Sexual and physical violence victimization among senior high school students in Ghana: Risk and protective factors

Abstract

Violence in all forms poses a concern because of associations with multiple adverse effects including injuries and mental health problems. There is however limited data on violence in general and youth violence in particular in Ghana. To explore the nature and scope of youth violence in Ghana, we used the nationwide Global School-based Health Survey, conducted among senior high school students in Ghana, to explore risk and protective factors at the individual, family, and environmental levels associated with sexual and physical violence victimization. A fifth of these students reported being forced to have sex in their lifetime while two out of five had been a victim of a physical attack in the year preceding the survey. In final multivariate analysis, for sexual violence victimization, history of sexual activity with or without condom use at last sex, feeling sad or hopeless, and being a victim of bullying and electronic bullying were identified as risk factors, while having friends who were not sexually active was protective. Independent risk factors for physical violence victimization were attempting suicide in the last year, alcohol use in the past month, and bullying other students in the past month. Parent respect for privacy just reached significance as a protective factor for physical violence victimization in the final model. Recognition of the magnitude of violence victimization among Ghanaian students and associated factors must be used to guide development and implementation of appropriate concrete measures to prevent and address the problem.

Full text

Sexual and physical violence victimization among senior high school
students in Ghana: Risk and protective factors
Sally-Ann Ohene
a
, Kiana Johnson
b
, Sarah Atunah-Jay
c
, Andrew Owusu
d
,
Iris Wagman Borowsky
b
,
*
a
World Health Organization Country Office, Accra, Ghana
b
Department of Pediatrics, Quillen College of Medicine, East Tennessee State University, TN, USA
c
Mercy Hospital eAllina Health, Coon Rapids, MN, USA
d
Department of Health and Human Performance, Middle Tennessee State University, TN, USA
article info
Article history:
Received 17 February 2015
Received in revised form
7 October 2015
Accepted 8 October 2015
Available online 19 October 2015
Keywords:
Ghana
Youth violence prevention
Victimization
Risk factors
Protective factors
Sexual violence
Physical violence
abstract
Violence in all forms poses a concern because of associations with multiple adverse effects including
injuries and mental health problems. There is however limited data on violence in general and youth
violence in particular in Ghana. To explore the nature and scope of youth violence in Ghana, we used the
nationwide Global School-based Health Survey, conducted among senior high school students in Ghana,
to explore risk and protective factors at the individual, family, and environmental levels associated with
sexual and physical violence victimization. A fifth of these students reported being forced to have sex in
their lifetime while two out of five had been a victim of a physical attack in the year preceding the survey.
In final multivariate analysis, for sexual violence victimization, history of sexual activity with or without
condom use at last sex, feeling sad or hopeless, and being a victim of bullying and electronic bullying
were identified as risk factors, while having friends who were not sexually active was protective. In-
dependent risk factors for physical violence victimization were attempting suicide in the last year,
alcohol use in the past month, and bullying other students in the past month. Parent respect for privacy
just reached significance as a protective factor for physical violence victimization in the final model.
Recognition of the magnitude of violence victimization among Ghanaian students and associated factors
must be used to guide development and implementation of appropriate concrete measures to prevent
and address the problem.
©2015 Elsevier Ltd. All rights reserved.
1. Introduction
Violence in all forms poses a concern because of associations
with multiple adverse effects, including injuries, mortality, and
psychosocial and economic costs (WHO, 2002, 2014a). It is esti-
mated that interpersonal violence contributed to 479,000 deaths
worldwide in 2011 (WHO, 2014b). Physical and sexual violence
among young people is particularly worrying given the link to
mortality, health risk behaviors, and negative reproductive health
outcomes (WHO, 2002; Blum and Nelson-Mmari, 2004; CDC, 2009;
Patton et al., 2009). Multiple studies demonstrate the association of
youth violence with school, social, mental, and physical health
problems (Resnick et al., 1997; Blum et al., 2003; Juvonen et al.,
2003; Borowsky et al., 2004; Resnick, 2004). These studies have
contributed to recommendations and strategies aimed at pre-
venting youth violence. Relative to high income countries, studies
on the various dimensions and mechanisms of youth violence in
low middle income countries, including many in sub-Saharan Af-
rica (SSA) such as Ghana, are limited. Yet evidence from a number of
reports points to a significant level of violence among youth in the
African sub-region, including Ghana (RB MOH, 2009; Swahn et al.,
2013; WHO, 2013a, 2013b). In a study reviewing Global School-
based Health Survey (GSHS) data from 4 sub-regions, SSA had the
second highest mean prevalence of any physical fighting among
adolescents (Swahn et al., 2013). Ghana ranked third after Djibouti
and Egypt among the 27 countries and cities studied, with 53.5% of
adolescents reporting a history of physical fighting (Swahn et al.,
2013). The 2008 GSHS data from Ghana also showed that 40.1% of
*Corresponding author. Division of General Pediatrics and Adolescent Health,
Department of Pediatrics, University of Minnesota, 717 Delaware St. SE, Suite 353,
Minneapolis, MN 55414, USA.
E-mail address: borow004@umn.edu (I.W. Borowsky).
Contents lists available at ScienceDirect
Social Science & Medicine
journal homepage: www.elsevier.com/locate/socscimed
http://dx.doi.org/10.1016/j.socscimed.2015.10.019
0277-9536/©2015 Elsevier Ltd. All rights reserved.
Social Science & Medicine 146 (2015) 266e275

senior high school students (SHS) reported being bullied in the 30
days preceding the survey (Owusu et al., 2011). In the 2008
nationwide Ghana Demographic Health Survey (GSS et al., 2009)
about 41% of female teenagers 15e19 years indicated that they had
been victims of sexual or physical violence. Among the women who
reported that they had been victims of sexual violence, 63% indi-
cated that they were 19 years old or younger at their first experi-
ence of sexual violence. Different prevention strategies work in
different contexts; thus it is key that appropriate approaches are
tailored to relevant settings (WHO, 2002). Very few studies have
assessed factors associated with physical and sexual violence
among youth in Ghana. The purpose of this study was therefore to
examine risk and protective factors associated with physical and
sexual victimization using a national school based survey of Gha-
naian high school students.
1.1. Conceptual framework of youth violence using the social
ecological model
To better understand violence and how it can be prevented, it is
useful to conceptualize the multiple factors that affect behavior
using the ecological model, a theoretical framework which has
been used to help understand the multiple dimensions of violence
(Bronfenbrenner, 1979) The model highlights the point that mul-
tiple linkages and interactions are at play between individual and
other contextual factors related to the family, peers, school and
other dimensions which contribute to risk and protective factors for
violence perpetration and victimization. At the individual level,
biological and personal factors such as substance use and prior
exposure to abuse are some characteristics of a person that might
put him or her at risk for violence involvement. Beyond the indi-
vidual are the factors related to other people with whom the youth
associates on a regular basis such as family and peers. For example a
youth's engagement in violence may be fanned by the endorsement
and similar behaviors of peers with whom he or she hangs out. The
next tier in the ecological model explores the effect on violence by
the environment in which youth find themselves such as the school
setting and neighbourhood. For example unbridled violence in
schools may be a risk factor for violence victimization among the
students. Considering the interaction of the factors at the respective
levels of the ecological model and the association with violence,
effective violence prevention programs stand to benefit from
evidence-based studies that explore various risk and protective
factors across the levels.
In the last decade, researchers have explored factors associated
with physical and sexual violence among youth in several African
countries (Erulkar, 2004; Rudatsikira et al., 2007; Brown et al.,
2009; Brieding et al., 2011; Swahn et al., 2012; Ybarra et al., 2012;
Sommera et al., 2013; Celedonia et al., 2013). Some of these
studies corroborate the risk and protective factors associated with
youth violence reported from other parts of the world. It is useful to
look at the factors associated with violence along the lines of the
ecological model levels. At the individual level, a study pooling data
from Namibia, Swaziland, Uganda, Zambia, and Zimbabwe, showed
that exposure to sexual or physical violence among 13e15 year old
students was associated with mental health problems, suicidal
ideation, current alcohol use, lifetime drug use, multiple sex part-
ners, and a history of sexually transmitted infection. With respect
to cigarette use, the more fights the adolescent had been involved
in, the higher the odds of current smoking (Brown et al., 2009).
Sexual violence victimization among female youth in Swaziland
was found to be associated with not attending school at the time of
survey completion and a history of emotional abuse as a child
(Breiding et al., 2011). Familywise, factors associated with a history
of sexual violence victimization among female youth in Swaziland
included a report of not being close to their biological mother and a
history of living with multiple caregivers (Breiding et al., 2011). A
study conducted among secondary school students in Uganda
showed that those who were victims of coercive and violent sex
were more likely to report lower levels of social support from their
families (Ybarra et al., 2012). Swahn and colleagues revealed that
parental neglect due to alcohol use was significantly associated
with violence victimization among youth living in the slums of
Kampala, Uganda (Swahn et al., 2012). School-level factors found to
be associated with forced sex among female students in 10
southern African countries included attendance in a school where
higher proportions of students reported drinking alcohol and
perpetrating forced sex (Andersson et al., 2012). On a broader so-
cietal perspective, in some sub-Saharan Africa settings community
and societal factors including gender inequality, cultural norms that
endorse males' perception that they cannot reign in their sexual
desires and socialization of females to think they are not in charge
of their sexuality facilitate coercive sexual experiences (Ybarra
et al., 2012).
1.2. The context of youth violence in Ghana
Ghana is a West African country with a population of about 25
million people, including 51% women and 11% youth ages 15e19
years. (GSS, 2012) More males than females are literate (80.2%
versus 68.5%) and economically active (54.7% versus 53.7%). The
country has a democratic government with a constitution that
entitles equal rights and safety for all. On a broad level, the legis-
lative framework in Ghana denounces violence, as shown by the
development and enactment of a number of policies and laws
intended to reduce violence. (GDHS, GSS et al., 2009) These include
the flagship Domestic Violence (DV) Act 732, establishment of a
unit within the Ghana Police Service for violence prevention and
support of victims, and laws restricting access and use of weapons.
(Pool et al., 2014; WHO, 2014b) Notwithstanding this legal back-
drop, socio-economic and gender inequalities and cultural norms in
the socialization of males and females continue to create an at-
mosphere conducive to violence, with the scale usually tipped
against women and children (Morris, 2012). Cultural norms are
such that victims of sexual violence may be portrayed as deserving
of the attack because of behavior or dress, while the perpetrator
may be seen as a jealous partner showing love (Morris, 2012). These
cultural perspectives persist; a study among youth in Ghana
showed that respondents leaned towards acceptance of violence
towards women (Glover et al., 2003). Even though Ghana Education
Service has taken several actions to abolish corporal punishment in
schools, including revision of the teachers' handbook and making
teachers aware of the negative consequences, corporal punishment
is still widely used as a disciplinary measure. It is estimated that
more than 80% of children in Ghana have experienced caning in
school. (UNICEF, 2014a) Factors contributing to this practice include
traditional beliefs, poverty, and personal challenges faced by stu-
dents as well as teachers. (UNICEF, 2014b) Some surveys report an
attitude of tolerance of the practice among parents, teachers, and
even students, with some studies drawing attention to the use of
physical punishment to make boys tougher, while enforcing female
submission (Dunne et al., 2006;EPC, 2013). Studies show that
corporal punishment and its acceptance increase the risk of further
violence and adverse socio-economic outcomes. (Dunne et al.,
2006; Pickett et al., 2015; UNICEF, 2014b).
In the present study, we used the nationwide Global School-
based Health Survey, conducted among senior high school stu-
dents in Ghana, to explore risk and protective factors associated
with sexual and physical violence victimization. We assessed
multiple risk and protective factors at the individual, family, peer
S.-A. Ohene et al. / Social Science & Medicine 146 (2015) 266e275 267

and school levels to provide a broader perspective of factors
contributing to our variables of focus. We hypothesized that risk
and protective factors unique to each type of violence victimization
and that cross-cut both physical and sexual violence victimization
at individual, peer, family and school levels would emerge.
2. Methods
2.1. Study design and sample
The study utilized data from the 2012 Ghana Global School-
based Student Health Survey (GSHS) (WHO, 2014c). GSHS is a
cross-sectional survey conducted in interested WHO member
states to assess the behavioral risk factors and protective factors in
multiple areas among youth in school. The United States Centers for
Disease Control and Prevention (CDC) provides technical assis-
tance. Data is collected by means of a self-administered question-
naire. More information on the methodology and the topics
addressed by the survey are available on the WHO website (WHO
2014d). The conduct of this survey in several countries in the
different regions of the world and the analyses of the data has
facilitated various collaborations across continents. Partnership for
the 2012 Ghana GSHS included WHO, CDC, Middle Tennessee State
University (MTSU) and Ghana Education Service (GES). The par-
ticipants in the Ghana 2012 GSHS consisted of a sample of students
in senior high schools (SHS) in Ghana. Following WHO GSHS
implementation guidelines, by means of a two-stage cluster sam-
pling design, 25 schools representing all 10 geographic regions in
Ghana were selected. Selection of schools was based on a proba-
bility proportional to the size of enrollment. In the second stage,
senior high school classes in the selected schools were randomly
chosen. Each student in selected classes was eligible to participate.
All students in a selected school had an equal probability of selec-
tion. A numerical weight was applied to each student record to
enable generalization of results to the eligible population. A total of
1984 students, 1065 (49.8%) male and 908 (50.2%) female, were
surveyed. Most students were 18 years or older (n ¼1062, 55.5%).
The students were relatively equally split across the four Senior
High School grade levels. Two-thirds of those surveyed were
boarding students.
2.2. Ethical considerations
The 2012 Ghana GSHS questionnaire was pilot tested to ensure
adequate comprehension of the survey items. Policies laid out by
the Ghana Education Service (GES) regarding consent procedures
for participation in surveys were followed. Official permission in
writing was requested and obtained fromGES, the selected schools,
and classroom teachers. Parental consent was sought and partici-
pation by students was entirely voluntary and anonymous. The
Institutional Review Board from the institution of the principal
investigator gave approval for the study.
2.3. Measures
2.3.1. Dependent variables
With the focus of our study being sexual and physical violence
victimization, the two dependent variables examined commensu-
rate with this focus. The question “Have you ever been forced to
have sexual intercourse when you did not want to?”was used to
assess whether a person had been at the receiving end of sexual
violence. Those who answered “yes”were classified as victims of
sexual violence. This variable used in our study features in the
Youth Risk Behavior Survey as one of the items used to assess
sexual assault. (CDC, 2015) The survey questionnaire did not
include an assessment of the possible items within the spectrum of
sexual victimization such as unwanted touching, kissing, unwanted
attempted sex, or pressured/coerced sex; hence the exclusive use of
this variable to assess sexual violence victimization. For physical
violence, the following question was used: “During the past 12
months, how many times were you physically attacked?”Partici-
pants responding one or more times were categorized as being a
victim of physical violence. This categorizationwas utilized because
we were interested in assessing factors associated with any
victimization, and it also allowed us to conduct analyses that were
parallel to our assessment of sexual violence victimization, which
was a dichotomous variable in the dataset eyes, no. There were 20
missing responses for the physical victimization item and 11 for the
sexual victimization item.
2.3.2. Independent variables
Selection of independent variables was informed by previous
studies of risk and protective factors in adolescents (Table 1)
(Resnick et al., 1997; Blum, 2007; Brown et al., 2009; Andersson
et al., 2014). Measures were assessed as dichotomous or contin-
uous based on the survey response options and the distribution of
responses across options. Demographic variables assessed were age
and gender.
2.4. Data analysis
Analyses were performed using SPSS 22. Sample weights were
used in all analyses so results are generalizable to the population.
Preliminary analyses were conducted to determine the prevalence
of risk and protective factors among youth who reported physical
or sexual violence. The primary analyses were performed in three
steps to determine factors most strongly associated with being
sexually or physically attacked among youth. First, bivariate ana-
lyses (chi-square tests and independent samples t-tests) were used
to examine relations between the independent variables and being
sexually or physically attacked. Variables that demonstrated sig-
nificant differences between those who reported sexual or physical
violence and those who did not at or below the 0.05 level (p .05)
were then entered into logistic regression models at the second
step. The second step involved creating chunk-wise logistic
regression models separately for risk factors and for protective
factors for sexual violence and for physical violence. Three logistic
regression models were tested for each dependent variable, sexual
violence and physical violence. The risk factors models included
age, sex, and risk factors significant in the bivariate analysis. The
protective factors models included age, sex, and significant pro-
tective factors in the bivariate analysis. All variables from these two
logistic regression models (both risk and protective factors) were
then entered simultaneously into final logistic regression models to
determine factors associated with sexual or physical violence
among youths. Demographic variables (age and sex) were included
in all logistic regression models. Continuous variables entered into
the logistic regression models were calculated on a 0 to 1 scale to
make interpretations of odds ratios for the variables more
comparable.
3. Results
3.1. Univariate analysis
In all 358 (18.6%) young people reported ever being forced to
have sexual intercourse when they did not want to. A total of 737
(38.0%) youth reported being physically attacked one or more times
during the past 12 months comprising 290 (14.1%) who had been
attacked once, 232 (12.1%) attacked 2 or 3 times, 79 (4.2%) attacked
S.-A. Ohene et al. / Social Science & Medicine 146 (2015) 266e275268

4 or 5 times, 47 (2.5%) and 89 (4.5%) attacked 8 or more times.
3.2. Bivariate analysis
Bivariate findings are presented in Tables 2 and 3; categorical
variables in Table 2, continuous variables in Table 3. For each risk
and protective factor in Table 2, the number and percent of re-
spondents with and without each of the factors who reported
sexual violence victimization and physical violence victimization
are listed.
Young women were more than twice as likely as young men to
report sexual violence (p <.001) (Table 2). Report of physical
violence did not differ significantly by gender. Age was not signif-
icantly associated with report of sexual or physical violence
(Table 3). There was a significant positive correlation between
report of sexual and physical violence and most individual level risk
factors examined (Tables 2 and 3). More than one in four youth who
reported each of the following eever having sex regardless of
condom use at last sex, ever being electronically bullied, lifetime
cigarette smoking, and lifetime alcohol use ealso reported being
victims of sexual violence. More than half of youth who reported
each of the following ebullying victimization, perpetrating
bullying, attempting suicide, lifetime cigarette smoking, drinking
alcohol in the past 30 days, and lifetime drug use ealso reported
being victims of physical violence.
Family level risk factors positively associated with physical
violence included food insecurity and parental alcohol and tobacco
use. Parental alcohol use was also positively associated with report
of sexual violence. Being a boarding student was not a risk factor for
being a victim of sexual or physical violence. None of the individual
Table 1
Independent variables.
Variable Description of variable
Risk factors
Individual level
Sexual activity and use of
condom
The last time you had sexual intercourse, did you or your partner use a condom (I have never had sex/Yes, used condom at last sex/No
did not use condom at last sex)
Felt lonely During the past 12 months, how often have you felt lonely? (5-point scale from never to always)
Felt worried During the past 12 months, how often have you been so worried about something that you could not sleep at night? (5-point scale from
never to always)
Felt sad or hopeless During the past 12 months, did you ever feel so sad or hopeless almost every day for two weeks or more in a row that you stopped doing
your usual activities? (yes/no)
Suicide plan During the past 12 months, did you make a plan about how you would attempt suicide? (yes/no)
Suicide attempt During the past 12 months, how many times did you actually attempt suicide? (0/1 or more times)
Victim of bullying During the past 30 days, on how many days were you bullied? (dichotomized to 0 days/1 or more days)
Witnessed bullying During the past 30 days, how often did you see other students being bullied? (5-point scale from never to always)
Electronically bullied Have you ever been electronically bullied? (count being bullied through e-mail, chatrooms, instant messaging, websites, or texting)
(yes/no)
Bullied others During the past 30 days, how often did you bully other students (dichotomized to never/rarely to always)
Lifetime cigarette smoking How old were you when you first tried a cigarette? (dichotomized to I have never smoked cigarettes/age first tried)
Lifetime alcohol use How old were you when you had your first drink of alcohol other than a few sips? (dichotomized to I have never had a drink of alcohol
other than a few sips/age first had)
Alcohol past 30 days During the past 30 days, on how many days did you have at least one drink containing alcohol? (dichotomized to 0 days/1 or more days)
Lifetime drug use How old were you when you first used drugs? (dichotomized to I have never used drugs/age first used)
Truancy During the past 30 days, on how many days did you miss classes or school without permission? (dichotomized to 0 days/1 or more days)
Family level
Parental tobacco use Which of your parents or guardians use any form of tobacco? (neither/my mother or female guardian, my father or male guardian, both,
I don't know)
Parental alcohol use Which of your parents or guardians drink alcohol? (neither/my mother or female guardian, my father or male guardian, both, I don't
know)
Food insecurity During the past 30 days, how often did you go hungry because there was not enough food in your home? (5-point scale from never to
always)
School level
Boarding student During this school term are you a boarding or day student? (boarding student/no)
Protective factors
Individual level
Self-perceived health How do you describe your health in general? (5-point scale from poor to excellent)
Physical activity During the past 7 days, on how many days were you physically active for a total of at least 60 min per day? (analyzed as continuous
variable in days; range 0e7 days)
Family level
Care from both parents Who is most responsible for taking care of your needs? (dichotomized to both parents/other)
Parents respect privacy During the past 30 days, how often did your parents or guardians go through your things without your approval? (5-point scale from
never to always)
Parents check homework During the past 30 days, how often did your parents or guardians check to see if your homework was done? (5-point scale from never to
always)
Parents understand problems During the past 30 days, how often did your parents or guardians understand your problems and worries? (5-point scale from never to
always)
Parental monitoring During the past 30 days, how often did your parents or guardians really know what you were doing with your free time? (5-point scale
from never to always)
Peer level
Friends who don't drink How many of your friends frequently drink 5 or more drinks on one occasion? (dichotomized to none/a few, some, most, all)
Close friends How many close friends do you have? (analyzed as continuous variable; range 0e3þ)
Friends not sexually active How many of your friends have had sexual intercourse? (dichotomized to none/a few, some, most, all)
School level
School mates kind and helpful During the past 30 days, how often were most of the students in your school kind and helpful? (5-point scale from never to always)
School teaches lessons on
alcohol refusal
During this school year, were you taught in any of your classes how to tell someone you did not want to drink alcohol? (yes/no)
S.-A. Ohene et al. / Social Science & Medicine 146 (2015) 266e275 269

Table 2
Percentage of Ghanaian youth reporting sexual or physical violence according to demographic, risk, and protective factors (Chi-square).
Sexual violence Physical violence
Categorical variables N (%) X
2
N (%) X
2
Demographics
Sex
Female 232 (27.5) 69.00*** 351 (39.1) 1.95
Male 120 (11.3) 380 (36.0)
Risk factors
Individual level
Sexual activity and use of condom
No, never had sex 159 (12.8) 75.51*** 432 (34.8) 3.05
Yes, used condom at last intercourse 95 (31.1) 112 (38.2)
Yes, did not use condom at last intercourse 81 (25.5) 122 (38.5)
Victim of bullying
Yes 184 (23.0) 14.05*** 398 (50.1) 104.59***
No 144 (14.0) 280 (27.3)
Electronically bullied
Yes 97 (25.5) 17.55** 184 (48.8) 27.16**
No 256 (16.3) 537 (34.4)
Bullied others
Yes 110 (24.9) 16.90*** 243 (55.5) 78.35***
No 248 (16.3) 487 (32.2)
Felt sad or hopeless
Yes 202 (22.7) 22.02** 401 (45.3) 42.05**
No 143 (14.5) 326 (31.0)
Suicide plan
Yes 102 (23.4) 11.09* 214 (49.2) 33.63**
No 247 (16.4) 507 (33.9)
Suicide attempt
Yes 101 (23.2) 9.58** 249 (57.5) 96.62***
No 254 (16.7) 479 (31.6)
Lifetime cigarette smoking
Yes 53 (27.7) 14.00*** 108 (56.8) 38.56***
No 271 (16.8) 547 (33.9)
Lifetime alcohol use
Yes 162 (28.7) 63.34*** 245 (43.7) 24.49***
No 148 (12.9) 360 (31.5)
Alcohol past 30 days
Yes 69 (29.6) 22.80*** 127 (55.5) 38.37***
No 270 (16.7) 555 (34.30)
Lifetime drug use
Yes 53 (21.0) 3.12 127 (51.0) 30.38***
No 260 (16.5) 518 (33.0)
Truancy
Yes 149 (23.9) 21.41** 281 (45.3) 23.62**
No 202 (15.2) 447 (33.9)
Family level
Parental tobacco use
Yes 99 (21.0) 3.65 222 (47.3) 25.88**
No 256 (17.1) 510 (34.3)
Parental alcohol use
Yes 143 (21.9) 9.75* 283 (43.7) 15.99***
No 212 (16.1) 450 (34.4)
School level
Boarding student
Yes 235 (18.9) 3.00 477 (38.4) 0.93
No 103 (15.7) 234 (36.1)
Protective factors
Family level
Care from both parents
Yes 156 (19.5) 1.89 298 (37.5) 0.03
No 195 (17.1) 431 (37.8)
Peer level
Friends who don't drink
Yes 212 (16.1) 9.75** 450 (34.4) 15.99***
No 143 (21.9) 283 (43.7)
Friends not sexually active
Yes 96 (10.9) 56.40*** 287 (32.7) 15.70*
No 257 (24.1) 438 (41.4)
School level
School teaches lesson on alcohol refusal
Yes 169 (18.5) 0.356 339 (37.4) 0.019
No 187 (17.9) 391 (37.6)
N(%) ¼number and percent of respondents with and without each of the factors who reported sexual violence victimization and physical violence victimization.
*p <.05; **p <.01; ***p <.001.
S.-A. Ohene et al. / Social Science & Medicine 146 (2015) 266e275270

level factors assessed were protective. At the family level, perceived
parental understanding of the youth's problems was protective
against both sexual and physical violence. Parents checking
homework, and parental monitoring were protective against sexual
violence only, while reporting that parents were less likely to go
through one's things without approval (parent respect for privacy)
was protective against physical violence only. Being cared for by
both parents was not protective against either violence victimiza-
tion variables. At the peer level, having friends with prosocial be-
haviors (friends who don't drink, friends who are not sexually
active) was protective against both sexual and physical violence.
Having friendly school mates was protective against sexual
violence; no other school level variables were significantly associ-
ated with either violence victimization variable.
3.3. Multivariate analysis
In logistic regression models that included gender, age, and risk
factors significant in bivariate analyses, we found that report of
sexual violence remained significantly associated with ever having
sex with (odds ratio, OR: 3.17; 95% confidence interval, 95% CI:
2.01e4.98) or without (OR: 3.26; 95% CI: 2.14e4.95) condom use at
last sex, being a victim of bullying (OR: 1.37; 95% CI: 1.02e1.84), and
electronic bullying (OR: 1.44; 95% CI: 1.02e2.02). (Table 4) Report of
physical violence remained significantly correlated with being a
victim of bullying (OR: 1.38; 95% CI: 1.05e1.85), bullying perpe-
tration (OR: 2.04; 95% CI: 1.67e2.48), attempting suicide (OR: 2.20;
95% CI: 1.30e3.70), and alcohol use in the previous thirty days (OR:
2.18; 95% CI: 1.21e3.97). In models that included gender, age, and
protective factors significant in bivariate analyses, we found that
only report of friends not being sexually active remained protective
(OR: 0.39; 95% CI: 0.28e0.54) against sexual violence victimization.
Parent respect for privacy (OR: 0.73; 95% CI: 0.59e0.93), friends not
drinking (OR: 0.63; 95% CI: 0.48e0.81), and friends not sexually
active (OR: 0.75; 95% CI: 0.61e0.92) were protective against phys-
ical violence victimization.
When all the risk and protective factors in these two initial
models, along with age and gender, were combined in a final
multivariate model, sexual activity with (OR: 2.43; 95% CI:
1.55e3.81) or without (OR 2.51; 95% CI: 1.60e3.90) condom use at
last sex, being a victim of bullying (OR: 1.56; 95% CI: 1.12e2.17), and
being electronically bullied (OR: 1.58; 95% CI: 1.10e2.27) remained
significant risk factors, and friends not sexually active (OR: 0.53;
95% CI: 0.34e0.84) remained protective for sexual violence
victimization. Parental monitoring nearly reached significance as a
protective factor in this final model for sexual violence victimiza-
tion (OR: 0.82; 95% CI: 0.66e
1.0 0 ; p ¼.06). Drinking alcohol in the
past 30 days (OR: 2.25; 95% CI: 1.20e4.20), attempting suicide (OR:
2.10; 95% CI: 1.25e3.54), and bullying others (OR: 1.96; 95% CI:
1.55e2.45) remained risk factors for physical violence victimiza-
tion. Parent respect for privacy just reached significance as a pro-
tective factor for physical violence victimization (OR: 0.83; 95% CI:
0.69e1.0 0 ; p ¼.05). (Table 4).
4. Discussion
This study examined risk and protective factors associated with
sexual and physical violence victimization among Ghanaian senior
high school students. A fifth of these students reported being forced
to have sex in their lifetime while two out of five had been a victim
of a physical attack in the year preceding the survey. In our final
multivariate analyses, a number of factors were found to be
significantly associated with sexual and physical violence victimi-
zation. For sexual violence victimization having friends who were
not sexually active was protective while history of having had sex
regardless of condom use at last sex, feeling sad or hopeless, being a
victim of bullying, and being electronically bullied were identified
as risk factors. Risk factors for physical violence victimization were
attempting suicide, drinking alcohol in the past 30 days, and
bullying others in the past month; a measure of parent respect for
youth privacy just reached the level of significance as a protective
factor.
The overall magnitude of sexual violence victimization among
our study population falls within the range of what has been re-
ported in studies from other African countries (Erulkar, 2004; Blum,
2007; Brown et al., 2009; Breiding et al., 2011). The finding of 28% of
Table 3
Percentage of Ghanaian youth reporting sexual and physical violence according to demographic, risk, and protective factors (T-tests).
Sexual violence Physical violence
Yes No Yes No
Continuous variables Range/skewness N m (sd) N m (sd) t-statistic N m (sd) N m (sd) t-statistic
Age (years) 12e18þ358 16.65 (0.73) 1608 16.60 (0.78) 1.06 736 16.64 (0.76) 1221 16.59 (0.79) 0.82
ŧ
Risk Factors
Individual level
Witnessed bullying 1 (never) e5 (always)/0.195 350 3.14 (1.23) 1599 2.98 (1.22) 2.24* 725 3.22 (1.19) 1215 2.89 (1.23) 5.76***
Felt lonely 1 (never) e5 (always)/0.168 357 2.96 (1.02) 1611 2.71 (2.96) 4.06** 732 2.83 (1.11) 1227 2.70 (1.10) 2.45*
Felt worried 1 (never) e5 (always)/0.102 357 2.52 (1.13) 1613 2.91 (1.03) 6.47** 733 2.76 (1.08) 1227 2.48 (1.13) 5.35***
Family level
Food insecurity 1 (never) e5(always)/0.066 357 2.59 (1.13) 1613 2.38 (1.16) 3.09** 735 2.58 (1.13) 1226 2.32 (1.16) 4.81***
ŧ
Protective factors
Individual level
Self-perceived health 1 (poor) e5 (excellent)/0.752 349 2.26 (1.15) 1594 2.16 (1.09) 1.59 726 2.21 (1.16) 1209 2.15 (1.07) 1.21
ŧ
Physically active (days) 0e7/0.662 348 3.78 (2.51) 1603 3.61 (2.50) 1.10 730 3.69 (2.39) 1212 3.62 (2.57) -.58
ŧ
Family level
Parents respect privacy 1 (never) e5 (always)/0.626 351 2.29 (1.41) 1600 2.22 (1.35) 0.89 728 2.45 (1.38) 1214 2.10 (1.33) 5.58***
Parents check homework 1 (never) e5 (always-0.118 350 2.81 (1.63) 1602 3.15 (1.61) 3.51* 729 3.14 (1.59) 1215 3.06 (1.64) 1.04
Parents understand problems 1 (never) e5(always)/0.338 351 3.13 (1.39) 1582 3.39 (3.13) 3.13** 724 3.26 (1.35) 1201 3.39 (1.40) 2.03*
ŧ
Parental monitoring 1 (never) e5 (always)/0.204 344 2.93 (1.48) 1591 3.20 (1.43) 3.17** 722 3.13 (1.41) 1205 3.17 (1.46) 0.63
Peer level
Close friends 0e3±0.176 353 1.57 (1.03) 1605 1.76 (1.04) 3.17** 729 1.69 (1.04) 1219 1.75 (1.05) 1.11
School level
School mates kind and helpful 1 (never) e5 (always)/0.184 350 3.13 (1.12) 1597 3.29 (1.22) 2.34* 730 3.21 (1.21) 1210 3.30 (1.20) 1.55
ŧ¼using Equal variances not assumed * p<.05; **p<.01; ***p<.001 m emean, sd estandard deviation.
S.-A. Ohene et al. / Social Science & Medicine 146 (2015) 266e275 271

Table 4
Odds ratios for physical and sexual violence victimization.
Sexual violence Physical violence
Odds ratio Confidence interval Odds ratio Confidence interval
a. Logistic regression results for risk factors N¼1415 N ¼1358
Sex 0.29*** (0.21, 0.41) 0.82 (0.61, 1.08)
Age 0.95 (0.82, 1.10) 0.97 (0.84, 1.11)
Individual level
No condom used at last sex ee
Yes condom use vs. never had sex 3.17*** (2.01, 4.98) ee
No condom use vs. never had sex 3.26*** (2.14, 4.95) ee
Felt lonely 1.16 (0.91, 1.50) 1.03 (0.88, 1.15)
Felt worried 1.15 (0.93, 1.42) 1.06 (0.93, 1.21)
Felt sad or hopeless 1.37 (0.98, 1.92) 1.54 (0.95, 2.48)
Suicide plan 1.47 (0.89, 2.42) 1.18 (0.72, 1.95)
Suicide attempt 1.28 (0.88, 1.88) 2.20* (1.30, 3.70)
Victim of bullying 1.37* (1.02, 1.84) 1.38* (1.05, 1.85)
Witnessed bullying 1.02 (0.82, 1.26) 1.12 (0.97, 1.30)
Electronically bullied 1.44* (1.02, 2.02) ee
Bullied others 1.22 (0.76, 1.94) 2.04*** (1.67, 2.48)
Lifetime cigarette smoking 1.50 (0.51, 1.21) 1.19 (0.71, 2.02)
Alcohol past 30 days 1.27 (0.81, 2.80) 2.18** (1.21, 3.97)
Lifetime drug use ee 1.69 (0.98, 2.94)
Truancy 1.01 (0.82, 1.25) 1.23 (0.77, 1.99)
Family level
Parental alcohol use 1.05 (0.68, 1.62) 1.08 (0.75, 1.61)
Parental tobacco use ee 1.30 (0.75, 2.25)
Food insecurity 0.97 (0.74, 1.27) 1.04 (0.90, 1.21)
School level
Boarding Student ee 1.13 (0.78, 1.64)
b. Logistic regression results for protective factors N¼1829 N ¼1866
Sex 0.33*** (0.26, 0.42) 0.86 (0.70, 1.07)
Age 1.05 (0.94, 1.18) 1.02 (0.85, 1.22)
Family level
Parents respect privacy ee 0.73** (0.59, 0.93)
Parents check homework 0.90 (0.74, 1.10) ee
Parents understand problems 0.92 (0.73, 1.16) 0.88 (0.75, 1.02)
Parental monitoring 0.88 (0.77, 1.01) ee
Peer level
Friends who don't drink 0.91 (0.67, 1.23) 0.63** (0.48, 0.81)
Close friends 0.93 (0.82, 1.06) ee
Friends don't have sex 0.39* (0.28, 0.54) 0.75* (0.61, 0.92)
School level
School mates kind and helpful 0.94 (0.82, 1.08) ee
c. Final logistic regression model with all risk and protective factors N¼1363 N ¼1328
Sex 0.26*** (0.18, 0.36) 0.78 (0.58, 1.06)
Age 0.93 (0.81, 1.06) 0.98 (0.86, 1.12)
Risk factors
Individual level
No condom used at last sex ee
Yes condom use vs. never had sex 2.43*** (1.55, 3.81) ee
No condom use vs. never had sex 2.51*** (1.6, 3.90) ee
Felt lonely 1.15 (0.88, 1.50) 1.00 (0.86, 1.16)
Felt worried 1.14 (0.93, 1.40) 1.06 (0.93, 1.21)
Felt sad or hopeless 1.38* (1.02, 1.87) 1.52 (0.93, 2.55)
Suicide plan 1.56 (0.92, 2.65) 1.16 (0.72, 1.86)
Suicide attempt 1.28 (0.88, 2.02) 2.10* (1.25, 3.54)
Victim of bullying 1.56** (1.12, 2.17) 1.34 (0.96, 1.87)
Witnessed bullying 1.00 (0.81, 1.24) 1.12 (0.97, 1.29)
Electronically bullied 1.58* (1.10, 2.27) ee
Bullied others 1.24 (0.78, 1.97) 1.96* (1.55, 2.45)
Lifetime cigarette smoking 1.34 (0.66, 2.68) 1.20 (0.71, 2.03)
Alcohol past 30 days 1.27 (0.83, 1.95) 2.25* (1.20, 4.20)
Lifetime drug use ee 1.63 (0.72, 2.63)
Truancy 1.09 (0.88, 1.35) 1.24 (0.75, 2.03)
Family level
Parental alcohol use 1.10 (0.74, 1.62) 1.13 (0.78, 1.62)
Parental tobacco use ee 1.36 (0.79, 2.32)
Food insecurity 0.94 (0.76, 1.17) 1.04 (0.89, 1.19)
School level
Boarding Student ee 1.09 (0.73, 1.60)
Protective factors
Family level
Parents respect privacy ee 0.83* (0.69, 1.00)
Parents check homework 0.80 (0.65, 1.03) ee
Parents understand problems 1.07 (0.87, 1.32) 0.97 (0.82, 1.13)
Parental monitoring 0.82 (0.66, 1.00) ee
S.-A. Ohene et al. / Social Science & Medicine 146 (2015) 266e275272

girls reporting sexual violence victimization in the present study
using the 2012 Global School-based Health Survey (GSHS) was
higher than the 17% reported for 15e19 year old girls in the 2008
Ghana Demographic and Health Survey (GDHS) (GSS et al., 2009)
and the 18% found among female teenagers in two towns in Ghana
(Bingenheimer et al., 2014). The difference may be due to mea-
surement in different study populations: students in the present
study, in and out of school adolescents in the 2008 GDHS, and
adolescents from two towns in Ghana.
In concurrence with other studies, we found that more female
than male respondents reported sexual violence victimization;
however, the sex difference was greater in our study. (Brown et al.,
2009; Andersson et al., 2012) One study of street youth in Ghana
also showed a highly significant difference between the pro-
portions of females reporting sexual abuse compared to their male
counterparts (Oppong et al., 2014). The dynamics of gender
inequality in adolescent relationships with limited communication
about sexual matters and males having the upper hand in decision-
making may be a contributory factor to this female vulnerability
(Moore et al., 2007). Unfortunately, socio-cultural perceptions and
poor law enforcement, especially in Africa, create an environment
in which sexual violence is downplayed, ignored, or even perceived
to be a show of love (Moore et al., 2007; Erulkar, 2004; Sommera
et al., 2013; de Vries et al., 2014). Societal norms that project the
use of violence on women as a show of masculinity and control over
women as an expression of love, as well as the acceptance of this
negative behavior by women, complicates the terrain and the quest
to prevent sexual violence (Ybarra et al., 2012; De Vries et al., 2014).
The study also points out that it cannot be assumed that only fe-
males are victims of sexual violence. At least one in ten males also
reported sexual violence victimization, highlighting the need to
elicit relevant history among males. In view of the magnitude of
sexual violence victimization and its ramifications, including the
risk of both sexual violence victimization and perpetration among
youth (Ybarra et al., 2012), it is important to promote awareness
about the problem among parents, clinicians who attend to young
people, school authorities, and adolescent health advocates so that
appropriate steps can be taken to prevent and address it.
The administration of the GSHS in other areas of the world have
revealed reports of violence victimization varying from 18% in
Cambodia to 57% in Dar es Salam, Tanzania (Tozija, 2007;
Nyandindi, 2008; DOHT, 2008; RB MOH, 2011; Prak and Chher,
2013), and the rates reported in our study fall within this range.
Contrary to the expectation that more males would report being
victims, physical violence victimization was about the same across
gender. The figures obtained in our study are similar to that of the
2008 Ghana Demographic Health Survey, in which 32.3% of female
respondents and 34.2% of male respondents aged 15e19 years re-
ported being victims of physical violence (GSS et al., 2009). Both
boys and girls in Ghana are exposed to corporal punishment in
schools. Teachers were cited as the most common perpetrator of
physical violence among adolescent girls in Ghana (UNICEF, 2014b).
Not assessed in this study, mechanisms of violence and violence
severity may differ by gender.
Our study showed the association between sexual violence
victimization and some individual level risk factors such as being a
victim of bullying and sexual activity with or without condom use
at last sex. Risky sexual behavior puts one at risk for transmission of
sexually transmitted infections, HIV and unwanted pregnancy,
conditions that have been linked to sexual violence in other studies
(WHO, 2002; Blum, 2004; Ohene et al., 2005; Campbell, 2008;
Ybarra et al., 2012; Roman and Frantz, 2013). Blum (2004) in his
report on the health of young people in a global context highlighted
the myriad of behavioral and psychological problems, depression,
low self-esteem and other mental health problems that sexual
abuse and coercion could lead to. Even though more victims of
sexual violence in our study reported mental health symptoms, in
the multivariate model feeling sad and hopeless was the only sig-
nificant variable. Song and colleagues in their study of Chinese
adolescents reported that more female sexual violence victims
exhibited internalizing symptoms, while their male counterparts
exhibited more externalizing symptoms (Song et al., 2014). Future
studies should evaluate findings among male and female victims of
sexual violence in Ghana.
For physical violence victimization, individual level risk factors
that were significant in the final multivariate model were suicide
attempt, alcohol use, and bullied others. These findings are
consistent with other studies. (Randall et al., 2014; Brown et al.,
2009;Nansel et al., 2003). The association between alcohol use
and violence involvement, both as a victim and perpetrator, is
evident worldwide (WHO, 2006; Hazemba et al., 2008; Brown
et al., 2009). It is clear that interventions targeting access to
alcohol, such as higher taxation and restrictions on alcohol sales
times and locations, reduce consumption levels and violence
(WHO, 2009). Ghana is in the process of developing its alcohol
policy. Considering the current voluntary restrictions on alcohol-
associated sponsorship of youth events (WHO, 2004), the revision
presents a timely opportunity to step up restrictions and regula-
tions on access to alcohol as a violence prevention intervention.
Evaluating the impact of such strategies on youth violence is key.
Unlike other studies which have shown connection to family
and school to be protective against physical and sexual violence
victimization, most of the family, peer, and school level variables in
our study did not show a clear cut significant protective association
in the final models. (Rudatsikira et al., 2007; Resnick et al., 1997;
WHO, 2007) In a study on factors associated with early sexual
debut, living with both parents was also not found to be protective
for Ghanaian adolescents (Stephenson et al., 2014). In their study of
factors associated with sexual violence among girls in Swaziland,
Brieding and colleagues showed that those who reported being
close to their biological mothers were less likely to be victims of
sexual violence compared to those who were not close to their
mothers (Breiding et al., 2011). Thus, the nature of the relationship
with a parent may be paramount.
Table 4 (continued )
Sexual violence Physical violence
Odds ratio Confidence interval Odds ratio Confidence interval
Peer level
Friends who don't drink 0.91 (0.67, 1.23) 0.89 (0.71, 1.13)
Friends don't have sex 0.53** (0.34, 0.84) 0.98 (0.69, 1.39)
Close friends 0.97 (0.80, 1.17) ee
School level
School mates kind and helpful 1.01 (0.85, 1.21) ee
*p <.05; **p <.01; ***p <.001.
S.-A. Ohene et al. / Social Science & Medicine 146 (2015) 266e275 273

The only peer level factor that showed a significant correlation
with violence victimization in the final regression model was
having friends who were not sexually active, which was protective
against sexual violence victimization. In a study of clinic-attending
adolescents, age at consensual sex was higher among adolescent
girls without a history of forced sex compared to their counterparts
with such a history. Young people who have not been victims of
sexual violence victimization may delay sexual initiation and have
like-minded friends with similar desires to defer initiation of sexual
activity. (Ohene et al., 2005) Positive peer support is associated
with fewer health risk behaviors (Saewyc and Tonkin, 2008).
Among youth in British Columbia who had experienced lifetime
abuse (physical, sexual or both), having peers with prosocial norms,
such as disapproval of beating someone up and alcohol and drug
use, was associated with lower levels of health risk behaviors,
including violence involvement (Saewyc et al., 2006). Schools and
community groups in Ghana should develop strategies to promote
and support positive peer support.
Given this study's finding, there is much that needs to be done to
address violence victimization among youth. In addition to the
critical need to institute primary prevention programs in schools,
there must be systems in place for identifying victims of violence,
assessing risk and protective factors, and offering the necessary
counseling and support to forestall other health compromising
consequences. Ghana does not currently have a National Action
Plan to address Youth Violence. Addressing this gap at the national
level could provide the necessary backing for concrete steps to
prevent and roll out interventions targeted at youth violence
(WHO, 2014b).
This study used the 2012 Global School-based Student Health
Survey and as such, the findings preclude generalization to out-of-
school adolescents in Ghana. Secondly, the variables analyzed were
based on participants' self-report, with no means of verification.
Consequently, reporter bias, especially under-reporting of sexual
victimization due to its sensitive nature, cannot be excluded
(Bingenheimer, 2014). Due to the cross-sectional design of the
study, attributing causality or directionality of the risk and pro-
tective factors examined is not possible. The context and mecha-
nism of violence victimization were beyond the scope of this study,
but would have provided a better understanding of the contexts in
which violence occurs and more insight into possible preventive
measures and interventions to reduce the risk of others falling
victim. The content of the available questionnaire limited the scope
of the range of independent variables that could be analyzed such
as some aspects of the school environment or socio-economic
status. Nevertheless, the study used nationally representative
data to explore factors associated with sexual and physical violence
victimization, variables that hitherto have infrequently been stud-
ied among Ghanaian youth.
5. Conclusion
This study brings to the fore the issue of violence victimization
in senior high schools. In envisaging violence prevention ap-
proaches, it is important to understand and take into consideration
the contribution of the complex interplay of multiple factors at
work as highlighted by the ecological framework. It is therefore
imperative for stakeholders, including parents, clinicians caring for
adolescents, and school authorities, to be alerted to the magnitude
of the problem and associated risk and protective factors. Parent
teacher association meetings are possible fora for disseminating
information about the prevalence of violence victimization and its
associated risk and protective factors. The Ministry of Education in
Ghana should take steps to create a culture in schools for pre-
venting and recognizing the problem of violence victimization and
its associated negative health behaviors and provide concrete in-
terventions to improve the well-being of adolescents.
Acknowledgments
We thank the students, teachers, and Ghana Education Service
for their participation and assistance in the Global School-based
Student Health Survey. KJ was supported by a National Research
Service Award (NRSA) in Primary Medical Care, grant no.
T32HP22239 (PI: Borowsky), Bureau of Health Workforce, Health
Resources and Services Administration, Department of Health and
Human Services. SAO is a WHO staff member. The content of this
article does not necessarily represent WHO policies.
References
Andersson, N., Paredes-Solı

s, S., Milne, D., Omer, K., Nobantu Marokoane, N.,
Laetsang, D., Cockcroft, A., 2012. Prevalence and risk factors for forced or
coerced sex among school-going youth: national cross-sectional studies in 10
southern African countries in 2003 and 2007. BMJ Open 2012 (2), e000754.
http://dx.doi.org/10.1136/bmjopen-2011-000754.
Bingenheimer, J.B., Reed, E., 2014. Risk for coerced sex among female youth in
Ghana. Int. Perspect. Sex. Reprod. Health 40 (4), ,184e195. http://dx.doi.org/
10.1363/4018414.
Blum, R.W., Halc
on, L., Beuhring, T., Pate, E., Campell-Forrester, S., Venema, A., 2003.
Adolescent health in the Caribbean: risk and protective factors. Am. J. Public
Health 93 (3), 456e460.
Blum, R.W., 2007. Youth in Sub-Saharan Africa. J. Adolesc. Health 41, 230e238.
Blum, R.W., Nelson-Mmari, K., 2004. The health of young people in a global context.
J. Adolesc. Health 35, 402e418.
Borowsky, I.W., Ireland, I., 2004. Predictors of future fight-related injury among
adolescents. Pediatrics 113, 530e536.
Breiding, M.J., Reza, A., Gulaid, J., Blanton, C., Mercy, J.A., Dahlberg, L.L., Dlaminid, N.,
Bamrahe, S., 2011. Risk factors associated with sexual violence towards girls in
Swaziland. Bull. World Health Organ. 89, 203e210. http://dx.doi.org/10.2471/
BLT.10.079608.
Bronfenbrenner, U., 1979. The Ecology of Human Development: Experiments by
Nature and Design. Harvard University Press, Cambridge, MA.
Brown, D.W., Riley, L., Butchart, A., Meddings, D.R., Kann, L., Harvey, A.P., 2009.
Exposure to physical and sexual violence and adverse health behaviours in Af-
rican children: results from the Global School-based Student Health Survey. Bull.
World Health Organ. 87, 447e455. http://dx.doi.org/10.2471/BLT.07.047423.
Campbell, R., 2008. The psychological impact of rape victims. Am. Psychol. 63,
702e717.
Celedonia, K.L., Wilson, M.L., El Gammal, H.A., Hagras, A.M., 2013. Physical fighting
among Egyptian adolescents: social and demographic correlates among a na-
tionally representative sample. PeerJ 1, e125. http://dx.doi.org/10.7717/
peerj.125.
Centers for Disease Control and Prevention, 2009. Adverse childhood experiences
study. In: Publications on Major Findings by Health Outcomes. CDC, Atlanta, GA.
Retrieved July 10, 2014 from. www.cdc.gov/nccdphp/ACE/outcomes.htm.
Centers for Disease Prevention and Control (CDC ), 2015. YRBSS Data, Question-
naires, &Documentation. Retrieved August 10, 2015, from: http://www.cdc.
gov/healthyyouth/data/yrbs/data.htm.
De Vries, H., Eggers, S.M., Jinabhai, C., Meyer-Weitz, A., Sathiparsad, R., Taylor, M.,
2014. Adolescents' beliefs about forced sex in KwaZulu-natal, South Africa.
Arch. Sex. Behav. 43, 1087e1095. http://dx.doi.org/10.1007/s10508-014-0280-8.
Demographic and Health Survey 2008. Accra, Ghana: GSS, GHS, and ICF Macro.
Department of Health, Thailand (DOHT), 2008. Global School-based Student Health
Survey Thailand 2008 Report. Retrieved January 10, 2015, from: http://www.
who.int/chp/gshs/GSHS_FINAL_Report_Thailand_2008.pdf.
Dunne, M., Leach, F., Humphreys, S., 2006. Gender violence in schools in the
developing world. Gend. Educ. 18 (1), 75e98.
End Corporal Punishment (EPC), 2013. Global Initiative to End All Corporal Punish-
ment of Children. Country Report for Ghana. Retrieved August 10, 2015, from:
http://www.endcorporalpunishment.org/progress/country-reports/ghana.html.
Erulkar, A.S., 2004. The experience of sexual coercion among young people in
Kenya. Int. Fam. Plan. Perspect. 30 (4), 182e189 .
Ghana Statistical Service (GSS), 2012. Census 2010 Summary Report of Final Results.
Retrieved August 10, 2015, from: http://www.statsghana.gov.gh/docfiles/
2010phc/Census2010_Summary_report_of_final_results.pdf.
Ghana Statistical Service (GSS), Ghana Health Service (GHS), and ICF Macro. 2009.
Ghana.
Glover, E.K., Bannerman, A., Pence, B.W., Jones, H., Miller, R., Weiss, E., Nerquaye-
Tetteh, J., 2003. Sexual health experiences of adolescents in three Ghanaian
towns. Int. Fam. Plan. Perspect 29 (1), 32e40.
Hazemba, A., Seter Siziya, S., Muula, A.S., Rudatsikira, E., 2008. Prevalence and
correlates of being bullied among in-school adolescents in Beijing: results from
the 2003 Beijing Global School-Based Health Survey. Ann. General Psychiatry 7,
6. http://dx.doi.org/10.1186/1744-859X-7-6.
S.-A. Ohene et al. / Social Science & Medicine 146 (2015) 266e275274

Juvonen, J., Graham, S., Schuster, M.A., 2003. Bullying among young adolescents:
strong, weak, troubled. Pediatrics 112, 1231e1237. http://dx.doi.org/10.1542/
peds.112.6.1.
Moore, A.M., Awusabo-Asare, K., Madise, N., John-Langba, J., Kumi-Kyereme, A.,
2007. Coerced first sex among adolescent girls in Sub-Saharan Africa: preva-
lence and context. Afr. J. Reprod. Health 11 (3), 62e82.
Morris, J.N., 2012. The domestic violence act: Ghana's bright future. Retrieved
August 8, 2015 from http://repositories.lib.utexas.edu/handle/2152/19643.
Nansel, T.R., Overpeck, M.D., Haynie, D.L., Ruan, W.J., Scheidt, P.C., 2003. Relation-
ship between bullying and violence among US youth. Arch. Pediatr. Adolesc.
Med 157 (4), 348e353.
Nyandindi, U.S., 2008. Tanzania Global School-based Student Health Survey Report
2008. Retrieved January 10, 2015, from: http://www.who.int/entity/chp/gshs/
TANZANIA_GSHS_FINAL_REPORT_2008.pdf?ua¼1.
Ohene, S.A., Halcon, L., Ireland, M., Carr, P., McNeely, C., 2005. Sexual abuse history,
risk behavior and sexually transmitted diseases: the impact of age at abuse. Sex.
Transm. Dis. J. 32 (6), 358e363.
Oppong Asante, K., Meyer-Weitz, A., Petersen, I., 2014. Subst. Abuse Treat. Prev.
Policy 9, 45.
Owusu, A., Hart, P., Oliver, B., Kang, M., 2011. The association between bullying and
psychological health among senior high school students in Ghana, West Africa.
J. Sch. Health 81 (5), 231e238. http://dx.doi.org/10.1111/j.1746-
1561.2011.00590.x.
Patton, G.C., Coffey, C., Sawyer, S.M., Viner, R.M., Haller, D.M., Bose, K., Vos, T.,
Ferguson, J., Mathers, C.D., 2009. Global patterns of mortality in young people: a
systematic analysis of population health data. Lancet 374 (9693), 881e892.
http://dx.doi.org/10.1016/S0140-6736(09)60741-8.
Pickett, W., Elgar, F.J., 2015. Reducing school violence in Africa: learning from
Uganda. Lancet Glob. Health 3 (7). http://dx.doi.org/10.1016/S2214-109X(15)
00031-5 e344-5.
Pool, M.S., Otupiri, E., Owusu-Dabo, E., de Jonge, A., Agyeman, C., 2014. Physical
violence during pregnancy and pregnancy outcomes in Ghana. BMC Pregnancy
Childbirth 14, 71.
Prak, P.R., Chher, T., 2013. Global School-based Student Health Survey (GSHS)
Cambodia 2013 Country Report. Retrieved January 25, 2015, from: http://www.
who.int/chp/gshs/Cambodia_2014_GSHS_Report.pdf?ua¼1.
Randall, J.R., Doku, D., Wilson, M.L., Peltzer, K., 2014. Suicidal behaviour and related
risk factors among school-aged youth in the Republic of Benin. PLoS ONE 9 (2),
e88233. http://dx.doi.org/10.1371/journal.pone.0088233.
Republic of Benin Ministry of Health (RB MOH), 2011. Global School-based Student
Health Survey Benin 2009 Final Report. Retrieved January 10, 2015, from:
http://www.who.int/chp/gshs/Country_Report_GSHS_BENIN_2009.pdf?ua¼1.
Resnick, M.D., Bearman, P.S., Blum, R.Wm, Bauman, K.E., Harris, K.M., Jones, J.,
Tabor, J., Beuhring, T., Sieving, R.E., Shew, M., Ireland, M., Bearinger, L.H.,
Udry, J.R., 1997. Protecting adolescents from harm; findings from the national
longitudinal study on adolescent health. J. Am. Med. Assoc. 278, 823e832.
Resnick, M.D., Ireland, M., Borowsky, I., 2004. Youth violence perpetration: what
protects? What predicts? Findings from the national longitudinal study of
adolescent health. J. Adolesc. Health 35, 424.e1e424.e10.
Roman, N.V., Frantz, J.M., 2013. The prevalence of intimate partner violence in the
family:a systematic review of the implications for adolescents in Africa. Fam.
Pract. 30, 256e265. http://dx.doi.org/10.1093/fampra/cms084.
Rudatsikira, E., Siziya, S., Kazembe, L.N., Muula, A.S., 2007. Prevalence and associ-
ated factors of physical fighting among school-going adolescents in Namibia.
Ann. General Psychiatry 6, 18. http://dx.doi.org/10.1186/1744-859X-6-18.
Saewyc, E.M., Tonkin, R., 2008. Surveying adolescents: focusing on positive devel-
opment. Paediatr. Child Health 13 (1), 43e47.
Saewyc, E., Wang, N., Chittenden, M., Murphy, A., McCreary Centre Society, 2006.
Building Resilience in Vulnerable Youth. Retrieved 25 January 2015 from. http://
www.mcs.bc.ca/pdf/vulnerable_youth_report.pdf?Document¼promoting_
healthy_bodies_web.pdf.
Sommera, M., Likindikokib, S., Kaaya, S., 2013. Boys' and young men's perspectives
on violence in Northern Tanzania. Cult. Health &Sex. 15 (6), 695e709. http://
dx.doi.org/10.1080/13691058.2013.779031.
Song, Y., Ji, C., Agardh, A., 2014. Sexual coercion and health-risk behaviors among
urban Chinese high school students. Glob. Health Action 7, 24418.
Stephenson, R., Simon, C., Finneran, C., 2014. Community factors shaping early age
at first sex among adolescents in Burkina Faso, Ghana, Malawi, and Uganda.
J. Health Popul. Nutr. Jun 32 (2), 161e175.
Swahn, M.H., Gressard, L., Palmier, J.B., Kasirye, R., Lynch, C., Yao, H., 2012. Serious
violence victimization and perpetration among youth living in the slums of
Kampala, Uganda. West. J. Emerg. Med. 13 (3), 253e259.
Swahn, M.H., Gressard, L., Palmier,J.B., Yao, H., Haberlen, M., 2013. The prevalence of
very frequent physical fighting among boys and girls in 27 countries and cities:
regional and gender differences. J. Environ. Public Health 215126. http://
dx.doi.org/10.1155/2013/215126.
Tozija, F., 2008. 2007/2008 Global School-based Student Health Survey Results
Macedonia. Retrieved January 10, 2015, from: http://www.who.int/chp/gshs/
Macedonia_GSHS_2007_Report_English.pdf?ua¼1.
United Nations Children's Fund, 2014a. UNICEF Ghana. In Ghana, an Effort to
Abolish Violent Discipline in Schools. Retrieved August 10, 2015 from: http://
www.unicef.org/education/ghana_75533.html.
United Nations Children's Fund (UNICEF2014b), 2014. Hidden in Plain Sight: A
Statistical Analysis of Violence against Children. UNICEF, New York. Retrieved
May 18, 2015 from: http://www.unicef.org/publications/index_74865.html.
World Health Organization, 2002. World Report on Violence and Health. World
Health Organization, Geneva, Switzerland. Retrieved July 10, 2014, from: http://
www.who.int/violence_injury_prevention/violence/world_report/en/full_en.
pdf.
World Health Organization, 2004. Global Status Report: Alcohol Policy. World
Health Organization, Geneva, Switzerland. Retrieved February 4. 2015, from:
http://www.who.int/substance_abuse/publications/policy_ghana.pdf.
World Health Organization, 2006. Youth Violence and Alcohol. World Health Or-
ganization, Geneva, Switzerland. Retrieved January 24, 2014, from: http://www.
who.int/violence_injury_prevention/violence/world_report/factsheets/fs_
youth.pdf.
World Health Organization, 2007. Helping Parents in Developing Countries Improve
Adolescents' Health. World Health Organization, Geneva, Switzerland.
Retrieved January 25. 2015, from: http://www.who.int/maternal_child_
adolescent/documents/9789241595841/en/.
World Health Organization, 2009. Preventing Violence by Reducing the Availability
and Harmful Use of Alcohol. World Health Organization, Geneva, Switzerland.
Retrieved February 4. 2015, from: http://whqlibdoc.who.int/publications/2009/
9789241598408_eng.pdf?ua¼1.
World Health Organization, 2013a. Global School-based Student Health Survey.
Namibia: 2013 Fact Sheet. Retrieved January 10, 2015, from: http://www.who.
int/chp/gshs/2013_Namibia_Fact_Sheet.pdf.
World Health Organization, 2013b. Global School-based Student Health Survey.
Swaziland: 2013 Fact Sheet. Retrieved January 10, 2015, from: http://www.who.
int/chp/gshs/2013_Swaziland_GSHS_FS.pdf?ua¼1.
World Health Organization, 2014a. Addressing the Global Challenge of Violence, in
Particular against Women and Girls. World Health Organization, Geneva,
Switzerland. Retrieved January 10, 2015, from: http://apps.who.int/gb/ebwha/
pdf_files/WHA67/A67_22-en.pdf (last accessed January 2015).
World Health Organization, 2014b. Global Status Report on Violence Prevention
2014. World Health Organization, Geneva, Switzerland. Retrieved January 10,
2015, from. http://www.who.int/violence_injury_prevention/violence/status_
report/2014/en/.
World Health Organization, 2014c. Global School-based Student Health Survey
(GSHS) Purpose and Methodology. Retrieved July 10, 2014, from: http://www.
who.int/chp/gshs/methodology/en/.
World Health Organization, 2014d. Global School-based Student Health Survey
(GSHS). Retrieved July 10, 2014, from: http://www.who.int/chp/gshs/en/.
Ybarra, M.L., Bull, S.S., Kiwanuka, J., Bangsberg, D., Korchmaros, J., 2012. Prevalence
rates of sexual coercion victimization and perpetration among Uganda ado-
lescents. AIDS Care 24 (11), 1392e140 0. http://dx.doi.org/10.1080/
09540121.2011.648604.
S.-A. Ohene et al. / Social Science & Medicine 146 (2015) 266e275 275