Development of a Parenting Support Program to Prevent Abuse of Adolescents in South Africa

Abstract

Purpose:

Violence against children increases in adolescence, but there is a research and practice gap in research-supported child abuse prevention for the adolescent years. A pilot program for low-resource settings was developed in collaboration with nongovernmental organizations, government, and academics in South Africa, using research-supported principles.

Method:

This study used a pre-post design to test initial effects of a 10-session parenting program with 60 participants (30 caregiver–adolescent dyads) in high-poverty rural South Africa. Areas requiring further testing and adaptation were also identified.

Results:

Pre-post findings show medium to large program effects in reducing child abuse and adolescent problem behavior, as well as large effects in improvements of positive parenting, and perceived parent and adolescent social support.

Discussion:

There is potential to reduce child abuse, improve parenting, and reduce adolescent problem behavior in rural South Africa through parenting programs. Further development, testing and longer term follow-up are required to ascertain potential for scale-up.

Keywords

child abuse violence abuse prevention parenting stress parenting psychosocial aspects South Africa

Child abuse rates are disproportionately elevated in sub-Saharan Africa (Burton, Ward, & Artz, 2015) and are exacerbated by family-level stressors such as poverty and HIV/AIDS (Meinck, Cluver, & Boyes, 2015). Adolescence is a particularly high-risk stage for abuse, with rates of violence victimization increasing not only within the home but also in community settings (Finkelhor, Turner, Ormrod, & Hamby, 2009; Finkelhor, Turner, Shattuck, & Hamby, 2013). Evidence shows severe, current and long-term adverse effects of adolescent abuse on physical and mental health, education, employment, and sexual health (MacMillan & Hagan, 2004; Thornberry, Ireland, & Smith, 2001). Consequences of adolescent abuse may be particularly acute in sub-Saharan Africa, with increased risks of homicide and HIV infection (Jewkes et al., 2006; Richter et al., 2013).

Worldwide, less than 10% of abused children access any child protection services, with even lower access to effective preventative programs (Finkelhor, Ormrod, Turner, & Hamby, 2011). While this is true globally, there remains a particularly large research and treatment gap between developed and developing countries. A recent systematic review of reviews identified that, among current interventions, behavioral parenting programs showed greatest effectiveness in reducing child abuse (Mikton & Butchart, 2009). Notably 99.4% of studies were in high-income countries, and almost no research-supported programs focused on adolescents.

Although evidence is drawn largely from high-income countries, a recent systematic review found that these parenting principles transport well across diverse countries, cultures, and systems (Gardner, Montgomery, & Knerr, 2015). Other factors, however, work against transportability, as many programs with the strongest research support require qualified professionals (Olds et al., 1997) or have associated costs of purchase, training, and materials, thus reducing opportunities for large-scale implementation in low- and middle-income countries (LMICs). A systematic review of parenting programs in LMICs found a small number of programs showing improvements in parental knowledge of child development and in parent–child communication, but to date only three studies (in Chile, Turkey, and Iran) have investigated the effects on abuse or harsh parenting, all with infants or children aged under 6 (Knerr, Gardner, & Cluver, 2013).

There is a clear need for research-supported child abuse prevention interventions that are scalable and affordable in LMIC. In response, an international collaboration has been established to develop and rigorously test these. Partners include the World Health Organization (WHO), United Nations Children’s Fund, (UNICEF), and academics from the global South and North. Initial testing in South Africa, with government and nongovernmental organization (NGO) collaboration, will be followed by multicountry adaptation and evaluation. Programs focus on three distinct developmental stages: infancy, younger children, and older children/adolescents; and utilize principles demonstrated in systematic reviews as common to effective child abuse prevention programs, such as positive parenting and collaborative learning approaches (Barlow, Johnston, Kendrick, Polnay, & Stewart-Brown, 2006). Interventions are explicitly designed for low-literacy populations, to be implemented by nonprofessional staff, with no requirements for electricity or equipment such as videos and will be freely accessible under creative commons or WHO/UNICEF licenses.

Development of the adolescent program follows the UK Medical Research Council’s framework for designing and evaluating complex social interventions (Craig et al., 2008). Phase 1 included the review of existing evidence and community participatory research, leading to Phase 2: pilot studies of program acceptability, feasibility, and effects on primary and secondary outcomes (Campbell et al., 2007). This article reports outcomes from the first pilot pre-post test of the program, with the explicit aim of informing further adaptation and improvement of intervention components. Subsequent to this will be pre-post testing of a second version of the program and a third stage of adaptation and improvement, and we anticipate that this will be followed by a large-scale randomized controlled trial. This iterative approach aims to maximize research-supported adaptation of the core parenting principles to the needs of vulnerable families in South Africa (Figure 1).

Figure 1.

Hypothesized theory of change.

Method

Participants

Participants (N = 60) were 30 adolescent–caregiver dyads (adolescents aged 10–17). Given high rates of caregiving by nonbiological parents in South Africa, henceforth “parent” refers to biological and nonbiological primary caregivers of children. All lived in two high-poverty, deep rural communities of South Africa’s Eastern Cape province and were identified as in need of parenting support by a local community-based organization, the “Keiskamma Trust” who provides support services to orphaned and vulnerable children. No eligibility exclusions were made regarding factors such as parental literacy, prior history of mental health, domestic violence, or adolescent characteristics.

Inclusion criteria used a “pragmatic” approach, in order to reflect real-world service provision in rural areas of Africa. In the context of no available social services data to indicate which families had confirmed history of abuse, known indicators of parenting stress and behavior problems were used. Families were identified by the NGO as having expressed challenges with their adolescents, or with adolescent behavior, or families where the NGO or community suspected violence. Also following a pragmatic approach, there were no exclusions for severity of circumstances nor for mental or physical health problems or any other cause. Thirty caregiver–adolescent dyads were nominated and invited to participate in a 10-session, 5-week parenting support program, the Sinovuyo Caring Families Teen Programme (“Sinovuyo” is a Xhosa word meaning “we have happiness or joy”). Following other parenting programs and given high levels of stigma around child abuse, the intervention was presented in the community as aimed at reducing parenting stress and improving adolescent outcomes (Parra-Cardona et al., 2009).

Procedures

The study used a pre-post design with standardized questionnaires. Ethical protocols were approved by the Universities of Cape Town and Oxford. Written informed consent was obtained from all participants and, given low literacy levels, consent procedures were also read aloud. Parents and adolescents were interviewed face to face, separately and in private, by interviewers trained in working with vulnerable families, prior to and in the 2 weeks after completing the program. No incentives were offered, apart from certificates of participation. Confidentiality was maintained, except if participants were at risk of significant harm or requested assistance. If participants reported severe abuse, rape, or other significant harm, immediate referrals were made to child protection, health, and HIV/AIDS services, with follow-up support (n = 4, all of whom continued in the program). All research materials were translated into Xhosa and checked by back translation.

Primary Outcome Measures

Violent/abusive discipline was measured using the International Society for Prevention of Child Abuse and Neglect child and parent version of the International Child Abuse Screening Tool (ICAST-C and ICAST-P; 26 items). This has been field tested in eight countries, reviewed internationally using the Delphi process, and successfully used in developing countries (Runyan et al., 2009; Zolotor et al., 2009). Reliability (internal consistency) was α = .81 parent report (ICAST-P) and α = .74 adolescent report (ICAST-C).

Adolescent behavior problems were measured using the rule-breaking (17 items) and aggressive behavior (18 items) scales of the Child Behavior Checklist with established reliability and validity in multiple countries (Weisz, Sigman, Weiss, & Mosk, 1993). Reliability for rule-breaking behavior was α = .82 parent report and α = .62 adolescent report. Reliability for aggressive behavior was α = .85 parent report and α = .54 adolescent report.

Secondary/Linked Outcome Measures

Positive parenting was measured using child and parent report of the widely used and well-validated Alabama Parenting Questionnaire ([APQ] 32 items; Frick, 1991), with previous use in South Africa (Lachman, Cluver, Boyes, Kuo, & Casale, 2013; Madalane, 2014). Items were summed to create an overall assessment of positive parenting. The reliability of the APQ in this study was α = .75 for parent report and α = .63 for child report.

Social support was measured using the Medical Outcome Study Social Support Survey (Sherbourne & Stewart, 1991), comprising subscales of emotional, tangible, and affectionate support and positive social interaction (19 items). Reliability of the parent scale was α = .93 and adolescent scale α = .95.

Sociodemographic Measures

Sociodemographic measures used items modeled on the South African census (Statistics South Africa, 2001) and included parent and child home language, age, gender, level of education, parent marital status, employment status, nationality, source of income, relationship of child to caregiver, school attendance, household size, formal/informal housing, and household structure. Poverty was measured using an index of access to the eight highest socially perceived necessities for children corroborated by >80% of the population in the nationally representative SA Social Attitudes Survey (Pillay, Roberts, & Rule, 2006). Reliability for adult report was α = .96 and for adolescent report was α = .59.

Data Analysis

Data analysis was conducted using SPSS 20.1. α reliability coefficients were calculated for each scale and subscale. To examine the effects of the intervention on adolescent and parent outcome measures, a series of paired t-tests comparing baseline and posttest scores were employed (Field, 2009). Due to the large number of tests performed (n = 10), we used Bonferroni adjusted levels of p > .005 to determine statistically significant results (.05/10). We also calculated the Cohen’s d effect sizes for all of the outcome measures (small effect was 0.2–0.5, a medium effect was 0.5–0.8, and large effect was 0.8 or higher; Cohen, 1988). All participants were reinterviewed at follow-up, regardless of attendance levels, and analyses used an intention to treat approach, whereby outcome analyses include all participants present at the time of preintervention regardless of extent of program attendance, completion or adherence to the program protocol. As all 60 participants were included in the follow-up, no imputation was required.

Parenting Program

The pilot “Sinovuyo Caring Families Teen Programme” was developed in collaboration with South African NGOs and government departments. The design used (1) extensive literature review, (2) community consultation, and (3) expert consultations with developers of existing adolescent programs. Manual development was led by a national NGO, Clowns Without Borders South Africa, together with the Universities of Oxford and Cape Town, and in consultation with NGOs including the WHO, UNICEF South Africa, the Regional Psychosocial Support Initiative, and the South African national government departments of Social Development and Basic Education. Community consultations were undertaken with parents and with an advisory group of 20 adolescents from low-income South African communities.

The program uses social learning and parent management training principles, with group-based parent, adolescent, and joint parent–adolescent sessions. It utilizes a collaborative learning approach, with activity-based learning, role-play, and home practice (Webster-Stratton, 1998). Sessions include establishing special time for parents and adolescents, specific and immediate praise, dealing with stress and anger, establishing rules and responsibilities, and responding to crises (Figure 2). Simple stress reduction activities derived from mindfulness-based stress reduction were incorporated into the weekly group sessions and home practice assignments for parents and teens (Kabat-Zinn, 2013). These activities included a body relaxation exercise intended to relieve stress-related physical pain as well as a brief breath-awareness meditation to help increase socioemotional regulation skills. Unlike parenting programs for younger children that only include parents, evidence from high-income countries suggests that including both adolescents and parents is necessary for this developmental stage, with a combination of joint sessions where skills (such as problem solving) can be practiced together, and separate sessions where developmentally appropriate responses (such as addressing anger) can be established (Rotheram-Borus, Lee, Lin, & Lester, 2004).

Figure 2.

Session contents.

This study aimed to test research-supported program effects in real-world conditions relevant to sub-Saharan African contexts. Therefore, the manualized pilot program was implemented in one of South Africa’s three poorest provinces (National Development Agency, 2014), in two very-low-income rural communities with limited infrastructure. Adaptations included using role-plays and acted-out scenarios instead of video materials commonly used in parenting programs in high-income countries. Implementation took place in isiXhosa, the local language. A local-language observer attended all sessions and noted participant engagement and facilitator delivery, as well as challenges to implementation fidelity, and this information was subsequently used to improve manual design.

Two sets of parenting and adolescent groups were held in two rural villages. When available, groups took place in local early childhood centers, which had no electricity or running water. On days in which these were used for political or community meetings, groups took place in community huts or under trees. A simple group meal was provided in each session. Sessions were led by lay community workers from Clowns Without Borders South Africa and the Keiskamma trust. Staff had no formal qualifications but were experienced in conducting parenting programs and were given a week’s intensive training on the program, including collaborative learning techniques, modeling praise, and problem solving skills.

Results

Demographics

All 30 recruited dyads completed pre- and posttest interviews, and there was no attrition from the research study. Flow of participants through each stage of the study is illustrated in Figure 3. Rates of attendance were high to acceptable, at 86% for adolescents and 63% for parents. All participants were South African and isiXhosa speaking. Half reported past-week food insecurity and 40% lived in informal (shack) housing. Parents were aged 30–79 (mean age 46), 97% female, half were married, 87% unemployed, 90% had not completed secondary school, and 47% were chronically ill. Only 40% were the biological parent of the adolescent, reflecting high rates of orphaning and migration. Other caregivers were grandmothers (23%), aunts (17%), other relatives (13%), and siblings (7%). Adolescents were aged 12–16 (mean age 13), half female, and all were enrolled in school. Sixty percent were maternally or paternally orphaned, half of these due to HIV/AIDS (Table 1).

Figure 3.

Flow of participants through each stage.

Table 1.

Sociodemographic Characteristics of Participants.

Demographic Variables	Mean (Range) or %	Standard Deviation
Adolescent variables
Age	13.3 (12–16)	1.1
Female gender	50
School enrollment	100
Xhosa first language	100
South African	100
Orphaned	60
Orphaned by AIDS	33
Parent/caregiver variables
Age	45.9 (30–79)	14.9
Female gender	97
Employed	13
High school completed	10
Married	56
Biological parent of adolescent	40
Xhosa first language	100
South African	100
Past-year chronic illness	47
Household variables
Basic necessities score (out of 8)	4.2 (1–8)	2.0
1+ days in past week with insufficient food	50
Informal (shack/mud) housing	40
Number of people in household	6 (2–14)	2.7
Number of adolescents in household	4 (1–10)	1.9

Primary Outcomes

Results are summarized in Table 2. Comparisons between baseline and posttest scores showed reductions in violent or abusive discipline for both parent (t = 4.18, df = 16, and p = .001) and adolescent report (t = 2.39, df = 29, and p = .024). Although adolescent report of abuse was no longer significant once the Bonferroni correction for multiple testing was applied (i.e., p < .005), results showed a large effect size in reducing child abuse from adult report (d = −1.09 and 95% CI [−1.80, −0.32]) and a medium effect size from teen report (d = −0.47 and 95% CI [−0.99, 0.06]). Parent report of child behavior problems using the CBCL showed reductions in adolescent rule-breaking behavior (t = 3.21 df = 28, and p = .003) with a medium effect size (d = −0.35 and 95% CI [−0.86, 0.17]) and adolescent aggressive behavior (t = 3.07, df = 28, and p = .005) with a medium effect size (d = −0.41 and 95% CI [−0.92, 0.12]). Adolescent report showed no differences in either behavior subscales, primarily due to a floor effect with low levels of reported behavior problems at baseline and follow-up.

Table 2.

Study Variables, Reliability, Within Groups Paired t-Test, and Effect Sizes.

			Mean (SD)
Scale	Respondent	Reliability^a	Baseline	Posttest	Paired t-Test^b	df	p-Value	Effect Size^c [95% CI]
Primary outcomes
Violent/abusive discipline	Adult	.81	7.94 (7.72)	1.63 (2.83)	4.18**	15	.001	−1.09 [−1.80, −0.32]
Violent/abusive discipline	Adolescent	.74	23.53 (4.52)	21.87 (2.11)	2.39*	29	.024	−0.47 [−0.99, 0.06]
Adolescent aggressive behavior	Adult	.85	6.39 (6.52)	4.18 (3.99)	3.07**	28	.005^†	−0.41 [−0.92, 0.12]
Adolescent aggressive behavior	Adolescent	.54	5.44 (3.23)	4.69 (2.32)	1.29	28	.208	−.27 [−0.78, 0.25]
Adolescent rule-breaking behavior	Adult	.82	4.04 (5.36)	2.32 (4.47)	3.21^**	28	.003^†	−0.35 [−0.86, 0.17]
Adolescent rule-breaking behavior	Adolescent	.62	1.51 (2.47)	1.45 (1.90)	0.15	28	.885	−0.03 [−0.55, 0.48]
Secondary outcomes
Positive parenting	Adult	.75	117.75 (14.27)	132.13 (13.20)	−4.49***	23	.000^†	1.05 [0.43, 1.63]
Positive parenting	Adolescent	.63	118.24 (12.99)	127.38 (13.98)	−3.85**	23	.001^†	0.68 [0.08, 1.25]
Mediator outcomes
Perceived social support	Adult	.93	80.83 (16.15)	91.93 (7.23)	−3.69	28	.001^†	.89 [0.34, 1.41]
Perceived social support	Adolescent	.95	74.59 (19.91)	88.52 (13.15)	−5.19	28	.000^†	.83 [0.28, 1.35]

Note. ^aCronbach’s α coefficient for variables at baseline.

^b t-Value for paired t-tests comparing scores from baseline and posttest.

^cCohen’s d effect sizes based on comparisons between baseline and posttest scores.

*p < .05. **p < .01. ***p < .001.

^†Significant effect based on Bonferroni corrections for multiple testing.

Secondary/Linked Outcomes

Positive parenting using the APQ showed improvements in both parent (t = −4.49, df = 23, and p < .001) and adolescent report (t = −3.85, df = 23 and p = .001), with large effects for parent report (d = 1.05 and 95% CI [0.43, 1.63]) and medium effects for child report (d = 0.68 and 95% CI [0.08, 1.25]). Results also showed large effects for both increased perceived access to social support for parents (t = −3.69, df = 28, p = .001, d = 0.89, and 95% CI [0.34, 1.41]) and adolescents (t = −5.19, df = 28, p < .001, d = 0.89, and 95% CI [0.34, 1.41]).

Discussion and Applications for Practice

One of the greatest challenges for social workers in sub-Saharan Africa is how to address the alarming levels of child abuse that arise in contexts of poverty and disease. Social work services are understaffed and underresourced, especially in rural areas. It is for these reasons that it is essential that we develop programs that can prevent and reduce child abuse and that can be implemented by local, community NGOs.

This study represents a first pilot testing stage in such a program, with the aim of progressing toward a resource that can be used widely and for free. Of course, this is a noncontrolled study and so we need to interpret the findings with caution. But overall, these initial findings show associations with reduced abuse, reduced adolescent behavior problems, improved parenting, and improved social support, with no negative effects. This suggests that the program is suitable for further development and more rigorous testing using randomized controlled trial methods.

This study also aimed to identify what aspects of the program we need to improve and how we should adapt the research methods in the future. In particular, we need to think about how we use and interpret standardized scales developed in high-income settings in different cultural and economic contexts. Adolescent self-report of rule-breaking and aggressive behavior was substantially lower than parent report, although this disparity is found consistently and internationally in studies of child problem behavior (Achenbach, McConaughy, & Howell, 1987). An important implication for future research is the need to develop and validate measures of parenting and abuse that are culturally appropriate and psychometrically reliable for sub-Saharan Africa. This is part of a much wider need for consolidated research on child maltreatment in LMICs, where definitions of abuse remain contested, and sample sizes are often of nonrepresentative samples, such as university students, or very small numbers (Meinck, Cluver, Boyes, & Mhlongo, 2014).

With the main outcome variable—reductions in violent discipline, there were discrepancies between parent and teen report. It is possible that the larger reduction in the parent report may have been partly due to social desirability: At posttest, they certainly knew that this would not be regarded as a favorable outcome. However, even though adolescents reported a smaller overall reduction in violence, results indicate a medium effect of the program, which implies that parents had in fact at least made some changes in their disciplinary practices. Studies suggest that any reduction in family conflict may be beneficial for young people (Beauchaine, Webster-Stratton, & Reid, 2005). In addition, the large reported effect sizes for positive parenting suggest that the program may reduce the risk of child maltreatment by improving parent–child interaction as a protective factor in South Africa (Meinck et al., 2014). It is important to reflect on what effect sizes mean in the context of violent discipline and parenting: when designing interventions for practice we need to consider the extent of change and the practical implications of program effects. Further, we conducted the posttest within 2 weeks of the end of the program: it is possible that either parents continued to improve over time or that they failed to maintain this level of change. Future studies should include a longer term follow-up as well as an immediate posttest, to explore this question further.

There are a number of limitations to this study. First, causality cannot be determined from pretest–posttest studies such as this one. Second, the study took place in rural areas of South Africa with Xhosa-speaking participants and cannot be generalized to other ethnic groups or to urban areas. Third, follow-up was limited to 2 weeks postintervention, and subsequent studies should include longer term follow-up to determine longevity of effects (Aos et al., 2011). Fourth, reliability of some measures was low, and to date no measures for these outcome variables have been validated in sub-Saharan Africa. Although we used scales used previously in South Africa, and small pilot sample sizes often have lower α coefficients, it is important to interpret with caution the findings of scales that are standardized in different cultural contexts. Fifth, social services and law enforcement data are not available or reliable in rural South Africa, due to extremely low reporting rates and severe resource constraints (Sumner et al., 2015). Thus, it is not possible to validate self-report measures that always contain a risk of social desirability bias. While all community-based child abuse studies in LMICs rely on self-report, it is important to consider measurement risks in interpreting findings. Sixth, it would be of value for future research to investigate whether improved effects are achieved through a longer program with more components, which characterizes many successful programs with younger children in high-income countries. Finally, it will be important for future research with larger samples to elucidate the contributions of different mediator and moderator effects on outcomes of parenting programs in low- and middle-income settings (Gardner, Hutchings, Bywater, & Whitaker, 2010).

Despite these limitations, this study has valuable learning for social work practice. It suggests that it is possible to run a child abuse prevention program for families and their adolescents in a rural African setting. Findings also suggest that this is feasible in real-life service conditions led by local NGO staff and with no technological resources and no participant exclusion criteria. The initial findings suggest that—even in very low-resourced conditions reflecting our real-world services—there may be positive effects of such a program. For this particular program, further adaptation and testing will be needed to truly establish whether it can reduce child abuse in South Africa, and in other LMICs. In so many contexts, social work is only able to respond to child abuse when it has become too severe to allow children to remain in their families—yet preventing abuse from occurring in the first place is an important tenet of the profession. It is essential that social work research provides the profession with realistic programs to prevent and reduce child abuse, and opportunities to work with communities to support highly vulnerable families.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the European Research Council under the European Union’s Seventh Framework Program [FP7/2007-2013, grant agreement 313421], the John Fell Fund (UK, www.admin.ox.ac.uk/pras/jff) [grant number 103/757], the Leverhulme Trust (UK, ) [grant number PLP-2014-095], the University of Oxford’s Economic and Social Research Council Impact Acceleration Account, and the Canadian International Development Agency. NGO partners were Clowns Without Borders South Africa and the Keiskamma Trust.

References

Achenbach

T. M.

McConaughy

S. H.

Howell

C. T.

(1987). Child/adolescent behavioral and emotional problems: Implications of cross-informant correlations for situational specificity. Psychological Bulletin, 101, 213–232.

Aos

Cook

T. D.

Elliott

D. S.

Gottfredson

D. C.

Hawkins

J. D.

Lipsey

M. W.

Tolan

(2011). Commentary on Valentine, Jeffrey, et al.: Replication in prevention science. The advisory board of blueprints for violence prevention. Prevention Science, 12, 121–122; discussion 123–125. doi:10.1007/s11121-011-0219-4

Barlow

Johnston

Kendrick

Polnay

Stewart-Brown

(2006). Individual and group-based parenting programmes for the treatment of physical child abuse and neglect. Cochrane Database of Systematic Reviews, 3, 1–20.

Beauchaine

T. P.

Webster-Stratton

Reid

M. J.

(2005). Mediators, moderators, and predictors of 1-year outcomes among children treated for early-onset conduct problems: A latent growth curve analysis. Journal of Consulting and Clinical Psychology, 73, 371–388. doi:10.1037/0022-006X.73.3.371

Burton

Ward

C. L.

Artz

(2015). The optimus study on child abuse, violence and neglect in South Africa. Cape Town, South Africa: The Centre for Justice and Crime Prevention.

Campbell

N. C.

Murray

Darbyshire

Emery

Farmer

Griffiths

… Kinmonth

A. L.

(2007). Designing and evaluating complex interventions to improve health care. British Medical Journal, 334, 455–459.

Cohen

(1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum.

Craig

Dieppe

Macintyre

Michie

Nazareth

Petticrew

(2008). Developing and evaluating complex interventions: The new Medical Research Council guidance. British Medical Journal, 337, a1655.

Field

(2009). Discovering statistics using SPSS. London, England: Sage.

10.

Finkelhor

Ormrod

Turner

Hamby

(2011). School, police, and medical authority involvement with children who have experienced victimization. Arch Pediatr Adolesc Med, 165, 9–15. doi:10.1001/archpediatrics.2010.240

11.

Finkelhor

Turner

Ormrod

Hamby

S. L.

(2009). Violence, abuse, and crime exposure in a national sample of children and youth. Pediatrics, 124, 1411–1423. doi:10.1542/peds.2009-0467

12.

Finkelhor

Turner

H. A.

Shattuck

Hamby

S. L.

(2013). Violence, crime, and abuse exposure in a national sample of children and youth: An update. JAMA Pediatr, 167, 614–621. doi:10.1001/jamapediatrics.2013.42

13.

Frick

P. J.

(1991). The Alabama parenting scale. Tuscaloosa: University of Alabama.

14.

Gardner

Hutchings

Bywater

Whitaker

(2010). Who benefits and how does it work? Moderators and mediators of outcome in an effectiveness trial of a parenting intervention. J Clin Child Adolesc Psychol, 39, 568–580. doi:10.1080/15374416.2010.486315

15.

Gardner

Montgomery

Knerr

(2015). Transporting evidence-based parenting programs for child problem behavior (Age 3-10) between countries: Systematic review and meta-analysis. J Clin Child Adolesc Psychol, 1–14. doi:10.1080/15374416.2015.1015134

16.

Jewkes

Dunkle

Koss

M. P.

Levin

J. B.

Nduna

Jama

Sikweyiya

(2006). Rape perpetration by young, rural South African men: Prevalence, patterns and risk factors. Social science & medicine, 63, 2949–2961. doi:10.1016/j.socscimed.2006.07.027

17.

Kabat-Zinn

(2013). Full Catastophe Living: using the wisdom of your body and mind to face stress, pain and illness. 2nd edition, revised. London: Random House.

18.

Knerr

Gardner

Cluver

(2013). Improving positive parenting skills and reducing harsh and abusive parenting in low- and middle-income countries: A systematic review. Prevention Science: The Official Journal of the Society for Prevention Research. doi:10.1007/s11121-012-0314-1

19.

Lachman

J. M.

Cluver

L. D.

Boyes

M. E.

Kuo

Casale

(2013). Positive parenting for positive parents: HIV/AIDS, poverty, caregiver depression, child behavior, and parenting in South Africa. AIDS Care. doi:10.1080/09540121.2013.825368

20.

MacMillan

Hagan

(2004). Violence in the transition to adult-hood: Adolescent victimization, education, and socioeconomic attainment in later life. Journal of Research on Adolescence, 14, 127–158.

21.

Madalane

M. S.

(2014). Properties of the Alabama parenting questionnaire in an isiXhosa-speaking sample (Unpublished master’s thesis). Cape Town, South Africa: University of Cape Town.

22.

Meinck

Cluver

L. D.

Boyes

M. E.

(2015). Household illness, poverty and physical and emotional child abuse victimisation: Findings from South Africa’s first prospective cohort study. BMC Public Health, 15, 444. doi:10.1186/s12889-015-1792-4

23.

Meinck

Cluver

L. D.

Boyes

M. E.

Mhlongo

E. L.

(2014). Risk and protective factors for physical and sexual abuse of children and adolescents in Africa: A review and implications for practice. Trauma, Violence, and Abuse. doi:1524838014523336

24.

Mikton

Butchart

(2009). Child maltreatment prevention: A systematic review of reviews. Bulletin of the World Health Organization, 87, 353–361.

25.

National Development Agency. (2014). State of poverty and its manifestation in the nine provinces of South Africa. Pretoria, South Africa: Human Sciences Research Council,.

26.

Olds

D. L.

Eckenrode

Henderson

C. R. J.

Kitzman

Powers

Cole

… Luckey

(1997). Long-term effects of home visitation on maternal life course and child abuse and neglect. Journal of the American Medical Association, 278, 637–643.

27.

Parra-Cardona

Holtrop

Cordova

Escobar-Chew

Horsford

Tams

… Fitzgerald

(2009). Latino immigrants’ call to integrate cultural adaptation with best practice knowledge in a parenting intervention. Family Process, 48, 211–231.

28.

Pillay

Roberts

Rule

(Eds.). (2006). South African social attitudes: changing times, diverse voices. Cape Town, South Africa: HSRC Press.

29.

Richter

Komarek

Desmond

Celentano

Morin

Sweat

… Coates

(2013). Reported physical and sexual abuse in childhood and adult HIV risk behaviour in three African countries: Findings from project accept (HPTN-043). AIDS and Behavior. doi:10.1007/s10461-013-0439-7

30.

Rotheram-Borus

M.-J.

Lee

Lin

Y.-Y.

Lester

(2004). Six-year intervention outcomes for adolescent children of parents with the Human Immunodeficiency Virus. Archives of Pediatric and Adolescent Medicine, 158, 742–748.

31.

Runyan

D. K.

Dunne

M. P.

Zolotor

A. J.

Madrid

Jain

Gerbaka

… Youssef

R. M.

(2009). The development and piloting of the ISPCAN Child Abuse Screening Tool—Parent version (ICAST-P). Child Abuse & Neglect, 33, 826–832. doi: https://dx-doi-org.web.bisu.edu.cn/10.1016/j.chiabu.2009.09.006

32.

Sherbourne

Stewart

(1991). The Medical Outcomes Survey (MOS) social support survey. Social Science and Medicine, 32, 705–714.

33.

Statistics South Africa. (2001). Census 2001: Household questionnaire. Pretoria, South Africa: Statistics SA.

34.

Sumner

S. A.

Mercy

A. A.

Saul

Motsa-Nzuza

Kwesigabo

Buluma

, … Prevention. (2015). Prevalence of sexual violence against children and use of social services—Seven countries, 2007-2013. MMWR Morb Mortal Wkly Rep, 64, 565–569.

35.

Thornberry

T. P.

Ireland

T. O.

Smith

C. A.

(2001). The importance of timing: The varying impact of childhood and adolescent maltreatment on multiple problem outcomes. Dev Psychopathol, 13, 957–979.

36.

Webster-Stratton

(1998). Parent training with low-income families: Promoting parental engagement through a collaborative approach. In Lutzker

J. R.

(Ed.), Handbook of child abuse research and treatment (pp. 183–211). New York, NY: Plenum Press.

37.

Weisz

Sigman

Weiss

Mosk

(1993). Parent reports of behavioral and emotional problems among children in Kenya, Thailand, and the United States. Child Development, 64, 98–109.

38.

Zolotor

A. J.

Runyan

D. K.

Dunne

M. P.

Jain

Péturs

H. R.

Ramirez

… Isaeva

(2009). ISPCAN Child Abuse Screening Tool Children’s Version (ICAST-C): Instrument development and multi-national pilot testing. Child Abuse & Neglect, 33, 833–841. doi: https://dx-doi-org.web.bisu.edu.cn/10.1016/j.chiabu.2009.09.004