International Review of Social Sciences and Humanities

Vol. 3, No. 1 (2012), pp. 36-50

www.irssh.com
ISSN 2248-9010 (Online), ISSN 2250-0715 (Print)

Child Sexual Abuse among Urban Secondary School

Pupils: Impact of Family Characteristics and Family
Structure
Pesanayi Gwirayi
Department of Educational Foundations, Midlands State University
P Bag 9055, Gweru, Zimbabwe
E-mail: [email protected]

(Received: 28-10-11 / Accepted: 15-2-12)

Abstract
This study investigated the impact of family characteristics and family structure as risk factors
for sexual abuse among urban secondary school pupils in Zimbabwe. It adopted a survey
design as the operational framework for data gathering. Data were collected from three
secondary schools in the Gweru district of Zimbabwe. The sample comprised 268 secondary
pupils (50 % female; mean age=15.42, SD=1.376). Data were gathered by administering the
Child Abuse Screening Tool Children’s Version (ICAST-C). Of the 268 respondents who
completed the questionnaire, 151 reported having been sexually abused in one or more ways
before the age of 18 years, giving an overall prevalence rate of 56.3 %. Overall, the results
show that the most important factors that were found to be consistently associated with child
sexual abuse were parental absence, father’s absence, mother's education, parental domestic
violence and parental substance abuse. Risk of any child sexual abuse was not associated with
the existence of siblings, household size, parental occupation and father’s education. Policies
should focus on supporting single mothers and parents who both work outside of the home to
reduce the likelihood that children are left vulnerable and at risk of being sexually abused.

Keywords: child sexual abuse; family characteristics; family structure; secondary pupils;
Zimbabwe.

1. Introduction
Child sexual abuse (CSA) is an issue of veritable concern worldwide. It is associated with
numerous deleterious consequences such as physical injury, teenage pregnancy, sexually
transmitted infections (including HIV and AIDS), emotional trauma, poor school
performance, rejection by family and society, family disharmony, poor parenting and abusive
behaviour in later life (Casey & Nurius, 2005; Spies, 2006; WHO, 2004; Wurtele, 2009; Zink
et al.,2009). For instance, a study by Stirpe and Stermac (2003) reported that about one-third
of violent and nonviolent offenders were survivors of child sexual abuse. In addition, research
shows that more sexual offenders than nonsexual offenders significantly reported being
survivors of childhood sexual abuse (Leung et al., 2010).
Socioeconomic status, employment, education level, marital relationships, family structure
and substance abuse, have been reported to be associated with the perpetration of CSA (Bahal
et al., 2010; Black, et al., 2001; Elbedour et al., 2006; Finkelhor et al., 2009; Holmes, 2007;
International Review of Social Sciences and Humanities, Vol. 3, No. 1 (2012), 36-50 37

Hunter, 2006; King et al., 2004; Leung et al., 2010; Priebe & Svedin, 2009; Strega et al.,
2008; Yen et al., 2008).
Several studies have confirmed the link between parental level of education and increased risk
of being sexually abused (Chaffin et al., 1996; Elbedour et al., 2006; Finkelhor, 1980; Leung
et al., 2010; Pinheiro, 2006; Yen et al., 2008). For instance, in a study that was conducted in
the US, Chaffin et al. (1996) found that families with lower education levels were more likely
to be significant features of cases encountered in child protection agencies, probably
reflecting their increased likelihood of being detected and disproportionately reported. This
finding is further supported by studies which have reported that mothers of those sexually
abused are less likely to have finished high school (Finkelhor, 1979; Leventhal, 1998). In
another study, Zuravin and DiBlassio (1992) also found that parental low education level was
linked to increased risk for CSA. Although parental level of education seems to be a strong
predictor of CSA, contradictory results have also been reported (Chen et al., 2004, 2006).
However, as Elbedour et al. (2006) argue, parental successful academic experiences may play
a protective role in preventing sexual victimisation. A possible explanation could be that
educated parents are likely to make use of suggested prevention programmes and other
available resources.
A review of socio-demographic risk factors of CSA based on a retrospective community
study in the USA (Black, et al., 2001) revealed that parental occupational status was
moderately related to having a child who was sexually abused. In the same vein,
unemployment has been reported as a risk factor for sexual abuse (Townsend & Dawes,
2004). One prospective study suggested that children who had experienced CSA often grew
up with parents who were unemployed (Horwitz et al., 2001). Unemployment may affect risk
factors through the stress of reduced material resources, a sense of powerlessness in the
unemployed parent, or through increased parent-child contact (Belsky, 1980). In the same
vein, maternal employment brings significant stresses into the parent-child relationship and
has implications in terms of childcare arrangements, but may also act as a protective factor
through a range of social-psychological benefits (Sidebotham & Heron, 2006). In a study that
was conducted among female high school students in Israel, higher occupational status of the
father was associated with higher likelihood of sexual abuse (Elbedour et al., 2006). Such a
finding is as important as it is puzzling. However, it is not clear whether the findings indicate
that daughters of fathers with high occupational status actually experience higher rates of
sexual abuse or whether they are more likely to report such abuse, for example, because they
are more liberated and perhaps empowered and therefore, are less likely to fear the
consequences of self-disclosure such as family dishonour .
Extant studies have reported an association between parental absence and history of CSA. For
instance, being raised without one’s father, or living apart from one’s mother for a significant
part of one’s childhood, have been associated with CSA (Cole, 1995; Collings, 1991;
Finkelhor, 1980, 1984). Living with one biological parent has been found to be related to
sexual abuse, consistent with other studies (Holmes & Slap, 1998). In a study that was
conducted in South Africa (Jewkes, Levin, Mbananga, & Bradshaw, 2002) adolescents who
were not living with a biological father reported an increased risk of teenage pregnancy. This
finding suggests that parental figures can play an important role by providing greater control
and restrictions on youth risk behaviour, which may lead to a lesser likelihood of incest and
increased protection from potential perpetrators (Jewkes et al., 2002). In the literature, there is
substantial agreement concerning family structure in that not living with both biological
parents places a child at high risk of CSA (Bahal et al., 2010; Fergusson et al., 1997;
Fergusson & Mullen, 1999; Holmes & Slap, 1998; Madu, 2002). Nevertheless, Ricket and
colleagues (2004) found no association between CSA and living together with both, one, or
no parent. However, evidence contrary to Ricket and colleagues’ (2004) finding seems to be
overwhelming. For example, several studies have shown children to be at higher risk of CSA
when they live in single-parent (Berger, 2004; Dubowitz, 1999; Lauritsen, 2003) or
reconstituted families (Turner, Finkelhor, & Ormstad, 2007). In the same vein, several studies
have reported that children living with divorced or single parents or stepfamilies were at two
to three times higher risk of CSA (Bahal et al., 2010; Black, et al., 2001; Brown et al., 1998;
Pesanayi Gwirayi 38

Sidebotham & Heron, 2006). In addition, youth in stepfamilies have been reported to be at
risk, relative to other family structures, of sexual assault and parental assault, with a variety of
other predictors controlled (Finkelhor & Asdigian, 1996). In reconstituted families, the
stepfather appears to double the risk for CSA for girls (Brown et al., 1998). The behaviour of
stepfathers in this context may be attributed to a lack of emotional attachment to the girl child
whom they may not see as their biological offspring.
Several studies that were conducted in Zimbabwe have also reported a link between parental
absence and CSA (Chiroro et al., 2006; Dzikira, 2007; Feist-Price et al., 2002). For instance,
Dzikira (2007) reported that children whose parents were cross-borders were more likely to
report having been sexually abused than those whose parents were present most of the time.
In another study, Chiroro et al. (2006) reported that girls who were brought up by their
brothers-in-law were at risk of being sexually abused by them. This suggests that the cultural
practice whereby the brother-in-law plays with his young sisters-in–law, (as a way of
preparing her for marriage) touching their body parts especially breasts, still exists in some
sections of the Zimbabwean society.
Several studies have linked parental substance abuse to CSA (Berger & Waldfogel, 2000;
Finkelhor, 1979; Horwitz et al., 2001; Hunter, 2006; Walsh et al., 2003; Velleman et al.,
2008; Velleman & Reuber, 2007). One prospective study suggested that children who had
experienced CSA often grew up in families where parents used drugs or alcohol (Horwitz et
al., 2001). In addition, a retrospective study that was conducted in Canada (Walsh et al.,
2003), reported that respondents with sexual abuse increased by twofold among those
reporting parental substance abuse histories. Yen et al. (2008) argue that an adult who is
abusing substances sometimes does not make rational choices in life. Research shows that in
cases where the substance abuser is a parent, chemical dependence is likely to lead to
parenting styles that are detrimental to children (Berger, 2005; Hampton, Senatore, & Gullota,
1998). This suggests that parents who abuse substances may face tremendous obstacles in
providing the safe home environment and warm care-giving that have been identified as
buffers against CSA. This implies that, parental substance abuse may create an environment
that is favourable for the perpetration of CSA by incapacitating parents from giving the
necessary monitoring and supervision.
Strong family relationships may act as a buffer against the perpetration of CSA. Several
studies suggest that sexual abuse often takes place in non-nurturing environmental and
familial contexts (Brown et al., 1999; Meyerson, Long, Miranda, & Marx, 2002) such as poor
parent-child relationships and family stress (Brown et al., 1998; Yen et al. 2008). In a study
that was conducted by Alexander (1992), it was noted that insecure interpersonal attachment
in a family, such as between a daughter and a father who seems to reject her, may increase the
risk of CSA in that family. This finding has far reaching implications on intra-familial sexual
abuse in that, on one hand adult perpetrators who are not strongly attached to their children
may have poor control of their impulses toward them. On the other hand, non-offending
parents who are not attached to their children may not notice that the child is being sexually
abused. As a result, children may be more willing to submit to sexual abuse if it seems a way
to build attachments they otherwise lack (Yuen, 2007). For instance, children may end up
seeking advice from their peers, sugar daddies or mummies, who are likely to mislead and
expose them to sexual abuse. Holt et al’s. (2008) review that was conducted within an 11-year
framework (1995-2006) established that children and adolescents living with domestic
violence are at increased risk of experiencing sexual abuse and behavioural problems and of
increased exposure to the presence of other adversities in their lives. As Finkelhor et al.
(2009) further argue, in families characterised by considerable chaos and lack of interpersonal
attachment, multiple, ongoing problems and/or conflicts, children may be poorly supervised
and subjected to a considerable amount of dislocation that exposes them to victimisation in
different contexts. Related to the foregoing observations is that children without secure
parental relationships may also be less likely to disclose sexual abuse experiences to their
parents or other adults, thereby perpetuating the problem.
International Review of Social Sciences and Humanities, Vol. 3, No. 1 (2012), 36-50 39

1.1 Purpose of the Study

The purpose of the present study was to investigate the prevalence of CSA among secondary
school pupils in Gweru, Zimbabwe, in connection to family structure. Although it is known
that a number of factors are associated with the perpetration of sexual abuse, very few studies
have investigated these factors among urban secondary schools in Zimbabwe, despite the
existence of research suggesting that the urban environment predisposes children to sexual
abuse. The few studies that are available have either targeted females (Chiroro et al., 2006;
Gwirayi & Shumba, 2008) or have been conducted in big cities such as Harare (Khan, 1995)
and Bulawayo (Meursing et al., 1995). The identification of risk factors has serious
implications for both preventive and intervention strategies.

2. Method

2.1 Research Design

The centrality of the research design in any research endeavour is highlighted by Hall (1996,
p. 17) who says that “the test question for any academic discussion which is based on research
findings is whether the conclusions are justified by the research design”. This study adopted
the descriptive survey strategy because survey research has the capacity to collect data for
describing populations that are too large to observe directly (Babbie & Mouton, 2005). Using
this design, this study was able to establish factors, which exposed secondary school pupils in
Gweru urban district, to the various forms of sexual abuse.

2.2 Participants
The sample comprised 268 pupils attending Gweru district day secondary schools (50 %
female; mean age=15.42, SD=1.376). The identification of these participants was based on
three premises. First, local literature shows that secondary school pupils are vulnerable to
sexual abuse (Makoni, 2006; Masvingise, 2009; Thompson, 2009). Second, as adolescents,
secondary school pupils are a high risk population group as they are more likely to be
sexually active. Third, these pupils were preferred because they were believed to be mature
enough to have the language to express their perspectives in English.

2.3 Sampling Procedure

A two-stage random sampling procedure was used. In the first stage, three day secondary
schools were randomly selected from a cluster of seven in Gweru district. In the second stage,
stratified random sampling was used to select Form 2 and 4 pupils at each participating school
(Marvasti, 2004).

2.4 Measures
The Child Abuse Screening Tool Children’s Version (ICAST-C) (Zolotor et al., 2009), which
is a new, multi-national and multi-cultural child abuse and research tool, was adopted.

CSA
Respondents were asked to indicate their frequency of experiencing the nine forms of sexual
abuse (contact and noncontact) on a four point scale ranging from “many times” to “not in the
past year but this has happened”. If respondents answered “no” to all nine statements, they
Pesanayi Gwirayi 40

were coded 0 (no sexual abuse); if they answered “yes” to any of the nine statements, they
were coded as 1 (sexually abused), consistent with previous studies (Gault-Sherman et al.,
2009; Raviv et al., 2010; Tyler, Johnson & Brownridge, 2008).

Predictors of CSA
Parental education Respondents were asked to provide information about their parents’ (both
father and mother) level of education. Responses were coded as 0 (primary), 1 (secondary), 2
(college), 3 (university) and 4 (this does not apply to me I don’t have a father (mother).
Parental occupation Respondents were also asked to provide information about their parents’
(both father and mother) occupation. Responses were coded as 0 (unemployed), 1 (working
full time), 2 (working part time), 3 (doing own business) and 4 (does not apply to me, I don’t
have a father (mother).
Parental presence A dichotomous variable was used to measure parental presence as follows:
Do you live with your parents? Mother Yes/No Father Yes/No
Parental substance abuse Has one of your parents used drugs and/or alcohol and then
behaved in a way that frightened you? Yes/No
Domestic violence Have you seen adults in your home shouting and yelling at each other in a
way that frightened you?
Have you seen adults in your home hit, kick, slap, punch each other or hurt each other
physically in other ways? Responses to both questions were coded as 0 (never), 1 (sometimes)
and 2 (many times).
Household size How many rooms are there in your present house?
Number of siblings In total, how many brothers and sisters you have including yourself?

2.5 Data Collection Procedure

The research instrument was administered personally by the researcher in order to ensure a
high response rate (Sarantakos, 2005). On the agreed dates with the school heads, and with
the cooperation of the class teachers, the researcher distributed the questionnaires to the pupils
in their classrooms. Data were collected during free periods in order to minimise interruption
of other school activities. Participants were required to sit separately so that they could feel
free to complete the questionnaire without sharing their answers.

2.6 Ethical Considerations

Before collecting data, the researcher sought for permission from the Ministry of Education
Sport and Culture, which was granted. In addition, the researcher also asked for informed
consent from all the participants (Sarantakos, 2005). The researcher explained the purpose of
the study to the participants and allowed them to ask questions about the research before
asking them to complete the questionnaire (Babbie & Mouton, 2005). It was also made clear
to the participants that taking part in the study was voluntary, and that those who chose to
participate were free to withdraw their participation at any stage if they so wished.
The school counsellor was available during the administration of questionnaires in order to
provide counselling and follow-up in the event of any pupil showing emotional distress
(Yuen, 2007). In addition, all participants were given information pamphlets on Gweru based
organisations which provide professional assistance on issues relating to CSA.

2.7 Data Analysis

The Odds Ratio (OR) in cross-sectional studies such as this is a measure of the strength of the
association between a dichotomous exposure (CSA/No CSA) and risk factors of sexual abuse.
The Confidence Interval (95 %) is the range within which the true OR is likely to fall. If the
International Review of Social Sciences and Humanities, Vol. 3, No. 1 (2012), 36-50 41

interval does not extend below 1.0, then the OR is statistically significant (equivalent to
p<.05). A correlation matrix was also used to analyse the association between CSA and risk
factors.

3. Results

3.1 Prevalence of CSA

Of the 268 respondents who completed the questionnaire, 151 reported having been sexually
abused in one or more ways before the age of 18 years, giving an overall prevalence rate of
56.3 %.

3.2 CSA and Family Characteristics

Results on family characteristics associated with CSA are summarised in Tables 1, 2 and 3.
The multiple logistic regression analysis and the correlation matrix together showed that five
factors were found to be significantly correlated with sexual abuse; not living with both
parents, not living with father, mother’s level of education, parental domestic violence and
parental alcoholism.
Contrary to popular belief, the multiple logistic regression analysis (Table 1 and Table 2) did
not report any associations between CSA and variables such as father’s education, parental
occupation, and child’s gender. First, the analysis of the correlates of CSA was done by
separating the logistic regression model for sexual abuse by gender.

3.2.1 Risk Factors for CSA by Gender of Respondents

Separate gender models (Table 1) revealed two statistically significant correlates of sexual
abuse. Both were reported in the logistic model consisting only of males. Male respondents
whose mothers had achieved secondary education had greater odds (5.00 times) of
experiencing sexual abuse compared to those whose had minimal or higher education.
Parental alcoholism elevated the odds of male adolescents experiencing sexual abuse by
almost three times.

Table 1 Ordinal Logistic Regression Models for Sexual Abuse Separated by Gender

______________________________________________________________________

Variable in model Females odds ratios Male odds ratios

(95 % CIs) (95 % CIs)

Father’s education

No education 2.33 (.99-5.49) 0

Primary .66 (.13-3.45) .63 (.16-2.49)

Secondary 1.06 (.38-3.01) 2.06 (.72-5.91)

College/University .82 (.34-2.01) .40 (.17-.96)

Mother’s education
Pesanayi Gwirayi 42

No education .75 (.08-7.21) 1.00 (.15-6.77)

Primary .94 (.44- 8.45) .53 (.17-1.70)

Secondary .95 (.36-2.47) 5.00 (1.21-20.61)

College/University .69 (.24-2.00) .30 (.67-1.31)

Mothers’ occupation

Unemployed 1.07 (.39-2.95) 1.13 (.38-3.35)

Working full time .97 (.24-4.00) 1.57 (.40-6.14)

Working part time 1.30 (.31- 5.39) .58 (.15-2.26)

Doing own business .59 (.21-1.65) 1.18 (.39-3.64)

Father’s occupation

Unemployed 1.89 (.41-8.85) 2.74 (.64-11.82)

Working full time .22 (.23-2.12) .28 (.028-2.92)

Working part time 1.60 (.10-24.70) 5.00 (.34-72.77)

Doing own business 1.91 (.35- 10.54) .31(.67-1.41)

Live with mother

No 1.39 (.64-3.03) .92 (.42-2.01)

Yes .72 (.33-1.57) 1.09 (.50-2.39)

Live with father

No 1.13 (.57-2.23) 1.95 (.97-3.93)

Yes .89 (.45-1.75) .51 (.25-1.03)

Live with both parents

No 1.23 (.62-2.43) 1.68 (.84-3.37)

Yes .81 (.41-1.61) .59 (.30-1.19)

Parental substance abuse

No .04 (.41-2.60) .34 (.12-.99)

Yes .96 (.38- 2.42) 2.93 (1.01-8.52)

______________________________________________________________________
International Review of Social Sciences and Humanities, Vol. 3, No. 1 (2012), 36-50 43

3.2.2 Risk Factors for CSA by Respondents’ Level of Education

Models separated by the respondents’ level of education showed three statistically significant
variations on correlates of sexual abuse (Table 2). The logistic regression model consisting
only of Form 2 respondents showed that the odds of experiencing sexual abuse were higher
among those who did not live with both parents than those with both parents (OR=2.22,
CI=1.10-4.48). Furthermore, Form 2 respondents who did not stay with their fathers had
greater odds (2.34 times) of experiencing sexual abuse compared to those who stayed with
their fathers. The logistic regression model consisting only of those in Form 4 showed that
respondents whose mothers had obtained secondary education (OR=3.91, CI=1.18-12.99)
were more likely to have been sexually abused than those whose mothers had either lower or
higher levels of education.

Table 2 Ordinal Logistic Regression Models for Sexual Abuse Separated by

Respondents’ Level of Education

______________________________________________________________________

Variable in model Form 2 odds ratios Form 4 odds ratios

(95 % CIs) (95 % CIs)

Father’s education

No education 0 1.00 (.10-9.61)

Primary 1.00 (.19-5.33) .40 (.99-1.62)

Secondary 1.59 (.52-4.91) 1.75 (.60-5.09)

College/University .31 (.12-.81) .86 (.36-2.05)

Mother’s education

No education .70 (.049-10.01) .63 (.09-4.40)

Primary .85 (.24-3.00) 1.23 (.30-5.05)

Secondary .84 (.30-2.33) 3.91 (1.18-12.99)

College/University .62 (.21-1.80) .48 (.13-1.81)

Mothers’ occupation

Unemployed .41 (.14-1.19) 2.69 (.91-7.95)

Working full time 1.50 (.34-6.56) 1.00 (.27-3.72)

Working part time 1.56 (.35-6.94) .47 (.12-1.76)

Doing own business 1.15 (.71- 1.84) .71 (.39- 1.33)

______________________________________________________________________
Pesanayi Gwirayi 44

Table 2 (Continued). Ordinal Logistic Regression Models for Sexual Abuse Separated by
Respondents’ Level of Education

______________________________________________________________________

Variable in model Form 2 odds ratios Form 4 odds ratios

(95 % CIs) (95 % CIs)

Father’s occupation

Unemployed 2.35 (.47-11.78) 2.10 (.50-8.84)

Working full time .36 (.25-.53) .42 (.78-2.31)

Working part time 3.00 (.97-9.30) 1.50 (.18-12.46)

Doing own business .33 (.57-1.92) 1.20 (.27-5.31)

Live with mother

No 1.94 (.92-4.08) .65 (.28-1.50)

Yes .52 (.25-1.09) 1.55 (.69-3.60)

Live with father

No 2.34 (1.17- 4.69) 1.04 (.51-2.11)

Yes .43 (.21-.86) .96 (.47-1.95)

Live with both parents

No 2. 22 (1.10-4.48) 1.06 (.53-2.14)

Yes 5 (.22-.91) .94(.48-1.91)

Parental substance abuse

No .78 (.27-2.23) .60 (.24-1.48)

Yes 1.29 (.45-3.68) 1.68 (.68-4.18)

______________________________________________________________________

Pearson correlation coefficients, means and standard deviations of the variables of interest are
presented in Table 3. From Table 3, it can be observed that respondents, who reported
parental domestic violence were more likely to be exposed to sexual abuse than those who
scored less on parental domestic violence. Household size and number of siblings were not
correlated with sexual abuse.

Table 3 Means, Standard Deviations, and Correlation Coefficients among the Variables
Included in the Analyses.

1 2 3 4 5 6 7 8
______________________________________________________________________
International Review of Social Sciences and Humanities, Vol. 3, No. 1 (2012), 36-50 45

1 Family interaction

2 Household size .01

3 Number of siblings .02 .24**

4 Form -.08 -.17** .02

5 Domestic violence -.21** .03 .01 .20**

6 Neighbourhood -.05 .03 -.06 .10 .15*

7 Sexual abuse -.07 .00 -.03 .12 .27** .15*

8 age -.14* -.16* .02 .80** . 22** .08 . 13*

Mean 6.09 5.72 3.90 3.00 1.37 7.87 2.73 15.42

Standard deviation 2.33 2.82 1.66 1.00 1.83 3.57 3.98 1.38

______________________________________________________________________

**p< 0.01 * p< 0.05

Correlation coefficients are calculated based on pair wise deletion of the missing cases.

4. Discussion
The principal objective of this study was to determine the prevalence of CSA in connection to
family characteristics. An overall prevalence rate of 56.3 % was reported. This rate is way
above those that have been found in developed as well as other African countries. However, it
is comparable to rates that have been reported in African countries such as Nigeria 55 %
(Olley, 2008) and South Africa 54.2 % (Madu & Peltzer, 2000).
The presence of both parents as a buffer against the perpetration of CSA is unequivocal from
the research findings of this study. It was established that, the odds of experiencing sexual
abuse were higher among Form 2 respondents who did not stay with both parents than those
with both parents. This finding could be attributed to the age of the survivors relative to their
Form 4 counterparts. In this regard, the finding from this study is consistent with numerous
earlier studies (Bahal et al., 2010; Dzikira, 2007; King et al., 2004; Leung et al., 2010, Madu,
2002; UNICEF, 2010) which reported that not living with both biological parents places a
child at high risk of CSA.
Consistent with research (Jewkes et al., 2002; Strega et al., 2009), this study established that,
Form 2 respondents who did not stay with their fathers had greater odds of experiencing
sexual abuse compared to those who stayed with their fathers. While this finding is consistent
with available research in Zimbabwe, concerning the vulnerability of children in this age
group (Khan, 1995; Rudd, 2004), it also buttresses the protective role of fathers against CSA.
As Jewkes et al. (2002) earlier reported in a South African study, parental figures especially
fathers can play an important role by providing greater control and restrictions on youth risk
behaviour, which may lead to a lesser likelihood of incest and increased protection from
potential perpetrators.
While one-parent homes may confer risk by increasing total parental absence (Holmes, 2007),
the situation may be aggravated by parental employment. Parents may be temporarily
available due to employment outside the home (Leung et al., 2010; Townsend & Dawes,
2004). Consistent with this finding, this study established that, some survivors of CSA were
living on their own because their parents were either away on work-related business, were
Pesanayi Gwirayi 46

cross-border traders, or had migrated to the Diaspora, leaving them on their own. Due to lack
of parental guidance, supervision, and protection, some children ended up being exposed to
sexual abuse. By and large, this study confirms existing literature in reporting family structure
as a risk factor for the perpetration of sexual abuse.
This study established that parental occupation is not a risk factor of sexual abuse. While this
finding is consistent with some studies (Chen et al., 2004, 2006) it is contrary to others (Black
et al., 2001; Elbedour et al., 2006; Horwitz et al., 2001; Holmes & Slap, 1998; Townsend &
Dawes, 2004; Velleman et al., 2008). In contemporary Zimbabwe, unemployment has reached
phenomenal proportions. Specifically, in Gweru where the study was conducted, the
economic meltdown has led to massive retrenchments in big companies. It is rather puzzling
that this has not compounded children’s vulnerability given that they have been reported to
supplement meagre family resources by engaging in sexual activities.
This study established startling findings on mother’s education as a risk factor for CSA. The
odds of experiencing sexual abuse were greater among male respondents whose mothers had
achieved secondary education compared to those whose mothers had minimal education. This
finding contradicts previous studies that have reported that mothers of those sexually abused
are less likely to have finished high school (Finkelhor, 1979; Leventhal, 1998), and that little
parental education predisposes children to sexual abuse (Bahal et al., 2010; Leung et al.,
2008; Olley, 2008; Yen et al., 2008). Unfortunately, there are no studies from Zimbabwe to
either confirm or refute this finding. One possible explanation could be that, mothers with
secondary education had better employment opportunities compared with those with no
education or with primary education alone. In this regard, their working outside the family
could have exposed their children to sexual abuse through lack of maternal protection and
supervision, thereby diminishing maternal availability. It could not be confirmed whether
male respondents exposed themselves to CSA by being more adventurous in the absence of
parental monitoring and supervision. This finding is important in that it showed that, even
mother's higher level of education can also indirectly predict children's exposure to sexual
abuse.
In this study, it emerged that there was a statistically significant relationship between
reporting presence of parental domestic violence and exposure to CSA. This finding is
consistent with several previous studies (Finkelhor et al., 2009; Holt et al., 2008). For
instance, a review that was conducted by Holt et al. (2008) reported that children and
adolescents living with domestic violence are at increased risk of experiencing sexual abuse.
The foregoing view is further corroborated by Finkelhor et al. (2009) who established that in
families characterised by parental domestic violence, children may be poorly supervised and
subjected to a considerable amount of dislocation that exposes them to sexual abuse.
The deleterious effects of parental substance abuse on children's lives can be hardly disputed.
It emerged in this study that, parental substance abuse increased the odds of male adolescents
experiencing sexual abuse by almost three times. Although this finding is consistent with
research in confirming an association between parental substance abuse and CSA (Hunter,
2006; Velleman et al., 2008; Velleman & Templeton, 2007; Yen et al., 2008), few studies
have reported the influence of gender in their investigations (Walsh et al., 2003). Not
surprisingly, there are no known studies that were conducted in Zimbabwe, against which the
finding from this study can be compared. It therefore remains difficult to explain why male
respondents were at a greater risk of experiencing sexual abuse in the presence of parental
substance abuse. In this study, it could not be confirmed whether boys more than girls were
more likely to be mischievous by engaging in risky sexual behaviour in their unsafe home
environments, devoid of parental monitoring and supervision, in the wake of parental
substance abuse. Also, what could have compounded the complication in this study was that
no distinction was made between parental, paternal and maternal substance abuse.
International Review of Social Sciences and Humanities, Vol. 3, No. 1 (2012), 36-50 47

Conclusion
This study established a high prevalence rate of CSA which was associated with a wide range
of family characteristics, and mediated by respondent's gender and level of education.
Overall, the results show that the most important factors that were found to be consistently
associated with CSA were parental absence, father’s absence, mother's education, parental
domestic violence and parental alcoholism. Risk of any CSA was not associated with the
existence of siblings, household size, parental occupation and father’s education. Policies
should focus on supporting single mothers and parents who both work outside of the home to
reduce the likelihood that children are left vulnerable and at risk of being sexually abused.

References
[1] P.C. Alexander, Application of attachment theory to the study of sexual abuse,
Journal Consulting and Clinical Psychology, 60(2) (1992), 185-195.
[2] E. Babbie and J. Mouton, The Practice of Social Research, (2005), Cape Town:
Oxford University Press Southern Africa.
[3] K. Bahali, R. Akcan, A.Y. Tahiroglu and A. Avci, Child sexual abuse: Seven years
in practice, Journal of Forensic Science, 55(3) (2005), 633-636.
[4] J. Belsky, Etiology of child maltreatment: An ecological integration, American
Psychologist, 35(4) (1980), 320-335.
[5] L.M. Berger, Income, family structure and child maltreatment risk, Child and Youth
Services Review, 26 (2004), 725-748.
[6] L.M. Berger, Income, family characteristics and physical violence toward children,
Child Abuse and Neglect, 29 (2005), 107-133.
[7] L.M. Bergner and J. Waldfogel, Parental cocaine exposure: Long-run effects and
policy implications, Social Service Review, 74 (2000), 28-54.
[8] D.A. Black, R.E. Heyman and S.A.M. Smith, Risk factors for child sexual abuse,
Aggression and Violent Behaviour, 6 (2-3) (2001), 203-229.
[9] J. Brown, P. Cohen, J.G. Johnson and S. Salzinger, A longitudinal analysis of risk
factors for child maltreatment: Findings of a 17-year prospective study of officially
recorded and self-reported child abuse and neglect, Child Abuse and Neglect, 22(11)
(1998), 1065-1078.
[10] J. Brown, P. Cohen, J.G. Johnson and E.M. Smailes, Childhood abuse and neglect:
Specificity and effects on adolescent and youth adult depression and suicidality,
Journal of the American Academy of Child & Adolescent Psychiatry, 38(12) (1999),
1490-1496.
[11] E.A. Casey and P.S. Nurius, Trauma exposure and sexual victimization risk:
Comparisons across single, multiple incident and multiple perpetrator
victimizations, Violence against Women, 11(2005), 505-530.
[12] M. Chaffin, K. Kelleher and J. Hollenberg, Onset of physical abuse and neglect:
Psychiatric, substance abuse, and social risk factors from prospective community
data, Child Abuse and Neglect, 20(3) (1996), 191-203.
[13] J.Q. Chen, M. Dunne and P. Han, Child sexual abuse in China: A study of
adolescents in four provinces, Child Abuse and Neglect, 28(11) (2004), 1171-1186.
[14] J.Q. Chen, M.P. Dunne and P. Han, Child sexual abuse in Henan Province, China:
Associations with sadness, suicidality and risk behaviours among adolescent girls,
Journal of Adolescent Health, 38(5) (2006), 544-549.
[15] P. Chiroro, T.G. Viki, A. Frodi, T. Muromo and A. Tsigah, Nature and prevalence of
childhood sexual abuse among high school girls and college students in Zimbabwe,
Journal of Psychology in Africa, 16(1) (2006), 17-26.
[16] J. Cole, Parental absence as risk factor in the sexual exploitation of children, Acta
Criminologica, 8 (1995), 35-41.
Pesanayi Gwirayi 48

[17] S.J. Collings, Childhood sexual abuse in a sample of South African university males:
Prevalence and risk factors, South African Journal of Psychology, 21(1991), 153-
158.
[18] H. Dubowitz, The families of neglected children, In M. E. Lamb (Ed), Parenting and
Child Development in Nontraditional Families, (1999), Mahwah, NJ: Lawrence
Erlbaum Publishers.
[19] W. Dzikira, Forms and Causes of Child Abuse in Kadoma Urban Secondary Schools,
(2007), Bachelor of Education Dissertation: Midlands State University.
[20] S. Elbedour, B. Abu-Badgers, A.J. Onwuegbuzie, A. Abu-Rabia and S. El-Aassam,
The scope of sexual, physical and psychological abuse in a Bedouin-Arab
community of female adolescents: The interplay of racism, urbanization, polygamy,
family honour and the social marginalisation of women, Child Abuse and Neglect,
30(3) (2006), 215-229.
[21] S. Feist-Price, E. Mpofu, R. Zimmerman and P. Cupp, Theory-based school HIV
prevention in Zimbabwe: A prospective study, Zimbabwe Journal of Educational
Research, 4 (2002), 224-229.
[22] D.M. Fergusson, L.J. Horwood and M.T. Lynskey, Childhood sexual abuse,
adolescent sexual behaviors and sexual revictimisation, Child Abuse and Neglect,
21(8) (1997), 789-803.
[23] D.M. Fergusson and P.E. Mullen, Childhood Sexual Abuse, An Evidence-based
Perspective, (1999), Thousand Oaks, CA: Sage.
[24] D. Finkelhor, Sexually Victimized Children, (1979), New York: Free Press.
[25] D. Finkelhor, Risk factors in the sexual victimization of children, Child Abuse and
Neglect, 4(4) (1980), 265-273.
[26] D. Finkelhor, Child Sexual Abuse: New Theory and Research, (1984), New York:
Free Press.
[27] D. Finkelhor and N.L. Asdigian, Risk factors for youth victimization: Beyond a
lifestyles theoretical approach, Violence and Victims, 11(1996), 3-20.
[28] D. Finkelhor, R. Ormrod, H. Turner and M. Holt, Pathways to poly-victimization,
Child Maltreatment, 14(4) (2009), 316-329.
[29] M. Gault-Sherman, E. Silver and I.D. Sigfusdottir, Gender and the associated
impairments of child sexual abuse: A national study of Icelandic Youth, Social
Science and Medicine, 69 (2009), 1515-1522.
[30] P. Gwirayi and A. Shumba, Childhood sexual abuse experiences of female student
teachers in Zimbabwe, Journal of Psychology in Africa, 18(2) (2008), 283-286.
[31] V. Hall, Dancing on the Ceiling: A Study of Women Managers in Education, (1996),
London: Paul Chapman Publishing Ltd.
[32] R.L. Hampton, V. Senatore and T.P. Gullotta, Substance Abuse, Family Violence and
Child Welfare, (1998), Thousand Oaks, CA: Sage Publications.
[33] W.C. Holmes, Men’s childhood sexual abuse histories by one-parent versus two-
parent status of childhood home, Journal of Epidemiology and Community Health,
61(2007), 319- 325.
[34] W.C. Holmes and G.B. Slap, Sexual abuse of boys: Definition, prevalence,
correlates, sequelae and management, Journal of American Medical Association,
280(21) (1998), 1855-1862.
[35] S. Holt, H. Buckley and S. Whelan, The impact of exposure to domestic violence on
children and young people: A review of the literature, Child Abuse and Neglect, 32
(2008), 797-810.
[36] A.V. Horwitz, C.S. Widom, J. McLaughlin and H.R. White, The impact of childhood
abuse and neglect on adult mental health: A prospective study, Journal of Health and
Social Behaviour, 42(2) (2001), 184-201.
[37] S.V. Hunter, Understanding the complexity of child sexual abuse: A review of the
literature with implications for family counselling, The Family Journal, 14 (2006),
349-359.
International Review of Social Sciences and Humanities, Vol. 3, No. 1 (2012), 36-50 49

[38] R. Jewkes, J. Levin, N. Mbananga and D. Bradshaw, Rape of girls in South Africa,
Lancet, 359 (2002), 319-320.
[39] R. Jewkes, L. Penn-Kekana and H. Rose-Junius, If they rape me, I can’t blame them:
Reflections on gender in the social context of child rape in South Africa and
Namibia, Social Science and Medicine, 61 (2005), 1809-1820.
[40] N. Khan, Patterns of child sexual abuse in Zimbabwe: An overview, Zimbabwe
Journal of Educational Research, 7(2) (1995), 181-208.
[41] G. King, A.J. Flisher, F. Noubary, R. Reece, A. Marais and C. Lombard, Substance
abuse and behavioural correlates of sexual assault among South African
adolescents, Child Abuse and Neglect, 28(6) (2004), 685-698.
[42] J.L. Lauritsen, How Families and Communities Influence Youth Victimization,
(2003), Washington, DC: Office of Juvenile Justice and Delinquency Prevention.
[43] P. Leung, Jr R.L. Curtis and S.C. Mapp, Incidences of sexual contacts of children:
Impacts of family characteristics and family structure from a national sample,
Children and Youth Services Review, 32 (2010), 650-656.
[44] J.M. Leventhal, Epidemiology of sexual abuse of children: Old problems, new
directions, Child Abuse and Neglect, 22(6) (1998), 481-491.
[45] S.N. Madu, The relationship between perceived parental physical availability and
child sexual, physical and emotional abuse among high school students in the
Northern Province, South Africa, The Social Science Journal, 39 (2002), 639-645.
[46] S.N. Madu and K. Peltzer, Risk factors and child sexual abuse among secondary
school students in the Northern Province of South Africa, Child Abuse and Neglect,
24(2) (2000), 259-268.
[47] A.B. Marvasti, Qualitative Research in Sociology, (2004), London: Sage.
[48] K. Meursing, T. Vos, O. Coutinho, M. Moyo, S. Mpofu, O. Oneko, V. Mundy, S.
Dube, T. Mahlangu and F. Sibindi, Child sexual abuse in Matabeleland, Zimbabwe,
Social Sciences and Medicine, 41(12) (1995), 1693-1704.
[49] L.A. Meyerson, P.L. Long, R. Miranda and B.P. Marx, The influence of childhood
sexual abuse, physical abuse, family environment, and gender on the psychological
adjustment of adolescents, Child Abuse and Neglect, 26(4) (2002), 387-405.
[50] B.O. Olley, Child sexual abuse as a risk factor for sexual risk behaviours among
socially disadvantaged adolescents in Ibadan, Nigeria, Vulnerable Children and
Youth Studies, 3(3) (2008), 243-248.
[51] P.S. Pinheiro, The UN Secretary General’s Report on Violence against Children:
The Way Forward, (2006), New York: UN.
[52] G. Priebe and C.G. Svedin, Prevalence, characteristics and associations of sexual
abuse with socio-demographics and consensual sex in a population-based sample of
Swedish adolescents, Journal of Child Sexual Abuse, 18(1) (2009), 19-39.
[53] T. Raviv, H.N. Taussig, S.E. Culhane and E.F. Garrido, Cumulative risk exposure
and mental health symptoms among maltreated youth placed in out-of-home care,
Child Abuse and Neglect, 34 (2010), 742-751.
[54] V.I. Rickert, C.M. Wiemann, R.D. Vaughan and J.W. White, Rates and risk factors
for sexual violence among an ethnically diverse sample of adolescents, Archives of
Pediatrics and Adolescent Medicine, 158 (2004), 1132-1139.
[55] C. Rudd, Case studies of child sexual abuse in Zimbabwe, In L.M. Richter, A. Dawes
and C. Higson-Smith (Eds), Sexual Abuse of Young Children in Southern Africa,
(2004), 395-410, Cape Town: Human Sciences Research Council.
[56] S. Sarantakos, Social Research (3rded), (2005), London: Macmillan.
[57] P. Sidebotham and J. Heron, Child maltreatment in the children of the nineties: A
cohort study of risk factors, Child Abuse and Neglect, 30(5) (2006), 497-522.
[58] D. Silverman, Interpreting Qualitative Data in Methods for Analysing Talk, Text and
Interaction, (2001), London: Sage.
[59] G.M. Spies, The effect of sexual abuse on a child, In G.M. Spies (Ed), Sexual Abuse:
Dynamics, Assessment and Healing, (2006), Pretoria: Van Shaik
Pesanayi Gwirayi 50

[60] S. Strega, C. Fleet, L. Brown, L. Dominelli, M. Callahan, et al., Connecting father

absence and mother blame in child welfare policies and practice, Children and Youth
Services Review, 30(2008), 705-716.
[61] L. Townsend and A. Dawes, Individual and contextual factors associated with the
sexual abuse of children fewer than 12: A review of recent literature, In L.M.
Richter, A. Dawes and C. Higson-Smith (Eds.), Sexual Abuse of Young Children in
Southern Africa, (2004), Cape Town: Human Sciences Research Council.
[62] H.A. Turner, D. Finkelhor and R. Ormrod, The effect of lifetime victimization on the
mental health of children and adolescents, Social Science and Medicine, 62 (2007),
13-27.
[63] K.A. Tyler, K.A. Johnson and D.A. Brownridge, A longitudinal study of the effects
of child maltreatment on later outcomes among high-risk adolescents, Journal of
Youth Adolescence, 37 (2008), 506-521.
[64] UNICEF, Inequities in the Fulfillment of Children’s Rights in South Africa: A
Statistical Review, (2010), Pretoria: Author.
[65] R. Velleman and L. Templeton, Understanding and modifying the impact of parental
substance misuse on children, Advances in Psychiatric Treatment, 13 (2007), 79-89.
[66] R. Velleman, L. Templeton, D. Reuber, M. Klein and D. Moesgen, Domestic abuse
experienced by young people living in families with alcohol problems: Results from
a cross-European study, Child Abuse Review, 17 (2008), 387-409.
[67] C. Walsh, H.L. MacMillan and E. Jamieson, The relationship between parental
substance abuse and child maltreatment: Findings from the Ontario Health
Supplement, Child Abuse and Neglect, 27(12) (2003), 1409-1425.
[68] WHO, Child Sexual Abuse: Salient Health Emergency, Regional Committee for
Africa, (2004), Fifty-fourth Session, Brazzaville, Republic of Congo, 30 August – 3
September.
[69] S.K. Wurtele, Preventing sexual abuse of children in the twenty-first century:
Preparing for the challenges and opportunities, Journal of Child Sexual Abuse, 18
(2009), 1-18.
[70] C. Yen, M. Yang, M. Yang, M. Su, M. Wang and C. Lan, Childhood physical and
sexual abuse: Prevalence and correlates among adolescents living in rural Taiwan,
Child Abuse and Neglect, 32 (2008), 429-438.
[71] C.W. Yuen, Child maltreatment in Malaysia: Prevalence, correlates and probable
health and behavioural consequences, Unpublished PhD Thesis, (2007), Queensland
University of Technology.
[72] T. Zink, L. Klesges, S. Stevens and P. Decker, The development of a sexual abuse
severity score: Characteristics of childhood sexual abuse associated with trauma
symptomatology, somatisation and alcohol abuse, Journal of Interpersonal Violence,
24(3) (2009), 537-546.
[73] A.J. Zolotor, D.K. Runyan, M.P. Dunne, D. Jain and H.R. Peturs, et al., ISPCAN
Child abuse screening tool children’s version (ICAST-C): Instrument development
and multi-national pilot testing, Child Abuse and Neglect, 33 (2009), 833-841.
[74] S.J. Zuravin and F.A. DiBlassio, Child-neglecting, adolescent mothers: How do they
differ from their non-maltreating counterparts? Journal of Interpersonal Violence, 7
(1992), 471-489.