Unlocking the Power of Parenting: Unraveling How Family Atmo - sphere and Parenting Styles Impact the Pivotal Role in Bullying Behavior

Bullying at school negatively affects the class environment as well as all young people who participate in the bullying. One of the most important factors related to bullying is family. In the current study, the effect of family atmosphere and parenting style on different roles in bullying (bullies, victims, defenders) was examined in 155 adolescents (aged 13-15). The participants completed three self-reported scales measuring bullying, parenting style, and family atmosphere. For data analysis, Bayesian logistic regression was used. It was found that the family of bullies had less cohesion and control, more conflicts and fathers who demonstrated desultory parenting. Victims had higher family control, less family cohesion, fewer fathers with positive parenting and slightly higher positive parenting of mothers. Defenders had families with higher expressivity and organization, fewer conflicts, more fathers with positive parenting, fewer fathers with directive parenting and fewer mothers with autonomous parenting. Defenders were also predicted by the hostile parenting of fathers. Knowledge about the factors related to family atmosphere and parenting styles can be very useful in preparing bullying prevention and intervention programs.

Bullying is an undesirable social-pathological phenomenon that occurs at almost every school to a certain degree (Aluede et. al., 2008;Frisén et al., 2008;Hong & Espelage, 2012;Hellström et al., 2015;Thornberg, 2015).The prevalence of bullying is relatively high with the National Center for Educational Statistics ( 2016) putting it at 20.8% in terms of the 6 months period before the interview.Modecki et al. (2014) conducted a comprehensive meta-analysis of 80 studies on bullying referring to the recall period immediately preceding data collection, revealing a mean prevalence rate of 35% for traditional bullying involvement.Bullying is a behavior characterized by maliciousness, deliberate and often repeated aggressive conduct (Aluede et al., 2008), which is manifested in the form of physical, verbal or indirect bullying (Olweus, 1995).Bullying is the problem of a group in which individuals play different roles: bully, victim, defender, motivator and assistant as well as outsiders who do not belong to the group (Salmivalli, 1999;Sutton & Smith, 1999).Bullying is a risk factor for serious mental health problems in both victims (Gini & Pozzoli, 2013;Gini & Espelage, 2014;Van Geel et al., 2014;Nielsen et al., 2015;Reed et al., 2015) and aggressors (Ttofi et al., 2012;Copeland et al., 2013), and these problems tend to prevail even into adulthood.Bullying represents a multidimensional phenomenon, arising from the complexity of family and peer relationships, school community and culture (Swearer & Doll, 2001).It can also be considered a manifestation of defense mechanisms, which can be a reaction to existential questions as well as identity confusion (Vindišová, 2010).Several predictors of bullying have been identified (Cook et al., 2010;Atik & Güneri, 2013;Kljakovic & Hunt, 2016).One of the most important predictors of bullying is family (Papanikolaou et al., 2011;Gómez-Ortiz et al., 2015;Nocentini et al., 2019).To a certain extent, children's experience with the family environment determines their ability to adapt to the school environment and influences the way in which they get on with classmates.
There is also evidence that different roles in bullying are related to different risk or protective factors.For bullies, there is a typical absence of a positive approach and warmness (Bowes et al., 2009).Bullies perceive their family as less cohesive, with a lower level of parental control and tendency to use physical punishment (Espelage et al., 2000).In their family, conflicts and hostility are dominant.For victims, overprotective parenting is typical especially from the mothers (Georgio, 2008).Extreme parental involvement, characterized by intense closeness in a parent-child relationship, could be a problem in higher levels of victimization, especially in boys (Dake et al., 2003).It seems that there have been few studies looking at the specific type of family atmosphere (e.g., Espelage et al., 2000;Bowes et al., 2009) and mothers' and fathers' parenting styles separately (e.g., Schaffer et al., 2008) for children with different roles in bullying.This raises the question as to whether there are specific characteristics in family atmosphere and parenting styles of fathers and mothers for the bullies, victims and defenders.
The aim of the present study was to explore family atmosphere (cohesion, expressiveness, conflicts, organization, and control), parenting styles of fathers and mothers (positive parenting, directive parenting, hostile parenting, autonomous parenting, and desultory parenting) in the context of the different roles in bullying (bullies, victims, and defenders).

Participants
The participants of the research were 160 adolescents aged between 13 and 15 (M = 13.93,SD = 0.41).However, five adolescents could not be included in the research because they did not provide any data on bullying.The total number of participants was therefore 155 adolescents (78 boys and 77 girls).About 53% of participants had been involved in bullying, out of whom 10% were in the position of aggressors, 17% in the position of victims, and 27% in the position of defenders.All the aggressors were boys and the boys also represented 77% of the victims.78% of the defenders were girls and approximately the same number of children had not been involved in bullying (53% of them being girls).

Procedure
Pupils and parents confirmed participation in the study with informed consent and were also informed that the study would not publish information about the name or location of the school.The research was also carried out with the consent of the headteachers at the schools.The average time needed to complete the questionnaire was 25 minutes.Participation in the research was anonymous and voluntary.Data collection was carried out at six different primary schools (eight different classes) in various regions in Slovakia.

Measures
The role in bullying was measured by the Olweus Bully/Victim Questionnaire (Olweus, 2006).For the purpose of our research 12 items focused on analyzing the different roles in bullying in the class (aggressors, victims, defenders, not involved in bullying) were selected.Participants reported how often they had engaged in the behavior described in each item on a five-point scale.It is a self-report scale for analyzing different roles in bullying.The bullying is confirmed (Solberg & Olweus, 2003) if it has occurred in the last two or three months and not only if it is happening right now or has happened in the last week.
The parenting styles of mothers and fathers were explored by ADOR (Adolescents about Parents, in Slovak or Czech) (Matějček & Říčan, 1983).The questionnaire contains 100 items that are equally distributed with respect to exploring the parenting styles of fathers and mothers.The self-report questionnaire is for pubescents and adolescents aged from 13 to 18 and examines the following factors: 1) Positive interest vs. hostility, 2) Directiveness vs. autonomy, 3) Desultoriness in upbringing.The validity and reliability of the questionnaire were verified on the Czech and Slovak populations (Matějček & Říčan, 1983), e.g., test-retest reliability values for mothers' parenting were r = 0.70-0.83and for fathers' parenting r = 0.71-0.86.
Family Environment Scale (Hargašová & Kollárik, 1992) is a self-reported scale for the description of family atmosphere.It is based on Moos' conception of the family system (Moos & Moos, 1976) and has been developed for those older than 9 years in the Czech and Slovak populations.The questionnaire consists of 100 items and 10 subscales.For every item, children have to state whether it is true or false.In the current research, a two-dimension version (5 subscales, 45 items) was used.In particular, the dimension of the relationships (Cohesion, Expressiveness, Conflicts) and the dimension of family system stability (Organization, Controls).The validity and reliability of the questionnaire were verified on the Czech and Slovak populations (Hargašová & Kollárik, 1992), e.g., test-retest reliability values r = 0.53-0.83.
By employing two simple questions, we inquire about the participants' school experience: 1. "How do you like school?" with potential responses ranging from "I don't like school very much" to "I like school very much."Responses were coded from 1 to 5.
2. "How many good friends do you have in your class?" with possible answers varying from "None" to "I have 6 or more good friends in my class."Responses were coded from 1 to 5.

Data Analysis
Number of missing data was about 3.4%.The missing values were imputed by using multiple imputation methods with a minimum of 5 imputations (Rubin, 1987) and at least 10 iterations per imputation (Reiter & Raghunathan, 2007).The multiple imputation was carried out using the R package MICE (van Buren & Groothuis-Oudshoorn, 2011).For the models' estimation, a Bayesian logistic regression model was selected with the aim of avoiding the problem with perfect separation in the binary logistic regression models (Gelman et al., 2008).For the Bayesian logistic regression model, the R package arm (Gelman et al., 2008) was used.Data and analytical code are available at https://osf.io/kwxr9/.

Regression Models and Rationale of Selected Predictors
We have opted to test three distinct regression models, focusing on bullies, victims, and defenders as separate outcomes.Our intention is to examine the unique impact of family atmosphere and parenting styles on these different roles in bullying, without conflating their interrelatedness.Both family atmosphere and parenting styles play critical roles in predicting bullying behaviors, with previous research supporting their significance (e.g., Baldry & Farrington, 2000;Lereya et al., 2013).Moreover, we recognize that parenting styles may differ when considering the perspectives of both mothers and fathers, which is why we have included parenting styles of both parents in our regression models (e.g., Gómez-Ortiz, 2015).To ensure robust analysis, we have incorporated two control variables in all regression models.These factors are the number of friends the participants have in their class and their affection for the school environment, reflecting their enjoyment of school.Existing studies have identified these variables as valuable indicators of various roles in bullying (e.g., Eslea et al., 2003;Buhs et al., 2009;Markkanen et al., 2021).Importantly, these control variables are unrelated to family factors and directly associated with the school environment.

Descriptive Statistics
Table 1 shows the descriptive information about the family atmosphere of the sample in subgroups according to the role in bullying.
Table 2 shows descriptive information about parenting styles and correlations of the sample in subgroups according to the role in bullying.

Regression Analyses
The following tables show the results of the Bayesian logistic regressions using different roles in bullying as the criterion and family atmosphere (Table 3) and parenting styles (Table 4), and number of friends and enjyoing school as the predictors.The differences in significant predictors are shown in the supplementary materials (see figures for differences in predictors).Overall, concerning the values of Nagelkerke R squared, regression models for bullies and defenders indicate a strong relationship between predictors and outcomes, while for victims, it shows a moderate relationship in the context of family atmosphere (Table 3).In the context of parenting styles, all three regression models demonstrate a strong relationship between predictors and outcomes (Table 4).

Discussion
In the present study, as many as 52.9% of the pupils had been involved in bullying; 9.7% of them had been the bullies, 16.7% the victims and 26.5% were defenders.It was found that bullies are from families characterized by less family cohesion and control, with more conflicts and with fathers' desultory parenting.
The family of victims is characterized by less family cohesion, higher family control, fewer fathers' positive parenting and with slightly higher mothers' positive parenting.Victims also have fewer good friends in the class.A characteristic family for defenders is one with higher family expressivity and organization, fewer conflicts, more fathers' positive parent-ing and fewer fathers' directive parenting as well as fewer mothers' autonomous parenting.Defenders are also predicted by fathers' hostile parenting and also have more good friends in the class.

Family Atmosphere and Parenting Styles in Bullies
According to the current findings, the families of bullies are characterized by less cohesion, higher conflict rates and less control.Similar findings have been reported by Cook et al. (2010).Parental violence and blaming are more frequent in families of bullies and bullies perceived parents more negatively (Braithwaite, 1989).Neglect, abuse and maladaptive behavior from parents can contrib- ute to the more negative family atmosphere of bullies (Lereya et al., 2013).Olweus (1995) has claimed that parents did not give their children enough love and attention, took a more negative attitude, did not set firm boundaries in their behavior and tolerated violent and bullying behavior.According to Espelage et al. (2000), aggressors perceive their families as less cohesive with a low level of parental control and tendency to use physical punishment.Higher family cohesion acts as a protective factor of aggressive behavior (Duggings et al., 2016).In the family of bullies, an absence of affection and warmth is typical (Bowes et al., 2009).From the point of view of parenting style, it was found that bullies experienced higher desultory parenting by fathers.Desultory parenting has a significant negative impact as it is a manifestation of indifference, even covert hostility in relation to the adolescent.Children who labeled their fathers as strict, controlling or had a lack of presence in their lives were more involved in bullying behavior (Kawabata et al., 2011).On the other hand, children who perceived a positive parental style involving emotional sensitivity and warmth were less involved in bullying.

Family Atmosphere and Parenting Styles in Victims
It was found that families of victims are characterized by less cohesion and higher control.In the context of Eastern Europe, Říčan et al. (1993) also found higher control and hostility in the families of victims.In the context of parenting, it was found that victims perceived a lower level of fathers' positive parenting and a slightly higher level of mothers' positive parenting.The lack of positive interest, encouragement, listening and understanding of the problems and difficulties that the child experiences with the father are related to victimization.The fathers of these children are more distant, cold and it is difficult for these children to identify with them.Absence and passivity in parenting on the part of the father are risk factors for victimization (Fosse & Holen, 2002).In addition, Flouri and Bechanan (2002) have stated that victims do not have a friendly and close relationship with their fathers.The victims also tend to experience greater indifference and abuse from their fathers.On the other hand, higher positive parenting and a supportive family (Shaheen et al., 2019) are protective factors of bullying.The current study found that victims had a higher level of positive parenting from their mothers.This finding combined with a higher level of control in the family can indicate the higher protectivity of mothers.However, having an overprotective mother is a risk factor for being bullied because it can limit the ability to self-defend and effectively cope with victimization attacks (Georgio, 2008).

Family Atmosphere and Parenting Styles in Defenders
In defenders, it was found that their family typically has a higher level of expressivity and organization, lower level of conflicts, a higher level of fathers' positive parenting, lower level of fathers' directive parenting, and lower level of mothers' autonomous parenting.The results of the regression model indicate that there is also an effect of fathers' hostile parenting on the defender.As far as it is known, the defenders have the least-known role in bullying.Indeed, several studies have directed their attention towards defenders, exploring various aspects that influence their roles.Some researchers, such as Huitsing et al. (2014) and Burger et al. (2022), have investigated sociodemographic predictors that may impact defenders' behavior.Additionally, other studies, like the one conducted by Nickerson et al. (2008), have focused on examining personality characteristics, such as empathy or attachment style, as potential factors influencing defenders' involvement in bullying dynamics.Unlike the aforementioned studies that explored sociodemographic predictors or personality characteristics, our study focused on the family or parental context.By doing so, we have revealed alternative aspects that play a crucial role in shaping defenders' actions within the context of bullying dynamics.This unique approach allowed us to gain valuable insights into the influence of family and parental factors on defenders' behavior, contributing to a more comprehensive understanding of their role in the bullying phenomenon.Mulvey et al. (2018) found that positive family management is characteristic of the family of defenders.The current study also found that the defenders' family is typically well organized, has clear rules, open to sharing feelings and has greater harmony (fewer conflicts).Similarly, Mazzone and Camodeca (2019) found that less functioning families decreased defending behavior and increased aggressive behavior in bullying.Regarding parenting styles in defenders there has been no systematic research.Fathers in case of defenders are more positive, less directive and less hostile.It means there is less enforcement of parental orders, control of friends, and punishment.Fathers are more supportive and are a good role model for their children.Defenders also refer to less autonomous parenting of mothers.Given the higher level of expressivity and organization, as well as less conflict in the family, lower autonomous parenting could indicate that mothers of defenders communicate to them what is right and what is wrong more often than parents of bullies and victims.Fewer autonomous parenting could be helpful in learning how to behave in conflict situations.
There are also several limitations in the present findings which must be considered in the case of potential generalizing the research.Firstly, there were 155 respondents in the study and, when considering the division into groups based on the individual roles in bullying, the representativeness of the groups is thus limited.Absolutely, considering the sample size in our current study and the number of predictors per regression model, it is crucial to interpret the results with caution.While the findings provide valuable insights, the limited sample size may influence the statistical power and generalizability of the results.Therefore, it is advisable to replicate these findings in future studies with larger and more diverse samples to ensure the robustness and reliability of the conclusions.Replication studies with larger samples can help validate the patterns observed in our study and provide a more comprehensive understanding of the relationships between these predictors and outcomes in the context of bullying dynamics.Secondly, as the study design was cross-sectional, it is not possible to provide adequate comments on the potential causal validity of the identified relationships.Thirdly, the individual forms of bullying were not distinguished in the study; bullying was understood as a complex phenomenon and included direct as well as indirect forms of bullying behavior.Fourth, the research used self-report questionnaires, which may have caused answering in terms of social desirability.Despite the limitations, the present study provides interesting and thought-provoking research findings regarding the different roles in bullying specifically focused on family atmosphere and parenting styles of mothers and fathers.Knowledge about the factors related to family atmosphere and parenting styles can be very useful for school psychologists in preparing bullying prevention and intervention programs.Knowledge about factors related to the family atmosphere and parenting styles can be very useful for school psychol-ogists in the preparation of bullying prevention programs and intervention programs, which will include lectures and discussions with parents.Parenting styles of fathers and mothers have different importance in different roles in bullying.Family climate and parenting styles (fathers and mothers separately) are important factors in bullying for both aggressors and victims (e.g., Cerezo et al., 2018).An equally important protective role is played by discussing bullying with children (Parker, 2015), which can be used to encourage children to seek help (Cross et al., 2012).At the same time, knowledge about parental factors can be useful for the educational process in the creation of responsibility of adolescents, which can lead to minimizing the risks of inclination to some forms of inappropriate adolescent behavior (Martinkovič, 2017), such as bullying.Since the main aim of education now is to produce a flexible workforce, the content of education and the teaching process is primarily set to acquire the so-called key competences.It is also necessary to focus on the ethical aspects of educating young people, as part of the ongoing curriculum reforms, which could help prevent various forms of bullying (Jahelka, 2014).Lester et al. (2017) found that including family in preventive and interventive programs focused on bullying increased child-parent communication about bullying.These authors have also stated that it is important to work on these programs with both mothers and fathers.The social support as well as professional help should be adjusted to the subjective needs of each affected individual (Vindišová, 2012), and should be more comprehensive throughout the overall school culture and teacher-student relationships (Brestovanský, 2019;2020).According to the present study, the parenting styles of fathers and mothers have different importance in different roles in bullying.Family climate and parenting styles (of fathers and mothers sep-arately) are important factors in bullying for both the bullies and victims (e.g., Cerezo et al., 2018).However, it seems that there is a research gap in exploring which factors related to family are important for defenders.They play a very important role in bullying from the point of view of bullying intervention and prevention.Thus, further research should be focused on examining the effect of fathers' and mothers' parenting on bullying separately, as well as exploring the family of defenders.

Table 1
Descriptive statistics of the family atmosphere in subgroups according to the role in bullying

Table 2
Descriptive statistics of parenting style in subgroups according to the role in bullying

Table 3
Bayesian logistic regression results using different roles in bullying as the criterion and family atmosphere, number of friends and enjoying school as predictors