- Email: [email protected]
Preventing violence against children in schools: Contributions from the Be Safe program in Sri Lanka
Child Abuse & Neglect 76 (2018) 129–137
Contents lists available at ScienceDirect
Child Abuse & Neglect journal homepage: www.elsevier.com/locate/chiabuneg
Preventing violence against children in schools: Contributions from the Be Safe program in Sri Lanka
MARK
⁎
Steven Lama,b, , Christine Zwarta, Inem Chahala, David Lanea, Harry Cummingsa,c a b c
Harry Cummings and Associates, 96 Kathleen St, Guelph, Ontario, Canada Department of Population Medicine, University of Guelph, 50 Stone Rd E, Guelph, Ontario, Canada School of Environmental Design and Rural Development, University of Guelph, 50 Stone Rd E, Guelph, Ontario, Canada
AR TI CLE I NF O
AB S T R A CT
Keywords: Sri Lanka Program evaluation School-based intervention Childhood violence Violence prevention Dose response
Violence against children is a global public health issue with serious social, economic, physical, and emotional impacts. This study evaluates the effectiveness of a school-based program aimed to prevent and respond to physical, sexual, and psychological violence against children in Sri Lanka from the perspective of parents. A cross-sectional retrospective study design was used. A total of 835 parents of children who participated in the program were surveyed across seven districts in Sri Lanka. Dose-response analyses were conducted to assess for correlations between program exposure and perceived prevention of violence against children. Low to moderate correlations were found between exposure to the program and perceived child safety in schools, school policies, and in the community. The findings provide preliminary evidence of program effectiveness; however, more efforts are needed to validate and sustain outcomes. Implications for future violence prevention programming, along with the use of dose-response evaluations, are discussed.
1. Introduction Prevention of violence against children is a global public health and development priority (United Nations Economic and Social Council, 2016; United Nations Secretary-General’s Study, 2006). Article 19 of the United Nations Convention on the Rights of the Child defines violence against children as “all forms of physical or mental violence, injury and abuse, neglect or children negligent treatment, maltreatment or exploitation, including sexual abuse” (Child Rights International Network, 2017). It is well documented that children exposed to violence are at risk for impaired physical, emotional, and social development, with broader social and economic impacts (Fang, Brown, Florence, & Mercy, 2012; Fang et al., 2015; Korbin, 2003; Stark and Landis, 2016). Children in humanitarian settings are presumed to face an increased risk of exposure to violence (Stark & Landis, 2016). A humanitarian setting is one in which an event (e.g. armed conflict, natural disaster, political unrest) or series of events has resulted in a critical threat to the health, safety, security, or well-being of a population (The Sphere Project, 2011). Sri Lankan children, in particular, are especially vulnerable to violence, given the aftermath of the Asian tsunami in 2004 which devastated a large part of the country, along with the long-lasting civil war which came to an end in 2009 (Catani, Schauer, & Neuner, 2008; Catani et al., 2010; United Nations, 2011). For example, Catani et al. (2008) found a relationship between war violence and violent behaviour inflicted on children. Furthermore, studies have suggested an unsettling high rate of child sexual abuse in Sri Lanka, with children of low socio-economic status being the most susceptible (de Zoysa, 2002). A recent study in 2013 suggested that violence against children
⁎
Corresponding author at: 96 Kathleen St Guelph, ON, N1H 4Y3, Canada. E-mail address: [email protected] (S. Lam).
http://dx.doi.org/10.1016/j.chiabu.2017.10.016 Received 16 May 2017; Received in revised form 20 October 2017; Accepted 30 October 2017 0145-2134/ © 2017 Elsevier Ltd. All rights reserved.
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
has considerably increased in Sri Lanka, especially towards the end of the last decade (Arachchige, Shiroma, & Shirajanie, 2013). Despite the alarming evidence of violence against children in humanitarian settings, efforts to prevent and respond to the violence remain scarce (Hundeide & Armstrong, 2011; Hyder & Mac Veigh, 2007; Stark and Landis, 2016), with limited efforts focusing on the Sri Lankan context (Catani et al., 2010, 2008). There is an urgent need to provide children with the means to protect themselves. Children spend most of their time in school, and school plays an important role in the learning and development of children. As such, schools present an opportunity to help children develop the knowledge and skills necessary to keep them safe. To this end, the Canadian Red Cross Society, in partnership with the Ministry of Education of Sri Lanka, and endorsement from the Sri Lanka Red Cross Society, implemented a school-based violence against children prevention program called Be Safe. Be Safe was implemented in primary schools across the country from 2008 to 2014 and targeted boys and girls between the ages of 5 and 9. The goal of the program was to create and maintain environments that are safe from violence and abuse for children in 10% of schools under the Ministry of Education by 2014. Although Be Safe has officially ended in 2014, the relationship between program activities and reduced violence against children have yet to be determined. To support evidence-based programming and development of approaches to respond to violence against children, a rigorous evaluation was warranted. Dose-response evaluation represents an approach to examine the link between program exposure (or dose) and program outcomes. This approach has been widely-used in the evaluation of clinical trials, and more recently, in the evaluation of program interventions in controlled research contexts (Grieco, Jowers, Errisuriz, & Bartholomew, 2016; Jørgensen et al., 2015; Kim et al., 2016). However, dose-response evaluations have not been well applied to large-scale interventions in complex humanitarian environments. This study aims to evaluate the effectiveness of the Be Safe program in Sri Lanka using a dose-response approach from the perspectives of parents. Specifically, the objectives are to: 1) determine the extent of perceived prevention of violence against children; and 2) determine if variability in program exposure is associated with variability in level of outcomes achieved. This study adds to the knowledge base of effectiveness of violence against children prevention programs in Sri Lanka, with implications for violence against children prevention programming in other humanitarian contexts internationally. Furthermore, this study advances evaluation methodologies that seek to understand the relationship between program exposure and program outcomes in long-term and complex programming environments. 1.1. Overview of the Be safe program Engaging teachers to provide education early in life is an important strategy for preventing child abuse and neglect (Fortson, Klevens, Merrick, Gilbert, & Alexander, 2016; World Health Organization, 2016). Be Safe is a personal safety, school-based program for children ages 5–9, with a focus on preventing sexual abuse (Canadian Red Cross, 2016). Through simple key messages and childfriendly learning resources, Be Safe introduces young children to the concepts of safe and unsafe touching, children’s rights, adult’s responsibilities to protect children from harm, and safe and supportive friendships. Going further, Be Safe introduces simple and personal safety rules (e.g. say NO, get away, tell someone you trust, keep on telling until someone helps you). This program also provides detailed guidelines for educators, information for parents, and templates and tools to help schools communicate with parents and assist administrators to implement Be Safe. There are seven types of resources: ‘Asha the Elephant’, ‘Be Safe! lesson cards’, ‘Be Safe! classroom poster’, ‘Be Safe! booklet Your Body Belongs to You’, ‘Be Safe! song cassette and CD’, ‘Be Safe! bookmark’, and ‘Be Safe! leaflet for parents and adults’. The resources used by teachers aimed to build capacity and skills of children. For instance, the ‘Be Safe! lesson cards’ involved story-telling, interactive games, and role-playing to teach children about how everyone has a different and a unique body, which deserves to be safe and treated with respect. These lesson cards also helped to familiarize the children with concepts such as body ownership, how to determine unsafe environments, avoid unsafe situations, and seek help. The lesson cards took around 20–30 min to teach and were delivered in all classrooms over a period of 10–12 weeks. The classroom sizes ranged from 20–30 children depending on the district and size of community. The teacher adapted the sessions as appropriate to consider the needs of different ages and cultural backgrounds of children. 1.2. Development, implementation, and monitoring Be Safe was developed from 2005 to 2007 and was piloted in early 2007. The Ministry of Education was involved in the development phase from the very beginning. During this phase, parents, teachers, and children, along with representatives from nearly 70 organizations working in child protection, were consulted. After incorporating recommendations from the pilot, a memorandum of understanding between Canadian Red Cross and Ministry of Education was signed in 2008. The program ran from 2008 to 2014 in all 25 districts of Sri Lanka, and in 2238 schools. Building on this program, the Ten Steps program was added which focused on providing support for schools to create policies, structures and other mechanisms to protect children from violence. Ten Steps was implemented from 2010 to 2014 in eight districts in 223 schools. A total of 186,545 parents and teachers, and 125,027 students were reached by these two programs. Around 23% of schools in Sri Lanka were reached by these programs, which is more than double the program’s target goal of creating safe environments in 10% of schools by 2014. A total of 150 master trainers were recruited by in-service advisors from 97 Zonal Offices in nine provinces in Sri Lanka. The Canadian Red Cross trained the master trainers in basic counselling skills. Then, each master trainer trained two to four teachers from each project school. Upon successful training, the teachers were equipped to integrate Be Safe elements into their classrooms and to deliver the Be Safe messages to their students. However, teachers were instructed not to start training children until their parents were educated about the program. Awareness building sessions for parents were conducted by teachers during or after school hours. 130
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
A master trainer was almost always present during such sessions. A typical awareness building session lasted one hour and introduced parents to diverse types of violence against children and provided an overview of the Be Safe program. Such sessions allowed for an opportunity to build support for the program from parents. The master trainers were responsible for monitoring the program. They visited their trainees at least twice − once at the beginning of the lesson card delivery and again towards the end. The teachers submitted monitoring reports at the end of their lesson card delivery to the master trainers, who then consolidated the data and submitted quarterly reports to the Canadian Red Cross. 2. Methods 2.1. Evaluation design The purpose of this study is to evaluate outcomes of the school-based violence against children prevention program on students, parents, teachers, and school environments in Sri Lanka. When determining the appropriate evaluation design, the research team considered a range of contextual and limiting factors. These included: 1) the large-scale and large-time frame implementation of the program across all 25 districts in Sri Lanka; 2) the complex humanitarian environment of Sri Lanka; 3) the lack of baseline data or control group for longitudinal analysis or pre/post-test comparisons; and 4) the implementation of this final evaluation one year after program completion. The evaluation of interventions in complex environments remain challenging, as is the case for evaluation efforts where conventional methods (e.g. randomized control trials) are unethical or impractical to implement (Deering et al., 2011). As such, a cross-sectional retrospective approach was used. This approach assesses a sample of schools at one point in time, and relies on the memory of participants to recall program activities and program outcomes. Ethical guidelines of the Tri-Council Policy Statement regarding program evaluations were followed, including obtaining informed consent and ensuring privacy rights of all participants. 2.2. Participants A sample of children, parents, and teachers that participated in the Be Safe program had taken part in the program evaluation. For the purposes of this study, however, we only report on the perspectives of parents, who have best knowledge on the development outcomes of their children. We recognize that children and teachers also represent an important perspective of program outcomes. Children and teacher perspectives of the program may be presented in future publications. 2.3. Sampling and data collection A cluster sampling approach was used to select schools for this study. First, seven out of the 25 districts in Sri Lanka were randomly selected. Then, seven to eight schools were randomly selected among these districts. In total, the number of schools involved in the study was 54. The Be Safe and Ten Steps programs reached 125,027 students (63,198 males and 62,738 females) across the country. For a simple random sample design, a sample of 384 students would be needed to achieve a 95% confidence level. Pure random sampling, however, was not logistically possible. A cluster sampling design was used in which 835 parents of children who participated in the program were surveyed (around 15–16 parents from each school). This increase in sample size reflected a design effect of 2.18 which was needed to adjust for the cluster sampling strategy. Prior to data collection, the appropriate zonal education authorities were contacted and informed of the work and the presence of the research teams. In each school, The Be Safe and/or Ten Steps teacher was contacted with the assistance of the principal. The teacher helped to identify the students who had participated in the program. The students, the teacher, and the principal, then assisted the researchers in locating parents. The parents were surveyed in their homes. The research team trained two teams of seven surveyors, one fluent in Tamil and the other in Sinhalese. Each team had a senior team leader. A local Sri Lankan consultant supported both teams in data collection. A pretest of the data collection tools was carried out in Hambantota district on April 22, 23, and 24, 2015. Surveys were then revised, data entry templates were established, and the Tamil and Sinhalese translations were finalized. The survey team surveyed the remaining sample. Throughout the data collection process, the questionnaire data were continuously monitored for data quality and completeness as it was uploaded onto an online survey database (SurveyMonkey). 2.4. Measure The cross-sectional survey was conducted from April to July 2015. The survey asked for demographic information and respondent characteristics. The survey also assessed for degree of program exposure, along with perceptions of program outcomes surrounding child safety in school, school policies, and the community. For example, respondents were asked, “to your knowledge, did Be Safe influence your child’s sense of safety at his/her school?” (Table 1). All answers were designed using a Likert scale (e.g. strongly disagree, somewhat agree, mostly agree, strongly agree). To determine the dose-response relationship, dosage variables were defined as: 1) the number of years in the Be Safe program; 2) the number of program elements remembered by parents; and 3) the number of program objectives remembered by parents. Information on the number of years of program involvement came from discussions with the teacher prior to meeting with the parent, and was then confirmed by the parent. Parents were asked to recall Be Safe objectives and Be Safe elements that they remember from 131
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
Table 1 Questions and statements from surveys that assess for perceived outcomes of the Be Safe program. Effectiveness questions
Survey Question Number
Be Safe program outcomes for school safety The program increased the student’s sense of safety at my child's school It is okay for teachers to use physical punishment against children to maintain discipline I am aware of rules prohibiting violence among students at my child’s school The students are informed about the rules for prohibiting physical punishment The program influenced how informed parents and students are about the rules for prohibiting physical punishment School’s that participated in the program make sure that there is someone in the school to talk to if the rules about physical punishment are violated
Q. Q. Q. Q. Q. Q.
2b 2e 2g 2h 2i 2k
Exposure to the Be Safe program Did your children share with you or talk about the elements of Be Safe? (check all that apply) Did your children share with you or talk about the following Be Safe objectives? (check all that apply)
Q. 5 Q. 6
Be Safe program outcomes for children Do you feel that the program is beneficial for boys? Do you feel that the program is beneficial for girls? How beneficial was Be Safe for younger (5, 6, 7 years old) children? How beneficial was Be Safe for older (8, 9 years old) children? The program taught my child who to go to with concerns regarding violence.
Q. Q. Q. Q. Q.
7 8 9a 9b 11c
discussions with their children. Be Safe objectives referenced main themes on how to be safe, while Be Safe elements referred to tools used in the classroom to teach material to the children. 2.5. Data analysis The analytical process began with exporting survey data from Survey Monkey into Excel 2007 for the preliminary data cleaning and analysis. Then, the data was uploaded into IBM SPSS Statistics 22.0 (IBM Corp., Armonk, NY, USA) for further testing and analysis. All statistical analyses were performed using this software. Parent questionnaire responses were grouped according to the following variables: 1) variables describing demographic information and respondent characteristics (e.g. age, gender); 2) variables describing program outcome (e.g. child safety, school policies, safety in community); and 3) program dosage variables. Several quantitative analysis tests were conducted. Descriptive statistics including means and percentages were calculated for demographic information, respondent characteristics, and program outcome variables. Cross tabulations between these variables were also generated. Spearman Rank-Order Correlation was used to measure the strength of association between two variables. For example, it was used to determine whether an increase in program exposure (e.g. parent’s recall of Be Safe objectives) was associated with program outcome (e.g. the program increased the student’s sense of safety at my child’s school). 3. Results A total of 835 parents were surveyed across seven districts (Hambantota, Nuwara Eliya, Badulla, Kegalle, Trincomale, Anuradhapura, Matale) in Sri Lanka. Respondents mainly included the child’s mother (65.6%) or father (17.4%). In some cases where the parents were not available, the grandmother was surveyed (6.2%). Around 94% of household heads were male and 6% were female. The average age for the head of the household was 43 years old, and the average highest level of education for the head of the household was secondary education (grade 10). Regarding ethnicity, 67.2% of respondents were Sinhala, 30.4% were Tamil, and 2.3% were Muslim. On average, families had three children. Over half of the children were male (53.7%). Almost all parents (95%) were involved in the Be Safe program while 5% were involved in both Be Safe and Ten Steps. Very few parents (1.7%) participated in the program in 2014, while most parents (30.4%) participated in the program in 2013. Most participants participated in the program in 2013 (30.4%) and 2012 (23%), or earlier (31.5%). 3.1. Program exposure As a measure of program exposure, the survey asked parents to provide information regarding Be Safe objectives, Be Safe elements, and number of years that their children participated in the program. Of the 12 Be Safe objectives, 45% of parents did not recall any Be Safe objectives discussed with their children. Over half of parents (55%) reported discussing at least one Be Safe objective with their child (Table 2), and among these parents, the objectives discussed included good and bad touches (19.8%), your body belongs to you (19.1%), Bodies are private (11.3%), and it is unsafe for someone to hit you (10%). Of the seven Be Safe elements, 37.3% of parents did not recall any Be Safe elements. Around 63% of parents recalled at least one Be Safe element, and among these parents, most remembered their children talking about the classroom poster (28.9%), followed by Asha the Elephant (23.7%), and the lesson cards (19.9%). Very few parents remembered seeing the take-home bookmark (2.3%) or the parent leaflet (3.3%). Most children and parents (82.4%) participated in the program for one year. 132
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
Table 2 Exposure of children and parents to the Be Safe program. Exposure Variable
Number (%)
Number of years the child has participated in the Be Safe programa 1 280 (82.4) 2 47 (13.8) 3 13 (3.8) 4 0 5 0 Number of Be Safe objectives recalled by parents 0 1 2 3 4 5 6 7 8 9 10 11 12
377 (44.9) 35 (4.2) 51 (6.1) 90 (10.7) 93 (11.1) 48 (5.7) 38 (4.5) 24 (2.9) 21 (2.5) 6 (0.7) 8 (1) 8 (1) 41 (4.9)
Number of Be Safe elements recalled by parents 0 1 2 3 4 5 6 7
313 (37.3) 125 (14.9) 108 (12.9) 148 (17.6) 79 (9.4) 36 (4.3) 13 (1.5) 18 (2.1)
a
Question regarding the first child (if families had more than one children).
3.2. Program effectiveness Descriptive statistics show that most parents believe that the program was beneficial for both genders, with more parents agreeing that the program was beneficial for girls (99.9%) than for boys (82.7%) (p < 0.05) (Table not shown). Spearman’s Rank Correlation was used to assess for the presence of a dose-response relationship between program exposure and perceived benefits of the program for girls and boys. A moderate correlation was found between the dosage variable “Be Safe objectives” and outcome variable “is the program beneficial for boys?” (r = 0.398) (Table 3). A small but significant correlation was found between the dosage variable “Be Safe objectives” and the outcome variable “is the program beneficial for girls?” (r = 0.117). The direction of the correlations suggests that parents with higher exposure to the program were more likely to agree that the program was beneficial for boys and girls. Most parents (91%) agreed that the program increased students' sense of safety at their child’s school children (Fig. 1). A significant correlation was found between the dosage variable “Be Safe elements” and the outcome variable “the program increased students' sense of safety at my child’s school” (r = 0.311). A significant correlation was also found between the “Be Safe objectives” and “Be Safe taught my child who to go to regarding violence and abuse” (r = 0.282). There was a small, non-significant correlation found between the dosage variable “Be Safe objectives” and the outcome variable “It is okay for teachers to use physical punishment against children to maintain discipline”. Several survey questions aimed to determine the effectiveness of school policies in creating safe environments for protecting children from violence. Three in four parents reported that they are aware of rules prohibiting violence among students at their child’s school, and there was a small significant correlation found between this outcome and the dosage variable “Be Safe objectives” (r = 0.136). However, when asked if the program influenced how informed parents and students are about the rules for prohibiting physical punishment, the extent of agreement was lower (70.4%) and there was no dose-response correlation found. Going further, there was no dose-response correlations found between dosage variables and parents' attitude towards physical punishment against children, students' awareness of rules for prohibiting physical punishment, and schools' efforts for ensuring there is someone students can talk to regarding rules of physical punishment. 3.3. Impact of year of program delivery and age of child The time of program delivery ranged from 2008 to 2014. There was no significant correlation found between the year of program delivery and measures of program outcomes, with one exception. In one instance, a small correlation was found between the year of program delivery and the outcome variable “the program increased the students' sense of safety at my child’s school” (r = − 0.119, 133
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
Table 3 Correlations between Be Safe program exposure and program outcomes. Effectiveness Questions
School safety for children The program increased the student’s sense of safety at my child's school It is okay for teachers to use physical punishment against children to maintain discipline I am aware of rules prohibiting violence among students at my child’s school The students are informed about the rules for prohibiting physical punishment The program influenced how informed parents and students are about the rules for prohibiting physical punishment Schools that participated in the program make sure that there is someone in the school to talk to if the rules about physical punishment are violated The program taught my child who to go to with concerns regarding violence Program benefits for different groups of children Do you feel that the program is beneficial for boys? Do you feel that the program is beneficial for girls? How beneficial was Be Safe for younger (5, 6, 7 years old) children? How beneficial was Be Safe for older (8, 9 years old) children? a
a
Dosage Variable Applied
Pearson Rank Correlation
Interpretation
Elements
(r = 0.311, p = 0.001, n = 622) N/A
Statistically significant, moderate correlation Not statistically significant
Not applicable
(r = 0.136, p = 0.001, n = 764) N/A
Statistically significant, low correlation Not statistically significant
Not applicable
N/A
Not statistically significant
Not applicable
N/A
Not statistically significant
Objectives
(r = 0.282, p = 0.001, n = 527)
Statistically significant, low correlation
Objectives
(r = 0.398, p = 0.001, n = 782) (r = 0.117, p = 0.003, n = 662) N/A (r = 0.133, p = 0.003, n = 484)
Statistically significant, moderate correlation Statistically significant, low correlation Not statistically significant Statistically significant, low correlation
Objectives Objectives
Objectives a
Objectives Objectives
Question regarding the first child (if families had more than one children).
Fig. 1. Parents' attitudes towards Be Safe program’s influence on their child’s safety at school.
p = 0.049, n = 272). The direction of the correlation suggests that parents of children who participated in Be Safe in later years were less likely to agree that the program increased their child’s sense of safety in the school. Apart from this correlation, the general trend suggests that parents' perceived outcomes of the program are consistent regardless of what year their child participated in the program. Almost all participants (99%) reported that the program was beneficial to both younger and older children. Regarding program benefits for younger and older children, there was no significant correlation found between dosage variables and perceived program benefits for younger children (r = 0.51, p = 0.259, n = 498). However, there was a small but significant correlation found between dosage variable “Be Safe objectives” and program benefit for older children (r = 0.133, p < 0.05, n = 484). This finding suggests that higher exposure to the program is correlated with an increase in parents' perception of program benefit for older children.
134
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
4. Discussion The purpose of this study was to determine the effectiveness of the Be Safe program. We explored the extent of program outcomes achieved, along with dose-response relationships between program exposures and program outcomes. Results show that most parents agree that the program increased their child’s sense of safety at the school, and agree that their child has someone to talk to regarding rules about physical punishment. Going further, there appears to be plausible correlations between increase in program exposure and increase in parents' perceived child safety in the school, school policies targeting violence prevention, and perceived child safety in the community. While correlations were found between program outcomes and program exposure, most correlations were low. In some cases, correlations were not statistically significant. For example, we did not find a statistically significant correlation between program exposure variables and perceived benefits of the program for younger children. Furthermore, no significant correlations were found regarding rules about physical punishment. There may be several reasons for the low correlations observed, including the lack of appropriately defined exposure variables. For instance, in many cases, correlations were found when applying the dosage variable “Be Safe objectives”, and rarely found using the other dosage variables (e.g. Be Safe elements, number of years in the Be Safe program). However, similarly defined dosage variables have been applied in other program intervention studies where strong correlations have been observed (Deering et al., 2011; Patel, Corter, Pelletier, & Bertrand, 2016). The low correlation could be due to the presence of confounding variables. Less than half of respondents (45%) did not remember their children discussing any of the Be Safe objectives, and 37.3% did not recall their children discussing any Be Safe elements. Most parents and students (82.4%) were involved in the program for one year. The low exposure could be due to several limitations of this study, including the reliance on parent’s self-reported knowledge as opposed to children’s self-reported knowledge, or the lack of parental involvement in the programming. The timing of the study is also a major factor in low exposure rates given that the program ended nearly two years ago for most parents at the time of the evaluation. Future evaluative research is needed to determine the reason for lack of intervention coverage. Although most parents perceived that the program was beneficial for both boys and girls, significantly more parents perceived that the program was beneficial for girls than boys (p < 0.05). However, results suggest that as exposure to the program increases, parents' perceived benefits of the program to both boys and girls increases. Other domestic abuse prevention programs have also assumed that the intervention would have more of an impact on girls than boys (Fox, Corr, Gadd, & Sim, 2016). This likely stems from the developing country context where prevalence of gender-based violence is high, especially against girls and women (Jayasuriya, Wijewardena, & Axemo, 2011; Martin et al., 2002). To allow for gender equality, future programming should emphasize the importance of violence prevention programs for both girls and boys, while engaging parents in the discussion. Despite low exposure to the program interventions and low correlations found, there is some evidence to suggest that increasing exposure to the program contributes to violence against children prevention outcomes. For example, parents that recalled more Be Safe objectives were more likely to perceive that the program benefited their child, and that their child knows who to go to regarding reporting abuse. Going further, although students participated in the program across a seven-year period (2008–2014), no correlations were found between the year of participation in the program and program outcome variables. This finding suggested that children who participated in early years had similar outcomes to children who participated in later years. Dose-response provides evidence of program effectiveness that goes beyond reporting of frequencies and counts of achieved outcomes. Continued prevention and program evaluation efforts are important for demonstrating evidence-based programming and in reaching long-term outcomes. There are many factors that contribute to violence against children. For example, studies have shown that the experience of mass trauma in the family (e.g. war, natural disasters) may contribute to violence against children in a post-conflict setting (Catani et al., 2008; Sriskandarajah, Neuner, & Catani, 2015). Going further, studies have shown a correlation between exposure to war and domestic violence (Catani et al., 2008; Yount, DiGirolamo, & Ramakrishnan, 2011). The impacts of violence against children, combined with traumatic experiences, points toward an urgent need for prevention and intervention. Treatment approaches should focus not only on the child level, but also on the family and community level (Catani et al., 2008). The high level of family violence indicate a need for interventions to promote non-violent parenting and to protect children living in post-war settings (Catani et al., 2008; Sriskandarajah et al., 2015; Tolan, Gorman-Smith, & Henry, 2006). Although parents were secondary targets of the Be Safe program, more efforts are needed to engage parents in programming activities to prevent children against violence. This need was also recently supported by the Global Partnership to End Violence Against Children in 2016 (End Violence Against Children: The Global Partnership, 2016). There are several important limitations of this study. First, this study was based on a cross-sectional survey that relied on the perspective of parents of children who participated in the Be Safe program (i.e. indirect program participants). Furthermore, it has been more than two years since the program has ended for most parents and children. Yet, this method may be the only method by which data for programs of this nature can be obtained, and significant correlations were observed. The extent of outcomes achieved and the strength of correlations may be higher if we examined self-reported data from children. However, the data collection on these outcomes would be challenging. Finally, the lack of a baseline or control group also makes it difficult to interpret the importance of the correlations. However, despite these limitations and challenges, we found correlations that suggest that the program is effective in preventing violence against children. 5. Conclusion This study evaluated the effectiveness of a violence against children prevention program using a dose-response approach. Study 135
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
findings suggested that an increase in exposure to the program was associated with improvements in parents' perceived child safety in the school, school policies surrounding violence prevention, and child safety in the community. However, the exposure to the program was relatively low, and low correlations were found. Researchers and practitioners may consider the usefulness of applying dose-response measures in program monitoring and evaluation efforts. While improvements were achieved across various violence against children prevention outcomes, there remains a large unfinished agenda for developing, applying, and evaluating efforts to reduce violence against children. Future research and evaluations need to consider the perspectives of children. Furthermore, experimental designs and pre-program measures should be considered early in the planning process to ensure that stronger evidence of program effectiveness can be collected. Conflict of interests None. Author contributions SL contributed to drafting the manuscript. HC collected the data and supported data analysis. CZ supported data analysis and report writing. DL supported data analysis. IC supported manuscript writing. All authors read and approved the final manuscript. Acknowledgements Thank you to the survey teams for supporting data collection, and thank you to Henry de Mel for supporting the survey teams in Sri Lanka. We thank all the study participants for sharing their time and information with us. Thank you to Rob Maclean, Kendra Siekmans, and David Kupp for providing helpful feedback on earlier drafts. Thank you to the two anonymous reviewers for their insightful comments. Finally, we would like to thank the Canadian Red Cross for funding this study. The funder had no role in the design and conduct of the study. References Arachchige, M., Shiroma, D., & Shirajanie, J. (2013). Violence against children and ensuring the justice for child victims through human rights and penal laws perspectives in Sri Lankan context? International Journal of Research In Social Sciences, 4(3), 111–120. Canadian Red Cross (2016). Program for young children: Be Safe! can help prevent sexual abuse of children. Retrieved March 21, 2016, from: http://www.redcross.ca/ how-we-help/violence–bullying-and-abuse-prevention/educators/child-abuse-and-neglect-prevention/program-for-young-children–be-safe–can-help-preventsexual-abuse-of-children. Catani, C., Schauer, E., & Neuner, F. (2008). Beyond individual war trauma: domestic violence against children in Afghanistan and Sri Lanka. Journal of Marital and Family Therapy, 34(2), 165–176. http://dx.doi.org/10.1111/j. 1752-0606.2008.00062.x. Catani, C., Gewirtz, A. H., Wieling, E., Schauer, E., Elbert, T., & Neuner, F. (2010). Tsunami, war, and cumulative risk in the lives of Sri Lankan schoolchildren. Child Development, 81(4), 1176–1191. http://dx.doi.org/10.1111/j. 1467-8624.2010.01461.x. Child Rights International Network (2017). Article 19: Protection from abuse and neglect. Retrieved April 14, 2017, from: https://www.crin.org/en/home/rights/ convention/articles/article-19-protection-abuse-and-neglect. de Zoysa, P. (2002). Child sexual abuse in Sri Lanka: The current state of affairs and recommendations for the future. Journal of Child Sexual Abuse, 11(2), 97–113. http://dx.doi.org/10.1300/J070v11n02_06. Deering, K. N., Boily, M.-C., Lowndes, C. M., Shoveller, J., Tyndall, M. W., Vickerman, P., & Alary, M. (2011). A dose-response relationship between exposure to a large-scale HIV preventive intervention and consistent condom use with different sexual partners of female sex workers in southern India. BMC Public Health, 11(Suppl. 6), S8. http://dx.doi.org/10.1186/1471-2458-11-S6-S8. End Violence Against Children: The Global Partnership (2016). A world in which every children grows up free from violence. Retrieved March 21, 2016, from: http:// www.end-violence.org/. Fang, X., Brown, D. S., Florence, C. S., & Mercy, J. A. (2012). The economic burden of child maltreatment in the United States and implications for prevention. Child Abuse & Neglect, 36(2), 156–165. http://dx.doi.org/10.1016/j.chiabu.2011.10.006. Fang, X., Fry, D. A., Brown, D. S., Mercy, J. A., Dunne, M. P., Butchart, A. R., & Swales, D. M. (2015). The burden of child maltreatment in the East Asia and Pacific region. Child Abuse & Neglect, 42(43123767), 146–162. http://dx.doi.org/10.1016/j.chiabu.2015.02.012. Fortson, B., Klevens, J., Merrick, M., Gilbert, L., & Alexander, S. (2016). Preventing child abuse and neglect: A technical package for policy, norm, and programmatic activities. national center for injury prevention and control. Atlanta, Georgia: Centers for Disease Control and Prevention. Retrieved from: http://www.cdc.gov/ violenceprevention/pdf/can-prevention-technical-package.pdf. Fox, C. L., Corr, M. L., Gadd, D., & Sim, J. (2016). Evaluating the effectiveness of domestic abuse prevention education: Are certain children more or less receptive to the messages conveyed? Legal and Criminological Psychology, 21(1), 212–227. http://dx.doi.org/10.1111/lcrp.12046. Grieco, L. A., Jowers, E. M., Errisuriz, V. L., & Bartholomew, J. B. (2016). Physically active vs. sedentary academic lessons: A dose response study for elementary student time on task. Preventive Medicine, 89, 98–103. http://dx.doi.org/10.1016/j.ypmed.2016.05.021. Hundeide, K., & Armstrong, N. (2011). ICDP approach to awareness-raising about children’s rights and preventing violence, child abuse, and neglect. Child Abuse Neglected, 35(12), 1053–1062. http://dx.doi.org/10.1016/j.chiabu.2011.09.008. Hyder, T., & Mac Veigh, J. (2007). Gender-based violence against children in emergencies: Save the Children UK’s response. Gender & Development, 15(1), 81–93. http://dx.doi.org/10.1080/13552070601179110. Jørgensen, T. S., Rasmussen, M., Aarestrup, A. K., Ersbøll, A. K., Jørgensen, S. E., Goodman, E., & Krølner, R. (2015). The role of curriculum dose for the promotion of fruit and vegetable intake among adolescents: Results from the Boost intervention. BMC Public Health, 15, 536. http://dx.doi.org/10.1186/s12889-015-1840-0. Jayasuriya, V., Wijewardena, K., & Axemo, P. (2011). Intimate partner violence against women in the capital province of sri lanka: Prevalence, risk factors, and help seeking. Violence Against Women, 17(8), 1086–1102. http://dx.doi.org/10.1177/1077801211417151. Kim, S. S., Rawat, R., Mwangi, E. M., Tesfaye, R., Abebe, Y., Baker, J., & Menon, P. (2016). Exposure to large-scale social and behavior change communication interventions is associated with improvements in infant and young child feeding practices in Ethiopia. PLoS One, 11(10), http://dx.doi.org/10.1371/journal.pone. 0164800. Korbin, J. E. (2003). Children, childhoods, and violence. Annual Review of Anthropology, 32(1), 431–446. http://dx.doi.org/10.1146/annurev.anthro.32.061002. 093345. Martin, S. L., Moracco, K. E., Garro, J., Tsui, A. O., Kupper, L. L., Chase, J. L., & Campbell, J. C. (2002). Domestic violence across generations: Findings from northern
136
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
India. International Journal of Epidemiology, 31(3), 560–572. http://dx.doi.org/10.1093/ije/31.3.560. Patel, S., Corter, C., Pelletier, J., & Bertrand, J. (2016). Dose-response relations between participation in integrated early childhood services and children’s early development. Early Childhood Research Quarterly, 35, 49–62. http://dx.doi.org/10.1016/j.ecresq.2015.12.006. Sriskandarajah, V., Neuner, F., & Catani, C. (2015). Predictors of violence against children in Tamil families in northern Sri Lanka. Social Science and Medicine, 146, 257–265. http://dx.doi.org/10.1016/j.socscimed.2015.10.010. Stark, L., & Landis, D. (2016). Violence against children in humanitarian settings: A literature review of population-based approaches. Social Science and Medicine. http://dx.doi.org/10.1016/j.socscimed.2016.01.052. The Sphere Project (2011). Humanitarian charter and minimum standards in humanitarian response. response, Vol. 1http://dx.doi.org/10.3362/9781908176202. Tolan, P., Gorman-Smith, D., & Henry, D. (2006). Family violence. Annual Review of Psychology, 57, 557–583. http://dx.doi.org/10.1146/annurev.psych.57.102904. 190110. United Nations Economic, & Social Council (2016). Progress towards the sustainable development goals. Retrieved March 22, 2017, from: http://www.un.org/ga/search/ view_doc.asp?symbol=E/2016/75&Lang=E. United Nations Secretary-General’s Study (2006). World report on violence against children. Retrieved March 21, 2016, from: https://www.unicef.org/lac/full_tex(3). pdf. United Nations (2011). Report of the secretary-General’s panel of experts on accountability in Sri Lanka. Retrieved March 21, 2016, from: http://www.un.org/News/dh/ infocus/Sri_Lanka/POE_Report_Full.pdf. World Health Organization (2016). Inspire: Seven strategies for ending violence against children. Geneva, Switzerland: World Health Organization. Yount, K. M., DiGirolamo, A. M., & Ramakrishnan, U. (2011). Impacts of domestic violence on child growth and nutrition: A conceptual review of the pathways of influence. Social Science and Medicine, 72(9), 1534–1554. http://dx.doi.org/10.1016/j.socscimed.2011.02.042.
137
Contents lists available at ScienceDirect
Child Abuse & Neglect journal homepage: www.elsevier.com/locate/chiabuneg
Preventing violence against children in schools: Contributions from the Be Safe program in Sri Lanka
MARK
⁎
Steven Lama,b, , Christine Zwarta, Inem Chahala, David Lanea, Harry Cummingsa,c a b c
Harry Cummings and Associates, 96 Kathleen St, Guelph, Ontario, Canada Department of Population Medicine, University of Guelph, 50 Stone Rd E, Guelph, Ontario, Canada School of Environmental Design and Rural Development, University of Guelph, 50 Stone Rd E, Guelph, Ontario, Canada
AR TI CLE I NF O
AB S T R A CT
Keywords: Sri Lanka Program evaluation School-based intervention Childhood violence Violence prevention Dose response
Violence against children is a global public health issue with serious social, economic, physical, and emotional impacts. This study evaluates the effectiveness of a school-based program aimed to prevent and respond to physical, sexual, and psychological violence against children in Sri Lanka from the perspective of parents. A cross-sectional retrospective study design was used. A total of 835 parents of children who participated in the program were surveyed across seven districts in Sri Lanka. Dose-response analyses were conducted to assess for correlations between program exposure and perceived prevention of violence against children. Low to moderate correlations were found between exposure to the program and perceived child safety in schools, school policies, and in the community. The findings provide preliminary evidence of program effectiveness; however, more efforts are needed to validate and sustain outcomes. Implications for future violence prevention programming, along with the use of dose-response evaluations, are discussed.
1. Introduction Prevention of violence against children is a global public health and development priority (United Nations Economic and Social Council, 2016; United Nations Secretary-General’s Study, 2006). Article 19 of the United Nations Convention on the Rights of the Child defines violence against children as “all forms of physical or mental violence, injury and abuse, neglect or children negligent treatment, maltreatment or exploitation, including sexual abuse” (Child Rights International Network, 2017). It is well documented that children exposed to violence are at risk for impaired physical, emotional, and social development, with broader social and economic impacts (Fang, Brown, Florence, & Mercy, 2012; Fang et al., 2015; Korbin, 2003; Stark and Landis, 2016). Children in humanitarian settings are presumed to face an increased risk of exposure to violence (Stark & Landis, 2016). A humanitarian setting is one in which an event (e.g. armed conflict, natural disaster, political unrest) or series of events has resulted in a critical threat to the health, safety, security, or well-being of a population (The Sphere Project, 2011). Sri Lankan children, in particular, are especially vulnerable to violence, given the aftermath of the Asian tsunami in 2004 which devastated a large part of the country, along with the long-lasting civil war which came to an end in 2009 (Catani, Schauer, & Neuner, 2008; Catani et al., 2010; United Nations, 2011). For example, Catani et al. (2008) found a relationship between war violence and violent behaviour inflicted on children. Furthermore, studies have suggested an unsettling high rate of child sexual abuse in Sri Lanka, with children of low socio-economic status being the most susceptible (de Zoysa, 2002). A recent study in 2013 suggested that violence against children
⁎
Corresponding author at: 96 Kathleen St Guelph, ON, N1H 4Y3, Canada. E-mail address: [email protected] (S. Lam).
http://dx.doi.org/10.1016/j.chiabu.2017.10.016 Received 16 May 2017; Received in revised form 20 October 2017; Accepted 30 October 2017 0145-2134/ © 2017 Elsevier Ltd. All rights reserved.
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
has considerably increased in Sri Lanka, especially towards the end of the last decade (Arachchige, Shiroma, & Shirajanie, 2013). Despite the alarming evidence of violence against children in humanitarian settings, efforts to prevent and respond to the violence remain scarce (Hundeide & Armstrong, 2011; Hyder & Mac Veigh, 2007; Stark and Landis, 2016), with limited efforts focusing on the Sri Lankan context (Catani et al., 2010, 2008). There is an urgent need to provide children with the means to protect themselves. Children spend most of their time in school, and school plays an important role in the learning and development of children. As such, schools present an opportunity to help children develop the knowledge and skills necessary to keep them safe. To this end, the Canadian Red Cross Society, in partnership with the Ministry of Education of Sri Lanka, and endorsement from the Sri Lanka Red Cross Society, implemented a school-based violence against children prevention program called Be Safe. Be Safe was implemented in primary schools across the country from 2008 to 2014 and targeted boys and girls between the ages of 5 and 9. The goal of the program was to create and maintain environments that are safe from violence and abuse for children in 10% of schools under the Ministry of Education by 2014. Although Be Safe has officially ended in 2014, the relationship between program activities and reduced violence against children have yet to be determined. To support evidence-based programming and development of approaches to respond to violence against children, a rigorous evaluation was warranted. Dose-response evaluation represents an approach to examine the link between program exposure (or dose) and program outcomes. This approach has been widely-used in the evaluation of clinical trials, and more recently, in the evaluation of program interventions in controlled research contexts (Grieco, Jowers, Errisuriz, & Bartholomew, 2016; Jørgensen et al., 2015; Kim et al., 2016). However, dose-response evaluations have not been well applied to large-scale interventions in complex humanitarian environments. This study aims to evaluate the effectiveness of the Be Safe program in Sri Lanka using a dose-response approach from the perspectives of parents. Specifically, the objectives are to: 1) determine the extent of perceived prevention of violence against children; and 2) determine if variability in program exposure is associated with variability in level of outcomes achieved. This study adds to the knowledge base of effectiveness of violence against children prevention programs in Sri Lanka, with implications for violence against children prevention programming in other humanitarian contexts internationally. Furthermore, this study advances evaluation methodologies that seek to understand the relationship between program exposure and program outcomes in long-term and complex programming environments. 1.1. Overview of the Be safe program Engaging teachers to provide education early in life is an important strategy for preventing child abuse and neglect (Fortson, Klevens, Merrick, Gilbert, & Alexander, 2016; World Health Organization, 2016). Be Safe is a personal safety, school-based program for children ages 5–9, with a focus on preventing sexual abuse (Canadian Red Cross, 2016). Through simple key messages and childfriendly learning resources, Be Safe introduces young children to the concepts of safe and unsafe touching, children’s rights, adult’s responsibilities to protect children from harm, and safe and supportive friendships. Going further, Be Safe introduces simple and personal safety rules (e.g. say NO, get away, tell someone you trust, keep on telling until someone helps you). This program also provides detailed guidelines for educators, information for parents, and templates and tools to help schools communicate with parents and assist administrators to implement Be Safe. There are seven types of resources: ‘Asha the Elephant’, ‘Be Safe! lesson cards’, ‘Be Safe! classroom poster’, ‘Be Safe! booklet Your Body Belongs to You’, ‘Be Safe! song cassette and CD’, ‘Be Safe! bookmark’, and ‘Be Safe! leaflet for parents and adults’. The resources used by teachers aimed to build capacity and skills of children. For instance, the ‘Be Safe! lesson cards’ involved story-telling, interactive games, and role-playing to teach children about how everyone has a different and a unique body, which deserves to be safe and treated with respect. These lesson cards also helped to familiarize the children with concepts such as body ownership, how to determine unsafe environments, avoid unsafe situations, and seek help. The lesson cards took around 20–30 min to teach and were delivered in all classrooms over a period of 10–12 weeks. The classroom sizes ranged from 20–30 children depending on the district and size of community. The teacher adapted the sessions as appropriate to consider the needs of different ages and cultural backgrounds of children. 1.2. Development, implementation, and monitoring Be Safe was developed from 2005 to 2007 and was piloted in early 2007. The Ministry of Education was involved in the development phase from the very beginning. During this phase, parents, teachers, and children, along with representatives from nearly 70 organizations working in child protection, were consulted. After incorporating recommendations from the pilot, a memorandum of understanding between Canadian Red Cross and Ministry of Education was signed in 2008. The program ran from 2008 to 2014 in all 25 districts of Sri Lanka, and in 2238 schools. Building on this program, the Ten Steps program was added which focused on providing support for schools to create policies, structures and other mechanisms to protect children from violence. Ten Steps was implemented from 2010 to 2014 in eight districts in 223 schools. A total of 186,545 parents and teachers, and 125,027 students were reached by these two programs. Around 23% of schools in Sri Lanka were reached by these programs, which is more than double the program’s target goal of creating safe environments in 10% of schools by 2014. A total of 150 master trainers were recruited by in-service advisors from 97 Zonal Offices in nine provinces in Sri Lanka. The Canadian Red Cross trained the master trainers in basic counselling skills. Then, each master trainer trained two to four teachers from each project school. Upon successful training, the teachers were equipped to integrate Be Safe elements into their classrooms and to deliver the Be Safe messages to their students. However, teachers were instructed not to start training children until their parents were educated about the program. Awareness building sessions for parents were conducted by teachers during or after school hours. 130
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
A master trainer was almost always present during such sessions. A typical awareness building session lasted one hour and introduced parents to diverse types of violence against children and provided an overview of the Be Safe program. Such sessions allowed for an opportunity to build support for the program from parents. The master trainers were responsible for monitoring the program. They visited their trainees at least twice − once at the beginning of the lesson card delivery and again towards the end. The teachers submitted monitoring reports at the end of their lesson card delivery to the master trainers, who then consolidated the data and submitted quarterly reports to the Canadian Red Cross. 2. Methods 2.1. Evaluation design The purpose of this study is to evaluate outcomes of the school-based violence against children prevention program on students, parents, teachers, and school environments in Sri Lanka. When determining the appropriate evaluation design, the research team considered a range of contextual and limiting factors. These included: 1) the large-scale and large-time frame implementation of the program across all 25 districts in Sri Lanka; 2) the complex humanitarian environment of Sri Lanka; 3) the lack of baseline data or control group for longitudinal analysis or pre/post-test comparisons; and 4) the implementation of this final evaluation one year after program completion. The evaluation of interventions in complex environments remain challenging, as is the case for evaluation efforts where conventional methods (e.g. randomized control trials) are unethical or impractical to implement (Deering et al., 2011). As such, a cross-sectional retrospective approach was used. This approach assesses a sample of schools at one point in time, and relies on the memory of participants to recall program activities and program outcomes. Ethical guidelines of the Tri-Council Policy Statement regarding program evaluations were followed, including obtaining informed consent and ensuring privacy rights of all participants. 2.2. Participants A sample of children, parents, and teachers that participated in the Be Safe program had taken part in the program evaluation. For the purposes of this study, however, we only report on the perspectives of parents, who have best knowledge on the development outcomes of their children. We recognize that children and teachers also represent an important perspective of program outcomes. Children and teacher perspectives of the program may be presented in future publications. 2.3. Sampling and data collection A cluster sampling approach was used to select schools for this study. First, seven out of the 25 districts in Sri Lanka were randomly selected. Then, seven to eight schools were randomly selected among these districts. In total, the number of schools involved in the study was 54. The Be Safe and Ten Steps programs reached 125,027 students (63,198 males and 62,738 females) across the country. For a simple random sample design, a sample of 384 students would be needed to achieve a 95% confidence level. Pure random sampling, however, was not logistically possible. A cluster sampling design was used in which 835 parents of children who participated in the program were surveyed (around 15–16 parents from each school). This increase in sample size reflected a design effect of 2.18 which was needed to adjust for the cluster sampling strategy. Prior to data collection, the appropriate zonal education authorities were contacted and informed of the work and the presence of the research teams. In each school, The Be Safe and/or Ten Steps teacher was contacted with the assistance of the principal. The teacher helped to identify the students who had participated in the program. The students, the teacher, and the principal, then assisted the researchers in locating parents. The parents were surveyed in their homes. The research team trained two teams of seven surveyors, one fluent in Tamil and the other in Sinhalese. Each team had a senior team leader. A local Sri Lankan consultant supported both teams in data collection. A pretest of the data collection tools was carried out in Hambantota district on April 22, 23, and 24, 2015. Surveys were then revised, data entry templates were established, and the Tamil and Sinhalese translations were finalized. The survey team surveyed the remaining sample. Throughout the data collection process, the questionnaire data were continuously monitored for data quality and completeness as it was uploaded onto an online survey database (SurveyMonkey). 2.4. Measure The cross-sectional survey was conducted from April to July 2015. The survey asked for demographic information and respondent characteristics. The survey also assessed for degree of program exposure, along with perceptions of program outcomes surrounding child safety in school, school policies, and the community. For example, respondents were asked, “to your knowledge, did Be Safe influence your child’s sense of safety at his/her school?” (Table 1). All answers were designed using a Likert scale (e.g. strongly disagree, somewhat agree, mostly agree, strongly agree). To determine the dose-response relationship, dosage variables were defined as: 1) the number of years in the Be Safe program; 2) the number of program elements remembered by parents; and 3) the number of program objectives remembered by parents. Information on the number of years of program involvement came from discussions with the teacher prior to meeting with the parent, and was then confirmed by the parent. Parents were asked to recall Be Safe objectives and Be Safe elements that they remember from 131
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
Table 1 Questions and statements from surveys that assess for perceived outcomes of the Be Safe program. Effectiveness questions
Survey Question Number
Be Safe program outcomes for school safety The program increased the student’s sense of safety at my child's school It is okay for teachers to use physical punishment against children to maintain discipline I am aware of rules prohibiting violence among students at my child’s school The students are informed about the rules for prohibiting physical punishment The program influenced how informed parents and students are about the rules for prohibiting physical punishment School’s that participated in the program make sure that there is someone in the school to talk to if the rules about physical punishment are violated
Q. Q. Q. Q. Q. Q.
2b 2e 2g 2h 2i 2k
Exposure to the Be Safe program Did your children share with you or talk about the elements of Be Safe? (check all that apply) Did your children share with you or talk about the following Be Safe objectives? (check all that apply)
Q. 5 Q. 6
Be Safe program outcomes for children Do you feel that the program is beneficial for boys? Do you feel that the program is beneficial for girls? How beneficial was Be Safe for younger (5, 6, 7 years old) children? How beneficial was Be Safe for older (8, 9 years old) children? The program taught my child who to go to with concerns regarding violence.
Q. Q. Q. Q. Q.
7 8 9a 9b 11c
discussions with their children. Be Safe objectives referenced main themes on how to be safe, while Be Safe elements referred to tools used in the classroom to teach material to the children. 2.5. Data analysis The analytical process began with exporting survey data from Survey Monkey into Excel 2007 for the preliminary data cleaning and analysis. Then, the data was uploaded into IBM SPSS Statistics 22.0 (IBM Corp., Armonk, NY, USA) for further testing and analysis. All statistical analyses were performed using this software. Parent questionnaire responses were grouped according to the following variables: 1) variables describing demographic information and respondent characteristics (e.g. age, gender); 2) variables describing program outcome (e.g. child safety, school policies, safety in community); and 3) program dosage variables. Several quantitative analysis tests were conducted. Descriptive statistics including means and percentages were calculated for demographic information, respondent characteristics, and program outcome variables. Cross tabulations between these variables were also generated. Spearman Rank-Order Correlation was used to measure the strength of association between two variables. For example, it was used to determine whether an increase in program exposure (e.g. parent’s recall of Be Safe objectives) was associated with program outcome (e.g. the program increased the student’s sense of safety at my child’s school). 3. Results A total of 835 parents were surveyed across seven districts (Hambantota, Nuwara Eliya, Badulla, Kegalle, Trincomale, Anuradhapura, Matale) in Sri Lanka. Respondents mainly included the child’s mother (65.6%) or father (17.4%). In some cases where the parents were not available, the grandmother was surveyed (6.2%). Around 94% of household heads were male and 6% were female. The average age for the head of the household was 43 years old, and the average highest level of education for the head of the household was secondary education (grade 10). Regarding ethnicity, 67.2% of respondents were Sinhala, 30.4% were Tamil, and 2.3% were Muslim. On average, families had three children. Over half of the children were male (53.7%). Almost all parents (95%) were involved in the Be Safe program while 5% were involved in both Be Safe and Ten Steps. Very few parents (1.7%) participated in the program in 2014, while most parents (30.4%) participated in the program in 2013. Most participants participated in the program in 2013 (30.4%) and 2012 (23%), or earlier (31.5%). 3.1. Program exposure As a measure of program exposure, the survey asked parents to provide information regarding Be Safe objectives, Be Safe elements, and number of years that their children participated in the program. Of the 12 Be Safe objectives, 45% of parents did not recall any Be Safe objectives discussed with their children. Over half of parents (55%) reported discussing at least one Be Safe objective with their child (Table 2), and among these parents, the objectives discussed included good and bad touches (19.8%), your body belongs to you (19.1%), Bodies are private (11.3%), and it is unsafe for someone to hit you (10%). Of the seven Be Safe elements, 37.3% of parents did not recall any Be Safe elements. Around 63% of parents recalled at least one Be Safe element, and among these parents, most remembered their children talking about the classroom poster (28.9%), followed by Asha the Elephant (23.7%), and the lesson cards (19.9%). Very few parents remembered seeing the take-home bookmark (2.3%) or the parent leaflet (3.3%). Most children and parents (82.4%) participated in the program for one year. 132
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
Table 2 Exposure of children and parents to the Be Safe program. Exposure Variable
Number (%)
Number of years the child has participated in the Be Safe programa 1 280 (82.4) 2 47 (13.8) 3 13 (3.8) 4 0 5 0 Number of Be Safe objectives recalled by parents 0 1 2 3 4 5 6 7 8 9 10 11 12
377 (44.9) 35 (4.2) 51 (6.1) 90 (10.7) 93 (11.1) 48 (5.7) 38 (4.5) 24 (2.9) 21 (2.5) 6 (0.7) 8 (1) 8 (1) 41 (4.9)
Number of Be Safe elements recalled by parents 0 1 2 3 4 5 6 7
313 (37.3) 125 (14.9) 108 (12.9) 148 (17.6) 79 (9.4) 36 (4.3) 13 (1.5) 18 (2.1)
a
Question regarding the first child (if families had more than one children).
3.2. Program effectiveness Descriptive statistics show that most parents believe that the program was beneficial for both genders, with more parents agreeing that the program was beneficial for girls (99.9%) than for boys (82.7%) (p < 0.05) (Table not shown). Spearman’s Rank Correlation was used to assess for the presence of a dose-response relationship between program exposure and perceived benefits of the program for girls and boys. A moderate correlation was found between the dosage variable “Be Safe objectives” and outcome variable “is the program beneficial for boys?” (r = 0.398) (Table 3). A small but significant correlation was found between the dosage variable “Be Safe objectives” and the outcome variable “is the program beneficial for girls?” (r = 0.117). The direction of the correlations suggests that parents with higher exposure to the program were more likely to agree that the program was beneficial for boys and girls. Most parents (91%) agreed that the program increased students' sense of safety at their child’s school children (Fig. 1). A significant correlation was found between the dosage variable “Be Safe elements” and the outcome variable “the program increased students' sense of safety at my child’s school” (r = 0.311). A significant correlation was also found between the “Be Safe objectives” and “Be Safe taught my child who to go to regarding violence and abuse” (r = 0.282). There was a small, non-significant correlation found between the dosage variable “Be Safe objectives” and the outcome variable “It is okay for teachers to use physical punishment against children to maintain discipline”. Several survey questions aimed to determine the effectiveness of school policies in creating safe environments for protecting children from violence. Three in four parents reported that they are aware of rules prohibiting violence among students at their child’s school, and there was a small significant correlation found between this outcome and the dosage variable “Be Safe objectives” (r = 0.136). However, when asked if the program influenced how informed parents and students are about the rules for prohibiting physical punishment, the extent of agreement was lower (70.4%) and there was no dose-response correlation found. Going further, there was no dose-response correlations found between dosage variables and parents' attitude towards physical punishment against children, students' awareness of rules for prohibiting physical punishment, and schools' efforts for ensuring there is someone students can talk to regarding rules of physical punishment. 3.3. Impact of year of program delivery and age of child The time of program delivery ranged from 2008 to 2014. There was no significant correlation found between the year of program delivery and measures of program outcomes, with one exception. In one instance, a small correlation was found between the year of program delivery and the outcome variable “the program increased the students' sense of safety at my child’s school” (r = − 0.119, 133
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
Table 3 Correlations between Be Safe program exposure and program outcomes. Effectiveness Questions
School safety for children The program increased the student’s sense of safety at my child's school It is okay for teachers to use physical punishment against children to maintain discipline I am aware of rules prohibiting violence among students at my child’s school The students are informed about the rules for prohibiting physical punishment The program influenced how informed parents and students are about the rules for prohibiting physical punishment Schools that participated in the program make sure that there is someone in the school to talk to if the rules about physical punishment are violated The program taught my child who to go to with concerns regarding violence Program benefits for different groups of children Do you feel that the program is beneficial for boys? Do you feel that the program is beneficial for girls? How beneficial was Be Safe for younger (5, 6, 7 years old) children? How beneficial was Be Safe for older (8, 9 years old) children? a
a
Dosage Variable Applied
Pearson Rank Correlation
Interpretation
Elements
(r = 0.311, p = 0.001, n = 622) N/A
Statistically significant, moderate correlation Not statistically significant
Not applicable
(r = 0.136, p = 0.001, n = 764) N/A
Statistically significant, low correlation Not statistically significant
Not applicable
N/A
Not statistically significant
Not applicable
N/A
Not statistically significant
Objectives
(r = 0.282, p = 0.001, n = 527)
Statistically significant, low correlation
Objectives
(r = 0.398, p = 0.001, n = 782) (r = 0.117, p = 0.003, n = 662) N/A (r = 0.133, p = 0.003, n = 484)
Statistically significant, moderate correlation Statistically significant, low correlation Not statistically significant Statistically significant, low correlation
Objectives Objectives
Objectives a
Objectives Objectives
Question regarding the first child (if families had more than one children).
Fig. 1. Parents' attitudes towards Be Safe program’s influence on their child’s safety at school.
p = 0.049, n = 272). The direction of the correlation suggests that parents of children who participated in Be Safe in later years were less likely to agree that the program increased their child’s sense of safety in the school. Apart from this correlation, the general trend suggests that parents' perceived outcomes of the program are consistent regardless of what year their child participated in the program. Almost all participants (99%) reported that the program was beneficial to both younger and older children. Regarding program benefits for younger and older children, there was no significant correlation found between dosage variables and perceived program benefits for younger children (r = 0.51, p = 0.259, n = 498). However, there was a small but significant correlation found between dosage variable “Be Safe objectives” and program benefit for older children (r = 0.133, p < 0.05, n = 484). This finding suggests that higher exposure to the program is correlated with an increase in parents' perception of program benefit for older children.
134
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
4. Discussion The purpose of this study was to determine the effectiveness of the Be Safe program. We explored the extent of program outcomes achieved, along with dose-response relationships between program exposures and program outcomes. Results show that most parents agree that the program increased their child’s sense of safety at the school, and agree that their child has someone to talk to regarding rules about physical punishment. Going further, there appears to be plausible correlations between increase in program exposure and increase in parents' perceived child safety in the school, school policies targeting violence prevention, and perceived child safety in the community. While correlations were found between program outcomes and program exposure, most correlations were low. In some cases, correlations were not statistically significant. For example, we did not find a statistically significant correlation between program exposure variables and perceived benefits of the program for younger children. Furthermore, no significant correlations were found regarding rules about physical punishment. There may be several reasons for the low correlations observed, including the lack of appropriately defined exposure variables. For instance, in many cases, correlations were found when applying the dosage variable “Be Safe objectives”, and rarely found using the other dosage variables (e.g. Be Safe elements, number of years in the Be Safe program). However, similarly defined dosage variables have been applied in other program intervention studies where strong correlations have been observed (Deering et al., 2011; Patel, Corter, Pelletier, & Bertrand, 2016). The low correlation could be due to the presence of confounding variables. Less than half of respondents (45%) did not remember their children discussing any of the Be Safe objectives, and 37.3% did not recall their children discussing any Be Safe elements. Most parents and students (82.4%) were involved in the program for one year. The low exposure could be due to several limitations of this study, including the reliance on parent’s self-reported knowledge as opposed to children’s self-reported knowledge, or the lack of parental involvement in the programming. The timing of the study is also a major factor in low exposure rates given that the program ended nearly two years ago for most parents at the time of the evaluation. Future evaluative research is needed to determine the reason for lack of intervention coverage. Although most parents perceived that the program was beneficial for both boys and girls, significantly more parents perceived that the program was beneficial for girls than boys (p < 0.05). However, results suggest that as exposure to the program increases, parents' perceived benefits of the program to both boys and girls increases. Other domestic abuse prevention programs have also assumed that the intervention would have more of an impact on girls than boys (Fox, Corr, Gadd, & Sim, 2016). This likely stems from the developing country context where prevalence of gender-based violence is high, especially against girls and women (Jayasuriya, Wijewardena, & Axemo, 2011; Martin et al., 2002). To allow for gender equality, future programming should emphasize the importance of violence prevention programs for both girls and boys, while engaging parents in the discussion. Despite low exposure to the program interventions and low correlations found, there is some evidence to suggest that increasing exposure to the program contributes to violence against children prevention outcomes. For example, parents that recalled more Be Safe objectives were more likely to perceive that the program benefited their child, and that their child knows who to go to regarding reporting abuse. Going further, although students participated in the program across a seven-year period (2008–2014), no correlations were found between the year of participation in the program and program outcome variables. This finding suggested that children who participated in early years had similar outcomes to children who participated in later years. Dose-response provides evidence of program effectiveness that goes beyond reporting of frequencies and counts of achieved outcomes. Continued prevention and program evaluation efforts are important for demonstrating evidence-based programming and in reaching long-term outcomes. There are many factors that contribute to violence against children. For example, studies have shown that the experience of mass trauma in the family (e.g. war, natural disasters) may contribute to violence against children in a post-conflict setting (Catani et al., 2008; Sriskandarajah, Neuner, & Catani, 2015). Going further, studies have shown a correlation between exposure to war and domestic violence (Catani et al., 2008; Yount, DiGirolamo, & Ramakrishnan, 2011). The impacts of violence against children, combined with traumatic experiences, points toward an urgent need for prevention and intervention. Treatment approaches should focus not only on the child level, but also on the family and community level (Catani et al., 2008). The high level of family violence indicate a need for interventions to promote non-violent parenting and to protect children living in post-war settings (Catani et al., 2008; Sriskandarajah et al., 2015; Tolan, Gorman-Smith, & Henry, 2006). Although parents were secondary targets of the Be Safe program, more efforts are needed to engage parents in programming activities to prevent children against violence. This need was also recently supported by the Global Partnership to End Violence Against Children in 2016 (End Violence Against Children: The Global Partnership, 2016). There are several important limitations of this study. First, this study was based on a cross-sectional survey that relied on the perspective of parents of children who participated in the Be Safe program (i.e. indirect program participants). Furthermore, it has been more than two years since the program has ended for most parents and children. Yet, this method may be the only method by which data for programs of this nature can be obtained, and significant correlations were observed. The extent of outcomes achieved and the strength of correlations may be higher if we examined self-reported data from children. However, the data collection on these outcomes would be challenging. Finally, the lack of a baseline or control group also makes it difficult to interpret the importance of the correlations. However, despite these limitations and challenges, we found correlations that suggest that the program is effective in preventing violence against children. 5. Conclusion This study evaluated the effectiveness of a violence against children prevention program using a dose-response approach. Study 135
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
findings suggested that an increase in exposure to the program was associated with improvements in parents' perceived child safety in the school, school policies surrounding violence prevention, and child safety in the community. However, the exposure to the program was relatively low, and low correlations were found. Researchers and practitioners may consider the usefulness of applying dose-response measures in program monitoring and evaluation efforts. While improvements were achieved across various violence against children prevention outcomes, there remains a large unfinished agenda for developing, applying, and evaluating efforts to reduce violence against children. Future research and evaluations need to consider the perspectives of children. Furthermore, experimental designs and pre-program measures should be considered early in the planning process to ensure that stronger evidence of program effectiveness can be collected. Conflict of interests None. Author contributions SL contributed to drafting the manuscript. HC collected the data and supported data analysis. CZ supported data analysis and report writing. DL supported data analysis. IC supported manuscript writing. All authors read and approved the final manuscript. Acknowledgements Thank you to the survey teams for supporting data collection, and thank you to Henry de Mel for supporting the survey teams in Sri Lanka. We thank all the study participants for sharing their time and information with us. Thank you to Rob Maclean, Kendra Siekmans, and David Kupp for providing helpful feedback on earlier drafts. Thank you to the two anonymous reviewers for their insightful comments. Finally, we would like to thank the Canadian Red Cross for funding this study. The funder had no role in the design and conduct of the study. References Arachchige, M., Shiroma, D., & Shirajanie, J. (2013). Violence against children and ensuring the justice for child victims through human rights and penal laws perspectives in Sri Lankan context? International Journal of Research In Social Sciences, 4(3), 111–120. Canadian Red Cross (2016). Program for young children: Be Safe! can help prevent sexual abuse of children. Retrieved March 21, 2016, from: http://www.redcross.ca/ how-we-help/violence–bullying-and-abuse-prevention/educators/child-abuse-and-neglect-prevention/program-for-young-children–be-safe–can-help-preventsexual-abuse-of-children. Catani, C., Schauer, E., & Neuner, F. (2008). Beyond individual war trauma: domestic violence against children in Afghanistan and Sri Lanka. Journal of Marital and Family Therapy, 34(2), 165–176. http://dx.doi.org/10.1111/j. 1752-0606.2008.00062.x. Catani, C., Gewirtz, A. H., Wieling, E., Schauer, E., Elbert, T., & Neuner, F. (2010). Tsunami, war, and cumulative risk in the lives of Sri Lankan schoolchildren. Child Development, 81(4), 1176–1191. http://dx.doi.org/10.1111/j. 1467-8624.2010.01461.x. Child Rights International Network (2017). Article 19: Protection from abuse and neglect. Retrieved April 14, 2017, from: https://www.crin.org/en/home/rights/ convention/articles/article-19-protection-abuse-and-neglect. de Zoysa, P. (2002). Child sexual abuse in Sri Lanka: The current state of affairs and recommendations for the future. Journal of Child Sexual Abuse, 11(2), 97–113. http://dx.doi.org/10.1300/J070v11n02_06. Deering, K. N., Boily, M.-C., Lowndes, C. M., Shoveller, J., Tyndall, M. W., Vickerman, P., & Alary, M. (2011). A dose-response relationship between exposure to a large-scale HIV preventive intervention and consistent condom use with different sexual partners of female sex workers in southern India. BMC Public Health, 11(Suppl. 6), S8. http://dx.doi.org/10.1186/1471-2458-11-S6-S8. End Violence Against Children: The Global Partnership (2016). A world in which every children grows up free from violence. Retrieved March 21, 2016, from: http:// www.end-violence.org/. Fang, X., Brown, D. S., Florence, C. S., & Mercy, J. A. (2012). The economic burden of child maltreatment in the United States and implications for prevention. Child Abuse & Neglect, 36(2), 156–165. http://dx.doi.org/10.1016/j.chiabu.2011.10.006. Fang, X., Fry, D. A., Brown, D. S., Mercy, J. A., Dunne, M. P., Butchart, A. R., & Swales, D. M. (2015). The burden of child maltreatment in the East Asia and Pacific region. Child Abuse & Neglect, 42(43123767), 146–162. http://dx.doi.org/10.1016/j.chiabu.2015.02.012. Fortson, B., Klevens, J., Merrick, M., Gilbert, L., & Alexander, S. (2016). Preventing child abuse and neglect: A technical package for policy, norm, and programmatic activities. national center for injury prevention and control. Atlanta, Georgia: Centers for Disease Control and Prevention. Retrieved from: http://www.cdc.gov/ violenceprevention/pdf/can-prevention-technical-package.pdf. Fox, C. L., Corr, M. L., Gadd, D., & Sim, J. (2016). Evaluating the effectiveness of domestic abuse prevention education: Are certain children more or less receptive to the messages conveyed? Legal and Criminological Psychology, 21(1), 212–227. http://dx.doi.org/10.1111/lcrp.12046. Grieco, L. A., Jowers, E. M., Errisuriz, V. L., & Bartholomew, J. B. (2016). Physically active vs. sedentary academic lessons: A dose response study for elementary student time on task. Preventive Medicine, 89, 98–103. http://dx.doi.org/10.1016/j.ypmed.2016.05.021. Hundeide, K., & Armstrong, N. (2011). ICDP approach to awareness-raising about children’s rights and preventing violence, child abuse, and neglect. Child Abuse Neglected, 35(12), 1053–1062. http://dx.doi.org/10.1016/j.chiabu.2011.09.008. Hyder, T., & Mac Veigh, J. (2007). Gender-based violence against children in emergencies: Save the Children UK’s response. Gender & Development, 15(1), 81–93. http://dx.doi.org/10.1080/13552070601179110. Jørgensen, T. S., Rasmussen, M., Aarestrup, A. K., Ersbøll, A. K., Jørgensen, S. E., Goodman, E., & Krølner, R. (2015). The role of curriculum dose for the promotion of fruit and vegetable intake among adolescents: Results from the Boost intervention. BMC Public Health, 15, 536. http://dx.doi.org/10.1186/s12889-015-1840-0. Jayasuriya, V., Wijewardena, K., & Axemo, P. (2011). Intimate partner violence against women in the capital province of sri lanka: Prevalence, risk factors, and help seeking. Violence Against Women, 17(8), 1086–1102. http://dx.doi.org/10.1177/1077801211417151. Kim, S. S., Rawat, R., Mwangi, E. M., Tesfaye, R., Abebe, Y., Baker, J., & Menon, P. (2016). Exposure to large-scale social and behavior change communication interventions is associated with improvements in infant and young child feeding practices in Ethiopia. PLoS One, 11(10), http://dx.doi.org/10.1371/journal.pone. 0164800. Korbin, J. E. (2003). Children, childhoods, and violence. Annual Review of Anthropology, 32(1), 431–446. http://dx.doi.org/10.1146/annurev.anthro.32.061002. 093345. Martin, S. L., Moracco, K. E., Garro, J., Tsui, A. O., Kupper, L. L., Chase, J. L., & Campbell, J. C. (2002). Domestic violence across generations: Findings from northern
136
Child Abuse & Neglect 76 (2018) 129–137
S. Lam et al.
India. International Journal of Epidemiology, 31(3), 560–572. http://dx.doi.org/10.1093/ije/31.3.560. Patel, S., Corter, C., Pelletier, J., & Bertrand, J. (2016). Dose-response relations between participation in integrated early childhood services and children’s early development. Early Childhood Research Quarterly, 35, 49–62. http://dx.doi.org/10.1016/j.ecresq.2015.12.006. Sriskandarajah, V., Neuner, F., & Catani, C. (2015). Predictors of violence against children in Tamil families in northern Sri Lanka. Social Science and Medicine, 146, 257–265. http://dx.doi.org/10.1016/j.socscimed.2015.10.010. Stark, L., & Landis, D. (2016). Violence against children in humanitarian settings: A literature review of population-based approaches. Social Science and Medicine. http://dx.doi.org/10.1016/j.socscimed.2016.01.052. The Sphere Project (2011). Humanitarian charter and minimum standards in humanitarian response. response, Vol. 1http://dx.doi.org/10.3362/9781908176202. Tolan, P., Gorman-Smith, D., & Henry, D. (2006). Family violence. Annual Review of Psychology, 57, 557–583. http://dx.doi.org/10.1146/annurev.psych.57.102904. 190110. United Nations Economic, & Social Council (2016). Progress towards the sustainable development goals. Retrieved March 22, 2017, from: http://www.un.org/ga/search/ view_doc.asp?symbol=E/2016/75&Lang=E. United Nations Secretary-General’s Study (2006). World report on violence against children. Retrieved March 21, 2016, from: https://www.unicef.org/lac/full_tex(3). pdf. United Nations (2011). Report of the secretary-General’s panel of experts on accountability in Sri Lanka. Retrieved March 21, 2016, from: http://www.un.org/News/dh/ infocus/Sri_Lanka/POE_Report_Full.pdf. World Health Organization (2016). Inspire: Seven strategies for ending violence against children. Geneva, Switzerland: World Health Organization. Yount, K. M., DiGirolamo, A. M., & Ramakrishnan, U. (2011). Impacts of domestic violence on child growth and nutrition: A conceptual review of the pathways of influence. Social Science and Medicine, 72(9), 1534–1554. http://dx.doi.org/10.1016/j.socscimed.2011.02.042.
137