha” (yes and no); ever watched pornography (yes and no); ever had sexual intercourse (yes and no); age of first sexual partner (same as me, older than me, much older than me, other, and don’t know), currently/last 12 months more than one sexual partner (yes and no) and discussed reproductive health issues (sex, contraceptives, and GBV) with family (yes and no). These questionnaires were adapted for use in previous studies in Ethiopia (Abeya et al., 2011; Bekele et al., 2012; CSA Ethiopia & ICF Macro, 2016; Philpart et al., 2009).

Statistical Analysis

Descriptive statistics were calculated. The variance inflation factors (VIF) were calculated to determine whether there were any multicollinearity issues. The model fitness was determined using Pearson’s goodness of fit test. Poisson regression with robust variance estimators was conducted to identify the factors associated with the outcome of interest. Prevalence ratios were calculated to estimate the prevalence of the relationships between explanatory variables and the outcome variable given the potential of odds ratios to overestimate the magnitude of the relationship for variables with a prevalence exceeding 10% as is the case for the reporting of GBV, our outcome variables (Barros & Hirakata, 2003; Lee et al., 2009; Petersen & Deddens, 2008). All variables associated with GBV perpetration, which had a statistically significant p-value <0.2 in the bivariate analysis were entered together in the Multivariable Poisson analysis. All variables with a p-value greater than or equal to 0.2 in the bivariate analysis were excluded. We also considered the findings of the literature review to select the variables entered to bivariate analysis. An adjusted prevalence ratio (Adj. PrR) with 95% CIs was computed to identify the independent factors for each dependent variable, and a p-value <0.05 was used as a measure of statistical significance. The data analysis was executed using STATA, version 16.

Results

Parents and Family Living Circumstances

Table 2 includes descriptive statistics for demographics and life circumstances. In this study, 69.92% of the respondents were aged between 16–19 years with a mean age of 17.28 years and 44.08% were attending Grade 9. This study revealed that most (83.27%) of the respondents’ parents were living together, 2.73% were divorced/separated, and 3.20% were not alive. More than half (52.35%) of the respondents’ mothers’ educational status was illiterate, and 37.31% of the respondents’ fathers had never attended formal education. Around 24% of respondents’ mothers were housewives, and 62.22% of the respondents’ fathers were farmers by occupation. The majority (84.49%) of participants reported receiving help from their parents if it was needed.

Table 2.

Family History Among Male High School Students in East Hararghe, Eastern Ethiopia, December 2018.

Variable	Frequency (N)	Percentage (%)
Age <15 16–19 >20	196 744 124	18.42 69.92 11.65
Living together	886	83.27
Divorced/separated	29	2.73
Only mother alive	91	8.55
Only father alive	24	2.26
Both of them not alive	34	3.20
Father’s education
Illiterate	397	37.31
Grade 1–8 complete	366	34.40
Grade 9–12 complete	135	12.69
Above Grade 12	118	11.09
Do not know	48	4.51
Mother’s education
Illiterate	557	52.35
Grade 1–8 complete	294	27.63
Grade 9–12 complete	101	9.49
Above Grade 12	62	5.83
Do not know	50	4.70
Family help, if needed
Yes	899	84.49
No	165	15.51
Enough money is given
Yes	507	47.65
No	557	52.35
Family income status
Better-off	481	45.21
Average	452	42.48
Poor	131	12.31
Witnessed violence as a child
Yes	261	24.53
No	803	75.47
Family control
Tight	640	60.15
Average	343	32.24
Loose/free	81	7.61
Mother’s occupation
Farmer	397	37.31
Government employee	94	8.83
Merchant	270	25.38
Housewife	255	23.97
No permanent employee	42	3.95
Other^a	6	0.56
Father’s occupation
Farmer	662	62.22
Government employee	161	15.13
Merchant	188	17.67
No permanent employee	35	3.29
Others^a	18	1.69

Note. ^a daily laborer, driver, self-employed, and janitor.

Among the study participants, 45.21% reported perceiving they were from a better-off family compared to their neighbors. Nearly one-quarter of the respondents had witnessed violence as a child, and 60.15% perceived their family’s control over them as strict.

Substance Use

Among the respondents, 83.36% reported ever chewing khat. More than a quarter (26.41%) of the respondents had ever smoked cigarettes and about a quarter had ever drunk alcohol. Nearly 14% of the respondents said that they had friends who had drunk alcohol in the past 12 months. Approximately one-quarter of the respondents had ever used shisha. Around 40% had ever watched pornography (Table 3).

Table 3.

History of Substance Use Among Male High School Students in East Hararghe, Eastern Ethiopia, December 2018.

Variable	Frequency (N)	Percentage (%)
Ever chewed khat
Yes	887	83.36
No	177	16.64
Frequency of chewing
Every day or nearly every day	399	44.98
Once or twice a week	145	16.35
One to three times a month	31	3.49
Occasionally/less than once a month	312	35.17
Ever smoked tobacco
Yes	281	26.41
No	783	73.59
Frequency of smoking tobacco
Every day or nearly every day	48	17.08
Once or twice a week	25	8.90
One to three times a month	38	13.52
Occasionally/less than once a month	170	60.50
Ever drank alcohol
Yes	266	25.00
No	798	75.00
Frequency of drinking alcohol
Every day or nearly every day	38	14.29
Once or twice a week	36	13.53
One to three times a month	44	16.54
Occasionally/less than once a month	148	55.64
Drunk alcohol in the past 12 months
Yes	97	9.12
No	967	90.88
Have peers who drank in the past 12 months
Yes	153	14.38
No	911	85.62
Ever used drugs
Yes	273	25.66
No	791	74.34
Frequency of using drugs
Every day or nearly every day	31	11.36
Once or twice a week	27	9.89
One to three times a month	36	13.19
Occasionally/less than once a month	179	65.57
Ever watched pornography
Yes	423	39.76
No	641	60.24
Frequency of watching pornography
Every day or nearly every day	47	11.11
Once or twice a week	34	8.04
One to three times a month	50	11.82
Occasionally/less than once a month	292	69.03

Sexual Experiences

Among the respondents, 44.08% had ever had sexual intercourse. Out of those who had had sexual intercourse, 20.30% had their first experience between the ages of 10–15 years, and 29.42% had sex with a partner who was older than them. Nearly 40% of the respondents started sexual intercourse based on their willingness. Out of those who had sex based on their desire, the reasons they reported were peer pressure (37.00%), a threat (23.03%), a false promise (18.55%), and family pressure/marital engagement (11.09%). Among the sexually active respondents, 78.46% had not used a condom during their first sexual intercourse, and approximately 40% had had two or more sexual partners in their lifetime. A total of 28% of the students said that they had had more than one sexual partner in the past 12 months. Of all the students, 31.04% had discussed reproductive health with their families (Table 4).

Table 4.

Sexual Experience Among Male High School Students in East Hararghe, Eastern Ethiopia, December 2018.

Variable	Frequency (N)	Percentage (%)
Ever had sexual intercourse (n = 1,064)
Yes	469	44.08
No	595	55.92
Age at first sexual intercourse (n = 469)
10–15	216	20.30
16–19	74	6.95
20–24	13	1.22
Don’t know	761	71.52
Age of first sexual partner (n = 469)
Same age as me	122	26.01
Older than me	138	29.42
Much older than me	31	6.61
Other^a	7	1.49
Don’t know	171	36.46
Willing at first sexual intercourse (n = 469)
Yes	187	39.87
No	282	60.13
Condom use during first sexual intercourse (n = 469)
Yes	101	21.54
No	368	78.46
Reasons for having first intercourse^b
Family pressure/marital engagement b	52	11.09
Peer pressure	174	37.10
A threat	108	23.03
A false promise	87	18.55
For financial support (money)	16	3.41
To pass an exam	29	6.18
Drunk	28	5.97
Other	25	5.33
Number of sexual partners during lifetime (n = 469)
One	285	60.77
Two	92	19.62
Three	34	7.25
Four or more	58	12.37
More than one current sexual partner (n = 469)
Yes	130	27.72
No	339	72.28
Discussed reproductive health with family(n = 1064)
Yes	330	31.02
No	734	68.98

Notes. ^aYounger than me and unspecified. ^bMultiple responses.

Associated Factors for Different Types of GBV Perpetration

Bivariate analysis showed that age, marital status, family support, witnessing violence as a child, chewing khat, smoking tobacco, drinking alcohol, have peers who drink alcohol, using illicit drugs such as shisha, watching pornography, being sexually active, and discussing reproductive health with the family were increased the risk of any type of GBV perpetrated (emotional, physical, and sexual) by male students in the last 12 months. Mother’s occupation increased the risk for emotional violence only. Conversely, family control, and family income status were the protective factors for any type of GBV perpetration (emotional, physical, and sexual) as well as mother education was protective factors for emotionally violent only. Childhood residence, academic performance, father’s occupation, and education were not associated with different types of GBV (Table 5).

Table 5.

PrR and 95%CI Bivariate Results for Factors Associated with Different Forms of GBV Committed by 1,109 Male High school Students During the Last 12 Months in East Hararghe, Eastern Ethiopia in December 2018.

Variables	Emotional Violence		Physical Violence		Sexual Violence		Gender–Based Violence
Variables	n (%)	PrR (95% CI)	n (%)	PrR (95% CI)	n (%)	PrR (95% CI)	n (%)	PrR (95% CI)
Age >15 ≤15	380 (43.78) 108 (55.10)	Ref. 1.26 (1.09–1.46)	378 (43.55) 109 (55.61)	Ref. 1.28 (1.10–1.48)	251 (28.92) 80 (40.82)	Ref. 1.41 (1.16–1.72)	469 (78.96) 125 (21.04)	Ref. 1.18 (1.04–1.33)
Academic performance Good and above Average Poor	213 (45.32) 267 (46.19) 8 (50.00)	Ref. 1.02 (0.89–1.16) 1.10 (0.67–1.82)	203 (43.19) 277 (47.92) 7 (43.75)	Ref. 1.11 (0.97–1.27) 1.01 (0.58–1.78)	142 (30.21) 185 (32.01) 4 (25.00)	Ref. 1.06 (0.88–1.27) 0.83 (0.35–1.96)	254 (54.04) 331 (57.27) 9 (56.25)	Ref. 1.06 (0.95–1.18) 1.04 (0.67–1.62)
Father’s occupation Farmer Government employee Merchant No permanent employee	289 (43.66) 84 (52.17) 91 (48.40) 24 (45.28)	Ref. 1.20 (0.99–1.42) 1.11 (0.93–1.32) 1.04 (0.76–1.41)	290 (43.81) 72 (44.72) 99 (52.66) 26 (49.06)	Ref. 1.02 (0.84–1.24) 1.20 (1.02–1.41) 1.12 (0.84–1.49)	201(30.36) 55 (34.16) 61 (32.45) 14 (26.42)	Ref. 1.13 (0.88–1.44) 1.07 (0.84–1.35) 0.87 (0.55–1.38)	352 (54.23) 92 (57.14) 112 (59.57) 31 (58.49)	Ref. 1.05 (0.91–1.23) 1.10 (0.96–1.26) 1.08 (0.85–1.37)
Educational status Grade 12 Grade 9 Grade 10 Grade 11	41 (46.59) 236 (50.32) 151 (43.64) 60 (37.27)	Ref. 1.08 (0.85–1.37) 0.94 (0.73–1.21) 0.80 (0.59–1.08)	41 (46.59) 237 (50.75) 147 (42.49) 61 (37.89)	Ref. 1.09 (0.86–1.39) 0.91 (0.71–1.18) 0.81 (0.60–1.10)	168 (35.8) 99 (28.61) 38 (23.60) 26 (29.55)	Ref. 1.21 (0.86–1.71) 0.97 (0.67–1.39) 0.80 (0.52–1.22)	45 (51.14) 280 (59.70) 185 (53.47) 84 (52.17)	Ref. 1.17 (0.94–1.45) 1.05 (0.83–1.31) 1.02 (0.79–1.31)
Residency Urban Rural	234 (48.55) 254 (43.64)	Ref. 0.90 (0.79–1.02)	233 (48.34) 254 (43.64)	Ref. 0.90 (0.79–1.03)	152 (32.57) 174 (29.90)	Ref. 0.92 (0.77–1.10)	279 (57.88) 315 (54.12)	Ref. 0.94 (0.84–1.04)
Father’s education Above Grade 12 Illiterate Grade 1–8 complete Grade 9–12 complete Do not know	63 (53.39) 164 (41.31) 171 (46.72) 68 (50.37) 22 (45.83)	Ref. 0.77 (0.63–0.95) 0.88 (0.72–1.07) 0.94 (0.74–1.20) 0.86 (0.60–1.22)	56 (47.46) 167 (42.07) 167 (45.63) 72 (53.33) 25 (52.08)	Ref. 0.89 (0.71–1.11) 0.96 (0.77–1.20) 1.12 (0.88–1.44) 1.10 (0.79–1.53)	40 (33.90) 124 (31.23) 110 (30.05) 44 (32.59) 13 (27.08)	Ref. 0.92 (0.69–1.23) 0.89 (0.66–1.19) 0.96 (0.68–1.36) 0.80 (0.47–1.36)	70 (59.32) 201 (50.63) 213 (58.20) 83 (61.48) 27 (56.25)	Ref. 0.85 (0.71–1.02) 0.98 (0.83–1.17) 1.04 (0.85–1.27) 0.95 (0.71–1.27)
Marital status No partner Married Girlfriend	67 (24.45) 20 (54.05) 401 (53.25)	Ref. 2.21 (1.54–3.18) 2.18 (1.75–2.71)	74 (27.01) 20 (54.05) 393 (52.19)	Ref. 2.00 (1.40–2.86) 1.93 (1.57–2.38)	25 (9.12) 17 (45.95) 289 (38.38)	Ref. 5.04 (3.02–8.40) 4.21 (2.86–6.18)	100 (36.50) 24 (64.86) 470 (62.42)	Ref. 1.78 (1.34–2.36) 1.71 (1.45–2.02)
Mother’s education Above Grade 12 Illiterate Grade 1–8 complete Grade 9–12 complete Do not know	36 (58.06) 235 (42.19) 139 (47.28) 47 (46.53) 31 (62.00)	Ref. 0.73 (0.58–0.92) 0.81 (0.64–1.04) 0.80 (0.60–1.08) 1.07 (0.79–1.45)	29 (46.77) 235 (42.19) 143 (48.64) 50 (49.50) 30 (60.00)	Ref. 0.90 (0.68–1.20) 1.04 (0.78–1.39) 1.06 (0.76–1.47) 1.28 (0.90–1.82)	23 (37.10) 163 (29.26) 95 (32.31) 32 (31.68) 18 (36.00)	Ref. 0.79 (0.56–1.12) 0.87 (0.61–1.25) 0.85 (0.55–1.32) 0.97 (0.59–1.59)	296 (53.14) 167 (56.80) 58 (57.43) 38 (61.29)	Ref. 0.87 (0.70–1.07) 0.93 (0.74–1.16) 0.94 (0.72–1.21) 1.14 (0.87–1.49)
Family help, if needed Yes No	396 (44.05) 92 (55.76)	Ref. 1.27 (1.08–1.48)	96 (58.18) 391 (43.49)	Ref. 1.34 (1.15–1.55)	261 (29.03) 70 (42.42)	Ref. 1.46 (1.19–1.79)	494 (54.95) 100 (60.61)	Ref. 1.10 (0.96–1.26)
Family income status Better-off Average Poor	232 (48.23) 185 (40.93) 71 (54.20)	Ref. 0.85 (0.73–0.98) 1.12 (0.94–1.35)	234 (48.65) 187 (41.37) 66 (50.38)	Ref. 0.85 (0.74–0.98) 1.04 (0.85–1.26)	160 (33.26) 120 (26.55) 51 (38.93)	Ref. 0.80 (0.65–0.97) 1.17 (0.91–1.50)	207 (34.85) 338 (56.90) 49 (8.25)	Ref. 0.88 (0.78–0.98) 1.00 (0.82–1.22)
Witness of violence No Yes	351 (43.71) 137 (52.49)	Ref. 1.20 (1.04–1.38)	345 (70.84) 142 (54.41)	Ref. 1.27 (1.10–1.45)	236 (29.39) 95 (36.40)	Ref. 1.24 (1.02–1.50)	429 (53.42) 165 (63.22)	Ref. 1.18 (1.06–1.32)
Family control Average Tight Loose/free	174 (50.73) 270 (42.19) 44 (54.32)	Ref. 0.83 (0.72–0.95) 1.07 (0.85–1.34)	170 (49.56) 275 (42.97) 42 (51.85)	Ref. 0.85 (0.74–0.98) 1.04 (0.85–1.26)	117 (34.11) 182 (28.44) 32 (39.51)	Ref. 0.80 (0.65–0.97) 1.17 (0.91–1.50)	207 (60.35) 338 (52.81) 49 (60.49)	Ref. 0.88 (0.78–0.98) 1.00 (0.82–1.22)
Mother’s occupation Farmer Government employee Merchant Housewife No permanent employee	177 (44.58) 54 (57.45) 141 (52.22) 91 (35.69) 25 (52.08)	Ref. 1.29 (1.05–1.58) 1.17 (1.00–1.37) 0.80 (0.66–0.98) 1.17 (0.87–1.57)	178 (44.84) 43 (45.74) 137 (50.74) 103 (40.39) 26 (54.17)	Ref. 1.02 (0.80–1.30) 1.13 (0.96–1.33) 0.90 (0.75–1.08) 1.21 (0.91–1.60	128 (32.24) 36 (38.30) 94 (34.81) 59 (23.14) 14 (29.17)	Ref. 1.19 (0.89–1.59) 1.08 (0.87–1.34) 0.72 (0.55–1.12) 0.90 (0.57–1.44)	216 (54.41) 56 (59.57) 163 (60.37) 130 (50.98) 20 (60.42)	Ref. 1.09 (0.91–1.32) 1.11 (0.97–1.27) 0.94 (0.81–1.09) 1.11 (0.87–1.42)
Ever chewed khat No Yes	43 (24.29) 445 (50.17)	Ref. 2.07 (1.58–2.70)	44 (24.86) 443 (49.94)	Ref. 2.01 (1.54–2.62)	20 (11.30) 311 (35.06)	Ref. 3.10 (2.03–4.73)	60 (33.90) 534 (60.20)	Ref. 1.78 (1.44–2.20)
Ever smoked No Yes	252 (32.18) 236 (83.99)	Ref. 2.61 (2.33–2.92)	265 (33.84) 222 (79.00)	Ref. 2.33 (2.08–2.62)	139 (17.75) 192 (68.33)	Ref. 3.85 (3.25–4.56)	348 (44.44) 246 (87.54)	Ref. 1.97 (1.80–2.16)
Ever drunk alcohol No Yes	271 (33.96) 217 (81.58)	Ref. 2.40 (2.15–2.69)	475 (34.46) 212 (79.70)	Ref. 2.31 (2.06–2.59)	149 (18.67) 182 (68.42)	Ref. 3.66 (3.10–4.33)	366 (45.86) 228 (85.71)	Ref. 1.87 (1.71–2.04)
Have peers who drink No Yes	390 (42.81) 98 (64.05)	Ref. 1.50 (1.30–1.72)	388 (42.59) 99 (64.71)	Ref. 1.52 (1.32–1.75)	255 (27.99) 76 (49.67)	Ref. 1.77 (1.47–2.15)	481 (52.80) 113 (73.86)	Ref. 1.40 (1.25–1.57)
Ever used drugs No Yes	257 (32.49) 231 (84.62)	Ref. 2.60 (2.33–2.91)	261 (33.00) 226 (82.78)	Ref. 2.51 (2.24–2.81)	135 (17.07) 196 (71.79)	Ref. 4.21 (3.55–4.99)	352 (44.50) 242 (88.64)	Ref. 1.99 (1.82–2.18)
Ever watched pornography No Yes	159 (24.80) 329 (77.78)	Ref. 3.14 (2.71–3.62)	164 (25.59) 323 (76.36)	Ref. 2.98 (2.59–3.44)	83 (12.95) 248 (58.63)	Ref. 4.53 (3.65–5.62)	240 (37.44) 354 (83.69)	Ref. 2.24 (2.01–2.49)
Ever had sexual intercourse No Yes	141 (23.70) 347 (73.99)	Ref. 3.12 (2.68–3.64)	160 (26.89) 327 (69.72)	Ref. 2.59 (2.24–3.00)	68 (11.43) 263 (56.08)	Ref. 4.91 (3.87–6.22)	224 (37.65) 370 (78.89)	Ref. 2.10 (1.87–2.35)
Discussed reproductive health with family No Yes	316 (43.05) 172 (52.12)	Ref. 1.21 (1.06–1.38)	165 (50.00) 322 (43.87)	Ref. 1.14 (0.99–1.31)	214 (29.16) 117 (35.45)	Ref. 1.22 (1.01–1.46)	385 (52.45) 209 (63.33)	Ref. 1.21 (1.08–1.34)

Note. PrR = Prevalence ratio; CI = Confidence interval; n (%) = Number and percentage of participants with each risk factors who perpetrated each type of GBV; Boldface numbers = statistically significant; 1 = Reference category

Associated Factors for Emotional Violence Perpetration

In the multivariable analyses, after adjusting for all variables, young male students who had a girlfriend (adj. PrR = 1.35, 95% CI = 1.09–1.67) had a higher prevalence of perpetrating emotional violence compared to males who did not have a girlfriend. Male students who had a mother who had completed Grade 9–12 (adj. PrR = 0.75, 95% CI = 0.59–0.96) had a 25% lower prevalence of perpetrating emotional violence compared to male students who had an illiterate mother. Male students who smoked tobacco (adj. PrR = 1.22, 95% CI = 1.03–1.44) and those who had watched pornography (adj. PrR = 1.92, 95% CI = 1.60–2.30) had a higher prevalence of perpetrating emotional abuse compared to young men who hadn’t smoked tobacco or watched pornography, respectively. Finally, male students who had had sex (adj. PrR = 1.78, 95% CI = 1.48–2.15) had a higher prevalence of perpetrating emotional violence compared to males who hadn’t had sex (Table 6).

Table 6.

PrR and 95% CI for Factors Associated with Different Forms of GBV Committed by 1,109 Male High School Students During the Last 12 Months in East Hararghe, Eastern Ethiopia in December 2018.

Variable	Emotional Violence adj. PrR (95% CI)	Physical Violence adj. PrR (95% CI)	Sexual Violence adj. PrR (95% CI)	GBV adj. PrR (95% CI)
Age >15 ≤15	Ref. 1.05 (0.92–1.20)	Ref. 1.12 (0.98–1.26)	Ref. 1.11 (0.94–1.32)	Ref. 1.09 (0.97–1.21)
Marital status No partner Married Girlfriend	Ref. 1.07 (0.78–1.48) 1.35 (1.09–1.67) ***	Ref. 1.05 (0.76–1.44) 1.25 (1.02–1.53) **	Ref. 1.89 (1.15–3.12) 2.11 (1.47–3.05) *	Ref. 1.08 (0.83–1.41) 1.22 (1.03–1.44) **
Mother’s education Above Grade 12 Illiterate Grade 1–8 complete Grade 9–12 complete Do not know	Ref. 0.84 (0.68–1.04) 0.85 (0.68–1.05) 0.75 (0.59–0.96) ** 0.96 (0.75– 1.23)	NS	NS	NS
Family help, if needed Yes No	Ref. 1.02 (0.90–1.16)	Ref. 1.12 (0.99–1.27)	Ref. 1.10 (0.94–1.29)	NS
Family income status Better-off Average Poor	Ref. 0.95 (0.84–1.07) 1.14 (0.97–1.34)	Ref. 0.93 (0.82–1.05) 1.01 (0.87–1.18)	Ref. 0.95 (0.81–1.12) 1.20 (0.98–1.46)	NS
Witnessed any violence No Yes	Ref. 1.02 (0.91–1.15)	Ref. 1.11 (0.98–1.25)	Ref. 1.05 (0.90–1.23)	Ref. 1.05 (0.95–1.16)
Family control Average Tight Loose/free	Ref. 0.92 (0.81–1.04) 0.92 (0.77–1.10)	Ref. 0.93 (0.82–1.05) 1.01 (0.87–1.18)	NS	Ref. 0.95 (0.86–1.04) 0.95 (0.81–1.11)
Mother’s occupation Farmer Government employee Merchant Housewife No permanent employee	Ref. 1.15 (0.97–1.36) 1.06 (0.94–1.21) 0.92 (0.78–1.08) 1.07 (0.82–1.39)	NS	NS	NS
Ever chewed khat No Yes	Ref. 1.21 (0.93–1.57)	Ref. 1.28 (1.00–1.65)	Ref. 1.32 (0.89–1.96)	Ref. 1.26 (1.02–1.55) **
Ever smoked cigarettes No Yes	Ref. 1.22 (1.03–1.44) **	Ref. 1.06 (0.90–1.25)	Ref. 1.18 (0.94–1.49)	Ref. 1.15 (1.00–1.31) ***
Ever drunk alcohol No Yes	Ref. 0.96 (0.82–1.12)	Ref. 1.02 (0.88–1.19)	Ref. 1.09 (0.87–1.36)	Ref. 0.94 (0.83–1.07)
Have peers who drink No Yes	Ref. 0.96 (0.86–1.08)	Ref. 1.01 (0.90–1.14)	Ref. 0.98 (0.83–1.15)	Ref. 1.03 (0.93–1.14)
Ever used shisha No Yes	Ref. 1.05 (0.90–1.22)	Ref. 1.17 (1.00–1.37)	Ref. 1.39 (1.10–1.77) ***	Ref. 1.07 (0.94–1.22)
Ever watched pornography No Yes	Ref. 1.92 (1.60–2.30) ***	Ref. 1.97 (1.64–2.37) ***	Ref. 1.91 (1.45–2.52) ***	Ref. 1.63 (1.42–1.88) ***
Ever had sexual intercourse No Yes	Ref. 1.78 (1.48–2.15) ***	Ref. 1.44 (1.21–1.73) ***	Ref. 2.08 (1.56–2.78) ***	Ref. 1.36 (1.19–1.57) ***
Ever discussed reproductive health with family No Yes	Ref. 1.07 (0.96–1.19)	NS	Ref. 1.01 (0.88–1.17)	Ref. 1.13 (1.03–1.24) ***

Notes. Ref: reference category; *** p < 0.001, **p < 0.05; adj. PrR = adjusted prevalence ratio; CI = Confidence intervals; Boldface numbers = Statistically significant for that outcome; NS = Statistically not significant for that outcome in bivariate analysis. For each outcome, variables included in the multivariable model were considered statistically significant at p < 0.2 in bivariate analysis. Adjusted for all variable included in the models.

Associated Factors for Physical Violence Perpetration

In the multivariable analyses, after adjusting for all variables, only three factors were found to be statistically significant risk factors for physical violence perpetration. Young male students who had a girlfriend (adj. PrR = 1.25, 95% CI = 1.02–1.53), who had watched pornography (adj. PrR = 1.97, 95% CI = 1.64–2.37) or who had had sex (adj. PrR = 1.44, 95% CI = 1.21–1.73) had a higher prevalence of perpetrating physical violence compared to young men who did not have a girlfriend, hadn’t watched pornography, and hadn’t had sex respectively (Table 6).

Associated Factors for Sexual Violence Perpetration

In the multivariable analyses, after adjusting for all variables, four factors remained statistically significant risk factors for sexual violence perpetration. Young male students who had a girlfriend (adj. PrR = 2.11, 95% CI = 1.47–3.05) or were married (adj. PrR = 1.89, 95% CI = 1.15–3.12) had a higher prevalence of sexual violence compared to males who had no girlfriend or were not married, respectively. Young males who had used illicit drugs, smoked shisha (adj. PrR = 1.39, 95% CI = 1.10–1.77), had watched pornography (adj. PrR = 1.91, 95% CI = 1.45–2.52) and had sex (adj. PrR = 2.08, 95% CI: 1.56–2.78) had a higher prevalence of perpetrating sexual violence compared to young males who hadn’t used smoked shisha, hadn’t watched pornography, and hadn’t had sex respectively (Table 6).

Associated Factors for any GBV Perpetration

In the multivariable analyses, after adjusting for all variables, marital status, use of substances, watching pornography, being sexually active, and discussing reproductive health with family are factors that were highly associated with any GBV perpetration. Young male students who had a girlfriend (adj. PrR = 1.22, 95% CI = 1.03–1.44) had a higher prevalence of perpetrating any GBV compared to men who hadn’t had a girlfriend. Use of khat and smoking tobacco were also associated with any GBV perpetration, with a young man who chewed khat (adj. PrR = 1.26, 95% CI = 1.02–1.55) and smoked tobacco (adj. PrR = 1.15, 95% CI = 1.00–1.31) having a higher prevalence of perpetrating any GBV compared to young men who hadn’t chewed khat and hadn’t smoked tobacco, respectively. Young male students who had watched pornography (adj. PrR = 1.63, 95% CI = 1.42–1.88) had a higher prevalence of perpetrating any GBV compared to young men who hadn’t watched pornography. Similarly, young male students who had had sex (adj. PrR = 1.36, 95% CI = 1.19–1.57) had a higher prevalence of perpetrating any GBV compared to young men who hadn’t had sex. Young male students who had discussed reproductive health with their family (adj. PrR = 1.13, 95% CI = 1.03–1.24) had a higher prevalence of perpetrating any GBV compared to young men who hadn’t discussed reproductive health with their parents (Table 6).

Discussion

The current study is one of the first school-based studies to investigate the individual and relationship factors associated with different types of GBV perpetration by males in Ethiopian school setting. It adds essential information to available literature concerning risk factors of any GBV perpetration (emotional, physical, or sexual violence). Additionally, no other published research in Ethiopia school setting has examined the risk factors for each type of GBV perpetration. GBV, whether emotional, physical, or sexual violence perpetration, had common factors, including marital status, watching pornography, and having sex. However, chewing khat and discussing of reproductive health with family were mainly associated with any GBV perpetration, whereas the use of shisha was peculiarly related to sexual violence perpetration. Mother education had decreased the risk of emotional violence perpetration.

In this study, it was found that male students who had a girlfriend or were married had a higher prevalence of perpetrating emotional violence, sexual violence, physical violence, and any type of GBV. This finding concurs with prior studies (Philpart et al., 2009). Moreover, intimate partners/boyfriends are the main perpetrators of violence against women (UNFPA & WAVE, 2014; WHO, 2005a). This might be because as couples become intimate, they spend more time together in private. This may give the opportunity for males to commit violence against females. Other explanations may be due to culture and gender norms that are deeply rooted in Ethiopian society, where once in a relationship, males might have an expectation that females are compliant, with regard to sex and in other ways (Kedir & Admasachew, 2010; Le Mat et al., 2019). This suggests that interventions for those in relationships could assist in preventing GBV.

Students who reported watching pornography had a higher prevalence of perpetrating emotional violence, sexual violence, physical violence, and any type of GBV. This finding confirms the results of previous studies around the world (Alemayehu, 2012; Carr & VanDeusen, 2004; Chirwa et al., 2018). This may be due to pornography making youth curious about certain acts that they might then go on to practice, potentially without consent (Alemayehu, 2012; Carr & VanDeusen, 2004; Le Mat et al., 2019). Other researchers have suggested that viewing any sexually explicit material diminishes morals and that specific types of pornography, such as that depicting violence against women, leads to increased violence against women in real life (Lim et al., 2016). Even in the case of non-violent pornography, there is the proposition that people view pornography as ‘real’ rather than fantasy and that this negatively affects attitudes and real-life sexual behavior, specifically when people’s sexual experience is limited, such as in adolescence (Lim et al., 2016). However, more investigation is needed to ascertain the nature of the relationship. This study suggests that educating young people about the potential influence of pornography may be very important in GBV prevention.

Consistent with another studies (Alemayehu, 2012; Chirwa et al., 2018), this study found students who had sex had a higher prevalence of perpetrating emotional violence, sexual violence, physical violence and any type of GBV. Having discussed reproductive health with family had also had a higher prevalence of perpetrating of any GBV perpetration. The possible explanation for this could be the sense of sexual entitlement and supremacy that males have which lets them decide when and how to have sex. This may originate from the dominant patriarchal culture in Ethiopia. Due to culture, society, and religion, males are expected to have sexual knowledge and be sexually active, whereas for females being sexually knowledgeable or active outside of marriage is considered taboo (Bekele et al., 2012; Decker et al., 2013; Gossaye et al., 2003; Jewkes & Abrahams, 2002; Kedir & Admasachew, 2010; Le Mat et al., 2019; Philpart et al., 2009; Worku & Mulugeta, 2002). Taken together, the results of past (Bekele et al., 2012; Decker et al., 2013; Kedir & Admasachew, 2010; Le Mat et al., 2019; Worku & Mulugeta, 2002) and our current research suggest that education around respectful and healthy relationships could be very helpful in preventing GBV.

Students who had used substances (khat, smoked shisha or tobacco) had a higher prevalence of perpetrating emotional violence, sexual violence, and any type of GBV except for physically violent. This result confirms previous studies conducted on this association (Alemayehu, 2012; Philpart et al., 2009; Wirtz et al., 2018). They may commit GBV acts believing substance use is an excuse for their victimizing behavior (Abbey et al., 2003; Abbey et al., 2004; Gossaye et al., 2003). However, it is impossible to predict whether these substances directly cause GBV or not. Further longitudinal study is needed to investigate the nature of relationship. The most common substances used in Ethiopia among young people is khat, alcohol, and shisha. Khat is a leafy green plant (Catha edulis) mostly planted in the eastern and south-western parts of Ethiopia. Young people mostly use this green leaf for recreational purposes, to change their mood and enhance stimulation. Shisha, traditionally known as “Gaya” in Ethiopia, is a sweetened and flavored substance (tobacco) smoked through a water pipe pot. The young people mostly used shisha as an accompaniment of khat for recreational purposes and social togetherness (Alemayehu, 2012; Philpart et al., 2009). Our results suggest that prevention of substance use might help to minimize the GBV.

Students who had a mother who completed Grade 9–12 had a lower prevalence of perpetrating emotional violence. This finding corroborates another study (Chirwa et al., 2018; Jewkes, 2002). This might be explained by educated mothers being less likely to accept traditional gender norms and which might translate into life skills training that incorporates an understanding of respectful relationships and the rights of women (Jewkes, 2002). Our results suggest that improving women’s access to education could assist to reduce GBV.

Overall, the findings of this study suggest that to reduce the long-term effects of GBV, education of women and other factors that improve the status of women might help; interventions in schools especially for students in relationships; and substance use prevention could all assist in preventing GBV. This study also revealed that making male students more responsible, educated, and aware is crucial to prevent GBV. Moreover, this study has used an adequate sample size, which helps to address the marked economic, social, and cultural discrepancies and could be used in the context of eastern Ethiopia for future decision-making.

Implications of the Study

This study used individual and relationship factors for GBV perpetration. Future research should include community-and society-level factors for different forms of GBV as well as involve out-of-school youths to establish risk factors using longitudinal data. Additionally, this research confirms that there is not a lot of research on emotional and physical violence perpetration. Further studies should be conducted among male high school students to better understand this public health problem and in order to facilitate prevention and intervention strategies. Future studies should investigate the culture and gender norm as a risk factors for GBV, as this may help to change social norms exist in society. Moreover, this study confirms the association between any type of GBV and risk factors. This suggests that school-based effective prevention and intervention programs should be urgently designed and targeted on key identified factors to mitigate GBV perpetration among youths in schools.

Strengths and Limitations of This Study

To the best of our knowledge, this is one of the first studies to identify the risk factors of physical, emotional, sexual, and any type of GBV by male high school students in Eastern Ethiopia. This study also used Poisson regression with robust variance estimators to identify factors associated with any type of GBV. However, the findings of this study should be interpreted with caution. The data were collected from male high school students so it may not be generalized to all male youths in Ethiopia or those who drop out of school. The respondents were more likely to over- or under-report their perpetration of GBV due to social desirability bias and recall bias. The study used a cross-sectional study design, which cannot establish a causal pathway or temporal relationship. This study did not consider males as potential victims.

Conclusion

The current study investigated many individual and relationship factors associated with any type of GBV perpetration. Marital status, mother’s education, chewing khat, smoking tobacco, watching pornography, having sex, and discussing reproductive health with family were all associated with different forms of GBV. This study suggests that effective preventions and interventions should be developed for schools and target key identified factors such as; improving mother education, prevention of substance use and interventions for those in relationships could all assist in reducing GBV. Pornography was strongly associated with GBV. More investigation is needed to ascertain the nature of the relationship. However, education authorities could discourage youth to view porn. Substance uses also strongly associated with any type of GBV. Further study is required to investigate the nature of relationship. We also found scant studies on emotional and physical violence perpetration. Future studies should include community- and society-level risk factors for different kinds of GBV, as well as involve out-of-school youths to establish causality using longitudinal data.

Footnotes

Acknowledgements

We want to acknowledge the University of Newcastle, Australia, and Haramaya University, Ethiopia for providing the necessary support and facilities during data collection. We appreciate the study participants, research assistants, and supervisors for their participation in fieldwork. We thank Hirbo Shore and Ibrahim Jemal for their assistance during fieldwork. We also thank Natalia Soeters for her assistance in language editing. We also highly appreciated Mrs. Peta Forder for her technical advice on statistics.

Authors’ Notes

All authors read, provided feedback, and approved the final manuscript.

Declaration of Conflicting Interests

Funding

The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: The University of Newcastle, Australia, facilitated the study as part of its research training program for PhD students. The University of Newcastle has no role in the design of the study, data acquisition, data analysis, or interpretations of the results of the study. This study didn’t receive any fund.

ORCID iD

Addisu Shunu Beyene

Author Biographies

Addisu Shunu Beyene, PhD, received his PhD from the University of Newcastle, Australia. His research focuses on the prevalence, risk factors, and consequences of GBV. He is interested in examining GBV Sexual and Reproductive Health, Non-Communicable Diseases and provide evidence for decision-making.

Catherine Chojenta, PhD, is a senior research fellow and senior lecturer at the University of Newcastle. She is a public health researcher with a particular focus on women’s health and wellbeing across the life course.

Deborah Loxton, PhD, is a professor of public health at the University of Newcastle. She is a co-director of research center for generational health and aging and co-director of Australian Longitudinal Study on Women’s Health. She has interest on examining the health and well-being of women who have lived with violent partners, and the impact of reproductive health options and choices on women.

References

Abbey

, Clinton-Sherrod

A. M.

, McAuslan

, Zawacki

, & Buck

P. O.

(2003). The relationship between the quantity of alcohol consumed and the severity of sexual assaults committed by college men. Journal of Interpersonal Violence , 18(7), 813–833. https://dx-doi-org-s.web.bisu.edu.cn/10.1177%2F0886260503253301

Abbey

, Jacques-Tiura

A. J.

, & LeBreton

J. M.

(2011). Risk factors for sexual aggression in young men: An expansion of the confluence model. Aggressive Behavior , 37(5), 450–464. https://dx-doi-org-s.web.bisu.edu.cn/10.1002%2Fab.20399

Abbey

, Zawacki

, Buck

P. O.

, Clinton

A. M.

, & McAuslan

(2004). Sexual assault and alcohol consumption: What do we know about their relationship and what types of research are still needed? Aggression and Violent Behavior , 9(3), 271–303. https://doi.org/10.1016/s1359-1789(03)00011-9

Abeya

S. G.

, Afework

M. F.

, & Yalew

A. W.

(2011). Intimate partner violence against women in western Ethiopia: Prevalence, patterns, and associated factors. BMC Public Health , 11(1), 913. http://bmcpublichealth.biomedcentral.com/track/pdf/10.1186/1471-2458-11-913?site=bmcpublichealth.biomedcentral.com

Abrahams

, Jewkes

, Hoffman

, & Laubsher

(2004). Sexual violence against intimate partners in Cape Town: Prevalence and risk factors reported by men. Bulletin of the World Health Organization , 82, 330–337. http://www.ncbi.nlm.nih.gov/pmc/articles/pmc2622838/

Alemayehu

(2012). Determinants of sexual violence among eastern Ethiopian secondary school students . Utrecht University,

Arnold

, Gelaye

, Goshu

, Berhane

, & Williams

(2008). Prevalence and risk factors of gender-based violence among female college students in Awassa, Ethiopia. Violence and Victims , 23(6), 787–800. https://doi.org/10.1891/0886-6708.23.6.787

Barros

A. J.

, & Hirakata

V. N.

(2003). Alternatives for logistic regression in cross-sectional studies: An empirical comparison of models that directly estimate the prevalence ratio. BMC Medical Research Methodology , 3(1), 21. https://doi.org/10.1186/1471-2288-3-21

Bekele

A. B.

, van Aken

M. A.

, & Dubas

J. S.

(2011). Sexual violence victimization among female secondary school students in Eastern Ethiopia. Violence and Victims , 26(5), 608–630. https://doi.org/10.1891/0886-6708.26.5.608

10.

Bekele

A. B.

, van Aken

M. A.

, & Dubas

J. S.

(2012). Prevalence and patterns of sexual violence perpetration among male secondary school students in eastern Ethiopia. Ethiopian e-Journal for Research and Innovation Foresight , 4(1), 107–128.

11.

Berhane

(2004). Ending domestic violence against women in Ethiopia. Ethiopian Journal of Health Development , 18(3), 131–132. https://doi.org/10.4314/ejhd.v18i3.9951

12.

Carr

J. L.

, & VanDeusen

K. M.

(2004). Risk factors for male sexual aggression on college campuses. Journal of Family Violence , 19(5), 279–289. https://psycnet.apa.org/doi/10.1023/B:JOFV.0000042078.55308.4d

13.

Ethiopia

Central Statistical Agency [CSA]

, & Macro

ICF

. (2016). Ethiopia demographic and health survey . Addis Ababa, Ethiopia.

14.

Chirwa

E. D.

, Sikweyiya

, Addo-Lartey

A. A.

, Alangea

D. O.

, Coker-Appiah

, Adanu

R. M.

, & Jewkes

(2018). Prevalence and risk factors of physical or sexual intimate violence perpetration amongst men in four districts in the central region of Ghana: Baseline findings from a cluster randomised controlled trial. PloS One , 13(3), e0191663. https://doi.org/10.1371/journal.pone.0191663

15.

Decker

M. R.

, Miller

, Illangasekare

, & Silverman

J. G.

(2013). Understanding gender-based violence perpetration to create a safer future for women and girls. The lancet global health , 1(4), e170–e171. https://doi.org/10.1016/s2214-109x(13)70085-8

16.

Demise

, Shinebaum

, & Melesse

(2002). The problems of female students at Jimma University, Ethiopia, with some suggested solutions. Ethiopian Journal of Health Development , 16(3), 257–266. https://doi.org/10.4314/ejhd.v16i3.9793

17.

Dibaba

(2007). Sexual violence against female youth in Jimma town: Prevalence, risk factors and consequences. Ethiopian Journal of Health Sciences , 17(1), 47–58.

18.

Fincham

, & Smith

R. L.

(2010). Transforming policy and practice for gender in education: A gender review of 2010 EFA global monitoring report [technical paper]. UNGEI.

19.

Fulu

, Jewkes

, Roselli

, & Garcia-Moreno

(2013). Prevalence of and factors associated with male perpetration of intimate partner violence: Findings from the UN Multi-Country Cross-Sectional Study on Men and Violence in Asia and the Pacific. The Lancet Global Health , 1(4), e187–e207. https://doi.org/10.1016/S2214-109X(13)70074-3

20.

Getinet

, Tesfaye

, Ayele

, Gadisa

, Enqusillasie

, & Lemma

, et al. (2007). HIV/AIDS behavioral surveillance survey (BSS) Ethiopia 2006: Round two . Addis Ababa: FMOH.

21.

Gossaye

, Deyessa

, Berhane

, Ellsberg

, Emmelin

, Ashenafi

, & Hogberg

(2003). Women’s health and life events study in rural Ethiopia. Ethiopian Journal of Health Development , 17(Second Special Issue), 1–49. https://doi.org/10.4314/ejhd.v17i5.9856

22.

Graaff

, & Heinecken

(2017). Masculinities and gender-based violence in South Africa: A study of a masculinities-focused intervention programme. Development Southern Africa , 34(5), 622–634. https://doi.org/10.1080/0376835X.2017.1334537

23.

Jewkes

(2002). Intimate partner violence: Causes and prevention. The Lancet , 359(9315), 1423–1429. https://doi.org/10.1016/s0140-6736(02)08357-5

24.

Jewkes

, & Abrahams

(2002). The epidemiology of rape and sexual coercion in South Africa: An overview. Social Science & Medicine , 55(7), 1231–1244. https://doi.org/10.1016/s0277-9536(01)00242-8

25.

Jewkes

, Fulu

, Roselli

, & Garcia-Moreno

(2013). Prevalence of and factors associated with non-partner rape perpetration: Findings from the UN Multi-country Cross-sectional Study on Men and Violence in Asia and the Pacific. The Lancet Global Health , 1(4), e208–e218. https://doi.org/10.1016/s2214-109x(13)70069-x

26.

Kedir

, & Admasachew

(2010). Violence against women in Ethiopia. Gender, Place & Culture , 17(4), 437–452. https://doi.org/10.1080/0966369X.2010.485832

27.

Le Mat

M. L.

, Kosar-Altinyelken

, Bos

H. M.

, & Volman

M. L.

(2019). Discussing culture and gender-based violence in comprehensive sexuality education in Ethiopia. International Journal of Educational Development , 65, 207–215. https://doi.org/10.1016/j.ijedudev.2018.08.004

28.

Lee

, Tan

C. S.

, & Chia

K. S.

(2009). A practical guide for multivariate analysis of dichotomous outcomes. Annals Academy of Medicine Singapore , 38(8), 714–719.

29.

Lim

M. S.

, Carrotte

E. R.

, & Hellard

M. E.

(2016). The impact of pornography on gender-based violence, sexual health and well-being: What do we know? Journal of Epidemiol Community Health , 70(1), 3–5. https://doi.org/10.1136/jech-2015-205453

30.

Mekuria

, Nigussie

, & Abera

(2015). Childhood sexual abuse experiences and its associated factors among adolescent female high school students in Arbaminch town, Gammo Goffa zone, Southern Ethiopia: A mixed method study. BMC International Health and Human Rights , 15(1), 21. http://bmcinthealthhumrights.biomedcentral.com/track/pdf/10.1186/s12914-015-0059-6?site=bmcinthealthhumrights.biomedcentral.com

31.

Petersen

M. R.

, & Deddens

J. A.

(2008). A comparison of two methods for estimating prevalence ratios. BMC Medical Research Methodology , 8(1), 9. https://doi.org/10.1186/1471-2288-8-9

32.

Philpart

, Goshu

, Gelaye

, Williams

, & Berhane

(2009). Prevalence and risk factors of gender-based violence committed by male college students in Awassa, Ethiopia. Violence and Victims , 24(1), 122–136. https://doi.org/10.1891/0886-6708.24.1.122

33.

Russell

, Cupp

P. K.

, Jewkes

R. K.

, Gevers

, Mathews

, LeFleur-Bellerose

, & Small

(2014). Intimate partner violence among adolescents in Cape Town, South Africa. Prevention Science , 15(3), 283–295. https://dx-doi-org-s.web.bisu.edu.cn/10.1007%2Fs11121-013-0405-7

34.

Straus

M. A.

, Hamby

S. L.

, Boney-McCoy

, & Sugarman

D. B.

(1996). The revised conflict tactics scales (CTS2) development and preliminary psychometric data. Journal of Family Issues , 17(3), 283–316. https://doi.org/10.1177%2F019251396017003001

35.

UNFPA WAVE. (2014). Strengthening health system responses to gender-based violence in Eastern Europe and Central Asia. A Resource Package . UNFPA and WAVE.

36.

UNGEI. (2014). End school-related gender-based violence (SRGBV). https://www.ungei.org/what-we-do/school-related-gender-based-violence

37.

Wilkins

, Tsao

, Hertz

, Davis

, & Klevens

(2014). Connecting the dots: An overview of the links among multiple forms of violence.

38.

Wirtz

A. L.

, Perrin

N. A.

, Desgroppes

, Phipps

, Abdi

A. A.

, Ross

, Kaburu

, Kajue

, Kutto

, Taniguchi

, & Glass

(2018). Lifetime prevalence, correlates and health consequences of gender-based violence victimisation and perpetration among men and women in Somalia. BMJ Global Health , 3(4), e000773. https://dx-doi-org.web.bisu.edu.cn/10.1136/bmjgh-2018-000773

39.

Worku

, & Mulugeta

(2002). Sexual violence among female high school students in Debark, northwest Ethiopia. East African Medical Journal , 79(2), 96–99. https://doi.org/10.4314/eamj.v79i2.8911

40.

WHO. (2005a). WHO multi-country study on women's health and domestic violence against women: Summary report of initial results on prevalence, health outcomes and women's responses . WHO.

41.

WHO. (2005b). WHO multi-country study on women's health and domestic violence against women: Initial results on prevalence, health outcomes and women's responses . WHO.