Preservice Teachers’ Knowledge and Attitudes Toward Bullying: A Systematic Review

Preservice Teachers’ Future Role in Addressing Bullying

Classroom teachers have an instrumental role in reducing bullying behavior and supporting victims (Colpin et al., 2021; Demol et al., 2020; Wachs et al., 2019; Yoon & Bauman, 2014). For instance, teachers can create antibullying classroom norms and encourage students to enforce those norms (Veenstra et al., 2014), which in turn helps prevent future bullying. Specifically, how teachers respond to bullying can influence their students’ willingness to report bullying (Demol et al., 2020), their students’ behavior (e.g., likelihood of bystander intervention; Hektner & Swenson, 2012), and the occurrence of bullying in the classroom over time (Veenstra et al., 2014). Teachers are likewise important to victims’ adjustment in bullying incidents. On the one hand, when teachers do not address bullying, victims feel unprotected, fear retaliation, stay silent, and continue being bullied (Yoon & Kerber, 2003). On the other hand, when teachers successfully intervene in bullying, their response can help victims feel more socially supported by their teachers (Flaspohler et al., 2009). As an illustration, Yoon and colleagues (2016) found that when victims are bullied, 53% of teachers reported that they would console the victim by providing emotional support that can help victims cope with their experience. Indeed, when victims perceive that their teacher does not tolerate bullying, they are more likely to seek help from adults (Blomqvist et al., 2020), which is a critical step to getting the support they need. Given the vital role that teachers play in antibullying efforts, it is important that teachers have the skills necessary to participate in those efforts. Existing evidence from systematic reviews that include in-service teacher samples are concerning: one review found that in-service teachers lacked the confidence to address cyberbullying specifically (Macaulay et al., 2018), whereas another review reported mixed findings depending on the study design, with teachers seeming to express less confidence in qualitative studies than in quantitative studies (Fischer et al., 2021). Thus, even though some in-service teachers report a high degree of confidence, this pattern is not universal for all teachers (see Fischer et al., 2021). If the goal is to have all in-service teachers ready and equipped to effectively deal with bullying, it is logical to address any discrepancies in knowledge and skills during initial teacher training programs so that they learn how to effectively deal with a situation many of their future students will experience.

Theoretical Foundation

The overarching aim of this systematic review was to investigate whether preservice teachers are ready to fulfill their role as critical agents in antibullying initiatives and intervene in bullying events in the school setting. A teacher’s decision to intervene in bullying is predicted by a number of factors outlined in two theories that served as the foundation for our review and informed our specific research questions: the Bystander Intervention Model (Latané & Darley, 1970) and the Theory of Planned Behavior (TPB; Ajzen, 2012). The Bystander Intervention Model (Latané & Darley, 1970) posits that a teacher’s intervening behavior will occur if they do the following: (a) notice the situation, (b) perceive the situation as an emergency requiring help, (c) accept responsibility for taking action, (d) know how to help, and (e) take action to help. Researchers have applied this model to explain teachers’ intervention behavior in bullying situations (e.g., Nickerson et al., 2014), which provides a framework for identifying specific training components that may increase the likelihood that teachers will intervene. For example, teachers may not notice bullying because they are unsure of what bullying is or the various forms it can take. If that is the case, providing training so that preservice teachers understand the hallmarks of bullying and its numerous forms may increase the chance that they will recognize bullying, thereby increasing the likelihood that they will intervene.

The TPB is conceptually similar to the Bystander Intervention Model in that both theories assume that behavior is guided by multiple components. The TPB asserts that behavior will be influenced by one’s intentions to engage in that behavior, which are guided in part by their attitudes toward a behavior, subjective norms for the behavior (i.e., is there an expectation or pressure to engage in the behavior), and their sense of self-efficacy (or confidence) to perform the necessary behavior (Ajzen, 2012). This theory has been applied to understand the degree to which school staff—including teachers—intervene in bullying situations (van Verseveld et al., 2019; Yoon & Bauman, 2014). Based on the theory’s tenets, teachers are expected to intervene in bullying if they have negative attitudes about bullying (i.e., they think bullying is bad), if they think others expect them to intervene, and if they feel confident in their ability to intervene. For this review, we focused on preservice teachers’ attitudes and confidence as subjective norms may be more relevant to assess once preservice teachers begin teaching full-time and are influenced by the social expectations for intervening within their school environment.

The complementary nature of these theories is evident when we consider how components of the TPB could influence teachers’ progression through the sequential steps of the Bystander Intervention Model. For example, attitudes about bullying presumably would influence the first two steps of the Bystander Intervention Model by providing a frame for cognitive processes like attention and perception (Fazio et al., 1994): an antibullying attitude (i.e., bullying is bad) can prime teachers to be more alert for cues that bullying is occurring (i.e., notice the bullying) and more likely to interpret the situation as an emergency warranting action. Attitudes can also relate to the third step of the Bystander Intervention Model: accepting responsibility to act. For instance, if a teacher has a normative belief about bullying (i.e., bullying is a normal part of growing up; it can help teach kids important skills), then a teacher may be less likely to think they are responsible for intervening. Afterall, if they believe that bullying can help kids learn social skills, then they would not want to intervene in that process and deny their students the (presumed) opportunity to learn from their experience. Lastly, the components of self-efficacy from the TPB and taking action to help from the Bystander Intervention Model are similarly complementary: feeling like one is capable of executing an action will influence whether or not they take action (Ajzen, 2012; Bandura, 1977, 2001). Some teachers may know how to help but lack confidence (i.e., self-efficacy) in their ability to do so, which diminishes the likelihood that they will take action to intervene in bullying.

Both theories recognize that an individual’s decision and ability to act in a situation will be driven by a number of factors. Following this logic, in order to understand whether preservice teachers are likely to intervene in bullying situations once they start teaching full-time, we must first assess the factors assumed to influence their intervention behavior. Specifically, we assessed preservice teachers’ (a) knowledge of bullying hallmarks, (b) beliefs and attitudes about bullying, (c) attitudes toward different forms of bullying, (d) sense of responsibility to deal with bullying, and (e) level of confidence to address bullying.

Research Questions

Our first research question asked, What do preservice teachers know about the definition of bullying in terms of the key hallmarks that delineate bullying from general aggressive behavior? As previously mentioned, the hallmarks of bullying were originally articulated by Olweus (1993) to include (a) intent to harm, (b) repetition, and (c) a power imbalance between the bully and victim. Recent variations of the definition exist (e.g., goal-directed behavior, perceived power imbalance; Gladden et al., 2014; Volk et al., 2014), but they all include the core components initially outlined by Olweus (Gaffney et al., 2019), which is the predominant definition used in bullying research (Volk et al., 2014) and therefore the one that is most likely to be utilized in training materials. Preservice teachers’ understanding of this definition determines their ability to notice bullying. If they do not understand what bullying is, it is unlikely they will interpret a situation as bullying, thereby reducing the likelihood that they will intervene. Given that in-service teachers themselves often struggle to define bullying (e.g., Oldenburg et al., 2016), we expected preservice teachers would similarly lack knowledge of bullying hallmarks.

Our second research question addressed preservice teachers’ beliefs and attitudes about bullying, which is central to the TPB and helps predict teachers’ intervening behavior (Ajzen, 2012). One’s attitude toward a behavior is assumed to be a product of the belief that the behavior in question causes some outcome and one’s evaluation of that outcome (Ajzen, 2012). Applied to bullying, a negative attitude towards bullying represents the belief that bullying causes harm and such harm is bad for youth. One’s attitude can also be influenced by the belief that bullying is just a natural part of growing up (Troop-Gordon & Ladd, 2015). Bullying-specific research supports Ajzen’s (2012) supposition that beliefs are related to teachers’ intervention responses (Troop-Gordon & Ladd, 2015), which in turn are related to students’ experiences with peer victimization (Kochenderfer-Ladd & Pelletier, 2008) and their bystander behavior (Hektner & Swenson, 2011). Given the extant evidence, it seems prudent to assess preservice teachers’ attitudes toward bullying as those attitudes can directly impact their future intervention responses and their students’ outcomes (e.g., Demol et al., 2020; Wachs et al., 2019). Absent formal training on the consequences of bullying, preservice teachers may assume that—although concerning—bullying is just a normal part of growing up. As such, we expected preservice teachers to hold normative beliefs about bullying.

Our third question was specific to preservice teachers’ attitudes toward different forms of bullying. Complicating efforts to tackle bullying in schools is that it can take many forms including physical (e.g., hitting), verbal (e.g., calling names), relational (e.g., excluding, manipulating relationships), social (e.g., spreading rumors, gossiping), cyber (e.g., via technology such as text messages or social media), or bias-based against students who are members of a certain social group or protected social class (e.g., students with disabilities, sexual minority youth). Not all forms of bullying are seen as equally serious among teachers. Typically, physical acts are perceived to be more harmful than other forms (e.g., Maunder et al., 2010; Mishna et al., 2005), which does not reflect victims’ own reports of harm from nonphysical forms (Owens et al., 2000). Given the existing evidence that attitudes differ by form of bullying (e.g., perceived seriousness of bullying; Yoon & Kerber, 2003), we expected preservice teachers to perceive physical forms of bullying as more serious than other forms of bullying.

Our fourth research question relates to the third step of the Bystander Intervention Model (Latané & Darley, 1970), which asks whether preservice teachers see themselves as responsible for intervening in bullying. Teachers are tasked with promoting their students’ success at school, part of which includes managing their students’ peer experiences—including bullying—that can significantly impact students’ adjustment (Farmer et al., 2011). As the primary authority figure in the classroom, we expected preservice teachers to acknowledge their responsibility for intervening in incidents of bullying.

Our fifth research question was inspired by the TPB’s emphasis on understanding self-efficacy or confidence to execute an action in order to predict behavior (Ajzen, 2012). We sought to understand preservice teachers’ level of confidence in their ability to deal with bullying among their future students. Those who are more confident tend to intervene more (Fischer & Bilz, 2019; Fischer et al., 2021), underscoring the importance of bolstering teachers’ sense of efficacy for addressing bullying. Unfortunately, recent reviews suggest heterogeneity in teachers’ degree of confidence (Fischer et al., 2021; Macauley et al., 2018). Given critiques from preservice teachers that they are not adequately prepared to deal with bullying (e.g., Fry et al., 2020), we expected preservice teachers to report low confidence for managing bullying.

Current Systematic Review

Interest in assessing preservice teachers’ understanding of bullying has grown in the past few decades, warranting a comprehensive overview of the evidence. Systematic reviews are integral to scientific advances because they synthesize research, provide insight into the state of evidence, highlight future directions, and articulate practical implications for education and policy (Davies, 2000). Our review was guided by the five research questions outlined above. In addition to these questions, we assessed two possible moderators: teacher gender and personal experience with bullying. These two factors can frame how teachers perceive and interpret bullying incidents according to tenets from social information processing model and social cognitive theory (Bandura, 2001; Crick & Dodge, 1994). Understanding whether there are differences in preservice teachers’ readiness to address bullying based on these factors is critical information for implementation efforts to change preservice teachers’ responses to bullying. Both factors have been studied in preservice and in-service teacher samples, suggesting there is sufficient evidence to assess patterns in this current systematic review. For instance, studies show that female and male teachers differ in their attitudes and tolerance toward bullying (W. Craig et al., 2000; Green et al., 2008). Based on the available evidence, we expected female teachers to hold more negative attitudes about bullying (i.e., belief that bullying is harmful). As for prior experience with bullying, research suggests that teachers who were victimized as a student have more negative attitudes about bullying and stronger commitment to helping compared to those who were not victimized (Mishna et al., 2005), leading us to expect previously victimized preservice teachers to hold more negative attitudes toward bullying.

Method

Search Strategy

Our review followed PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for conducting a systematic review (Liberati et al., 2009). Before beginning the review process, we registered our systematic review on Open Science Framework (https://osf.io/wy4sx). We used search terms under three concepts: (1) preservice teachers (e.g., trainee teachers, student teachers), (2) bullying/cyberbullying, and (3) knowledge and attitudes. Our key term search was conducted in October 2019 and yielded a total of 2,829 records from the following databases and searches: ERIC (504 records), Education Source (588 records), PsycINFO (801 records), Web of Science (233 records), PubMED (201 records), ProQuest (202 records), and Google Scholar (limited to the first 300 results; Haddaway et al., 2015; see online Supplemental Appendix SA for key term search). Records were entered into a reference management software to identify duplicates. A total of 589 duplicate records were identified by the software and subsequently removed, yielding a total of 2,240 records that were screened according to the inclusion and exclusion criteria.

Inclusion and Exclusion Criteria

To be included in our review, articles had to meet the following criteria: (a) published between 1993 and 2019 (to increase the likelihood that measures reflected bullying specifically, we included articles published after Olweus’s, 1993, widely cited and available book which outlined the three hallmarks of bullying); (b) published in English; (c) either peer-reviewed article, conference papers/presentation, or dissertation; (d) studies involving preservice teachers who are not already teaching full-time; (e) studies with measures that assessed bullying specifically; (f) studies asking specifically about bullying between school-aged (PK–12) children; and (g) studies with results that addressed at least one of the five specific research questions. Both quantitative and qualitative studies were allowed. Further, for reports in which preservice teachers were a subsample, the study could be included if results specific to preservice teachers could be determined. Studies were excluded for the following reasons: (a) involved in-service teachers already teaching full-time, (b) measures did not include the term bully/bullying, (c) did not involve bullying among school-aged children (e.g., workplace bullying), (d) full texts were not available, (e) study did not address any of the five research questions, or (f) results specific to the subsample of preservice teachers could not be determined based on analyses performed.

The titles and abstracts of all records were independently screened for eligibility by two reviewers and coded as (1) include, (2) unsure, or (3) exclude. Any discrepancies were discussed until mutual agreement was reached. If two coders could not reach consensus, the third coder was consulted to make the final decision. Interrater agreement was assessed with percent agreement between coders and with Cohen’s Kappa that estimates the agreement between two sources and accounts for levels of agreement due to chance alone (Viera & Garrett, 2005). Kappa values from .41 to .60 indicate moderate agreement, values from .61 to .80 represent substantial agreement, and values from .81 to 1.00 represent near perfect agreement (Viera & Garret, 2005). Average interrater agreement for the title and abstract screening phase was high (M = 97.7%), with substantial agreement found among the coders (M_κ = .71).

After initial screening, 2,153 records were excluded for failure to meet the inclusion criteria, leaving a total of 87 records that proceeded to the full-text screening phase. Full texts were independently coded by two reviewers as either (1) include or (2) exclude, and all reasons for exclusion were noted (average interrater agreement = 82.2%, M_κ = .63). All discrepancies were reconciled through mutual agreement. Of the 87 full-text reports screened, 49 reports failed to meet the inclusion criteria and were excluded (see Figure 1 for PRISMA flow diagram). This yielded a total of 39 studies in 38 reports from the initial search that met eligibility criteria.

OPEN IN VIEWER

Figure 1.

Flow diagram of the systematic review selection process studies included in quantitative synthesis (n = 42 studies in 41 reports).

A few exemplars of excluded studies are worth noting. Some reports did not fit our publication criteria, or results specific to preservice teachers were not available such as when studies had combined samples of in-service and preservice teachers (e.g., Ortiz-Bush & Schultz, 2016; Pečjak & Pirc, 2015). We followed recommendations to seek additional information (T. Li et al., 2021) and emailed authors two times, 2 weeks apart for results specific to preservice teachers. Some authors were able to provide the necessary information to meet our eligibility criteria (e.g., Begotti et al., 2017, 2018), but others either did not respond or were unable to provide the requested information for various reasons.

In accordance with recommendations to perform supplemental searches (Higgins et al., 2011), we also (a) performed a continuous search and (b) searched the reference list of included articles for other reports that may be eligible for inclusion. Our continuous search was performed in September 2020 using the same key terms and search strategy to identify any reports published between the first search and publication. A total of 239 articles were identified. Of those, 107 were duplicates, leaving 132 that were screened for inclusion independently by two coders (97% agreement). The majority were excluded (n = 129) leaving three reports that moved on to full-text screening. All three met inclusion criteria for the review. Combining the number of included studies from our initial (n = 39) and continuous search (n = 3) yielded a total of 42 studies in 41 reports. We then searched the reference lists of these included articles and identified 12 for additional screening. Of those 12, 5 were excluded after screening the titles and abstracts, and the remaining 7 were excluded after screening the full-text reports, meaning no additional studies met our eligibility criteria from our supplemental search of the references lists. In sum, a total of 42 studies from 41 reports were included in our narrative synthesis.

Data Extraction

Once the final number of studies was identified, two reviewers independently extracted data from all full-text articles. We coded for publication type (e.g., peer-reviewed article, dissertation) and extracted information about the location of the study (i.e., country). We extracted sample information including (a) sample size, (b) gender breakdown (e.g., % females), and (c) participant age as reported. Information about type of teacher preparation program (e.g., undergraduate, master’s), year of study (e.g., 1st year, 4th year), and preservice teachers’ intended grade to teach (e.g., primary, secondary, special education) were extracted. We also coded for whether preservice teachers in the sample had participated in teaching practicums or had been exposed to bullying topics in their courses. Relevant details from the methods, including the form(s) of bullying assessed and the use—and content—of any bullying definition provided to participants, were also coded. Specifically for studies with an intervention design, additional information from the methods about the length of the intervention, intervention activities, the presence of a control group, the use of randomization, and the length between the pretest and posttest assessment were extracted. For all studies, we extracted descriptive and inferential statistics (e.g., Ms, SDs, %, Fs) and the reported results from relevant statistical or qualitative analyses (see Supplemental Appendix SB, available in the online version of the journal). Inspection of interrater agreement of codes that could be assessed quantitatively indicated moderate to perfect agreement (κ range = .44 to 1.00; Viera & Garrett, 2005). Other criteria (e.g., themes reported in qualitative studies) were compared descriptively based on the content. All discrepancies were resolved through discussion until mutual agreement was achieved.

Risk for Bias Assessment

During the data extraction phase, we assessed risk for bias for all included reports using a previously published assessment framework (Earnshaw et al., 2018) based on recommendations from Cochrane Collaboration (Higgins et al., 2011) to evaluate three common sources of bias as being either low or high: confounding, selection, and information bias. We added a moderate risk category to account for the possibility that some studies may have a combination of risk, as found in the Quality Assessment Tool for Quantitative Studies (Evans et al., 2015). Given that our systematic review included both quantitative and qualitative studies, we determined that this assessment framework was the best option available to accommodate different methodologies while still providing a robust overview of risk for bias in included studies.

Confounding bias occurs when the observed outcomes can be attributed to other factors. Studies that did not consider any confounds (i.e., covariates or controls) in their analyses were coded as high risk, studies that considered one confound were considered moderate risk, and studies that considered multiple confounds were considered low risk. Selection bias can occur when the participants are unrepresentative of the larger population. Studies were coded as being low in selection bias if random selection was utilized and response rates were high (at least 80%). Studies with moderate risk for selection bias had either a 100% response rate for the convenience sample or the response rate was high (> 80%) and the author discussed whether the sample was representative of the larger population. Lastly, studies with low response rates were coded as high risk for selection bias. Information bias occurs when there is a concern over how outcomes were measured. Studies that used established, psychometrically valid, and reliable measures and demonstrated reliability in the study’s sample or used standard qualitative protocols for data collection and analysis were coded as low risk. Studies that modified an established measure or had questionable reliability of an established measure were coded as moderate risk. Studies that used a new measure not previously validated or that did not follow standard protocols for qualitative analysis were coded as high risk.

Two reviewers independently assessed risk for bias. Interrater agreement was between 85.4% and 92.7% across the three bias categories (Mκ = .80, substantial agreement; Viera & Garrett, 2005). Discrepancies were resolved through discussion until mutual agreement could be reached; otherwise, a third reviewer was brought in to make the final determination.

Data Synthesis

We conducted a narrative synthesis, specifically a thematic synthesis of themes identified from the results, which allowed us to coalesce patterns together across different methodologies to answer our research questions (Thomas et al., 2012; Thomas & Harden, 2008). First, results from studies were coded using a combination of deductive and inductive codes: deductive codes aligned with our a priori research questions and hypotheses (e.g., which hallmarks were included in preservice teachers’ definitions of bullying), whereas inductive codes emerged from the results and added depth to the analysis (e.g., preservice teachers’ issue with bullying hallmarks). Second, codes were combined into themes that described patterns in the data that addressed our research questions (e.g., Do preservice teachers know all three hallmarks of bullying?). Finally, we provided narrative descriptions of each theme wherein we synthesized results across all included studies to provide an overall description of preservice teachers’ readiness to address bullying (Thomas et al., 2012).

Results

Overview of Study Characteristics of Included Reports

A listing of all included studies by research question is presented in Table 1. Of the 41 reports describing the 42 included studies, the majority were published in peer-reviewed journals (82.9%, n = 34), followed by dissertations (14.6%, n = 6), and 1 conference paper/presentation. Most of the 42 studies had quantitative designs (78.6%, n = 33), followed by qualitative designs (14.3%, n = 6) and 3 (7.1%) mixed-method designs. Preservice teachers represented the entire sample in the majority of included studies (83.3%, n = 35) and were a subsample in the remaining 7 studies (16.7%). Across all the studies, every form of bullying was assessed in relation to our five research questions, with the most common being physical (n = 14), verbal (n = 13), relational (n = 13), and cyberbullying (n = 11). The least common was bias-based bullying forms (e.g., against sexual minority youth, students with disabilities). Studies originated from 14 different countries: Australia, Canada, China, Finland, Greece, Italy, Norway, Republic of Ireland, South Africa, Spain, Switzerland, Turkey, United Kingdom (specifically England, Scotland, Northern Ireland), and the United States of America (USA). The largest number of studies had samples from the USA (31.7%, n = 13 studies), followed by Canada (14.6%, n = 7 studies) and the United Kingdom (14.6%, n = 7 studies). Preservice teachers were enrolled in one of three types of programs (note that some samples included preservice teachers from two different programs): an undergraduate degree teacher preparation program (most common, n = 34 studies), a postgraduate (e.g., master’s) degree program (n = 14 studies), or an alternative certificate/ accreditation program (n = 3 studies). Preservice teachers were training to teach students at all levels: early childhood, elementary/ primary, middle or high school/ secondary, and special education. Most were preparing to teach at the middle school/ high school/ secondary level (43.4%), followed by the elementary/primary level (35.8%). However, only 28 out of 42 studies explicitly stated what grade level participants intended to teach.

OPEN IN VIEWER

Table 1

Summary of included articles that addressed research questions

Author	RQ1n = 9	RQ2n = 16	RQ3n = 18	RQ4n = 6	RQ5n = 20
Amanaki &Galanaki (2013)		X A			X A
Banas (2014)					X
Bauman & Del Rio (2006): Study 1			X
Bauman & Del Rio (2006): Study 2 a			X
Begotti et al. (2017)			X
Begotti et al. (2018)			X
Benitez et al. (2009)	X				X A
Beran (2005)		X A			X
Boulton et al. (2014)			X A
Cilliers & Chinyamurindi (2020)		X
K. Craig et al. (2011)		X A	X A		X A
W. Craig et al. (2000)			X A
Crooks et al. (2017)		X
Curb (2015)		X		X	X
Davis (2016)			X
Fry et al. (2020)	X		X		X
Garner (2017)		X			X
Garner et al. (2013)			X
Garner et al. (2017)		X
Gonzalez Vega (2016)	X		X	X
Huang et al. (2018)			X
Kahn et al. (2012)			X
Kokko & Pӧrhӧlӓ (2009)					X B
Kyriacou & Uhlemann (2011)				X
Kyriacou et al. (2013)				X
Laferriere (2019)	X				X
Lester et al. (2018)		X A B	X A B		X A B
Q. Li (2009)		X			X
Macaulay et al. (2019)	X
Maynes & Mottonen (2017)					X A
Murphy (2014)		X A			X
Nicolaides et al. (2002)		X			X A B
O’Haver (2012)			X
Psalti (2017)			X
Purdy & Mc Guckin (2015)					X
Purdy & Mc Guckin. (2014)		X			X
Raven & Jurkiewicz (2014)	X	X		X	X
Redmond et al. (2018)	X			X
Rose et al. (2018)	X		X B
Ryan et al. (2011)		X			X
Small et al. (2013)	X		X
Yilmaz (2010)		X A			X

Note. Articles that addressed research questions (RQs) are marked with an X. Moderation by preservice teachers’ gender is marked with an ^A; moderation by preservice teachers’ prior victimization experience is marked with a ^B.

a

Bauman & Del Rio’s (2006) Study 2 with supplemental sample.

What Do Preservice Teachers Know About the Hallmarks of Bullying?

A total of 9 studies (21.4% of the 42 included studies) assessed preservice teachers’ definitions of bullying, 2 studies using quantitative (i.e., survey) methods, 5 using qualitative (e.g., focus groups, interview) methods, and 2 intervention studies (1 quantitative, 1 mixed methods). No study examined moderation in preservice teachers’ definitions of bullying by participants’ gender or prior experience with victimization (see Table S1 in Supplemental Appendix SB available in the online version for detailed information about the sample, method, results, and a summary of the findings for studies addressing our first research question).

Definition of bullying includes all three hallmarks

Only one study found that preservice teachers considered all three hallmarks to be important to their definition of bullying. Rose and colleagues (2018) asked participants to rate how critical the different hallmarks of bullying were to their personal definition of bullying. The results for intent, repetition, and power imbalance indicated that all three hallmarks were moderately critical to preservice teachers’ personal definition.

Definition of bullying missing one or more hallmarks

The majority of studies (six out of nine) found that one or more of the bullying hallmarks were missing from preservice teachers’ definition of bullying (Fry et al., 2020; Gonzalez Vega, 2016; Macaulay et al., 2019; Raven & Jurkiewicz, 2014; Redmond et al., 2018; Small et al., 2013). Power imbalance was not included as part of the definition of bullying in three of the studies using interviews or focus groups (Gonzalez Vega, 2016; Macaulay et al., 2019; Raven & Jurkiewicz, 2014). Repetition and intent to harm were not included in preservice teachers’ definition in two studies (definitions solicited from interviews, focus groups, or coded from their discussion board posts; Gonzalez Vega, 2016; Raven & Jurkiewicz, 2014; Redmond et al., 2018).

Conflicting views on hallmarks

One theme that emerged from the evidence was that there is some disagreement among preservice teachers about the necessity of the repetition (Fry et al., 2020; Macaulay et al., 2019) and intent to harm hallmarks (Raven & Jurkiewicz, 2014) for the definition of bullying. One preservice teacher articulated this issue succinctly for repetition:

I would argue it can even start with one . . . because where do you draw the line otherwise from when the bullying begins and ends. If somebody’s saying I’ve been subjected to an incident of bullying, you wouldn’t say to that, to that person who would feel like they’re bullied immediately . . . well I’d have to wait two or three more times to see how it rests on me. (Macaulay et al., 2019, p. 5)

For intent to harm, Raven and Jurkiewicz (2014) reported that some preservice teachers believed that harm must be intentional in order for it be considered bullying whereas others felt that intent was not as important as the impact of harm.

Interventions to improve understanding

Two intervention studies showed improvement in preservice teachers’ knowledge of bullying following an intervention. Benitez and colleagues (2009) compared preservice teachers who took a semester-long course on bullying (i.e., intervention group) to preservice teachers enrolled in another elective (i.e., control group). They found that at posttest, participants in the intervention included significantly more hallmarks in their definition compared to those in the control, especially the hallmarks of intentionality and repetition. However, the authors reported that there was no difference in the inclusion of power imbalance from pretest to posttest for preservice teachers in the intervention group (Benitez et al., 2009). Laferriere (2019) compared students that received a normal lecture on bullying (i.e., control group) to students in a course that received the normal lecture plus a video game intervention (i.e., intervention group). Results showed a significant increase in the intervention group’s ability to identify all characteristics of bullying compared to the control group (Laferriere, 2019).

Summary

In sum, only one study’s results suggested that preservice teachers define bullying in terms of all the hallmarks of bullying (Rose et al., 2018). More studies reported that preservice teachers were missing one or more hallmarks in their definitions of bullying. Lastly, as evidenced in the two intervention studies, it is possible to improve preservice teachers’ knowledge of bullying hallmarks.

What Are Preservice Teachers’ Beliefs and Attitudes About Bullying?

Sixteen studies (35.7% of included studies) investigated preservice teachers’ beliefs and attitudes about bullying. Of the studies assessing attitudes and beliefs, 14 used quantitative methods to survey attitudes, 1 used qualitative methods (i.e., focus groups) and 1 study assessed the impact of an intervention on preservice teachers’ attitudes and beliefs. Six studies examined moderation in preservice teachers’ definitions of bullying by participants’ gender, and one study examined whether prior experience with victimization moderated their attitudes and beliefs (see Table S2 in Supplemental Appendix SB available in the online version for sample characteristics and a summary of findings for the studies addressing our second research question).

Preservice teachers believe bullying is normative

Six studies addressed whether preservice teachers held normative beliefs about bullying, and results from five of the six suggest that preservice teachers do believe, to some extent, that bullying is a normal part of growing up. In two studies, preservice teachers, on average, held normative beliefs about peer victimization by agreeing or strongly agreeing with statements such as “teasing other children is just part of growing up” and “fights . . . help them learn to stand up for themselves” (Garner, 2017; Garner et al., 2017). Likewise, preservice teachers tended to agree or strongly agree with statements capturing dismissive beliefs toward victimization such as “compared to all of the other problems boys/girls have, being picked on is relatively minor” (Garner, 2017; Garner et al., 2017).

Other evidence of normative beliefs came from studies reporting the percentage of teachers who endorsed normative statements. These studies reported that approximately one in three preservice teachers believe bullying is just a natural part of growing up (34%; Curb, 2014), “that most youth will get bullied at some point during their school years—it is just a part of being a kid” (33%; Crooks et al., 2017), and that youth will never be able to practice how to solve conflict on their own if adults intervene in every incident of bullying (27.9%; Crooks et al., 2017). A smaller percentage of preservice teachers believed that being bullied can help victims handle difficult situations in the future (7.1%; Crooks et al., 2017).

This normative belief about bullying was also captured in focus groups with preservice teachers. Raven and Jurkiewicz (2014) reported that whereas most of the preservice teachers in their sample agreed that bullying was wrong, some also believed that it was just a normal part of life. For example, one of their participants stated,

You can’t shelter these kids their entire life. It’s real life. It’s going to be said [sic]. It’s going to happen. And I feel like some of them, at one point in time in their life, need to hear something like that, just to realize the world’s not sunshine and rainbows. (Raven & Jurkiewicz, 2014, p. 68)

Preservice teachers do not believe bullying is normative

Only one study out of the six that assessed beliefs about whether bullying is a normal part of growing up found evidence in the opposite direction as those described above (Lester et al., 2018). In Lester and colleagues’ (2018) study, preservice teachers, on average, disagreed or strongly disagreed with statements such as “bullying is just a part of school life which should be accepted” and “students who are bullied usually deserve what they get.”

Preservice teachers are concerned about bullying

A common theme from the results reported across several studies is that preservice teachers are concerned about bullying and cyberbullying specifically. This was reflected in survey response averages and in the percentage of preservice teachers in the sample that agreed or disagreed with statements that reflected different attitudes towards bullying. For instance, based on the reported averages in one study, preservice teachers agreed or strongly agreed that bullying is concerning and a problem (K. Craig et al. 2011).

Most studies reported the percentage of their sample that agreed with statements assessing attitudes. The theme that emerged from these studies was that the majority of preservice teachers held negative attitudes about bullying and were concerned about bullying among their students (e.g., Amanaki & Galanaki, 2013). For instance, studies reported that a majority of preservice teachers (range from 64%–96.9%) agreed with the following statements: concerned about bullying, children are affected, there are long-term consequences of bullying for victims, and bullying was a serious problem at school (Beran, 2005; Curb, 2014). In line with these trends, three other studies found that more than 90% of preservice teachers saw bullying at school as a serious, important, and universal issue needing both attention and action (Crooks et al., 2017; Nicolaides et al., 2002; Purdy & Mc Guckin, 2014).

Four studies focused specifically on preservice teachers’ attitudes toward cyberbullying, and the pattern of results paralleled the pattern from studies on traditional bullying: a majority of preservice teachers were concerned about cyberbullying (77.3%–81%), believed children were affected (85.9%–94%), and believed that cyberbullying was a problem in schools (72%–80%; Ryan et al., 2010; Yilmaz, 2010). Another study found that 92% of preservice teachers believed cyberbullying was wrong and 90% believed it would be a serious issue for students (Cilliers & Chinyamurindi, 2020). Combined, most studies reported that a majority of preservice teachers held negative attitudes about bullying and cyberbullying, believing it to be a harmful problem that they were concerned about, but it is important to note that the percentages varied across studies.

Majority of preservice teachers not concerned about bullying

In contrast, a study by Q. Li (2009) found that only a minority—about one in three (31.9%)—of preservice teachers agreed that cyberbullying was a problem in school, and less than half of the sample (49.7%) were concerned about bullying.

Intervention to change attitudes

Only one study in this section had an intervention design. Murphy (2014) assessed preservice teachers’ beliefs and concerns about bullying before and after receiving an intervention modeled after the Steps to Respect program. Results indicate that after a 1-hour presentation, the program did not change participants’ attitudes toward bullying except for increasing their belief in cyberbullying’s impact (Murphy, 2014).

Moderation effects

Of the 16 studies that assessed attitudes, 6 assessed for moderation by preservice teachers’ gender, and 1 assessed moderation by prior experience with victimization. Four of the six studies reported significant gender differences, with females reporting more concern for bullying in general and cyberbullying specifically (Amanaki & Galanaki, 2013; Beran, 2005; K. Craig et al., 2011; Yilmaz, 2010), higher belief that children were affected by cyberbullying, and higher belief that cyberbullying was a problem in schools (Yilmaz, 2010) compared to male preservice teachers. Two studies found no gender differences in the level of concern about bullying (cyberbullying; Murphy, 2014) or in their attitudes toward teachers helping students being bullied (Lester et al., 2018). Only one study examined moderation by prior experience with victimization and the authors reported no significant differences, indicating that regardless of whether preservice teachers were bullied during secondary school, they held similar attitudes toward teachers helping bullied youth (Lester et al., 2018).

Summary

Combined, most studies reported that a majority of preservice teachers were concerned about bullying, held negative attitudes toward bullying, and recognized its harm and the serious problem it can be at school. However, some preservice teachers held normative beliefs that bullying and victimization is just a normal part of development. As for moderation patterns, the evidence indicates that female preservice teachers held more negative attitudes about bullying compared to male preservice teachers. Only one study assessed moderation by prior victimization and found no differences in preservice teachers’ attitudes, but replication of this finding is needed.

What Are Preservice Teachers’ Attitudes Toward Different Forms of Bullying?

A total of 18 studies (42.9% of included studies) assessed attitudes toward different forms of bullying. These studies assessed differences in perceived seriousness or harm of the different forms. Sixteen studies used quantitative methods and 2 employed qualitative (i.e., interview) methods to understand differences in preservice teachers’ perceptions. Five studies examined moderation by gender, and only two studies examined moderation by prior victimization experience (see Table S3 in Supplemental Appendix SB available in the online version for a summary of findings for studies relevant to this research question).

Perceived differences in seriousness by bullying form

Fourteen of the 18 studies (2 qualitative, 2 mixed-methods, and the rest quantitative designs), found that preservice teachers had different perceptions of the seriousness of bullying depending on the form of bullying. The most common pattern across these studies was that physical bullying was seen as more serious than other forms of bullying. Specifically, when compared to verbal and relational bullying, physical bullying situations were rated as significantly more serious than both of those forms (Bauman & Del Rio, 2006; W. Craig et al., 2000; Garner et al., 2013; Huang et al., 2018; O’Haver, 2012; Small et al., 2013). Physical bullying was also seen as more serious compared to cyberbullying (Boulton et al., 2014; K. Craig et al., 2011), biased-based bullying targeting sexual minority youth (i.e., homophobic bullying; K. Craig et al., 2011), and combined social and verbal bullying (e.g., making up rumors, teasing another student; Kahn et al., 2012). One study asked participants to rank order vignettes depicting different forms and found that the two physical bullying scenarios were more consistently ranked as the first and second most serious bullying situations (Bauman & Del Rio, 2006). The belief that physical bullying is more serious than other forms was likewise found in a qualitative study involving interviews with preservice teachers (Fry et al., 2000). As one participant argued, “Obviously it’s important that this form of bullying [ostracization] stops, but on the other hand, you’ve also got physical violence, where people are maybe in danger more so than being called a name” (p. 4).

It is worth noting that not all studies reporting differences in perceived seriousness found physical bullying to be the most serious form out of all forms assessed. Two studies reported that verbal bullying was seen by preservice teachers to be more serious than physical bullying: one quantitative (Psalti, 2017) and one qualitative (Gonzalez Vega, 2016). One preservice teacher interviewed as part of Gonzalez Vega’s (2016) study described it as follows: “The verbal bullying is the most serious because it hurts people more than the others. If you get a fist, you continue forward, but what they say hurts you in your mind and hurts you in a psychological way” (p. 81). Another study reported that physical and verbal bullying were seen as equally serious (although still more serious than relational and cyberbullying; Davis, 2016).

Following physical bullying, verbal bullying was typically viewed as more serious than relational bullying (Bauman & Del Rio, 2006; Boulton et al., 2014; W. Craig et al., 2000; Davis, 2015; Gonzalez, 2016; Huang et al., 2018; O’Haver, 2011; Psalti, 2017). However, two studies found no significant difference in perceptions of seriousness between the verbal and relational bullying (Garner et al., 2013; Small et al., 2013).

A few studies included cyberbullying as a form, but the results for perceived seriousness of cyberbullying compared to the other forms was mixed and no clear, consistent pattern emerged from the results across studies. For instance, compared to relational bullying specifically, some preservice teachers perceived cyberbullying to be more serious (Boulton et al., 2014), some saw it as less serious (K. Craig et al., 2011), whereas others saw no differences in seriousness between the two (Davis, 2016).

No differences in perceived seriousness by bullying form

Lastly, 4 of the 18 studies reported no overall differences between seriousness across all forms of bullying assessed. Specifically, preservice teachers rated similar levels of seriousness between overt (physical or verbal) and covert (social exclusion) bullying (Begotti et al., 2017, 2018); traditional and cyberbullying (Lester et al., 2018); or between physical, verbal, indirect, and sexual bullying (Rose et al., 2018).

Moderation effects

Five studies assessed whether perceptions of seriousness were moderated by preservice teachers’ gender. Four of the five studies found no significant gender moderation in ratings of seriousness or harm among forms of bullying (physical, verbal, relational, and cyberbullying; Boulton et al., 2014; physical, verbal, and relational; W. Craig et al., 2000; Garner et al., 2013; traditional and cyberbullying; Lester et al., 2018). Only one study found evidence of moderation effects: Female preservice teachers perceived homophobic (bias-based bullying against sexual minority youth) and cyberbullying to be more serious than male preservice teachers (K. Craig et al., 2011).

Only two studies examined whether preservice teachers’ prior experience with victimization moderated their ratings of seriousness by bullying form. Lester and colleagues (2018) found no significant differences in perceptions of harm for traditional versus cyberbullying depending on preservice teachers’ victim status as a student. Rose and colleagues (2018) asked preservice teachers to indicate how often they were victimized when they were in school, from which the authors created three groups: seldom victims, occasional victims, and frequent victims. They found no significant differences in perceptions of seriousness of physical, verbal, indirect, or sexual bullying forms by victim type (Rose et al., 2018).

Summary

Altogether, the results across the 18 studies that addressed this research question indicate that preservice teachers tend to perceive differences in the seriousness of bullying depending on the form of bullying. Of those that found differences in seriousness, physical bullying was often seen as the most harmful form. The pattern of results from studies assessing moderation effects suggest that preservice teachers usually view similar levels of seriousness across bullying forms regardless of their gender. Similarly, results from the limited studies assessing moderation by prior experience found no significant differences in perceptions of seriousness by preservice teachers’ victim status.

What Is Preservice Teachers’ Sense of Responsibility to Deal With Bullying?

A total of six studies (14.3% of included students) assessed preservice teachers’ perceptions of their level of responsibility as teachers to intervene in bullying. Half of the studies used quantitative methods and the other half used qualitative methods. No study examined moderation in sense of responsibility by gender or prior victimization experience (see Table S4 in Supplemental Appendix SB available in the online version for information about the studies addressing this research question).

Preservice teachers acknowledge their role in addressing bullying

Most studies reported that—at least to some degree—preservice teachers recognized their responsibility to deal with bullying at school. For example, one study found that 82% of preservice teachers disagreed or strongly disagreed with the statement, “dealing with bullying will not be one of my responsibilities as a teacher,” which indicates that a majority of preservice teachers recognize that managing bullying among their students will be part of their future responsibilities (Curb, 2014, p. 85). Similarly, two studies found that preservice teachers believed that teaching staff (e.g., head teachers, classroom teachers) should take a major role—and often the lead role—in dealing with bullies and victims (Kyriacou et al., 2013; Kyriacou & Uhlemann, 2011). Recognition of the responsibility to deal with bullying was also reported by preservice teachers in two qualitative studies (Gonzalez Vega, 2016; Raven & Jurkiewicz, 2014).

Preservice teachers not solely responsible

Redmond and colleagues (2018) coded preservice teachers’ discussion board posts about managing cyberbullying and found that some preservice teachers acknowledged the role and responsibility of teachers to deal with cyberbullying specifically, but a larger percentage of their posts emphasized the role and responsibility of the school. Additionally, preservice teachers in focus groups shared that part of their responsibility would be to bring in other personnel (e.g., social worker) who would then be responsible for addressing the bullying (Gonzalez Vega, 2016).

Summary

Collectively, the evidence indicates that most preservice teachers accept that they have some responsibility to intervene in bullying at school. However, some see other school personnel as also bearing responsibility to deal with bullying.

What Is Preservice Teachers’ Level of Confidence to Address Bullying?

A total of 20 studies (47.6% of included studies) provided evidence to address this research question, 2 of which used qualitative methods and 4 of which were intervention studies. Seven studies assessed moderation in confidence by gender, and only two studies examined moderation by prior victimization experience. Studies asked about confidence or self-efficacy for a number of tasks related to dealing with bullying such as identifying bullying, managing it, working with bullies, and so forth (see Table S5 in Supplemental Appendix SB for a summary of study findings concerning confidence in addressing bullying).

Majority of preservice teachers are not confident

The most common theme that emerged was that preservice teachers do not feel overly confident in their ability to address bullying in various ways. This was evident in the average response from preservice teachers in Garner’s (2017) study where preservice teachers tended to disagree that they felt confident to identify or manage bullying. Additionally, several studies found that a minority (i.e., less than half) of preservice teachers reported feeling confident to deal with bullying, identify it, or manage it (23.3%–41.9%; Amanaki & Galanaki, 2013; Beran, 2005), deal with bullies specifically (46%; Curb, 2015), identify or manage cyberbullying (11.1%–48.5%; Q. Li, 2009; Ryan et al., 2011; Yilmaz, 2010), or deal with biased-based bullying against students with disabilities (Purdy & Mc Guckin, 2014).

This lack of confidence to deal with bullying was echoed in qualitative studies wherein some preservice teachers shared that they do not feel confident to manage bullying or know what to do if a situation occurred (Purdy & Mc Guckin, 2015; Raven & Jurkiewicz, 2014). Specifically, preservice teachers indicated they lack the confidence to respond to instances of bias-based bullying such as against students with disabilities (Fry et al., 2020; Purdy & Mc Guckin, 2015) or for instances of bullying that can occur in science classrooms around controversial topics (e.g., bias-based bullying against people with certain religious beliefs, against different races; Raven & Jurkiewicz, 2014).

Preservice teachers have some degree of confidence

A few studies found that preservice teachers felt some degree of confidence to deal with bullying. Two studies found that preservice teachers reported being somewhat confident (i.e., above the theoretical mean) they could respond appropriately if a student was being bullied or bullying (Maynes & Mottonen, 2017) and reported some degree of confidence in their perceived skills to discuss, identify, manage, and prevent bullying (Lester et al., 2018). Another study found preservice teachers reported feeling somewhat confident to do the following: deal with a bullied student, talk with onlookers about their responsibility with bullying, help onlookers take a more active role in bullying, talk with a bullied students without blaming them for their victimization, work with the parents of victims and bullies, talk with bullies without blaming them, or make bullies stop bullying (Nicolaides et al., 2002).

Majority of preservice teachers are confident

A few studies found that a majority of the sample (> 50%) reported being confident to deal with bullying (65%; Purdy & Mc Guckin, 2014), identify bullies, or stop bullying (51.5%–66%; Curb, 2015; Yilmaz, 2010).

Interventions

Four studies involved interventions designed to increase preservice teachers’ confidence to address bullying. All four found that preservice teachers’ level of self-efficacy or confidence for bullying-related tasks increased from pretest to posttest. The lengths of the interventions ranged from a brief 1-hour presentation modeled after the Steps to Respect program (Murphy, 2014) to semester-long courses that included training elements such as a video game (Laferriere, 2019), an elective course covering bullying and school violence (Benitez et al., 2009), or authentic learning exercises (Banas, 2014). Only two of the studies included a control group (Benitez et al., 2009; Laferriere, 2019), and analyses revealed that the intervention group had an increase in bullying self-efficacy but the control group did not (Benitez et al., 2009; Laferriere, 2019).

Moderation effects

A total of eight studies assessed whether confidence differed by preservice teachers’ gender and the results were mixed. Three studies found that female preservice teachers reported lower levels of confidence, especially to identify bullying, manage bullying, or do something about the bullying (Amanaki & Galanaki, 2013; Beran, 2005; Yilmaz, 2010). The remaining six studies found no significant difference in confidence by preservice teachers’ gender (K. Craig et al., 2011; Lester et al., 2018; Maynes & Mottonen, 2017; Murphy, 2014; Nicolaides et al., 2002). Three studies assessed moderation in confidence by prior victimization experience, and all reported that confidence did not significantly differ depending on preservice teachers’ victimization status as a student (Kokko & Pӧrhӧlӓ, 2000; Lester et al., 2018; Nicolaides et al., 2002).

Summary

In sum, there is more evidence to suggest that preservice teachers lack confidence to deal with bullying in their classrooms. However, it is important to acknowledge that some studies found that preservice teachers reported feeling confident. The evidence for gender differences suggest that females and males tend to have similar levels of confidence; but when there were differences, females had lower levels of confidence compared to male preservice teachers. Prior victimization experience does not seem to be related to their level of confidence. Lastly, evidence suggests that interventions may help increase preservice teachers’ confidence to manage bullying.

Risk for Bias

Figure 2 presents a graph of the patterns in risk in the 41 reports. Approximately 40.5% of reports (n = 17) were coded as high risk for confounding bias, meaning that no relevant confounds were considered in analyses. A total of 15 reports (35.7%) were coded as low risk because they considered multiple variables potentially related to the outcomes of interest. Some exemplars include K. Craig et al. (2011), Crooks et al. (2017), Lester et al. (2018), and Rose et al. (2018), which considered demographic characteristics (e.g., gender, age), prior experience with bullying, type of preservice teacher program, teaching division (e.g., primary, secondary focus) and year in program (e.g., 1st year, 2nd year), among others. The most problematic source of bias was selection bias with 76.2% (n = 32; 80% of quantitative studies, 66.7% of qualitative studies) of reports being considered high risk. Most of these studies used convenience samples from teacher education programs, and many reported low response rates. Further, only nine studies descriptively discussed whether or not the study’s sample reflected the population of interest on demographic characteristics. With regard to information bias, approximately 35.7% of the studies (n = 15, all quantitative studies) posed high risk with most of these using measures created specifically for the current study. Roughly 33.3% of the studies (n = 14) were considered moderate risk with most modifying an existing measure. Thus, over two-thirds of the studies lacked robust evidence of validity and reliability.

OPEN IN VIEWER

Figure 2.

Risk for bias of included studies.

Discussion

This systematic review is the first to coalesce findings related to preservice teachers’ understanding of, attitudes toward, and readiness to address bullying. We sought to answer five related research questions guided by two theories: the Bystander Intervention Model (Latané & Darley, 1970) and the TPB (Ajzen, 2012). Generally, the evidence suggests that most preservice teachers do not define bullying in terms of the hallmarks of bullying. A majority of preservice teachers were concerned about bullying, but some believe bullying is a normal part of growing up. Many preservice teachers also viewed different forms of bullying as more serious than others, with physical bullying often viewed as the most serious. Lastly, most preservice teachers in these studies acknowledged their responsibility to intervene in bullying as a teacher, yet a majority of preservice teachers reported not feeling confident in their ability to do so.

Preservice Teachers’ Knowledge of the Hallmarks of Bullying

We found some support for our expectation that preservice teachers do not define bullying in terms of the three hallmarks of intent to harm, repetition, and power imbalance. When asked to define or discuss bullying, most studies reported that preservice teachers failed to reference all three hallmarks of bullying. This result suggests that preservice teachers may lack foundational knowledge of the key features of bullying. However, caution must be used when generalizing this finding given the high risk for bias found in these studies: seven out of the nine studies addressing this research question had high risk for selection bias. In contrast, no study had high risk for information bias, indicating that the methods used to collect preservice teachers’ definitions of bullying were less problematic. Thus, even though the representativeness of these findings is in question, the results do suggest there is cause for concern that preservice teachers may not fully understand what bullying is. Not understanding the key characteristics of bullying can theoretically hinder teachers’ ability to recognize bullying when it happens, which, according to the Bystander Intervention Model, will reduce the likelihood that teachers will intervene in bullying (Nickerson et al., 2014). It is important to note that this phenomenon is not limited to preservice teachers: Some in-service teachers likewise struggle to define bullying in terms of its hallmarks (Naylor et al., 2006).

Notably, the finding that preservice teachers did not fully understand the definition of bullying emerged from studies in which preservice teachers were asked to freely recall or discuss their definition of bullying either through interviews, focus groups, or by providing a response to an open-ended question. In contrast, the one study that found the hallmarks to be moderately critical to preservice teachers’ definitions provided participants with the hallmarks in a survey and asked participants to rate them, perhaps causing a priming effect (Rose et al., 2018). When confronted with a bullying episode in the classroom, preservice teachers need to be able to recall from memory the hallmarks of bullying so that they can immediately assess the situation and determine the appropriate course of action to stop the bullying and support the victim. It is concerning that these studies indicate such information is not part of some preservice teachers’ knowledge base, which suggests that they may need additional training to ensure they will be able to recall these critical facts when bullying occurs between their students.

One interesting theme that emerged from a few studies was debate about the necessity of different bullying hallmarks, specifically repetition and intent to harm. This finding is not unique to preservice teachers: In-service teachers have also excluded repetition in their definitions of bullying (Mishna et al., 2005). Further, whether repetition should be required in order for an act to be considered bullying or how it should be operationalized is an ongoing debate in the field, spurred, in part, by the rise in cyberbullying (e.g., Smith et al., 2013). Nevertheless, we argue that preservice teachers need to know the hallmarks of bullying so that they can effectively intervene in bullying situations, particularly in light of research that indicates that youth who are victims of bullying characterized by several hallmarks are at heightened risk for poor outcomes (Ybarra et al., 2014). Thus, if a teacher sees a student being repeatedly harmed by someone with more power, not intervening on behalf of the victim can have potentially devastating consequences (Hawker & Boulton, 2000; McDougall & Vaillancourt, 2015).

That said, we argue that teachers should respond to the plight of all victims, not just those whose victimization experiences include clear bullying hallmarks. This sentiment was articulated by a preservice teacher in Macaulay and colleagues’ (2019) study who questioned whether teachers need to know if the student is being victimized repeatedly before they intervene. Given that victims of general aggression still have problems—even if their victimization does not include all of the hallmarks of bullying (Ybarra et al., 2014)—preservice teachers should be ready to intervene in all victimization experiences; but their knowledge of the bullying definition is still critical to their ability to notice, assess, and manage bullying in their classrooms.

Preservice Teachers’ Beliefs and Attitudes About Bullying

We found partial support for our expectation that preservice teachers would hold normative beliefs about bullying: A few studies found that preservice teachers, on average, agreed with statements that bullying was a normal part of growing up, whereas other studies report that only some preservice teachers hold these views (approximately one in three; Crooks et al., 2017; Curb, 2014). Unfortunately, all but one of the studies had high risk for selection bias, indicating that these findings may not accurately reflect the population of preservice teachers. Further, information bias was also an issue for this group of studies, with 7 out of 16 having high risk for bias, indicating that caution must be exercised when interpreting preservice teachers’ attitudes toward bullying.

If a teacher believes that being bullied is normal because it is a useful process for teaching victims important coping skills, they may not intervene (Bauman, 2008) even if they recognize that the process will be painful for the victims. Indeed, the more teachers endorse normative beliefs, the more likely they are to tell victims to deal with the situation on their own (Troop-Gordon & Ladd, 2015). Further, normative beliefs are associated with a lower likelihood of reprimanding aggressive students, which is counter to most recommendations for effective practice when dealing with bullying (i.e., firm discipline methods; Ttofi & Farrington, 2011). Thus, it seems that more work is needed to counter these normative beliefs in preservice teachers so that they understand the need to support victims and directly address bullying incidents in their classrooms.

Other evidence of teachers’ attitudes about bullying revealed that a majority of preservice teachers are concerned about bullying and recognize that it is an important problem at school and that there can be significant consequences for victims. This pattern of concern was found for bullying in general and for cyberbullying specifically. Only one study found evidence that a majority of preservice teachers were unconcerned about cyberbullying (Q. Li, 2009). In contrast, a recent study by Cilliers and Chinyamurindi (2020) reported that over 90% of preservice teachers believed cyberbullying was an issue. We speculate that one possible explanation for this difference relates to when the studies were conducted. There have been tremendous changes in technology over the last decade, notably the ubiquitous use of smart phones and social media (Pew Research Center, 2018), which has made texting and social media the most common forms for cyberbullying (Kowalski et al., 2020). The ability of perpetrators to target their victims using technology has extended the reach of bullies: Whereas in the past, victims could experience some reprieve from bullying once they left the school grounds, the use of technology means that the torment can occur at any time (e.g., Tokunaga, 2010). Future preservice teachers will have come of age during a time when there is heightened awareness of cybervictimization’s impact (e.g., academic outcomes; Gardella et al., 2017) and may have their own personal experiences with this form of bullying. Such knowledge may lead to increased concern for cyberbullying specifically and bullying more generally among future preservice teachers, but more research is needed to ascertain whether that is the case.

There is some evidence that these two, seemingly contradictory, perspectives—bullying is normative but also concerning—are not mutually exclusive. One study reported that about one in four (27.9%) preservice teachers endorsed a normative belief, and yet 90% of that same sample indicated that they saw bullying as a serious and important issue for schools (Crooks et al., 2017). The fact that some preservice teachers endorse normative beliefs means we need to help them understand that the coping skills they think victims may gain from being bullied (e.g., ability to regulate their emotions or behavior in a stressful event; Compas et al., 2001) are unlikely to happen within the context of a victimization experience (e.g., Kanetsuna & Smith, 2002). Instead of thinking that bullying serves a purpose, we need teachers to strongly endorse antibullying attitudes because when students believe their teachers think bullying is wrong, bullying decreases, and victims are more likely to seek help from adults (Blomqvist et al., 2020; Saarento et al., 2015).

Preservice Teachers’ Attitudes Towards Different Forms of Bullying

As expected, we found evidence to support our hypothesis that preservice teachers would view different forms of bullying as more serious than others. The majority of studies reviewed for this research question revealed that preservice teachers’ perceptions of seriousness differed depending on the form of bullying. The most common pattern across studies reporting differences was that preservice teachers viewed physical forms as more serious than other forms (e.g., verbal, relational). After physical bullying, verbal bullying was typically viewed by preservice teachers as more serious than relational bullying. A few other forms of bullying were compared; however, given the variability in the results, no clear patterns emerged. For instance, cyberbullying was viewed as more serious, less serious, and equally serious to relational bullying (Boulton et al., 2014; K. Craig et al., 2011; Davis, 2016). A handful of studies reported no differences in perceived seriousness by bullying form (e.g., Begotti et al., 2017, 2018; Lester et al., 2018; Rose et al., 2018), but that pattern seems to be the exception rather than the rule. Our confidence in the generalizability of these findings is low given that most studies had high risk for selection bias. However, only 3 of the 18 studies had high risk for information bias (most used the original or modified version of the Bullying Attitudes Questionnaire; Bauman & Del Rio, 2006; Craig et al., 2000; Yoon & Kerber, 2003), giving greater confidence in the methods used to assess differences in seriousness by the form of bullying.

Judging a form of bullying to be serious indicates that the observer believes the resulting harm is significant and warrants immediate attention. Therefore, it is problematic if teachers discount the seriousness of other forms of bullying, as this may inhibit their intervention responses. For instance, a common finding in the included studies was that physical bullying was seen as more serious than relational bullying, which likely stems from the perception—albeit incorrect—that relational bullying is less harmful to victims (e.g., Bauman, 2008). Being victimized by relational bullying can be just as distressing for victims, and research suggests this form of victimization uniquely contributes to poorer outcomes beyond that of other forms (Chester et al., 2017). Indeed, students themselves report that social exclusion is particularly painful (Owens et al., 2000). We know from developmental research that peers and peer acceptance become increasingly important to youth, which means that any form of bullying that threatens their acceptance among peers has the potential to inflict harm (e.g., Adler & Adler, 1998). Thus, the finding that preservice teachers discount the seriousness of some forms of bullying suggests that either (a) preservice teachers lack knowledge about development or (b) they are not integrating what they do know about development to assess the harmful nature of bullying forms. If the goal is to have teachers intervene in all instances of bullying, regardless of form, then it is important to train future teachers both in development and in bullying to ensure that they view all forms as equally serious and as needing intervention.

Preservice Teachers’ Sense of Responsible for Intervening in Bullying

In line with our expectations, studies suggest that most preservice teachers recognize they have some responsibility to deal with bullying. That said, there are concerns with the fact that over half of the studies addressing perceptions of responsibility had high risk for confounding, selection, and/or information bias, which compels us to exercise caution in our interpretation. Keeping these risks in mind, it was encouraging to see that preservice teachers in these studies acknowledged that they will be responsible, to some extent, for addressing bullying incidents in their classrooms. Teachers are responsible not only for instruction and managing their classroom but also for promoting the social development of their students (e.g., Farmer et al., 2011), and the experience of bullying can obviously hinder positive social outcomes. However, it is important to note that bullying does not occur in a vacuum: It involves the peer group (Salmivalli, 2010), and it is part of larger peer social dynamic processes in the classroom that should be carefully monitored and managed by teachers to help students develop positive peer roles, reputations, and relationships (Farmer et al., 2011, 2019). For instance, teachers can create opportunities for a student with few friends to interact with prosocial peers in the classroom as a means to foster positive peer relationships. Unfortunately, some teachers may view managing these dynamics and their students’ peer experiences to be beyond their purview (Gest et al., 2014), preferring instead to focus on their responsibility to instruct. Such a stance seems counterintuitive in light of the evidence linking victimization experiences to academic problems (Gardella et al., 2017). Indeed, a myopic focus on academics alone is argued to undermine youths’ mental health and their development as well-adjusted citizens (Schonert-Reichl & Hymel, 2007). Thus, even though it is promising to find that preservice teachers generally recognize their responsibility to address bullying, our perspective is that teachers should not wait until bullying occurs before they involve themselves in the social lives of their students.

It is also worth highlighting the fact that even though preservice teachers did—in general—acknowledge some degree of responsibility to address bullying, there appears to be uncertainty over who is primarily responsible at school. When roles and responsibilities are unclear, preservice teachers may assume that someone else in the school will deal with it. Such ambiguity in roles can leave victims slipping through the cracks. This points to the need for clear policies that outline who is responsible and when teachers should involve other school personnel (counselors, school psychologists; Fry et al., 2020) and points to the urgent need to make sure all teachers are well-versed in these policies. To be clear, other stakeholders are also responsible for addressing bullying (e.g., Bellmore, 2016), and society at large has a role to play. But in the school system, teachers are the first line of defense.

Preservice Teachers’ Confidence to Address Bullying

Lastly, while evidence suggests that preservice teachers lack the confidence to deal with bullying, there was considerable variability in the confidence measures and questionable validity and reliability of those measures. Just under half of the studies on confidence had high risk for information bias, which reduces the certainty of these results. Some studies asked about confidence to deal with bullying more broadly, whereas other studies asked about confidence to handle specific tasks such as identifying bullying, managing bullying, working with victims or bullies, and so forth. Further, some studies asked about preservice teachers’ confidence to handle specific forms of bullying, including cyberbullying and bias-based bullying. There is evidence to suggest that confidence can vary depending on the form of bullying: Purdy and Mc Guckin (2014) found that preservice teachers reported more confidence to deal with general bullying incidents compared to bias-based bullying incidents, specifically against students with disabilities. It will be important for future research to take a more fine-grained approach to disentangle nuances in confidence for different skills needed to manage bullying in order to identify the specific actions that preservice teachers need assistance with to develop their sense of efficacy.

With recognition for this measurement handicap, the primary theme identified from the results was that a majority of preservice teachers lacked confidence in their ability to deal with bullying. Having a sense of confidence or efficacy for a behavior is critical to an individual’s ability to execute an action according to the TPB (Ajzen, 2012). Those who feel confident to intervene are more likely to do so (Fischer et al., 2021), which underscores the need to identify ways we can bolster preservice teachers’ confidence in managing this challenging peer experience. This lack of confidence is not unique to preservice teachers: Some in-service teachers also report lack of confidence in their role to address bullying (Boulton, 1997; Macaulay et al., 2018). It is worth investigating whether increased confidence during teacher preparation programs can be sustained once they become in-service teachers and the degree to which confidence translates into actual competence when dealing with bullying incidents themselves.

Preservice teachers themselves want this type of preparation. Although not a primary focus of this review, several of the studies reported on whether preservice teachers felt they were being adequately prepared. Multiple studies showed that preservice teachers believed their current preparation program was not sufficient in preparing them to deal with bullying (e.g., Beran, 2005). For instance, preservice teachers lamented that their “program focused too much on planning, theory, teachers’ certification preparation, and philosophy” (Gonzalez Vega, 2016, p. 89) or that programs were teaching them “to deliver lessons and that is not all that teaching is” (Fry et al., 2020, p. 7). It seems prudent to provide preservice teachers with comprehensive training that will prepare them to both teach lessons and foster their students’ holistic adjustment across multiple domains.

Evidence of Intervention Effects on Knowledge, Attitudes, and Confidence

A small number of studies (n = 4) used intervention designs to change preservice teachers’ knowledge, attitudes, or confidence levels. These intervention studies had problematic designs: Only two had comparison groups (Benitez et al., 2009; Laferriere, 2019), and none of the studies used randomized controlled trials. Given that randomization was not employed, which is the recommended method for intervention studies (Fraenkel et al., 2018), these results must be interpreted with caution. Future research with more rigorous designs is needed to examine whether training can increase preservice teachers’ knowledge of bullying and self-efficacy toward dealing with bullying. Longitudinal investigations in particular are critically needed to evaluate whether any changes are sustained once preservice teachers enter the teaching profession full time.

Moderation by Gender and Prior Victimization Experience

Several studies assessed whether attitudes, perceptions of seriousness by form of bullying, and confidence to deal with bullying differed depending on preservice teachers’ gender and/or their prior experience with victimization. It must be noted that all of these studies testing for gender differences involved majority female subsamples (range from 54% to 87.3% of the sample), which raises a concern about power to detect effects with the small sample sizes of male preservice teachers. That said, the overrepresentation of female preservice teachers reflects the gender breakdown of teachers from countries included in this review (e.g., Canada, Norway, Spain, USA; see World Bank Group, 2021a, 2021b). Nonetheless, it is important for studies on preservice teacher populations to strive for representative samples to ensure that perspectives of both male and female teachers are considered.

The results across studies that examined gender differences suggest that females were more likely to be concerned about bullying compared to male preservice teachers (Amanaki & Galanaki, 2013; Beran, 2005; K. Craig et al., 2011; Yilmaz, 2010). Gender does not appear to play a strong moderating effect on perceptions of seriousness by different forms of bullying given that the majority of studies testing for this found no differences between female and male preservice teachers (Boulton et al., 2014; W. Craig et al., 2000; Garner et al., 2013; Lester et al., 2018). As for confidence to deal with bullying, most studies found no gender differences, and those that found a difference reported that female preservice teachers were less confident than male preservice teachers to deal with bullying. The mixed evidence across these studies suggests that gender remains a relevant variable that should be examined, but the lack of cohesiveness in the results prevents us from making clear statements about the role that gender plays in preservice teachers’ knowledge of, attitudes toward, and confidence to deal with bullying. Further, there is a significant concern with the generalizability of these results given the high risk for selection bias across studies and the general overrepresentation of females in the samples.

Only a very small number of studies examined moderation by prior victimization, and the general pattern across the studies was that prior victimization experience was not related to the outcomes of interest in our research questions. Specifically, there were no differences in attitudes, perceptions of seriousness, or level of confidence depending on preservice teachers’ prior victimization experience (Kokko & Pӧrhӧlӓ, 2009; Lester et al., 2018; Nicolaides et al., 2002; Rose et al., 2018). However, it is important to note that the categories for victimization experience were not the same across all studies: Studies either looked at victim groups only related to frequency (Kokko & Pӧrhӧlӓ, 2000; Lester et al., 2018; Rose et al., 2018) or victim and bully groups (Nicolaides et al., 2002). Thus, more replication studies are needed using similar methods to capture prior victim status to elucidate the role of victimization experiences in preservice teachers’ perceptions of bullying.

Recommendations for Research

To allay concerns with research on preservice teachers’ readiness to deal with bullying, we make the following recommendations for research in this area. First, more effort should be made to recruit larger samples and ensure high response rates for generalizability purposes if random samples are not feasible. This may require coordination across multiple teacher preparation programs to distribute surveys to random samples to increase the generalizability of the results. These efforts are needed to reduce selection bias and ensure results are representative of the diverse population of preservice teachers.

Second, more attention must be paid to possible confounds when examining outcomes of interest. A few studies did this well (e.g., Lester et al., 2018; Rose et al., 2018); but overall, it is important that future work consider and report whether patterns differ depending on relevant factors. In addition to typical demographic or education characteristics, one relevant factor is preservice teachers’ prior knowledge of or exposure to bullying topics. When extracting data from each report, we noted whether study authors included any information about (a) whether preservice teachers were exposed to bullying topics during their preparation program and (b) whether preservice teachers were asked about their exposure to bullying topics in their preparation program. Only 14 of the 42 studies (33.3%) explicitly stated whether preservice teachers had been exposed to bullying topics during their training. For example, some authors reported that bullying topics were covered in a required education course, whereas other authors explicitly stated that preservice teachers had no formal training in bullying, as a course on bullying was not part of the curriculum (e.g., Beran, 2005; Nicolaides et al., 2002; Purdy & Mc Guckin, 2014; Rose et al., 2018). Similarly, only seven studies (16.7%) surveyed preservice teachers’ perspectives on how much training they had received (e.g., Lester et al., 2018). Formal training or exposure to bullying topics is an important confound to our conclusions about how much preservice teachers know about bullying and which training aspects are needed. Moving forward, we recommend that researchers address these possible confounds either by (a) asking preservice teachers if they took a course about bullying during their program, (b) asking whether a course on bullying is a required part of the curriculum, or (c) surveying preservice teachers about if and what they have learned about bullying in their courses.

Third, the structure of initial teacher preparation and other accreditation programs around the world may vary considerably. In order to draw parallels across studies with different program structures, we encourage researchers to provide more detailed descriptions of training programs. For instance, and related to the point above about confounds, information about when students begin internship or practicum courses, whether those internship courses involve observations versus primary instructor responsibilities, when those experiences occur during program course sequence, and how much experience they have in total is all vital and necessary to ascertain the role of program structure and experiences on preservice teachers’ readiness to manage bullying.

Fourth, given concerns with information bias, there is a pressing need to ensure the field is using theoretically and psychometrically strong measures (e.g., Bauman et al., 2021). Many studies used instruments with unknown properties or modified measures with unspecified properties on the current sample. If instruments are being used for design purposes in preservice teacher education, unreliable or invalid measurement of preservice teachers’ knowledge and attitudes toward bullying may result in misaligned curriculum decisions that negatively or minimally affect preservice teachers’ readiness to address bullying. If designed properly and used correctly, a valid and reliable assessment could be an integral component in designing an appropriate curriculum.

Fifth, we echo calls from other bullying researchers for more qualitative studies to gain a deeper understanding of preservice teachers’ knowledge and attitudes toward bullying (Spadafora et al., 2022). Qualitative research can produce more detailed, holistic descriptions of participants’ experiences, which can help shed light on nuanced results such as the contradictory finding in this review that some preservice teachers believing bullying to be both harmful and normative. In sum, in order for this research to compellingly convince policy-makers, funding agencies, and preparation programs to make changes, we recommend that future research incorporate these suggestions to improve the overall rigor and comprehensiveness of research on preservice teachers and bullying.

Limitations and Future Directions

Although we followed rigorous standards for systematic reviews, a few of the review’s limitations must be acknowledged. First, our narrative synthesis does not provide the level of specificity that would be provided by a meta-analytic review. A meta-analysis was not feasible given our specific research questions and our decision to include multiple methodologies. Thus, our analytic decision was the most appropriate in order to accommodate and coalesce findings from both quantitative and qualitative data. That said, future research would benefit from focused meta-analytic reviews on specific research questions. Second, we stress the need for caution in interpreting the results described herein given concerns with risk for selection, confounding, and information bias found in several studies. An important next step will be for replication studies using established measures to examine whether the pattern of results is consistent. Third, our review only focused on two moderators (gender and prior victimization experience); as such, we cannot speak to whether patterns in results would differ by other relevant factors. Future reviews should examine whether patterns differ depending on factors such type of preparation program.

Our review also suggests broader research directions related to antibullying training available for preservice teachers. More than 20 years ago, O’Moore (2000) issued a call for systematic changes to how we prepare teachers to handle bullying, but more research is needed to understand how programs have adjusted to meet this need. First, we need to examine how programs have—if at all—incorporated bullying content and training opportunities into their curricula, which may be overcrowded. Second, we need to examine which methods are most effective for training preservice teachers. Some studies have shown promising results using mixed-reality simulations during preparation programs (Theelen et al., 2019), and such methods have been used to train in-service teachers to respond to bullying (Pas et al., 2019). Whether this method is effective for preservice teachers antibullying readiness remains to be seen. Third, it is critical to assess whether training effects during preparation programs are sustained over time once preservice teachers enter the classroom. We are not suggesting that antibullying training only be limited to preparation programs; rather, ongoing professional development for teachers will likely be needed. We need more research to clarify the most effective and efficient training sequence that will give preservice teachers the foundational knowledge necessary for skill development on the job. Answers to these research questions will require rigorous study designs (i.e., randomized controlled trials) and commitment from key stakeholders (e.g., preparation programs, school districts), but these barriers are not insurmountable.

Conclusion

Preservice teachers will one day be in charge of a classroom full of students, many of whom are likely to be involved in bullying and victimization. The evidence summarized herein suggests that preservice teachers are not ready to intervene and manage bullying once they enter the teaching profession: Many lacked basic knowledge about bullying; some believed it to be normal part of growing up; and overall, they were not confident they could deal with it. There are consequences for inaction on this front. In the absence of formal training, teachers may not respond appropriately or respond at all. Inaction on the teacher’s part to a bullying incident can make it seem like they tolerate bullying, which is detrimental to antibullying efforts and victims’ adjustment (e.g., Yoon & Bauman, 2014). One can envision a scenario in which a student reports they are being bullied to a teacher, but the teacher fails to effectively intervene in the situation, leading to potentially devastating consequences for the victim, bystanders, and the classroom as a whole. Our review suggests more needs to be done to adequately prepare preservice teachers during their education programs so that they can effectively participate in antibullying efforts once they are in charge of a classroom. We owe it to our preservice teachers and to their future students to do so.

Supplemental Material

sj-pdf-1-rer-10.1177_00346543221094081 – Supplemental material for Preservice Teachers’ Knowledge and Attitudes Toward Bullying: A Systematic Review

Supplemental material, sj-pdf-1-rer-10.1177_00346543221094081 for Preservice Teachers’ Knowledge and Attitudes Toward Bullying: A Systematic Review by Molly Dawes, Colleen Gariton, Angela Starrett, Greysi Irdam and Matthew J. Irvin in Review of Educational Research