Abstract
Differences in bullying rates between schools could be explained by school efficacy. This study examined the relationships among teachers’ perceptions of principals’ practices, school climate, and school collective efficacy to prevent bullying. The sample comprises 403 Mexican elementary-school teachers; 35% were male, and 65% were female. The teaching experience ranged from 2 to 35 years (M = 13.2 years, SD = 9.1). Teachers answered self-report measures. A latent variable structural equation modeling (SEM) approach was used. SEM model indicated that principal’s bullying prevention was directly related to a positive school climate, but they did not influence teachers’ perceptions of school collective efficacy. Also, principals’ support for teachers’ antibullying practices positively affected school climate and school collective efficacy. Both principal involvement and support had an indirect relationship with school collective efficacy. Overall, findings suggest that the principal has a critical role in promoting teachers’ perceptions of school collective efficacy in bullying prevention.
Bullying is associated with severe short and long-term physical, psychological, and behavioral disorders (Bradshaw, 2015; Fullchange & Furlong, 2016; Holt et al., 2015; Schoeler et al., 2018; Wolgast & Donat, 2019). Unfortunately, the literature consistently demonstrates the prevalence of aggressive behaviors in schools worldwide (Craig et al., 2009; Modecki et al., 2014; Valdés-Cuervo et al., 2018). Schools have an ethical and legal responsibility to prevent bullying, ensuring students’ safety and human rights (Rosen & Cowan, 2012; Sercombe & Donnelly, 2013).
Scholars suggest that the school context is critical to understanding the differences in school bullying rates (Hymel & Swearer, 2015; Muijs, 2017; Pearce et al., 2011; Rose et al., 2015; Volungis & Goodman, 2017). In particular, the Dynamic Model of School Improvement (DASI; Creemers & Kyriakides, 2010; Kyriakides et al., 2014) emphasizes the importance of examining the influence of school-level variables on school learning environments in order to promote a school’s effectiveness toward bullying prevention. The emphasis given in DASI to school-level variables is in line with studies (Ertesvåg & Roland, 2015; Espelage et al., 2003; Hymel et al., 2015; Waasdorp et al., 2011) concluding that bullying is largely the result of norms, social control, and interpersonal relationships inside the school. Particularly, emerging research highlights that collective efficacy is a school ecological construct that may explain bullying rate differences (Aasheim et al., 2020; Barchia & Bussey, 2011; Gini et al., 2020; Williams & Guerra, 2011).
School Collective Efficacy
Collective efficacy emerged from Social Cognitive Theory (SCT; Bandura, 1986), which is defined as a group’s beliefs about their ability to work together to achieve a common goal (Bandura, 2000; Goddard et al., 2004). Sampson et al. (1997) suggest cohesion, trust, and social control are action-oriented components of collective efficacy. Cohesion and trust are forms of social capital that imply that group members share a sense of belonging to the community and recognize that collective objective mutually benefit each other (Hymel et al., 2015; Li, 2004; Sampson, 2006; Williams & Guerra, 2011). Social control reflects the group’s will to influence its members to achieve collective norms, protect other members, and the community’s good (Cho & Lee, 2018; Church et al., 2012; Goddard et al., 2004; Hirschi, 1969). Drawing on collective efficacy, bullying researchers believe that school members’ beliefs on their school’s efficacy to support students may be effective at either reducing or preventing bullying episodes (Hymel et al., 2015; Sapouna, 2010; Williams & Guerra, 2011).
Empirical studies have demonstrated that school members with higher perceptions of school collective efficacy tend to intervene better in bullying situations (Barchia & Bussey, 2011; Olsson et al., 2017; Sapouna, 2010; Wänström et al., 2019; Williams & Guerra, 2011). However, to our knowledge, there are no studies that examined how school-level variables encourage school collective efficacy to prevent bullying. Therefore, an examination of school variables associated with collective efficacy is essential for whole-school intervention programs.
Principals and School Collective Efficacy
Empirical research using the whole-school approach highlight that principals’ practices are critical to explain differences in bullying rates in schools (Castro Silva et al., 2017; Pepper & Hamilton Thomas, 2002; Rhodes et al., 2009). Moreover, these studies suggest that the effect of principals on bullying prevalence occurs through their influence on school efficacy to prevent bullying (Haataja et al., 2015; Marachi et al., 2007; Whitted & Dupper, 2005). Scholars posit that cooperation and trust among principals and teachers are essential for teachers’ positive perceptions of school efficacy that encourages positive behavior on students (Bryk et al., 2010; Cansoy & Parlar, 2018; Marachi et al., 2007; Skinner et al., 2014).
When principals focus on bullying prevention, it is reflected by the school’s effectiveness to handle bullying when it arises (Ansary et al., 2015; Gülşen & Gülenay, 2014; Lacey & Cornell, 2016; Olweus, 1993). This involvement may include establishing rules and actions to reduce bullying rates, such as supervision plans, meetings with parents, students, and teachers; and attempt to get all school members to commit to anti-bullying intervention strategies (Ansary et al., 2015; Dake et al., 2003; Menesini & Salmivalli, 2017).
Implementation of effective school anti-bullying strategies requires commitment from the entire school community. Thus, principal support is an essential component for teacher’s adherence and effort with the strategies (Ahtola et al., 2012; Leithwood et al., 2010; Olweus & Limber, 2010). Principal support involves any resources and processes to reinforce school staff effectiveness to handle bullying (Cohen et al., 2000; Robbins & Alvy, 2014). Supportive principals can foster a sense of responsibility and commitment to bullying prevention in teachers (American Educational Research Association, 2013; Kallestad & Olweus, 2003). Also, when teachers feel supported by the principal and others staff members, they feel more confident to handle bullying, thus increasing collective efficacy (Ahtola et al., 2012; Skinner et al., 2014).
Based on the literature, we believe principals’ involvement and support of teachers’ antibullying practices are related to school collective efficacy to prevent bullying (Ahtola et al., 2012; Ertesvåg & Roland, 2015; Hymel et al., 2015; Skinner et al., 2014). Thus, the present study aims to analyze the influence of principals’ practices on collective teacher efficacy to prevent bullying based on teachers’ perceptions.
The Mediate Influence of School Climate
Lezotte Miller et al. (1980) reminds us that “principals represent the organizational authority of the school, and in that regard, they serve to symbolize what the school stands for, how it will operate, and what is important. In general, they set the educational tone for the school” (p. 93). Hence, scholars believe that principals’ performance can affect school collective efficacy by promoting a positive social climate (Goddard et al., 2015; Hymel et al., 2015; Price, 2012). In this regard, Aladenusi and Ayodele (2014) found that the commitment of teachers to the school’s vision can be affected either positively or negatively by school climate.
Ajay and Mandakini (2013) suggest that school climate affects teachers’ performance, motivation, dedication, satisfaction, and well-being. Scholars report (see Aladenusi & Ayodele, 2014; Bellibas & Liu, 2018; Gülşen & Gülenay, 2014; Leithwood & Seashore-Louis, 2012) that positive school climate incites the commitment of teachers to the school’s vision. Additionally, empirical evidence shows that positive school social climates encourage teacher effort to comply with school social norms, which has related to successful school bullying prevention (Hymel et al., 2015; Konishi et al., 2017; Low & Van Ryzin, 2014; Malinen & Savolainen, 2016; Valdés-Cuervo & Carlos-Martínez, 2014; Wang et al., 2013).
Social Cognitive Theory (Bandura, 1997, 2000) suggests that the quality of school social climates influences people’s shared belief about their collective power to produce desired outcomes. Frequent, pleasant, and supportive relationships with school staff raise teachers’ perception of the school’s collective capacity as a whole (Bandura, 1997; Lim & Eo, 2014). Overall, the literature concludes that when teachers perceive school climate as caring and supportive, their bonds with the school community are stronger and, as a consequence, they are more likely to increase their perception of collective efficacy to cope with bullying (Barchia & Bussey, 2011; Hymel et al., 2015; Olsson et al., 2017; Williams & Guerra, 2011). However, despite its relevance, it remains unclear what school factors, especially which principal’s variables can foster school climate, and how they can mediate the effects of school collective efficacy to handle bullying.
Current Study
Despite a bulk of studies showing that bullying is an ecological-social problem (Espelage, 2014; Swearer & Hymel, 2015), bullying research has frequently focused on individual rather than whole-school factors. Many school-based antibullying programs are implemented, but most do not achieve the desired results (Evans et al., 2014; Langford et al., 2015; Ttofi & Farrington, 2011). Therefore, it is necessary to increase research focused on school-level factors that explain successful antibullying programs (Gulemetova et al., 2011; Spillane & Kenney, 2012), especially factors related to the principal’s role in school efficacy to prevent bullying. To the extent that the principal’s role in bullying prevention can be understood, it will help determine their functions and promote their effective involvement. However, no studies known by the authors have explored how the principal influences teachers’ perceptions of school collective efficacy to prevent bullying. As a result, there is a lack of clarity about school factors related to bullying prevention, especially the principal’s role.
In this context, the present study examined direct and indirect relationships between principal’s involvement in bullying prevention and support teachers’ antibullying practices, school climate, and school collective efficacy to handle bullying (see Figure 1). The following hypotheses are proposed for this intent:
Hypothesis 1a (direct relations principal’s involvement): A positive relationship between the principal`s involvement in bullying prevention, positive school climate, and the school collective efficacy in preventing bullying was anticipated.
Hypothesis 1b (direct relations principal’s support): A positive relationship between the principal’s support of teachers’ antibullying practices, positive school climate, and school collective efficacy to prevent bullying was expected.
Hypothesis 2a (indirect relations): Principal’s involvement and support indirectly affect school collective efficacy by the positive influence on school climate.

Hypothetical model depicting the relations between principal’s involvement, principal’s support, school climate, and school collective efficacy in preventing bullying.
Method
Participants
The teachers (n = 2,400) population came from elementary public schools from three cities in Sonora, Mexico. Like other public urban elementary schools in Mexico, these schools educate students with low and middle socioeconomic status (National Institute of Statistics and Geography, 2018). The study comprises 403 teachers selected by simple probabilistic sampling with replacement (p = .5, q = 95%, e = 5%). The sample included 141 males and 262 females. Aged ranged between 22 and 76 years (M = 36.53 years, SD = 10.1). The teaching experience of research participants ranged from 2 to 35 years (M = 13.2 years, SD = 9.1). Regarding teacher’s education, 74% had a bachelor’s degree, and 26% had post-graduate studies.
Measurement
Principal’s involvement
The scale was developed to measure teachers’ perceptions of principals’ involvement in bullying prevention. It comprises 10 items in Likert-type format (1 = Never to 5 = Always) grouped in three dimensions: (a) Schoolwide practices, principal’s activities to involve all school staff in bullying prevention (3 items, e.g., The principal organizes school meetings to handle bullying, Cronbach’s α = .80; McDonald’s ω = .81); (b) Supervision practices, surveillance activities to avoid bullying (3 items, e.g., The principal establishes supervision in school restrooms, α = .84; ω = . 86); and (c) non-permissive intervention, practices focused on stopping aggression and reasoning with students regarding misplaced aggression (4 items, e.g., In a serious situation of bullying, the principal intervenes to solve the problem by talking with the victim, the bully and parents, α = .88; ω = .89). The confirmatory factor analysis (CFA) revealed a good model fit to the data (χ2 = 51.94, df = 31, p = .011; AGFI = 0.96; CFI = 0.99; SRMR = 0.035; TLI = 0.98; RMSEA = 0.041, 90% CI [0.020, 0.060]).
Principal’s support
The Principal’s Support Scale (PSS; Littrell et al., 1994) was adapted to measure teacher’s perception of principals’ informational, affective, and instrumental support to anti-bullying practices (5 items, e.g., Principal gives me feedback to improve my skills to handle bullying, α = .85; ω = .87). The Likert-type response format was used (0 = never to 4 = always). The CFA results suggest that the model fit the data (χ2 = 11.98, df = 5, p = .035; AGFI = 0.96; CFI = 0.99; SRMR = 0.014; TLI = 0.98; RMSEA = 0.049, 90% CI [0.025, 0.055]).
School climate
We adapted two subscales from the Delaware School Climate Survey-Teacher/Staff (DSCS-T/S; Bear et al., 2014) that measure teachers’ perceptions of the quality of school relationships in bullying events. Items grouped in two dimensions: (a) Teacher-student relations that assess teacher and adult responsiveness to student bullying problems (3 items; e.g., Teacher listens to students when they have bullying problems; α = .75; ω = .76); and Teacher-teach relations that measure respect, caring, and cooperation in relationships between teachers (4 items, e.g., Teachers in this school give me advice about how to handle bullying situations, α = .74; ω = .77). Items respond in five-point Likert-type format (0 = never to 4 = always). The CFA indicates that the model fits the data (χ2 = 16.92, df = 11, p = .13; AGFI = 0.97; CFI = 0.99; SRMR = 0.016; TLI = 0.99; RMSEA = 0.035, 90% CI [0.012, 0.068]).
School collective efficacy
The Collective Teacher Efficacy Scale (CTES; Skaalvik & Skaalvik, 2007) was adapted to measures teachers’ perception of whole-school collective efficacy to prevent bullying. The scale comprises 7 items (e.g., Teaching staff of this school can solve bullying situations, α = .79, ω = .80) in five-point Likert-type response format (0 = none 0% to 4 = everyone 100%). The CFA showed the model fit the data (χ2 = 18.08, df = 11, p = .08; AGFI = 0.96; CFI = 0.99; SRMR = 0.015; TLI = 0.99; RMSEA = 0.040, 90% CI [0.02, 0.062]).
Control variable
The control variable includes teaching experience. This variable was measured through years of experience teaching in elementary school. This variable was controlled to account for the differences in the school collective efficacy related to this variable (Burger et al., 2015; Wolters & Daugherty, 2007; Yoon et al., 2011) rather than the principal’s involvement and support.
Procedure
The researchers gained permission from the Ethical Committee of the Food and Development Research Center to conduct the study. Later, a consent letter was sent to teachers in order to request their voluntary participation in the study. Only 8% of teachers refused to participate. Once approvals were gained, teachers were invited to respond to the instruments. Data collection took place in school classrooms during non-school hours.
Data Analysis
Missing values were less than 3% of the data. They were treated using the multiple imputation method, available in SPSS 25. The mean, standard deviations, and Pearson’s correlations between the variables were then calculated. Then, an unconditional random effects Anova was conducted to evaluate the school dependence of variables in the study. Results were used to calculate the intraclass correlation (ICC). The ICC < .10, indicated that differences in the variables were not dependent on school membership (Hox, 2010; Lai & Kwok, 2015).
Structural equation models were calculated using AMOS. The maximum likelihood estimation (ML) with Bollen-Stine and bias-corrected confidence intervals bootstraps (500 replicates with 95% CI) was used. The ML bootstrap is an AMOS procedure to deal with multivariate non-normality issues (Arbuckle, 2017; Byrne, 2016). The following statistical tests were used in order to evaluate the global goodness of fit for the model: (a) Chi-squared and associated probability (χ2 with p < .001), Bollen-Stine bootstrap p associate > .05, standardized root mean square residual (SRMR ≤ 0.05), Tucker-Lewis index (TLI ≥ 0.95), adjusted goodness of fit index (AGFI ≥ 0.95), comparative fit index (CFI ≥ 0.95) and root mean square error of approximation (RMSEA ≤ 0.05) with their confidence interval (Bollen & Stine, 1992; Byrne, 2016).
Results
Table 1 shows the means, typical deviations and correlations between the study variables. The results show that teachers’ perception of principals’ involvement and support positively correlates to school collective efficacy. In addition, teachers’ perception of school climate, which is the mediational variable in the model, is also positively associated with collective efficacy. Teachers teaching experiences have a negative correlation to principal’s support.
Mean, Standard Deviation and Correlation Between the Study Variables.
p < .05. **p < .01. ***p < .001.
Structural Model
The structural model results are presented in Figure 2. Fit values indices showed that the model adjusted to the data (χ2 = 132, df = 96, p = .008; Bollen Stine bootstrap = 0.07; TLI = .98; SRMR = .036; AGFI = .96; CFI = .98; RMSEA = .031, 90% CI [0.01, 0.04]). This model accounts for 35% of the variance in the teacher’s perceptions of school collective efficacy to prevent bullying.

Results of the structural model of the relations between principal’s involvement, support, school climate and school collective efficacy.
Regarding the direct relationships, results indicate that principal involvement in bullying prevention has a positive influence on school climate (β = .34, p < .001, 95% CI [0.18, 0.30]), but it does not influence the school collective efficacy (β = .10, p < .075, 95% CI [−0.07, 0.17]). For this part, principal support was positively associated to school climate (β = .24, p < .001, 95% CI [0.18, 0.29]) and school collective efficacy (β = .29, p < .001, 95% CI [0.15, 0.25]). Finally, school climate had a positive effect (β = .49, p < .001, 95% CI [0.42, 0.55]) on school collective efficacy.
Regarding indirect effects, results indicate that principal involvement had an indirect influence on teachers’ perceptions of school collective efficacy (β = .19, p < .001, 95% CI [0.07, 0.33]), through the positive effects of school climate. Also, principal’s support was indirectly associated to collective efficacy (β = 0.16, p = .039, 95% CI [0.08, 0.26]) by its positive influence on school climate.
Discussion
The study examines the direct and indirect relationships between principals’ involvement and support, school climate and teachers’ perceptions of school collective efficacy to prevent bullying. Overall, findings partially support our hypotheses about the direct and indirect relations proposed in the structural model. Below is a discussion of the direct and indirect relationships that were expected compared to what was found.
Principal Involvement (Direct Relations)
Findings partially supported the hypothesis about principal involvement in bullying prevention. Consistent with the literature, these results suggest that principal involvement positively influences school climate (Goddard et al., 2015; Hymel et al., 2015; Price, 2012). However, contrary to what was expected, results show that principals to be involved did not directly affect teachers’ perceptions of school collective efficacy. This result is contradictory to the current literature demonstrating the positive effects that principals have on school effectiveness (Cansoy & Parlar, 2018; Fancera & Bliss, 2011; Ross & Gray, 2006) and its role in the success of anti-bullying interventions (Ahtola et al., 2012; Dake et al., 2003; Muijs, 2017).
Principal’s Support (Direct Relations)
The study’s results validate the expected hypothesis about the positive influence of a principal’s support on school climate and school collective efficacy. These findings are consistent with the literature that shows principals’ influence in school social relationships and the effectiveness of school intervention programs to prevent bullying (Ansary et al., 2015; Dake et al., 2003; Menesini & Salmivalli, 2017; Olweus & Limber, 2010). Results showed that a principal’s informational, instrumental and emotional support of teachers’ antibullying interventions is essential to teachers’ perceptions of school efficacy to manage bullying events.
Mediate Influence of School Climate
The study found that school climate fully mediates the relationship between principals’ involvement and teachers’ perception of school collective efficacy to prevent bullying. This finding suggests that a principal’s involvement only influences teachers’ collective efficacy when generating a positive school social climate. In other words, if principals want to increase school collective efficacy to prevent bullying, their involvement should focus on fostering collaborative and supportive relationships among teachers for the “common good” which includes the care of students (Acosta et al., 2019; Bellibas & Liu, 2018; Goddard et al., 2015; Gülşen & Gülenay, 2014). The study result is in line with other studies that say that the principal has an essential effect on increasing the shared commitment to tasks and goals by improving climate (Aarons et al., 2014; Green et al., 2013).
Moreover, these findings suggest that school social climate partially mediates the relationship between principals’ support and school collective efficacy to prevent bullying. This finding is in line with previous research that suggests that teachers’ perceptions of school collective efficacy improve when principals promote positive relationships among teachers (Goddard et al., 2015; Hanson & Voight, 2014; Hymel et al., 2015; Price, 2012). Our findings confirm what other scholars (Tschannen-Moran et al., 1998) have already reported, teachers’ perception of collective efficacy is related to school context characteristics. Although further studies are necessary, a positive school climate increases the perception of kindliness and collaborative relationships among school staff members. These positive relationships encourage psychological variables related to collective efficacy, like well-being, sense of belonging, sharing and commitment of school goals and trust between teachers (Collie et al., 2012; Damanik & Aldridge, 2017; Hosford & O’Sullivan, 2016). Nevertheless, further research is needed to understand the mediational role of school climate and identify how their dimensions impact teachers’ perception of collective efficacy to handle bullying.
Conclusions
From a theoretical perspective, the study confirms the value of a dynamic school improvement model (DASI; Creemers & Kyriakides, 2010; Kyriakides et al., 2014) on bullying research. Consistent with the theory, the findings show that principals, directly and indirectly, influence the school’s collective efficacy in preventing bullying. In this regard, and in line with current research (Hymel et al., 2015; Trach et al., 2013; Williams & Guerra, 2011), data suggests that a principal’s involvement and support to encourage a positive social climate has a stronger influence on teachers’ perceptions of school collective efficacy than only the principal’s willingness to be involved in bullying prevention. These results indicate that effective principal anti-bullying practices should be to promote a positive relationship inside schools.
From a practical point of view, principals’ involvement and support are related to teachers’ perceptions of school collective efficacy to prevent bullying. In this regard, findings suggest that the effectiveness of whole-school anti-bullying interventions increases by the principal’s involvement and support of teacher practices to prevent bullying. Finally, the study showed that programs to prevent bullying must involve developing principal competencies to promote quality relationships among the participants in the educational process.
Limitations
Despite the present study contributing to the understanding of the principal’s role in promoting collective efficacy to prevent bullying, the results did have limitations that should be considered. The cross-sectional design used in this study does not allow establishing causal relationships. Therefore, longitudinal designs are recommended for the verification of this model. Likewise, teachers’ self-reported measures were used. Future research using observational techniques to collect data on these variables are highly encouraged.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work has been funded by the Research Strengthening Program of the Technological Institute of Sonora (Profapi 2021_0007).
