Introduction

The physical, social and emotional impacts of school bullying on all of those involved are well known and can be severe and long lasting (Carney 2008; Johnson 2009; Ttofi et al. 2012). While the prevalence of bullying varies internationally, a review by Juvonen and Graham (2014) suggests that 20–25% of young people are actively engaged in bullying others, are themselves victimized, or both. Other estimates are higher, with a recent meta-analysis of 80 studies measuring the bullying experiences of young people aged 12–18 years finding a mean prevalence rate of 36% for bullying involvement (Modecki et al. 2014). In Australia, approximately 10% of school students aged 8–14 years reported being bullied most days or more often, with 27% reporting being victimized frequently (i.e. every few weeks or more often) in the previous term at school; 9% reported bullying others frequently and 4% reported being frequent bully-victims in the previous term (Cross et al. 2009).

The prevalence and impact of bullying has given rise to interest in the effectiveness of school bullying prevention and intervention programs (Barbero et al. 2012; Ttofi and Farrington 2011). The application of social-ecological theory to the conceptualisation of bullying and victimization shows both bullying perpetration and victimization are reciprocally influenced by the individual, family, school, peer group, community and society (Bronfenbrenner 1995). It is, therefore, often recommended that families be included in efforts to combat bullying behaviour (Baldry 2003; Felix and Furlong 2008; Olweus and Limber 2009; Smith et al. 2008), with reviews of school bullying prevention and intervention programs showing the involvement of parents is a critical factor in the success of such programs and one of the most important elements related to a reduction in bullying behaviours (Barbero et al. 2012; Ttofi and Farrington 2009, 2011).

Family dynamics and relationships play a key role in children’s behaviour towards their peers (Cross and Barnes 2014). Studies have shown parenting styles, disciplinary approaches, parent–child communication, closeness of relationships, parental supervision, problem solving abilities, abuse and neglect are risk and protective factors related to child and adolescent involvement in bullying behaviours (Bowes et al. 2010; Georgiou 2008; Georgiou and Stavrinides 2013; Lereya et al. 2013; Roberts and Morotti 2000; Snyder et al. 2005; Spriggs et al. 2007; Stephens 2002). A meta-analysis of parenting factors and bullying found warm and affectionate relationships between parents and children, high parental involvement and support, and effective family communication and supervision were protective against victimization while maladaptive parenting, abuse and neglect were predictive of victimization (Lereya et al. 2013). Earlier studies found victimization to also be associated with over-protective, punitive and authoritarian parenting styles (Nickerson et al. 2010; Schwartz et al. 2000; Shields and Cicchetti 2001). Perpetration is associated with uninvolved, unempathic and hostile parents and power-assertive discipline styles (Bernstein and Watson 1997; Georgiou and Stavrinides 2008).

Family support and a positive home environment predicts positive adjustment, resiliency, fewer emotional and behavioural problems and fewer depressive symptoms in children who have been bullied (Bowes et al. 2010; Conners‐Burrow et al. 2009). Furthermore, a parent’s normative beliefs and their own personal experiences of bullying are predictive of strategies they employ to help their children attempt to resolve bullying situations (Cooper and Nickerson 2013; Kochenderfer-Ladd and Pelletier 2008). Parents may normalize types of bullying as less serious than others (Mishna 2008; Mishna et al. 2006; Sawyer et al. 2011), or put the responsibility for fixing the situation on the victim (Purcell 2012). Parents who historically experienced or witnessed bullying when in school are more likely to respond to bullying by discussing the situation with their child and offering suggestions for coping compared with parents who had no involvement in bullying as a child (Cooper and Nickerson 2013). Direct and indirect strategies and coping frameworks used by parents in bullying situations are aimed to comfort the child and to take action to address and resolve the situation (Harcourt et al. 2014). Direct strategies used by parents include involving the school, approaching the bully’s parents or transferring the child to another school (Brown 2010; Cassidy et al. 2012). Indirect strategies focusing on improving a child’s ability to deal with bullying on their own, include providing emotional support and promoting pro-social behaviour (Cassidy et al. 2012; Mishna 2008; Sawyer et al. 2011). In a recent study that examined how parents help their children cope with bullying situations, most parents self-reported they had discussed bullying with their child and had offered suggestions on coping strategies (Cooper and Nickerson 2013). The majority of parents told their child to get help from adults (e.g., family, parents and teachers) or to avoid the situation, though 42.3% of parents advised their child to fight back at least some of the time (Cooper and Nickerson 2013). This same study found the majority of parents would not advise their children to handle the situation alone, or tell their child to make fun of the bullying situation (Cooper and Nickerson 2013).

Existing research offers a limited understanding on the differences in mothers and fathers when interacting with their children with respect to bullying situations. Mothers and fathers typically use different parenting styles for their sons and daughters (McKinney and Renk 2008) with mothers perceived to be more open to direct communication than fathers (Cabrera et al. 2007; Miller-Day 2002; Tamis-LeMonda et al. 2004). Mothers and fathers may also differ in perceptions of a child’s social competence and in parenting qualities, which predict and model pro-social problem solving (Markulin 2009; Renk and Phares 2007; Taratuski 2010). Research has shown that fathers tend to focus on the negative behaviour rather than providing guidance (Taratuski 2010), their constructive problem solving is associated with social withdrawal (Miller et al. 2011), and fathers’ psychological control is associated with greater peer exclusion for shy males (Miller et al. 2011). Differences also exist in parental advice given to children on coping strategies with respect to bullying. Mothers tend to give advice, contact the school, seek professional advice, involve the child in self-esteem activities or self-defence classes, whereas fathers tend to normalize bullying or go straight to the authorities (Ford 2013).

The three-year Friendly Schools Friendly Families (FSFF) whole-school bullying prevention intervention was built on the previously trialled Friendly Schools Intervention (Cross et al. 2011), comprising training and whole-school, classroom and family resources to build the capacity of school teams to engage cohorts of Grades 2, 4 and 6 students and their families in awareness-raising and skill-building activities. The intervention resulted in significant reductions in bullying victimization and perpetration, and improvements in help-seeking (Cross et al. 2012). Results also suggested a whole-school capacity-building intervention in early and middle childhood can improve the likelihood and frequency of positive parent–child communication about bullying (Cross et al. 2016).

To date, many family responses encouraged by school bullying prevention and intervention programs do not distinguish between the impact of involving male and/or female parents, stepparents or carers (henceforth ‘parents’). This paper describes and evaluates the family component of the FSFF intervention to determine the extent to which this universal family intervention encouraged more than one parent in each student’s family to have more frequent and proactive conversations with their children about their social skills, ways to prevent bullying, resilient ways to cope if they were bullied, and how to help others who may be being bullied. It is hypothesized in this paper that over the 3 years of the FSFF study, both male and female parents who received the high dose FSFF intervention will be more likely to engage in proactive conversations with their children about ways to prevent and reduce harm from bullying, compared to male and female parents from the low dose group. It is also anticipated that female parents/carers will report more discussion with their children about bullying, and be more likely to recommend pro-social and help-seeking responses to bullying, than male parents/carers.

Method

Participants

This prospective group-randomized controlled trial, conducted in Perth, Western Australia, followed three different age cohorts of students recruited from 20 randomly selected Government primary schools (Cross et al. 2012). At baseline, three age cohorts of students, their teachers and parents were eligible to participate in the study; the students were aged 6–7 years (Grade 2 cohort), 8–9 years (Grade 4 cohort) and 10–11 years (Grade 6 cohort). The parents of each of these student cohorts are the focus of this paper.

Prior to randomization, schools were stratified by two potential confounders; school size (using a median split with larger schools >451 and smaller schools ≤451 total students) and socio-economic status (SES) (low, moderate and high SES groups using terciles from the Australian Bureau of Statistics Socio-economic Index for Areas: Scale of Disadvantage) (McLennan 1998). All 20 participating schools agreed to random assignment to one of the three experimental conditions: high dose (n = 7), moderate dose (n = 6) and low dose (n = 7).

The FSFF study was described to families using informed passive consent procedures, and parents were free to withdraw themselves or their child from the study at any time. The research protocol and school, student and parent consent procedures were approved by the Curtin University Human Research Ethics Committee and the Western Australian Department of Education.

The student, family and staff cohorts were exposed to high, moderate or low dose versions of a whole-school intervention and tracked for 3 years, until the three cohorts of students were in Grades 4, 6 and 8 (aged approximately 9, 11 and 13) respectively (Table 1). Both female and male parents of students were approached to complete a survey. This paper describes the data collected from female and male parents at baseline, and the 10 and 22 month post-tests from the parents of the Grades 2, 4 and 6 student cohort involved in the high, moderate and low dose groups.

Table 1 Parent participants by grade and dose
Full size table

Procedure

Given students’ families are central to understanding students’ bullying behaviour patterns, particularly the influence of family attitudes, expectations and perspectives (Hammer 1998), the FSFF project focussed on what attitudes and behaviour patterns related to bullying behaviour may emerge from the family and how these could be enhanced or modified, such as problematic responses to bullying as a bystander, target or perpetrator. Students were considered within a socio-ecological context, particularly in relation their family and the behaviour patterns they have learned at home and the influence of these on their bullying behaviour, as well as the influence of their peers, their school and community.

As mentioned previously, the FSFF intervention comprised three conditions: high, moderate and low. The high dose group receive training and resources for school teams to deliver, and whole-school, classroom, family and individual level strategies for each of 3 years of the FSFF project. The moderate and low dose groups did not receive any of the specific family education training or resources. As described earlier, the moderate dose schools received all other whole-school and classroom based training and resources and the low or comparison dose groups received the usual whole-school and classroom resources only, with no training (Cross et al. 2016, 2011, 2012).

The family education strategies delivered by schools systematically targeted parenting factors identified as protective against bullying behaviour. These factors included parent–child communication; parent modelling; parent bullying attitudes and beliefs; normative standards about bullying; family management techniques and parenting style; connectedness and cohesion. These resources used directed and self-help approaches to improve parents’ self-efficacy related to providing emotional support, attention, warmth and quality supervision, while also reinforcing the need for families to actively model and discourage aggressive behaviour. Parents were also given restorative strategies to use with their own children if they were engaging in sibling bullying. The educational materials focussed on helping parents to understand that bullying is not a normal part of growing up or ‘character building’, and that it can cause many harms for both perpetrators and targets. The resources also provided support and advice for parents if their child experienced bullying.

The family level strategies therefore worked in partnership with parents by building their awareness, attitudes and self-efficacy to role model, talk with and help their children to develop social competence and to prevent or respond to bullying. These strategies also aimed to encourage school and family communication and parents’ engagement with the school to reduce student bullying. At the start of each year of the study, school teams in the high dose condition received a 3-hour family training session and resources to establish key understandings and support the implementation of the family level strategies. The resources included six communication sheets containing self-help information and activities designed to stimulate parents’ discussions with their child about bullying-related issues. The activities were linked to the classroom curriculum and therefore, were implemented at home by the students with both parents. Students were encouraged to complete about 2 h of bullying prevention activities at home with their families in each of the 3 years of the project.

Key FSFF messages were also delivered to families in the high dose school in a variety of ways to maximize their exposure to the information. These included conducting family events such as a weekend family fun day at the school; school newsletter items; a parent information booklet; engaging with each school’s Parents’ and Friends’ committees; involving parents in developing and disseminating the school’s bullying policy; via school assembly items and motivating parents to be involved through family communication sheets, where students completed tasks at home with their family (e.g., family interviews); and lastly via student performance.

Measures

Discussions with child regarding bullying

How often a parent had discussed with their child what to do if they were being bullied was measured on a 5-point scale (1 = did not speak, 2 = spoke once, 3 = spoke 2–3 times, 4 = spoke more than four times, 5 = can’t remember how often discussed).

Advice given to child

Parents also recorded the advice they had given to their child with respect to being bullied: no advice—not needed; no advice—did not know what to say; ignore bullied; walk away; fight back; just accept it; tell students who bully to stop; make a joke of it; ask for help from other students; tell a teacher; and/or tell someone at home.

Perceived influence on child’s response to being bullied

The amount of perceived influence of a parent on a child’s response to being bullied was measured on a four point scale (1 = no influence, 2 = not sure of the amount of influence, 3 = some influence, 4 = a lot of influence).

Perceived influence on child’s decision whether to bully

The amount of perceived influence of a parent on a child’s decision to bully others was measured on a four point scale (1 = no influence, 2 = not sure of the amount of influence, 3 = some influence, 4 = a lot of influence).

Data Analyses

SPSS version 22 and Stata version 13 were used to analyse the data. Chi-square analysis were used to determine differences between mothers’ and fathers’ frequency of discussion with their child regarding bullying, and their level of perceived influence on a child’s response to being bullied and of their child’s decision to bully others, and advice given to their child for responding to bullying. Separate repeated measures mixed models were used to determine differences over time with respect to dose groups for mothers’ and fathers’ frequency of discussion with their child regarding bullying, perceived level of influence on a child’s response to being bullied and of their child’s decision to bully others by grade and dose. A random intercept was included in each regression model to account for the clustering of parents within schools.

Results

The following results are based on a total of 1429 mothers/stepmothers and fathers/stepfathers in the high and low dose groups where both male and female parents of a child completed a baseline survey (Table 1). A total of 499 Grade 2 parents, 472 Grade 4 parents and 458 Grade 6 parents completed the survey at baseline. Half of the parents were reporting on male children (50%). By Post 2, 60% of Grade 2 parents (n = 298), 65% of Grade 4 parents (n = 306) and 52% of Grade 6 (n = 238) parents completed surveys. At Post 2, 63% of parents in the low dose group completed surveys compared to 55% of parents in the high dose group.

Frequency of Discussion with School Regarding Child Being Bullied

At baseline, a significantly greater proportion of mothers than fathers had discussed bullying with their child at least once in Grade 2 (ages 7–8), Grade 4 (ages 9–10), and Grade 6 (ages 11–12) in both the high and low dose groups (Table 2). At Post 1, a significantly greater proportion of mothers than fathers in the low dose groups discussed bullying with their child at least once in Grade 2, Grade 4 and Grade 6, whereas in the high dose group, a significantly greater proportion of mothers than fathers of Grade 6 students only discussed bullying with their child at least once. At Post 2, a significantly greater proportion of fathers than mothers in both the high and low dose groups discussed bullying with their child at least once in Grade 2.

Table 2 Comparison of mother and father perceptions for high and low dose groups at baseline, Post 1 and Post 2
Full size table

Frequency of discussion with their child about bullying significantly increased over time for mothers (Post 1: β = 0.31, p < 0.001; Post 2: β = 0.12, p = 0.027) and fathers (Post1: β = 0.24, p < 0.001; Post 2: β = 0.14, p = 0.035) in the high dose group compared to parents in the low dose group (Table 3).

Table 3 Multi-level modelling results of mothers’ and fathers’ frequencies of discussion with their child and perceptions of perceived level of influence on child’s response to being bullied and on child’s decision to bully others
Full size table

Perceived Level of Influence on Child’s Response to Being Bullied

Approximately half of mothers and fathers perceived they had a lot of influence on their child’s response to being bullied irrespective of dose group (Table 2). There were no significant differences between mothers and fathers perceptions that they had a lot of influence on their child’s response to being bullied at baseline. At Post 1, a significantly higher proportion of mothers than fathers in the high dose group perceived they had a lot of influence on their child’s response to being bullied irrespective of Grade, while there were no significant differences in mothers and fathers perceived influence at Post 2. In comparison, at Post 2 a significantly higher proportion of mothers than fathers in the low dose group perceived they had a lot of influence on their child’s response to being bullied irrespective of Grade.

There were no significant differences with respect to mothers’ or fathers’ perceived influence on their child’s response to being bullied over time (Table 3). For mothers, perceived influence on their child’s response to being bullied remained constant over time (p > 0.05). However, for fathers, perceived influence on their child’s response to being bullied was significantly lower at Post 2 than baseline (β = −0.09, p = 0.008). There was no significant interaction between mothers’ or fathers’ perceived influence on their child’s response at Post 2 and dose group.

Perceived Level of Influence on Child’s Decision to Bully Others

The majority of mothers and fathers perceived they had a lot of influence on whether their child decided to bully others. There were no significant differences between mothers and fathers perceived level of influence on their child bullying others at any grade level at baseline, Post 1 or Post 2 (all p > 0.05).

Fathers perceived level of influence significantly decreased over time. Fathers within the high dose group had a significantly increased perception of the influence they have over their child’s decision to bully others at Post 1 (β = 0.10, p = 0.010) and Post 2 (β = 0.09, p = 0.008) compared to the low dose group (Table 3). Conversely, mothers in the high dose group (β = −0.09, p = 0.008) had significantly decreased perception of the influence they have over their child’s decision to bully others compared to mothers in the low dose group at Post 2.

Advice Given by Parent to Child if He or She was Bullied

The majority of mothers and fathers at baseline gave help-seeking (e.g., ask for help from another student, tell a teacher, tell someone at home), pro-social (e.g., tell bullies to stop, make a joke of it), and passive (e.g., just accept it, ignore bullies, walk away) advice to their child if they were bullied, irrespective of dose group (Table 4). A small percentage of mothers and fathers gave aggressive advice such as ‘fight back’. Advice did not significantly change over time for mothers or fathers (all p > 0.05).

Table 4 Comparison of mother and father advice given to their child for responding to bullying for high and low dose groups at baseline, Post 1 and Post 2 by grade
Full size table

In the high dose group at baseline, a significantly greater proportion of mothers than fathers gave pro-social, passive and help-seeking advice to their child who was being bullied, whereas a significantly greater proportion of fathers than mothers gave aggressive advice to their child irrespective of grade level (all p < 0.05) (Table 4). Similarly, in the low dose group at baseline, a significantly greater proportion of mothers than fathers gave pro-social and passive advice to their child who was being bullied, whereas a significantly greater proportion of fathers than mothers gave aggressive advice to their child irrespective of grade level (all p < 0.05). In the low dose group, there was no significant differences in the proportions of mothers and fathers giving help-seeking advice in Grades 2 and 6.

In both the high and low dose groups at Post 1, a significantly greater proportion of mothers than fathers gave pro-social advice to their child who was being bullied, whereas a significantly greater proportion of fathers than mothers gave aggressive advice to their child irrespective of grade level. In the low dose group at Post 1, a significantly greater proportion of mothers than fathers gave passive and help-seeking advice to their child who was being bullied.

In the high dose group at Post 2, similar proportions of mothers and fathers gave pro-social advice to their child who was being bullied in Grade 2 and 4, whereas a significantly higher proportion of mothers than fathers gave pro-social advice to their child who was being bullied in Grade 6. In the low dose group, a significantly higher proportion of mothers than fathers gave pro-social advice to their child who was being bullied in Grade 2. At Post 2 a significantly greater proportion of mothers than fathers gave passive advice to their child who was being bullied irrespective of dose group or Grade. In the high dose group at Post 2 a significantly greater proportion of fathers than mothers gave aggressive advice to their child only in Grade 4, whereas in the low dose group a significantly greater proportion of fathers than mothers gave aggressive advice to their child irrespective of grade level. In the high dose group, a significantly higher proportion of mothers than fathers gave help-seeking advice to their child who was being bullied in Grade 6 compared to the low dose group where a significantly higher proportion of mothers than fathers gave help-seeking advice to their child who was being bullied in Grades 4 and 6.

Discussion

The FSFF whole-school approach to bullying prevention actively engaged families in the delivery of social and emotional skill development and bullying prevention-related knowledge and skills, and encouraged them to deliver this content to their children. This paper focussed on the effectiveness of the FSFF family intervention to encourage both parents to provide consistent, positive non-aggressive messages to their children. This study also explored the frequency of mothers’ and fathers’ discussions about bullying with their child, types of coping strategies they encouraged their children to use if they were bullied, and their perceptions about their level of influence on their children’s response to being bullied or their child’s decision to bully others.

Mothers and fathers involved in the high dose group were found to discuss bullying more frequently with their child than mothers in the low dose group, with the frequency of discussion increasing as their children grew older. In the high dose group at baseline, mothers rather than fathers were more likely to frequently discuss bullying with their child. While this pattern continued at Post 1, by Post 2, larger proportions of fathers than mothers in the high dose group were discussing bullying with their child. These results may reflect greater involvement of mothers than fathers in managing children’s schedules and activities when they are younger (Parker 2015). Greater parent–child communication predicts less involvement in problem and socially undesirable behaviours (Crouter and Head 2002; Sullivan et al. 2004). While previous research found the majority of older children and younger adolescents (9–14 years) do not tell their parents they are being victimized, even when parents try to elicit information if they suspect something is wrong, (Fekkes et al. 2005; Hunter et al. 2004; Stavrinides et al. 2015), it is clear this discussion between parents and their children needs to begin before children experience difficulties. This is particularly important given a recent study found that younger adolescents who are victimized are less likely to tell their parents than older adolescents and often hide their experiences (Stavrinides et al. 2015). Stavrinides et al (2015) found parents were more aware of their child bullying others than when they were being victimized. Children’s disclosure of bullying involvement is protective against school victimization, with those who seek help from their parents feeling less stressed, more resilient and more likely to escape the victimization (Georgiou and Stavrinides 2013; Hunter and Borg 2006; Naylor et al. 2001). Hence, the involvement of both mothers and fathers discussing bullying with their child from as young an age as possible is recommended to encourage children’s help-seeking behaviour (Cross et al. 2012).

Parental care and support are the most effective ways by which victims cope with bullying (Naylor et al. 2001; Smith and Myron-Wilson 1998), with the majority of children seeking parental help or help from friends when being victimized (Cross et al. 2009). Coping strategies used as reported by targets of bullying include self-defence, standing up to the person bullying, seeking social support, pretending nothing happened, focusing on the positive, internalising and externalizing behaviours, and self-blame; with boys using more externalizing strategies and girls using more social support-seeking strategies (Tenenbaum et al. 2011). In this study, fathers were found to advise their children to be aggressive if bullied (i.e. fight back) more than mothers, whereas mothers were more likely to report giving pro-social, passive and help-seeking advice. Given differences between mothers’ and fathers’ pro-social and help-seeking advice became less evident over time in the high dose group for Grades 2 and 4, there is evidence to suggest the family intervention which recommended giving pro-social and help-seeking advice was effective. Many bullying prevention and intervention programs recommend targets tell someone, as seeking help from peers and adults is a more effective in reducing the likelihood of continued victimization (Slee and Murray-Harvey 2008). Hence, it is important for parents to talk with their children about the actions their children should take if they see others being bullied.

Father involvement has been shown to protect children from extreme victimization, and is more protective when the mother’s involvement is lower (Flouri and Buchanan 2002, 2003). The FSFF intervention had a significant positive effect on fathers’ perceptions of their influence on their child’s response to being bullied. The results from this study highlight the importance of deliberately involving fathers in interventions to help them proactively parent their children about how to prevent and respond adaptively and resiliently to bullying. This approach requires a family-centred intervention approach that reaches beyond the parent who typically responds to matters at school. In this study teachers actively encouraged students to complete the home activities sometimes with an adult parent who was male and at other times with a female parent. This process enabled the FSFF content to penetrate the families.

Limitations

The strengths of this study include its longitudinal design, random assignment to condition, appropriate comparison conditions, analyses that adjusted for clustering of individuals in schools, and sufficient power to detect moderate effects. These strengths were, however, moderated by several methodological limitations. Parent data were collected using self-report measures, which may not accord with children’s experiences and may be subject to social desirability biases. Future research might elucidate whether children’s perceptions of bullying-related parenting behaviours accord with parent reports of these behaviours. Further, while the analyses controlled for baseline differences, and despite the random sampling of schools parents, the high and moderate dose groups reported higher levels of education at baseline than those in the low dose group. Higher levels of parental education have been associated with lower victimization levels experienced by their children (Jansen et al. 2012). Intervention effects may have been spuriously inflated with educated parents more likely to engage with school initiatives. This study experienced high parent attrition rates over time reaching 59%: 55% for high dose and 63% for low dose. The parents who remained in the study (across dose groups) therefore may have been more engaged than those lost to follow-up. The high dose parents may have been affected by intervention fatigue as the home activities required their active involvement and their participation decreased slightly over time. It will be important for future research to determine the level of dose that is necessary to achieve change and is acceptable to parents. Lastly, the findings were tempered by a low dose ‘contaminated’ comparison group. This low dose was provided to simulate a usual dose bullying program in the comparison schools but may have mitigated the impact of the high and moderate dose intervention on parent behaviour. Flay et al. (2004) recommend that it is necessary to compare multi-component programs with the best school programs of known effectiveness to estimate the contribution of the new intervention over and above the currently-used school program.