Introduction

Bullying among students is a problem that many teachers face at some point in their careers. To distinguish bullying from other types of aggression, three characteristics are usually referred to. Bullying is proactive form of aggression that occurs repeatedly and is characterized by an asymmetric power relationship that makes it difficult for those affected to fight back (Olweus, 1993, 2013). From an educational perspective, it should be emphasized that bullying is a social phenomenon. The social-ecological model of bullying (Swearer et al., 2012) conceptualizes bullying based on Bronfenbrenner’s (1979) model of human development. This model suggests that an individual’s behavior is the result of reciprocal interactions between individual and environmental factors. The environmental factors are described as social systems with different degrees of proximity to the child (such as the peer group and the family at the micro level, school context at the meso level, and society or cultural norms at the macro level), all of which thus also contribute to the dynamics of bullying.

Considering that bullying occurs within the relatively stable group setting of a school class (Atria et al., 2007), it seems important to take a closer look at this meso system. The peer group has already been examined in more detail using the participant role approach. This involves empirically identifying—alongside those who bully and are bullied—other people involved in the bullying dynamic. This includes those who assist and reinforce the bully, those who defend the person being bullied, as well as passive bystanders (Salmivalli et al., 1996). However, teachers, “as socialization agents and key adults in the classroom” (Colpin et al., 2021, p. 782), also need to be recognized as being important actors at the classroom level. They shape peer relationships in the classroom and influence student group dynamics, as well as the classroom climate. At the same time, teachers also represent the school as an institution, its norms and beliefs, and the school culture (in dealing with bullying). Consequently, teachers play a crucial role in the bullying process, whether consciously or unconsciously.

The Role of Teachers in School Bullying

A growing body of literature acknowledges the importance of the role of teachers in counteracting bullying in schools. Recurring questions in this area to date have been concerned with, for example, what knowledge, attitudes, and beliefs about bullying teachers have (Bauman & Del Rio, 2005; Craig et al., 2011; Kochenderfer-Ladd & Pelletier, 2008; Mazzone et al., 2021; Nicolaides et al., 2002; Strohmeier et al., 2021), which actions teachers classify as bullying (Boulton, 1997; Craig et al., 2000a; Hazler et al., 2001), and how teachers intend to respond to bullying (Bauman et al., 2008; Burger et al., 2015; Dake et al., 2003; Dudziak et al., 2017; Özdemir et al., 2021; Strohmeier et al., 2021; Yoon et al., 2011, 2016). Other areas relevant to bullying research include how relevant the quality of the relationship between teachers and students is (e.g., Bouchard & Smith, 2017; Jungert et al., 2016; Longobardi et al., 2018; ten Bokkel et al., 2022; van Aalst et al., 2021; Wang et al., 2015) and what influence teachers have on the school climate (e.g., Kartal & Bilgin, 2009; Mucherah et al., 2018; Wang et al., 2013).

Teachers play a crucial role in resolving bullying and can, in the worst case scenario, contribute to the perpetuation of a bullying dynamic (Bilz et al., 2015; Burger et al., 2022; Wachs et al., 2019; Yoon & Bauman, 2014). The theory of planned behavior (Ajzen, 2012), or the transactional theory of coping (Lazarus & Folkman, 1987), have been used to frame teachers’ perceptions and responses to bullying (e.g., Gagnon et al., 2022; van Aalst et al., 2022; van Gils et al., 2023; Yoon & Bauman, 2014). From the perspective of teacher education, however, teachers’ awareness of violence and bullying and appropriate intervention strategies must be regarded as being important components of teachers’ professionalism. To systematize the factors that influence how and why teachers intervene in bullying situations, Bilz et al., (2017, see also Fischer & Bilz, 2019) have suggested a heuristic model outlining the conditions and consequences of teachers’ bullying interventions (see Fig. 1).

Fig. 1
figure 1

Model of conditions and consequences of teachers’ bullying interventions (Fischer & Bilz, 2019, p. 124) (first presented in German by Bilz et al., 2015, 2017)

Full size image

This model is, in turn, based on the COACTIV model of teachers’ professional competence described by Baumert and Kunter (2006, 2013), which defines teacher competence through a combination of knowledge, motivation, beliefs, and self-regulation. According to Bilz et al. (2017), these levels of teachers’ intervention competence, taken together with the school-related and situational context, constitute the essential conditions for how teachers respond to bullying incidents. The authors further divide teachers’ interventions in cases of bullying among students into a series of five steps: (1) awareness of the situation, (2) choice of intervention goals and (3) intervention strategies, (4) realization of the intervention, and (5) evaluation of goal attainment.

Since noticing and recognizing bullying can be understood as two discrete steps in an intervention and represent an important prerequisite for an appropriate response (Bilz et al., 2017), the present study focuses on the first step in this model: awareness of the situationFootnote 1.

Teachers’ Sensitivity to Bullying

In the context of this paper, sensitivity to bullying is understood not only as the ability of teachers to recognize bullying but also includes what prompted them to recognize it. It is also understood in reference to Thornberg and Jungert’s (2013) concept of basic moral sensitivity. The authors apply moral sensitivity as the “ability to recognize moral issues in complex situations” (Jordan, 2007, p. 325) and specifically to bullying as “a morally simple situation, […] in which a moral transgression is unambiguously wrong because of its inherent harm toward a person in a weaker or socially disadvantaged position in relation to the perpetrator or perpetrators.” Teachers are usually the first responsible contact person for students who have experienced or witnessed bullying (Wachs et al., 2019; Yoon & Bauman, 2014). Yet, research has repeatedly shown that, for teachers, bullying can often be hard to detect. The reasons for this include the fact that bullying often occurs in places that adults are absent from and have no control over (Vaillancourt et al., 2010), that teachers often overlook bullying (Craig et al., 2000b), and that teachers are less likely to recognize certain forms of bullying (Boulton, 1997; Hazler et al., 2001). For example, teachers are better able to recognize physical bullying than relational bullying, and they also perceive it as being more serious than other types of bullying because physical bullying and the consequences of physical bullying are more obvious (e.g., Bauman & Del Rio, 2006; Boulton, 1997; Chen et al., 2018).

However, past research on if and how teachers recognize bullying primarily sheds light on perceptual differences in what teachers and students consider bullying. Methodologically, these perceptual differences are often demonstrated by comparing prevalence data from student reports with teacher reports, whereby the level of agreement between bullying reports from students’ and teachers’ perspectives has been shown to be low (Bradshaw et al., 2007; Cornell & Brockenbrough, 2004; Demaray et al., 2013; Ladd & Kochenderfer-Ladd, 2002). Accordingly, if we want to know more about teachers’ awareness of and sensitivity to bullying, we also need to ask what specific aspects teachers recognize when they are confronted with bullying and what considerations they use when classifying the incident in question as bullying. It is these questions that are addressed in this study.

In most of the studies about teachers and bullying mentioned above (e.g., Bauman & Del Rio, 2005; Burger et al., 2015; Nicolaides et al., 2002; Yoon et al., 2016), participants were either informed about the research topic or were provided with a definition or description of bullying at the beginning of the study. This ensures that respondents share a common understanding of bullying. However, it is not possible to know whether or not the participants possessed any knowledge about or an understanding of bullying before the inquiry began, nor whether they would have been able to recognize the bullying incident and distinguish it from other dynamics without this definition. In other words, we cannot say whether the teachers were already sensitive to bullying or whether they were sensitized to bullying by the format of the study.

This shortcoming, which is due to the methodological logic of quantitative research designs, is also addressed in the present qualitative study. The title of the online qualitative survey (“Social Interaction among Students”) did not directly refer to bullying. In addition to this, participants were not presented with a definition of bullying, and demographic questions related to bullying were only asked after the vignette questions. To better frame teachers’ perceptions and responses to verbal and relational bullying, participants in this study were also presented with a separate vignette about physical aggression.

Research on Teachers and Bullying in Austrian Schools

An increasing number of international publications are currently addressing the topic of teachers and bullying. Nevertheless, context-specific aspects, such as bullying prevalence rates, anti-bullying programs, teachers’ roles in Austrian secondary schools, and teacher education on bullying, are of particular importance when considering this area of research.

For many years, according to the Health Behavior in School-aged Children (HBSC) surveys by the World Health Organization, Austria, ranked among the countries with the highest bullying rates (Craig & Harel, 2004; Currie et al., 2008; Molcho, 2012; Molcho et al., 2009; Walsh & Cosma, 2016). The latest HBSC survey, in the year 2017/18, reported that, in Austria, 8% of adolescents are victims of bullying and 7% of boys and 3% of girls had become involved in repeatedly bullying others over the previous few months (Inchley et al., 2020). However, since 2014, Austria has recorded the strongest decrease in bullying compared to the other countries surveyed. Felder-Puig et al. (2019) suggest that this decrease is the result of the inclusion of “social learning” and other social cohesion activities in the school curricula. In Austria, there are no specific guidelines, strategies, or prevention programs for bullying that are consistent across all types of schools, or even within one province. The Viennese Social Competence (ViSC) program is probably the best-known and most intensively researched program aimed at tackling violence in Austria (e.g., Gradinger & Strohmeier, 2018; Gradinger et al., 2015, 2016; Yanagida et al., 2019). However, it was only promoted in Austrian secondary schools between 2008 and 2013 (Strohmeier et al., 2012), and it was not implemented in all schools across the country.

Country-specific research relevant to the topic of teachers and bullying is thus limited in Austria. Studies that have been conducted in Austria, or with partly Austrian samples, have examined, for example, teachers’ intervention strategies in a hypothetical case of bullying (Burger et al., 2015), teachers’ opinions on (cyber)bullying prevention (Gradinger et al., 2017), teachers’ knowledge of and intervention strategies employed to deal with hate postings (Strohmeier & Gradinger, 2021), and how teachers profit from the ViSC program (Strohmeier et al., 2021). A recent quantitative study (Paljakka et al., 2021) examined the agreement between bullying reports from teacher, parental, and peer perspectives in Austria. The authors found that, overall, teachers reported the highest level of bullying compared to the two other groups of respondents, which may reflect teachers’ particular sensitivity to bullying behavior, although the reasons for their increased awareness remain to be explored.

In Austria, secondary school teachers are usually qualified for two subjects and, according to federal law (Schulunterrichtsgesetz (SchUG)), are entrusted in this context with independent teaching and educational work and are supposed to contribute to the learning success of the students as well as to carry out the performance assessment of the students. According to the Teaching and Learning International Survey (TALIS) 2018, Austrian teachers spend nearly half of their work time teaching in the classroom (Wallner-Paschon et al., 2019). The preparation time (planning, correction work) takes around 28% of their time, and around 20% of their weekly working hours are devoted to “other” tasks (teamwork and discussions with colleagues at school, advising students, professional development activities, discussions and cooperation with parents or guardians, participation in extracurricular activities, etc.). About 3 hours per week (6%) are spent on administrative tasks (participation in school administration, general administrative work including information exchange, paperwork, and other office work). In international comparison, Austrian teaching staff have significantly fewer administrative or pedagogical support staff, such as social workers, psychologists, dyslexia trainers, or speech therapists (Wallner-Paschon et al., 2019). While in the EU, there is on average one pedagogical support staff member for every 8 teaching staff members; this ratio is 1 to 19 in Austria.

The standard for class size in Austria is currently set at 25 students, but in practice, this varies, e.g., by school type, province, but also by region (urban–rural). Across all secondary school types, the average class size was in 2018/19 around 22.1 students (Statistik Austria, 2020) with an average of 9.3 students per teacher (full-time equivalent) (Vogtenhuber et al., 2021). In recent years, teachers have increasingly been expected to deal competently with the heterogeneity of their students, to be more responsive to different learning requirements and to promote students’ strengths (Vogtenhuber et al., 2021). Teachers—especially the head teacher of the class—are expected to have a comprehensive picture of what is going on in the classroom. This includes school performance or problems of the individual students, friendship relations, the class climate, and bullying.

When it comes to the Austrian teacher training curriculum, improvements are still needed in systematic education about bullying and its prevention (Burger et al., 2015). Within the framework of the teacher training program (bachelor’s and subsequent master’s degree program), the two teaching subjects are combined with educational science and pedagogical practice. In his diploma thesis, Hoffmann (2020) surveyed teacher educators at Austrian universities and showed that bullying was addressed in at least a few teacher education courses at almost all Austrian universities in the 2018/19 academic year. At the same time, in the opinion of the respondents, bullying was not addressed to a sufficient extent and teacher education students are likely to simply “miss” the few courses on bullying during their studies because they take one of the many courses offered concurrently.

To extend the knowledge of the predominantly quantitative body of research and to close a number of the gaps described above, this study adopts a qualitative approach. Without making participants aware of the topic of bullying in the study design, this study examines how teachers assess a vignette of a bullying scenario compared to a non-bullying scenario. It then aims to ascertain which aspects of the vignette in question or considerations about the situation given guided these teachers in their assessment.

The research questions for the study were as follows:

  1. 1.

    Do teachers recognize the bullying incident and how do they assess the non-bullying vignette in comparison?

  2. 2.

    What specific aspects do teachers recognize and what are their considerations when classifying this incident as bullying?

Methodology and Methods

Instrument

For this study, an online qualitative survey was developed that included four open-ended questions on each of the two vignettes, as well as closed-response demographic questions. According to Braun et al. (2021), online qualitative surveys are suitable for sensitive research topics where people might not want to participate in a face-to-face setting (e.g., sexual topics) and where social desirability is a major concern. In online surveys, participants feel more anonymous, which can facilitate disclosure and participation. Bullying can be considered just such a sensitive issue for teachers, not only because it involves aggressive behavior but also because teachers, in their role as pedagogical professionals, are expected to be knowledgeable about the issue and able to deal with it appropriately.

Using vignettes for data collection is a well-established approach in the investigation of teachers’ responses to bullying (e.g., Bauman & Del Rio, 2006; Burger et al., 2015; Craig et al., 2000a; Hazler et al., 2001, Yoon, et al., 2011; Shute et al., 2022). It allows situational characteristics to be kept constant, reducing potential observer effects, and evoking more realistic assessments than questionnaires or interviews. For this reason, two vignettes—one depicting a bullying scenario and the other a single case of physical aggression—were created based on scenarios used in the study conducted by Hazler et al. (2001) and in the classroom observations described by Bennewitz (2012).

Two consecutive pilot studies were performed with teacher education students to evaluate and improve the scenarios’ clarity and appropriateness. The pilot served to establish the construct validity of the scenarios for identifying the bullying and non-bullying scenario. Participants were asked to identify the main characteristics of bullying in the vignette and to name words and phrases that were relevant to their assessment. Based on these results, some of the wording was revised to ensure that the vignettes described the situation as concisely and accurately as possible. In addition, the clarity of the open-ended questions was also improved, and the decision was taken to use two vignettes, each with four open-ended questions. The effort involved in answering more than eight open-ended questions, it was felt, might have resulted in shorter or incomplete responses. As Braun et al. (2021) found, qualitative surveys often ask between four and 10 topic-based questions, and they also argue that the potential for participant disengagement and/or fatigue increases with survey length.

The vignettes depict typical situations from everyday life in school. Gender-neutral names and phrasing were used so that the vignettes applied to all genders. Consistent with the commonly used definition of bullying (Olweus, 1996), the first vignette meets the three main characteristics of bullying: social and emotional harm is done, it is repeated over a longer period of time, and there is an imbalance of power, wherein a student cannot defend themselves. Since school uniforms are not generally worn in Austria, clothing (e.g., as a status symbol, as an expression of individuality or religion) is a common topic among young people, and so this issue was included in the vignette.

Frankie has had different students sitting next to them for some time now. After the lesson finishes, Alex, Kit, and Jessie—who have been either ignoring Frankie or calling them names—walk past Frankie again. This time Alex says to Frankie: ‘No wonder nobody wants to sit next to you. You always wear such weird, old clothes.’ Frankie doesn’t answer but is clearly fighting back tears.

In order to include a contrasting scenario to contextualize the bullying scenario, the second vignette displays a single incident of physical aggression perpetrated by an older student on a younger student. It cannot be classified as bullying because the behavior is not repeated (at least up until this point) and the aggressive student is not acting proactively (with the intent to cause harm), but instead in response to a perceived provocation. The purpose of the second vignette was to have a comparison of how the participants perceive an incident that cannot be classified as bullying according to Olweus’ definition and to have a comparison of which characteristics of the scenarios are relevant for the respondents.

At break time, Andy is playing tag with some friends in the schoolyard. Andy accidentally bumps into Chris, an older kid, whom Andy hardly knows. Chris shouts at Andy, calling them an “idiot” and pushes them so hard that Andy falls and hurts themselves. After the incident, Andy has a grazed elbow and a dirty t-shirt.

Participants were asked the following survey questions about both vignettesFootnote 2:

  1. 1.

    What do you think is going on in the situation described?

  2. 2.

    Would you react to the observed situation?

  3. 3.

    What would you do?

  4. 4.

    Please explain why you would act like this. Did something influence your choice of action? If so, what?

  5. 5.

    Would you tend to avoid a particular response? Please explain what you would avoid and why.Footnote 3

Except for item (2), all questions were open-ended. Each question was given on a separate page of the online survey. Respondents had to provide their answers in writing with no length requirements or word limitations. Question two was designed as a filter question (yes/no/I do not know) and led to question (3) if the answer was “yes” or “I do not know,” or otherwise to question (4).

On the first page, a participant disclaimer was used to gain informed consent and explain participants’ confidentiality. Demographic data, including questions on anti-bullying programs or specific training on bullying, were collected after the open-ended questions on the last page of the survey. The survey’s title (“Social Interaction among Students”) also did not directly refer to bullying. This was to ensure, on the one hand, that participants were not attracted to participating in the study because of the topic of bullying and, on the other hand, that participants were not sensitized to bullying by the study format.

Participants

In total, 38 secondary school teachers from all nine federal states of Austria volunteered to participate in the study (six teachers each from Lower Austria, Upper Austria, and Burgenland; four each from Vienna, Carinthia, and Salzburg; three from Styria and Vorarlberg; and two from Tyrol). Most teachers completed their highest degree at a university (n = 25), six at a teacher-training college, and seven did not give any specific information on their training institution. The capital city of Vienna was the most frequent place in which teachers studied (n = 11). The remaining teachers were distributed among the other eight federal provinces of Austria. The age of the teachers (female = 22, male = 15) ranged from 25 to 63 years, with a mean of 48 years. The participants taught in different types of schools (academic secondary schools (Allgemeinbildende höhere Schulen) = 28, secondary schools (Mittelschule) = 8, schools for vocational education (Berufsbildende Schule) = 1) and in urban (n = 22) and rural areas (n = 15). Data regarding gender, age, and school type were not provided by one participant. Teaching experience ranged from 1 to 42 years (with a mean of 22 years). When asked whether anti-bullying strategies or programs existed in their schools, 21 participants agreed and mentioned, for instance, peer mediation, social learning courses with the students, and the provision of support within the school (in particular, from qualified staff members, in-school training sessions, etc.). When asked about specific training on the topic of bullying, 22 teachers said they had received no specific training, seven had participated in in-service training sessions, five had attended special courses during teacher education or school internships, and four had completed external training courses.

Procedure

Data Collection

Data were collected using a qualitative online survey between February and June 2019. The sampling strategy aimed at ensuring that teachers from all nine federal states of Austria were represented in the sample, with at least two teachers from each state. The sampling strategy was chosen to include teachers from different school contexts (school type, urban/rural areas, class sizes, etc.) as well as teachers who graduated from different universities or colleges of teacher education in order to improve the diversity and richness of the data in relation to the research questions.

From 1,477 secondary schools and academic secondary schools with an overall number of 51,563 teachers in service, a total of 65 schools from all nine federal states of Austria were systematically selected via the Austrian online school directory. The selection was made alphabetically and was based on the criteria of school type (secondary schools and academic secondary schools), federal state, and region (urban and rural). An email containing general information about the study and the link to the survey was sent to the principals of the selected schools. The principals were asked to forward the link to their teaching staff and thus support the research project. The principals were neither obliged to confirm that they had received the email nor that they had passed on the link. This meant that an overall response rate could not be calculated. In total, the link was opened 182 times (this also includes double clicks), 68 participants answered at least one question, and 38 participants (who were then included in the analysis) gave answers to all the open-ended questions. Since the participation was voluntary, a degree of selection bias cannot be ruled out. However, a possible selection bias is more likely to be related to the study title (“Social Interaction among Students”) and not to the topic of bullying, as this was not explicitly mentioned in the invitation or in the title of the study.

Data Analysis

The textual data collected from the online survey were imported into MAXQDA2020 program version 20.3.0. A qualitative content analysis was conducted according to the steps and rules described by Mayring (2014, 2015). The focus of this method is to systematically describe the meaning of data by assigning parts of it to categories of a coding frame (Schreier, 2013). These categories are developed based on theory, research questions, and the concrete material and are defined by construction rules and rules for assignment, which are revised and reviewed during the analytic process (Mayring, 2014). Categories therefore take the meaning of passages of the material to a higher level of abstraction. On the one hand, this leads to a loss of concrete information; on the other, it helps us to understand how “different parts of the material compare and relate to each other” (Schreier, 2013, p. 170). The procedure model of the analysis for this study (see Fig. 2) is based on the model presented by Mayring (2014, p. 54, Fig. 9). To maintain linguistic nuances, the original data were not translated for the analysis. However, the coding and categorization were carried out in English, and the anchor examples from the original data were subsequently translated into English for this article.

Fig. 2
figure 2

Model of procedure of the analysis based on the model presented by Mayring (2014, p. 54, Fig. 9)

Full size image

For each of the research questions, the answers to the most closely related survey questions were analyzed in the first coding circle. For both research questions, some categories were based on previous research results and were thus deductive, while all other categories were developed inductively from the material. The two vignettes were each analyzed separately. In the first coding cycle, each response of around half of the participants was coded by the researcher and assigned to one of the previously defined deductive categories. If they did not match, a new inductive category was developed. For each research question, a table was created that included the categories, category definitions, anchor examples from the material, notes, and coding rules when the category definition was not sufficient to differentiate from other categories.

For example, to answer the research question (1) “Do teachers recognize the bullying incident and how do they assess the non-bullying vignette in comparison?”, the answers given to survey question (1) “What do you think is going on in the situation described?” were coded first. Based on theoretical considerations and the research question, two categories were deductively formed for the bullying vignette (bullying, non-bullying), while the category probably bullying developed inductively from the material. Table 1 shows an example of the answers that led to the construction of the category probably bullying. The category definitions and level of abstraction were constantly revised and refined. In the second coding cycle, all 38 responses were analyzed again using the final category system. In a third coding cycle, the accuracy of the assigned categories was checked against the background of all the material of each participant and the assignment to the categories was adjusted if necessary. The category system developed in this way served as the basis for answering the research questions in the final analysis (see Tables 2 and 3).

Table 1 Category probably bullying
Full size table

Finally, the analyses were checked using the specific content-analytical quality criteria proposed by Mayring (2015). For the intra-coder check, coding began from the start; the results were compared with the preceding codings. The few differences found (e.g., concerning the length of the coding units) had no impact on the coding rules or the category system. For the inter-coder check, the category system and six randomly selected cases were given to two colleagues. Disagreement arose only about the coding of one paraphrase. As suggested by Mayring (2015), the coding in question was then discussed and all three coders agreed on one coding form and the coding rule was refined accordingly.

Qualitative content analysis also allows us to look at quantitative aspects (e.g., frequencies of the coded categories) of the analyzed data. Even though this was not the primary aim of the analysis, there were some places in which the reported frequencies appeared to help comprehension. According to Mayring (2014, p. 41), “registration of how often a category occurs may give added weight to its meaning and importance as well,” but it should be used cautiously and needs to be explained carefully.

Results and Discussion

This section presents the main findings of the analysis, discusses them in the light of the research questions, and relates the results to previous studies. The non-bullying vignette was primarily intended to contextualize the responses to the bullying vignette, and therefore, the related category systems are not explored here in any detail. Since a detailed discussion of the results cannot be carried out without referring to categories or interpreting respondent quotes and in order to avoid redundancies in this regard, it seemed more appropriate, in this paper, to combine the Results and Discussion sections into a single section.

Teachers’ Awareness of Bullying

Regarding research question (1)—“Do teachers recognize the bullying incident and how do they assess the non-bullying vignette in comparison?”—the category system developed for the bullying vignette is presented in Table 2. This shows the deductive (bullying, non-bullying) and inductive (probably bullying) categories, the category definitions, and the examples from the teachers’ responses.

Table 2 Categories of teachers’ assessment of the bullying vignette
Full size table

(Probably) Bullying

Without the teachers having been prepped for the topic beforehand, a large number of them described the vignette presented to them as (probably) bullying. When coding the answers, the categories bullying or probably bullying outweighed the category non-bullying. The frequencies of the coded categories are interesting when one considers previous research into bullying. Earlier studies (e.g., Bauman & Del Rio, 2006; Boulton, 1997; Chen et al., 2018; Hazler et al., 2001) have shown that teachers are more likely to classify behaviors involving physical assault as bullying and perceive them to be more severe than less visible forms of bullying. In contrast to these findings, a total of two out of three teachers (n = 26) in this sample recognized the socially exclusionary and verbal bullying (categories bullying n = 21 and probably bullying n = 5).

The participants not only recognized the verbal and socially exclusionary bullying incident but were also often able to label it as such. Of the 26 teachers who recognized bullying, 23 also used the term “bullying,” while none assessed the physical conflict in the non-bullying vignette as bullying. This is a surprising result. In the study by Hazler et al. (2001), which used very similar scenarios, 97% of the participants missed “the absence of the repeat characteristic critical to defining bullying” and misinterpreted situations involving physical attacks and an unfair power dynamic as bullying. Therefore, it was expected that the comparison vignette in the present study might be misclassified as (physical) bullying as well, but this was not the case. Compared to the bullying vignette, the responses to the non-bullying situation were described (and categorized in the analysis) as, for example, overreaction, aggression, accident, emotional response, misunderstanding, violence, and revenge. With regard to the first research question, the results show that the teachers in this sample were, in fact, sensitive to the differences between a bullying situation and a non-bullying situation. This finding seems encouraging, when taken together with the findings in the study by Paljakka et al. (2021), which found that teachers reported the highest levels of bullying overall compared to reports from the perspective of teachers, parents, and peers.

The answers to survey question (1) “What do you think is going on in the situation described?” provide information not only about whether bullying was recognized or not but also about teachers’ assumptions about how the situation might develop, for instance, whether Frankie might just not respond, withdraw, or even react aggressively. Furthermore, the answers reveal, in some cases, teachers’ assumptions about the motives for the bullying student(s). Some teachers believed that Frankie was being bullied because of their different clothing styles or because the bullying student wanted to “boost his self-esteem” (Teacher 3). These findings reflect those of Compton et al. (2014), who found, in their qualitative exploratory study, that teachers, parents, and peers believe that gaining power and status through bullying is an important motive for bullying, as is difference (in various facets). However, these responses also support Mazzone et al.’s (2021) observation that “teachers often make external attributions (i.e., bullying is related to individual characteristics of the target and the perpetrator and to family factors) and underestimate the role of social context.”

Non-bullying

For the bullying vignette, the responses of 12 teachers did not indicate that the vignette was recognized as such, while the responses to the contrasting vignette were coded as non-bullying for 32 participants (the remaining were coded under the categories students’ reactions or uncertain). One of the teachers who apparently did not recognize the bullying situation, for example, believed that the “student was excluded because of his clothes” (Teacher 38). This quote supports Mazzone et al.’s (2021) remark that external attributions obscure the role of social context. If one assumes that style of dress is the (only) determining reason for a verbal attack, then the full extent of the bullying will likely go undetected, for example, that this is not the first time that Frankie has been hurt by this group and that the style of dress only functions as a trigger for the verbal attack.

However, if we take a closer look at the non-bullying category, we can also see that the code students’ reactions was assigned to some of the responses. This code was used when assumptions were made about what might happen after the incident, how students might react, and if teachers generally offered suggestions about how the scenario might continue. For example:

Being questioned by another student[.] Approving laughter[.] Someone takes sides with the person being insulted[.] (Teacher 25)

A possible explanation for this is that the participants (mis)understood the survey question (1) “What do you think is going on in the situation described?” in the German phrasing (“Was passiert Ihrer Einschätzung nach in der beschriebenen Situation?”) as a question about how the situation would develop in the future and not about the present situation. So it is also possible that some of these teachers recognized the incident as bullying but answered in reference to how the situation might develop. Consequently, the number of teachers who did not recognize bullying (n = 12) must be interpreted with caution and may in fact be lower. This impression is reinforced, for some teachers, when taking into account the context unit—that is to say, the answers to the other questions about the bullying vignette. One teacher, for instance, explains in their response to survey question (3), about the reaction to the situation:

Talk to everyone involved. Especially with the person affected. Make him understand that he is not the ‘one to blame,’ just as the perpetrators often are not. Through repeated conversations, reflections, and appropriate consequences, I can quickly remedy such ‘issues’ using my experience. In the end, everyone involved usually learns something. (Teacher 36)

Another teacher justified her reactions as follows:

Because it can happen to anyone and, as a teacher, I also have a duty to protect the students. Verbal attacks are also attacks. Therefore, from my point of view, intervention is essential. However, I do not want to act in an overbearing way, but always want to come to a solution or find a possible ‘route to the goal’ together with the students. (Teacher 30)

These quotes illustrate these teachers’ sensitivity to what was happening in the vignette and also their willingness to intervene and improve the situation for everyone—whether bullying was identified or not.

Teachers’ Considerations When Classifying Bullying

To examine research question (2)—“What specific aspects do teachers recognize and what considerations do they make when classifying an incident as bullying?”—the answers the teachers gave were coded again. The resulting categories are presented in Table 3.

Table 3 Categories of teachers’ considerations when classifying the bullying vignette
Full size table

As can be seen from the table above, two categories emerged inductively from the material (doing harm, labelling). The two deductive categories (doing harm intentionally, power imbalance), on the other hand, reflect what was expected to be recognized as features of bullying if one were familiar with current definitions and conceptualizations. Surprisingly, the third key distinguishing feature of bullying—repetition of hurtful behavior—was not reflected in the data, which is why this deductive category was not coded and therefore removed from the category system.

Doing Harm (Intentionally)

Participants have recognized that negative actions such as humiliation, exclusion, disrespect, insult, or making fun of another student are taking place and that this makes the target student feel at least uncomfortable, if not hurt. When teachers mentioned a potential intention behind the bullying student’s behavior, the category doing harm intentionally was coded. Since the intention to harm was only explicitly named in part of the answers, the inductive category doing harm was included in the category system as well. For example, Teacher 7 described that “Humiliation, exclusion (isolation) […]” is taking place (doing harm) whereas Teacher 13 stated that “Frankie is deliberately humiliated” (doing harm intentionally).

The teachers’ answers show that the teachers in many cases remain on the descriptive level and are cautious in evaluating this kind of behavior. One possible reason for this could be that the data reflect what is also discussed in the literature regarding the definition of bullying; what Hellström et al. (2021, p. 9) refer to as the “fuzziness of intention coupled with the measurement problems of observing and measuring the intention.” Hellström et al. outline some of the controversies around the intention to harm as a criterion of bullying. They point, for instance, to the fact that students who bully (especially younger students) are sometimes unaware that their “harmless jokes” are hurtful, which is addressed, for example, by Teacher 19:

Children can be so hurtful without realizing the extent of it; sometimes it’s just about being the cool one.

On the other hand—in terms of measurability—it takes a sufficient degree of honesty and self-awareness on the part of the bullying student to admit that their behavior is intentional. It also requires third parties to have high subjective evaluation skills to judge this kind of behavior as intentional (Hellström et al., 2021). It is thus possible that teachers generally assume the presence of intentionality but do not want to impute intent hastily or unjustifiably. This is supported by, for example, responses to question (5) in which teachers express that they would prefer to avoid this kind of reaction:

Prejudgement—you don’t know all the background yet (even though it shouldn’t really be a justification). (Teacher 17)

However, to distinguish bullying from other unintentional or accidental harmful behaviors, the criterion of intent is fundamental. Olweus (2013) specified in his later definition of intent to harm that bullying students are likely to know or understand that their behavior will be perceived as at least unpleasant by the target, but that they do not let this stop them. While Naylor et al. (2006) found that only 25% of participating teachers stated intent to harm as a criterion for bullying, several other studies indicated that intention was seen by teachers as a key element of bullying (Compton et al., 2014; Macaulay et al., 2020; Mishna et al., 2005). It can therefore be assumed that teachers who have recognized that Frankie has been (actively) hurt are also aware that this behavior was not reactive. Rather, the bullying student accepted, if not intended, that Frankie would be hurt by their actions. This interpretation of the data is further supported by a comparison with the responses to the non-bullying vignette. For example, the teachers considered the behavior of Chris, who insults and pushes a younger student, as unintentional, reactive behavior, an overreaction, aggression, an accident, misunderstanding, or as an emotional response, but none of the participants considers it to be doing intentional harm. As Teacher 20 put it, “[t]he collision was not intended, the reaction was too violent, and the consequences probably not wanted.”

Power Imbalance

In terms of the subjectively perceived or objectively existent asymmetrical power relationship, a few teachers did mention the numerical superiority of the bullying students. However, in some of these cases, it remains open to speculation whether the teachers also perceived this numerical difference as representing an unequal power relationship (see Table 3). Three of the respondents, though, did more clearly describe the imbalance of power and the use of psychological power in the form of knowing another person’s vulnerability and using that knowledge to cause distress (Hellström et al., 2021). In the vignette, both Frankie’s appearance and the fact of changing seat neighbors were used to hurt Frankie, as is also pointed out, for example, in the following teacher quotes.

Making fun of another (supposedly weaker), recognizing another’s weak point (in this case sitting alone) and using it against them, demonstrating power over another. (Teacher 19)

Frankie is avoided and excluded by his classmates. One reason for this is his clothing. He is obviously hurt. (Teacher 20)

Previous research on teachers’ awareness of an asymmetrical power relationship as a feature of bullying has been inconsistent. Some studies (Compton et al. 2014; Mishna et al., 2005) found that, from teachers’ perspectives, a power imbalance is a pivotal component of bullying. In a qualitative study by Naylor et al. (2006), for example, around 70% of the teachers invoked abuse of power as being a criterion of bullying. In contrast, more recently, Macaulay et al. (2020) reported that trainee teachers were unaware that an imbalance of power was a defining feature of bullying. The results seem to reflect this ambiguous picture: seven out of the 26 teachers who recognized bullying addressed the question of a power imbalance in some way. However, the frequencies of the coded categories suggest that greater importance is ascribed to the harm (intentionally) inflicted on the victim (n = 12).

Nevertheless, it is important to mention that some teachers also perceived an asymmetric power relationship in the contrasting vignette. Here, the participants primarily recognized the physical (age) difference, but two teachers also explicitly referred to the older students’ demonstration of power.

Labelling

A large proportion of teachers (23 out of 26) who correctly identified the bullying incident used the term “bullying” in some way to describe how they would assess the situation. This indicates that, broadly, bullying is a phenomenon that teachers are already sensitized to. As mentioned earlier, contrary to what one might expect, none of the participants labelled the physical conflict in the contrasting vignette as bullying. The length of the answers that included labelling varied from a single word (“bullying”) to more detailed answers, such as the following:

The students are giving Frankie a hard time (bullying). Either Frankie doesn’t/can’t/won’t afford the most expensive fashion and/or the students only say that to humiliate him. In both cases, these kinds of statements fall under bullying. (Teacher 9)

This quote draws attention to the advantages and disadvantages of labels such as bullying. On the one hand, it shows that bullying is a phenomenon or at least a term that teachers are familiar with. If, on the other hand, teachers only use the term bullying without describing in more detail what they actually noticed, then we cannot say more precisely what teachers understand by the term bullying. It is also hard to say whether they are aware of common conceptualizations or definitions of bullying (e.g., that of Olweus) or understand what distinguishes bullying from other forms of aggression or violence. Although Teacher 9 seems to correctly assess that Frankie is being bullied by others, the above quote casts doubt on whether this teacher has a comprehensive understanding of current conceptions of bullying. The teacher seems to base the assessment solely on the hurtful statement and argues that such statements are bullying, regardless of whether Frankie is “really” badly dressed or if the students are just saying it to hurt Frankie. The other characteristics of bullying, such as a power imbalance or repetition, may have been only implicitly considered in this teacher’s assessment (number of students, use of psychological power), or at least may not have been perceived by this teacher as being worth mentioning (repetition).

Repetition

The repetitive nature of bullying—as one of its defining features for a long time—was not directly addressed by respondents and was therefore not included in the category system. Nevertheless, some possible explanations for this result should be briefly discussed. Repetition may be a “self-evident” feature of bullying for the respondents, and therefore, they did not feel it needed to be mentioned in their answers. Repetition could also be an aspect that teachers did not recognize or that was not decisive in classifying this incident as bullying, as previous studies suggest (Compton et al., 2014; Mishna et al., 2005; Naylor et al., 2006). Compton et al. (2014), for example, reported, from focus group discussions with teachers, that repetition was not mentioned as being a key component of bullying.

Even though none of the teachers in this study explicitly mentioned the repetition of the offending behavior in their answers to survey question (1)—“What do you think is going on in the situation described?”—a few teachers referred to a temporal dimension. Teachers 17 and 27 described the situation as the initial phase of bullying. In contrast to the teachers’ assessment, the vignette does not describe the emergence of bullying. The negative actions are described as having been going on for some time (Frankie has been changing who they sit next to for some time, and the bullying students have alternately ignored Frankie or called them names), and a new incident is described (“this time Alex says”). Another teacher refers to the repetition in the answers to survey question (3) about how to respond, as can be seen from the following quote:

I don’t find the idea of changing whom your sitting next to fundamentally questionable. After hearing this remark, I would (most likely) observe whether I notice these kinds of things again and only then react. (Teacher 18)

Again, the teacher apparently was not aware that the bullying had already started before the verbal attack described in the vignette. Since all of these teachers also used the label bullying in their answers, it can be assumed that their understanding of bullying is not consistent with the prevailing definition of bullying.

Hellström et al. (2021) highlight some of the recurring discussion points on repetition as a feature of bullying, such as arbitrariness when it comes to what counts as a repeated action and which timeframes are used. Since there is always a first incident of bullying, the definition by Gladden et al. (2014, p. 7) that bullying is repeated or “highly likely to be repeated,” or the proposed new definition of bullying by UNESCO (2020), which does not include repetition as a main characteristic anymore, seems to be helpful when it comes to applying the definition to everyday life in school. Thus, it would not be decisive to know whether the teachers knew that Frankie, in the vignette, had already been repeatedly insulted or not. In view of the realities of classroom life and the pedagogical practices involved, it seems more important that teachers are able to recognize bullying behaviors immediately and act accordingly than to apply characteristics of bullying on a theoretical level. Based on their findings, Macaulay et al. (2020, p. 595) similarly argue that “practitioners should acknowledge and respond to all incidents of bullying, irrespective of frequency.” A revised bullying definition without the repetition criterion would further imply that schools and the education system could not use “repetition of aggressive behaviours as the condition for taking action and lack of repetition as an excuse for doing nothing” (O’Higgins Norman et al., 2021, p. 4).

Finally, the insights gained on the awareness of the situation will be considered in conjunction with the model of the conditions and consequences of teachers’ bullying interventions (Bilz et al., 2017; Fischer & Bilz, 2019). When it comes to bullying interventions, contextual factors along with the intervention competence of the individual teacher (knowledge, motivation, beliefs, self-regulation) are considered to be decisive (Bilz et al., 2017; Fischer & Bilz, 2019). Although the answers to survey question (4)—“Please explain why you would act like this. Did something influence your choice of action? If so, what?”—were related to the motivation behind the planned intervention, they also, according to the heuristic model, provide information about what influenced the teachers’ awareness. An in-depth analysis of the conditions of the participants’ awareness and intended interventions (survey questions 2–5) is currently in progress and would go beyond the scope of this article. A preliminary analysis of this question suggests that participants’ awareness and interventions are closely related to situational aspects, personal and professional experiences, beliefs, and attitudes and only to a small extent to knowledge or training. This result seems to correspond with the teachers’ reports about anti-bullying measures in their schools and about bullying-specific education and training and how the teachers assessed the bullying vignette in relation to this education and training (see Table 4).

Table 4 Teachers’ assessment of the bullying vignette in relation to anti-bullying measures and bullying-specific training
Full size table

Of the 25 teachers who had taken part in either anti-bullying measures in their schools or bullying-specific training, or even both, 15 classified the bullying vignette as describing an incident of bullying. Of the 13 teachers who did not report any anti-bullying measures or training, 11 still recognized the bullying incident. This result underlines the relevance of situational aspects, experiences, and beliefs when it comes to bullying interventions. However, consistent with earlier research (Yoon & Bauman, 2014), this preliminary result suggests that it seems necessary for teacher education on bullying to give as much importance to working with experiences, beliefs, and attitudes as to imparting knowledge about bullying and how to deal with it properly. The reflection of situational aspects could be of particular relevance for teachers’ further in-service training and for working with case studies.

Limitations and Future Research Directions

The study deepens our understanding of teachers’ awareness of bullying and their considerations when classifying a situation as bullying. However, some limitations should be noted and taken into account when interpreting the results.

The teachers’ answers are based on hypothetical written vignettes. Facial expressions, tone of voice, and other important factors that influence how teachers perceive bullying cannot be communicated in written vignettes. Some of these shortcomings can be mitigated through the use of video vignettes (see Yoon et al., 2016), which were not available for this study. Nonetheless, a gap will always remain between what we can capture research-wise and how teachers perceive and respond to bullying in reality in school.

The analytical approach was open to categories that emerge inductively from the data. Nevertheless, prior knowledge and knowledge of the state of research may have narrowed the focus of the analysis. In addition to this, since the study was conducted by a single author, the inter-coder check for the quality control of the coding could only be carried out on parts of the data material. Therefore, the steps of the data analysis, the coding cycles, and the development of the category systems are described as transparently as possible.

In the survey, the two vignettes should have appeared in a randomized order to avoid any bias created by the order of the vignettes (Rasinski et al., 2012). Unfortunately, due to unexpected technical changes by the survey software provider (soscisurvey.de), all participants received the same survey version, with the bullying vignette presented first. Thus, it cannot be ruled out that the order of the vignettes might have led to a perception bias. Since the first vignette describes a bullying incident, one possibility could be that participants were consequently more likely to categorize the second vignette as bullying. However, this effect did not occur. Another possibility is that teachers would have perceived the bullying vignette to be less serious if they had been presented with the description of physical assault. This possibility remains open to speculation.

The results, as well as limitations of this study, offer some desiderata for future research. The method chosen seems promising in terms of gaining a better understanding of if and how teachers recognize bullying and how they subsequently respond to it. Further research that includes more bullying scenarios, possibly video vignettes, and a larger sample would be desirable to get an even more comprehensive overview of this. A mixed-methods approach with a combination of, for example, a qualitative online survey with face-to-face interviews or focus groups would encourage further questions. This approach would allow researchers to explore, in more detail, what considerations teachers rely on when identifying incidents of bullying and what conditions are most relevant to it. Despite all the methodological challenges, it would be particularly appealing, from a pedagogical point of view, to examine more closely the situational aspects and contextual factors at the micro level of the school class that are relevant to teachers’ bullying interventions.

Future research should not only investigate teachers as well as teacher education on bullying at the institutional level but also teacher educators themselves. An important question would be, for example, how the issue of bullying is actually embedded in the minds of teacher educators. In line with findings from existing research on teachers, we could expect that when it comes to dealing with bullying, for teacher educators, too, beliefs, attitudes, and experiences play at least as important a role as knowledge about bullying.

Implications for Practice

This study also reveals that one-third of the teachers surveyed did not recognize the bullying incident in the vignette and that the teachers’ considerations only partly reflect the current definition of bullying used in the research. To successfully counteract bullying, it is essential that we increase teachers’ sensitivity in this area. A stronger emphasis on this topic in teacher education is necessary. Based on the state of research for Austria (Burger et al., 2015; Hoffmann, 2020), this suggests that bullying needs to be better embedded, systematically, in teacher education curricula so that all prospective teachers are introduced to the topic of bullying during their studies. A commitment to the relevance of the topic is thus also required at the policy level in order to anchor the topic of bullying in the curricula at the next opportunity to adapt those.

Further some practical implications for teacher education can be identified. The study results underline the relevance of situational aspects, experiences, and beliefs when it comes to bullying interventions. In courses that teach the topic of bullying, the detour should therefore be taken via the personal relevance of the topic for the (pre-service) teachers in order to impart knowledge about bullying and how to deal with it properly. The use of bullying scenarios should be promoted in order to encourage reflection on the aspects of the situation that are relevant for a correct assessment of it. This format of work could be a good way for both pre-service and in-service teachers to gain or reinforce knowledge and confidence in dealing with bullying.

Conclusion

This qualitative study investigates teachers’ awareness of and sensitivity to bullying. The findings indicate that teachers were sensitive to a bullying incident when it was presented to them in the form of a vignette, especially when compared to a vignette describing a non-bullying incident. While the incident of physical violence in the second comparison vignette was not classified as bullying, a majority of respondents interpreted the verbal and socially exclusionary bullying in the first vignette as such.

The frequent use of the label bullying indicates that bullying is a well-known phenomenon, but at the same time, it does not necessarily imply a comprehensive knowledge of bullying on the part of the teachers. The things that teachers consider when classifying bullying are primarily based on the aspect of doing (intentional) harm and to a lesser extent on an imbalance of power. The criterion of repetition does not seem to be relevant for the teachers interviewed when evaluating bullying, which supports the use of a revised definition of bullying as recommended by UNESCO (2020).

However, the study also found that one-third of the teachers did not recognize the bullying incident and that teachers’ considerations only partially reflected the definition of bullying used in current research. There is therefore a need to intensify teacher education on bullying. Besides knowledge transfer, there should be a focus on working with experiences, beliefs, and attitudes, as well as bullying scenarios that can also cover situational aspects of bullying.