Teacher-to-Student Victimization: The Role of Teachers’ Victimization and School Social and Organizational Climates

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. P2, line 50: "yet most research on teacher-to-student victimization has been based on students’ reports (e.g., Chen & Wei, 2011)". The cited reference (Chen & Wei, 2011) is outdated and does not adequately support this claim.
2. Please provide the definition of Job Socialization.
3. There is likely a high correlation between school climateand the variables interpersonal conflict at work, job socialization, and trust in the principal, which could lead to multicollinearity in the regression model. For example, higher job socialization or higher trust in the principal would likely lead teachers to report a better school climate, while more interpersonal conflict at work would lead to reports of a worse school climate. It is recommended to re-analyze the logical relationship between these variables. Since the study does not provide a correlation matrix, the relationships between these predictor variables are unclear. If high correlations exist between school climate and interpersonal conflict at work, job socialization, and trust in the principal, potential solutions include: retaining only these three variables (interpersonal conflict at work, job socialization, trust in the principal) as predictors of the dependent variable, or retaining only the school climate variable as the predictor.
4. Questionnaires were administered during the 2019–2020 academic year. In recent years, due to the impact of the COVID-19 pandemic, have the variables measured in this study undergone changes?
5. The Cronbach’s alpha for the teacher-to-student victimizationscale was .52, which is low.
6. All other questionnaires measure aspects of the school or the teachers within the respondent's school (e.g., school climate, job socialization, trust in the principal). However, the itemsin Teachers’ victimization by students and parents seem to inquire about the respondent's own personal experiences. To maintain consistency, this scale should measure the situation for teachers within the respondent's school. The items should be revised, for example: "A parent or student threatened to complain about a teacher in my school."
7. Since the teacher-to-student victimizationitems do not ask respondents to report their own harmful actions towards students, but rather to report incidents of victimization by teachers/staff towards students within their school (e.g., "A teacher or staff member used a derogatory name, cursed or insulted a student in your school"), the respondent is acting in the role of an observer/bystander. Therefore, analyzing statistical differences (e.g., Chi-Square tests) based on the respondent's own demographic characteristics (like gender, age) is not meaningful for this particular variable.
8. Please explain the order in which the variables were entered into the regression equation. Why interpersonal conflict at workentered first and job socialization entered last? Also, why  victimization by students and parents was not included as a predictor variable in the regression equation?
9. What is the theoretical foundation of this study?

Author Response

1. P2, line 50: "yet most research on teacher-to-student victimization has been based on students’ reports (e.g., Chen & Wei, 2011)". The cited reference (Chen & Wei, 2011) is outdated and does not adequately support this claim.
   Reply: This citation has been replaced with a more recent publication by Chen and colleagues (2020[1]), demonstrating the exploration of teacher-to-student victimization based on students’ reports.
2. Please provide the definition of Job Socialization.

Reply: The definition of job socialization has been added to the beginning of the relevant section (p. 4 line 182): Novice teachers are immersed in a novel environment with both positive and negative influences on their professional identity, requiring them to adapt and gradually integrate into the school's culture (Güneş & Uysal, 2019[2]). Thus, a crucial responsibility for school leadership is facilitating teachers’ integration into the organization. This adjustment encompasses acquiring the skills necessary to perform the job effectively, accepting the established norms and practices of that particular organization, understanding how one’s role relates to the broader organizational structure, and fostering a sense of acceptance among colleagues (Gardner et al., 2022[3]).

1. There is likely a high correlation between school climate and the variables interpersonal conflict at work, job socialization, and trust in the principal, which could lead to multicollinearity in the regression model. For example, higher job socialization or higher trust in the principal would likely lead teachers to report a better school climate, while more interpersonal conflict at work would lead to reports of a worse school climate. It is recommended to re-analyze the logical relationship between these variables. Since the study does not provide a correlation matrix, the relationships between these predictor variables are unclear. If high correlations exist between school climate and interpersonal conflict at work, job socialization, and trust in the principal, potential solutions include: retaining only these three variables (interpersonal conflict at work, job socialization, trust in the principal) as predictors of the dependent variable, or retaining only the school climate variable as the predictor.

Reply: I appreciate the reviewer's concern regarding potential multicollinearity among the independent variables. I have assessed the correlations between these variables, and none exceed 0.66. Therefore, multicollinearity is unlikely to be a significant issue in the regression model. A correlation matrix has been included in Appendix A1 for reference. Further, as indicated on p. 8, multicollinearity was assessed using variance inflation factor values, and the model demonstrated no significant multicollinearity, with all values remaining below 3.04.

Additionally, it should be emphasized that while both are important predictors of teacher-to-student victimization, school social climate and organizational climate are theoretically distinct constructs, therefore I incorporated both into this study.

1. Questionnaires were administered during the 2019–2020 academic year. In recent years, due to the impact of the COVID-19 pandemic, have the variables measured in this study undergone changes?

Reply: The survey for this study was administered during the first semester of the 2019-2020 academic year, prior to the initial lockdown of schools in Israel in mid-March 2020. As such, no changes to the questionnaires were implemented to reflect the impact of the pandemic.

1. The Cronbach’s alpha for the teacher-to-student victimization scale was .52, which is low.

Reply: I acknowledge the reviewer's observation regarding the Cronbach's alpha for the teacher-to-student victimization scale, which was .52. While this value is lower than conventional thresholds, it may be attributable to the complexity and sensitivity of the subject matter, as well as the limited number of items included in the scale. This lower-than-ideal value may also be due to the heterogeneity of the scale, which assesses distinct forms of teacher-perpetrated violence rather than measuring a single, unified construct. The scale includes items ranging from relatively less severe actions (e.g., name-calling) to more severe forms of violence (e.g., physical harm), which may contribute to a lower internal consistency. Still, this scale encompasses a wide range of teacher-to-student perpetration, thus representing a valid measurement tool. This limitation has been clearly noted in the discussion section of the manuscript, suggesting that future research should consider refining the measure to enhance its reliability, possibly by examining different subtypes of teacher-to-student violence or employing subscales, to enhance reliability (p. 14, line 588).

6. All other questionnaires measure aspects of the school or the teachers within the respondent's school (e.g., school climate, job socialization, trust in the principal). However, the items in Teachers’ victimization by students and parents seem to inquire about the respondent's own personal experiences. To maintain consistency, this scale should measure the situation for teachers within the respondent's school. The items should be revised, for example: "A parent or student threatened to complain about a teacher in my school."

Reply: The study assumes that school social climate and the three organizational climate variables (trust in the principal, job socialization, and interpersonal conflict) are conceptualized as school-level constructs reflecting the overall atmosphere within a school. To illustrate, it is likely that a negative school climate, such as one resulting from a school principal treating a teacher disrespectfully, would be experienced by all teachers in that school. However, victimization is an interpersonal event that describes a personal experience, distinct from the general atmosphere within the school (although related). This approach, where school climate is measured as a school-level construct and victimization as an individual one, has been used extensively in prior research (e.g, Marchante et al., 2022[4]; Zhao et al., 2021[5]). For example, Huang et al. (2020[6]) examined teachers’ perceptions of school climate and teachers’ victimization by students. In this study, individual teachers were presented with items referring to their own experiences of victimization (e.g., “Has a student from this school threatened to injure you in the past 12 months?”), whereas school social and organizational climate items were phrased to reflect the general atmosphere at the school (e.g., “The school administration’s behavior toward the staff is supportive and encouraging”).

It should be highlighted that the decision to measure teacher-to-student victimization indirectly (e.g., “A teacher or staff member used a derogatory name, cursed, or insulted a student in your school”) was based on the understanding that teachers are unlikely to self-report their own perpetration of violence against students, due to social desirability bias and the severe legal and ethical consequences that such behavior entails.

Top of Form

 

7. Since the teacher-to-student victimization items do not ask respondents to report their own harmful actions towards students, but rather to report incidents of victimization by teachers/staff towards students within their school (e.g., "A teacher or staff member used a derogatory name, cursed or insulted a student in your school"), the respondent is acting in the role of an observer/bystander. Therefore, analyzing statistical differences (e.g., Chi-Square tests) based on the respondent's own demographic characteristics (like gender, age) is not meaningful for this particular variable.

Reply: I appreciate the reviewer's concern regarding the meaningfulness of examining statistical differences in teacher-to-student victimization based on respondents' demographic characteristics, given that the scale measures observations of such victimization by teachers. While I acknowledge that the teacher-to-student victimization items capture respondents’ perceptions as observers or bystanders, I believe that presenting these bivariate associations provides valuable context for the subsequent multivariate analyses. These initial analyses serve to establish foundational relationships and patterns within the data, which inform the development of the regression models. Furthermore, despite capturing observations, our results align with previous research (e.g., demonstrating greater teacher-to-student victimization in Arabic-language schools), suggesting that these observer reports may reflect relevant underlying realities. If the editor and reviewer ultimately believe that these results should be omitted from the manuscript, I will certainly comply; however, I believe that their inclusion adds value to the discussion

1. Please explain the order in which the variables were entered into the regression equation. Why interpersonal conflict at work entered first and job socialization entered last? Also, why victimization by students and parents was not included as a predictor variable in the regression equation?

Reply: Thank you for raising the question about the order in which variables were entered into the regression equation. Firstly, victimization by students and parents was not included as a predictor variable due to the absence of a significant association between teachers' victimization and teacher-to-student victimization in the bivariate analyses (as noted on p. 8). In the regression analysis, I first entered participants’ background characteristics to account for demographic influences. Subsequently, I entered social climate and organizational climate separately in successive steps. This hierarchical approach was chosen to not only observe the unique contributions of social and organizational climate beyond background factors but also to delineate the distinct predictive power of organizational climate relative to social climate.

9. What is the theoretical foundation of this study?

Reply: The theoretical foundation for this study is described on p. 3 (line 98): “Building on the ecology of human development model (Bronfenbrenner, 1977), school violence researchers have articulated schools at the center of a larger ecological system and suggested exploring school victimization by considering multiple risk and protective factors both inside and outside schools (Astor & Benbenishty, 2019). Because schools are positioned at the heart of the ecological system, the model suggests that school violence is shaped by the student body and family demographics and characteristics, the neighborhood, the broader community, and the larger cultural and ethnic environment. Still, the school’s internal context, including the social and organizational climate, moderates and mediates outside influences and helps shape the experiences and behaviors of the school community, including students, staff, and parents. The model also highlights the interrelationships among all members of the school community, including instances of mutual victimization (Astor & Benbenishty, 2019).”.

 

[1] Chen, J. K., Wu, C., & Wei, H. S. (2020). Personal, family, school, and community factors associated with student victimization by teachers in Taiwanese junior high schools: A multi-informant and multilevel analysis. Child Abuse & Neglect, 99, 104246.‏

[2] Güneş, Ç., & Uysal, H. H. (2019). The relationship between teacher burnout and organizational socialization among English language teachers. Journal of Language and Linguistic Studies, 15(1), 339-361.

[3] Gardner, D. G., Huang, G. H., Pierce, J. L., Niu, X., & Lee, C. (2022). Not just for newcomers: Organizational socialization, employee adjustment and experience, and growth in organization‐based self‐esteem. Human Resource Development Quarterly, 33(3), 297-319.‏

[4] Marchante, M., Coelho, V. A., & Romão, A. M. (2022). The influence of school climate in bullying and victimization behaviors during middle school transition. Contemporary Educational Psychology, 71, 102111.‏

[5] Zhao, Z., Liu, G., Nie, Q., Teng, Z., Cheng, G., & Zhang, D. (2021). School climate and bullying victimization among adolescents: A moderated mediation model. Children and Youth Services Review, 131, 106218.‏

[6] Huang, F. L., Eddy, C. L., & Camp, E. (2020). The role of the perceptions of school climate and teacher victimization by students. Journal of interpersonal violence, 35(23-24), 5526-5551.‏

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,
Thank you for the opportunity to review your article. Below, I offer constructive feedback to enhance its clarity, depth, and impact. I hope these suggestions assist in refining your manuscript and strengthening its overall contribution.

Title
Teacher-to-Student Victimization: The Contribution of Teachers’ Victimization, School Climate, Interpersonal Conflict at Work, Trust in the Principal, and Job Socialization
– The title is overly long and cumbersome, making it hard to follow. Consider rephrasing or shortening it to increase clarity and reader engagement.

Abstract
– While the abstract mentions a sample of 214 teachers from Israel (line 11), there is no detail about the type of schools or participant demographics that might give more context to the findings.
– The abstract mentions the sample and analysis but does not clearly state the study design.
– While the statistical analyses are described in the abstract, the measurement instruments used to assess key variables are not mentioned. Including at least the names or types of tools would improve transparency.

Introduction
– Lines 47–50: The authors mention that most research relies on student reports but do not critique or justify in detail why teacher reports are more appropriate. Consider elaborating on the added value of teacher perspectives.

Methodology
– There is no mention of ethical approval or institutional review board (IRB) oversight for a study involving human participants. This should be clearly stated, including the name of the approving body and reference number, if applicable.
– The authors mention that this is part of a "mixed-methods" project and describe the current study as using “quantitative measures,” but do not state whether it is cross-sectional, correlational, survey-based, etc. The study design should be explicitly named in a separate section.
– Line 273: The authors rightly point out the ethical constraints around directly assessing teacher-perpetrated violence. However, further justification of the validity and limitations of using indirect school-level teacher perception data for this purpose would be helpful.

Results
– Please revise the caption of Figure 1 to correctly reflect that the outcome is teacher-reported teacher-to-student victimization. Referring to 'students’ victimization' implies direct student reporting, which is not the case.
– The outcome is labelled in various ways: "teacher-to-student victimization", "students’ victimization", "reports of violence". This can confuse readers. Suggestion: consider using consistent terminology when referring to the dependent variable across the text, tables, and figure captions.
– The results state that the model shows excellent discrimination, but the AUC value is only shown in Figure 1. Please report the AUC value in the text and provide a brief interpretation.
– The regression results show that trust in the principal and job socialization were not statistically significant predictors. This should be noted and discussed briefly in the results section to provide a balanced interpretation.

Discussion
– The section is titled “Multivariate Prediction of Student-to-Teacher Victimization” but the whole paper (including this section) is about teacher-to-student victimization.
– Line 569: The Cronbach’s alpha for the key outcome variable was only .52. This should be explicitly acknowledged as a major limitation, not just a ‘moderate’ one, as it substantially affects the reliability of findings.
– Phrases like “job socialization reduces the risk…” or “trust decreases victimization” suggest causal relationships, which are not supported by the study design. Suggestion: the cross-sectional nature of the study does not allow for causal conclusions. Consider rephrasing causal language to reflect associations or correlations more cautiously.

Conclusion
– There is no conclusion section. I strongly recommend adding a standalone conclusion (even if brief) that effectively distils the key messages of the paper, reinforces its contributions, and outlines practical or research implications.

Author Response

1. Title
   Teacher-to-Student Victimization: The Contribution of Teachers’ Victimization, School Climate, Interpersonal Conflict at Work, Trust in the Principal, and Job Socialization
   – The title is overly long and cumbersome, making it hard to follow. Consider rephrasing or shortening it to increase clarity and reader engagement.

Reply: I appreciate the reviewer's concern regarding the length of the manuscript title. While considering the reviewer's suggestion, I briefly explored the possibility of shortening the title to "Teacher-to-Student Victimization: The Contribution of Teachers’ Victimization, School Social and Organizational Climates." However, I believe that this abbreviated title would reduce the informativeness of the title, as it is important for the journal's readership to readily understand the variables included in the study. Therefore, I maintain the preference for a more detailed and informative title that captures all of the variables under investigation. Nonetheless, if the reviewer and the editor still deem it necessary to shorten the title, I will, of course, comply.

2. Abstract
   – While the abstract mentions a sample of 214 teachers from Israel (line 11), there is no detail about the type of schools or participant demographics that might give more context to the findings.

- The abstract mentions the sample and analysis but does not clearly state the study design.

– While the statistical analyses are described in the abstract, the measurement instruments used to assess key variables are not mentioned. Including at least the names or types of tools would improve transparency.

Reply: The requested information has been added to the abstract. The relevant sentence reads as follows (p.1, line 11): “To address this gap, this cross-sectional study used a sample of 214 teachers from six Hebrew-language and four Arabic-language middle and high schools across Israel (69.2% female; 61.2% older than 41 years), to predict teachers’ reports on teacher-to-student victimization based on teachers’ victimization by students and parents, school social climate, and three dimensions of the school organizational climate: interpersonal conflict at work, trust in the principal, and job socialization:”. To maintain brevity, the names of the measurement scales were not included in the abstract. However, this information can be readily found in the Methods section.

3. Introduction
   – Lines 47–50: The authors mention that most research relies on student reports but do not critique or justify in detail why teacher reports are more appropriate. Consider elaborating on the added value of teacher perspectives.
   Reply: The importance of accounting for teachers’ perspectives on the studied phenomenon has been elaborated. This part reads as follows (p. 2, line 51): “Incorporating multiple perspectives from all members of the school community is valuable in its own right, as it reflects the unique viewpoint of each group. Incorporating accounts from teachers who have either victimized students or witnessed such actions by their colleagues is essential to expanding this line of inquiry. Such perspectives can improve understanding of the traits exhibited by teachers more inclined to mistreat students, as well as the personal and interpersonal dynamics contributing to these behaviors, viewed from the teachers' own vantage point. This insight would prove valuable in the development of effective and accessible intervention strategies for educators (Astor & Benbenishty, 2019).”
4. Methodology
   – There is no mention of ethical approval or institutional review board (IRB) oversight for a study involving human participants. This should be clearly stated, including the name of the approving body and reference number, if applicable.

Reply: To maintain anonymity during the peer review process, the Institutional Review Board statement has been removed from the manuscript. This information, however, is included in the full manuscript and has been provided separately to the journal.

– The authors mention that this is part of a "mixed-methods" project and describe the current study as using “quantitative measures,” but do not state whether it is cross-sectional, correlational, survey-based, etc. The study design should be explicitly named in a separate section.

Reply: The study design (cross-sectional) is now explicitly stated both in the abstract and the Methods section (p. 6, line 254: “This research relied on data from a larger cross-sectional mixed-methods study on the rate, scope, antecedents, and consequences of violence directed at teachers in Israel.“).  
– Line 273: The authors rightly point out the ethical constraints around directly assessing teacher-perpetrated violence. However, further justification of the validity and limitations of using indirect school-level teacher perception data for this purpose would be helpful.

Reply: Thank you for this important feedback. A limitation regarding the measurement of teacher-to-student victimization has been added to the discussion section, along with recommendations for future research to utilize multiple sources of information to more accurately capture this phenomenon. This addition appears as follows in the manuscript (p. 14, line 597): “Teacher-to-student victimization was measured indirectly in this study because caregiver reports likely underestimate their abusive behavior (e.g., Afifi et al., 2015[1]; Tonmyr et al., 2018[2]). This approach addresses concerns about socially desirable responding by teachers in disclosing self-perpetrated violence against students. However, because this indirect measurement may introduce limitations to measurement validity, it is recommended that future research examine teacher-to-student victimization using multiple perspectives, including those of both students and teachers within the same school. Incorporating these diverse viewpoints could provide a more valid and comprehensive understanding of this phenomenon.”.

5. Results
   – Please revise the caption of Figure 1 to correctly reflect that the outcome is teacher-reported teacher-to-student victimization. Referring to 'students’ victimization' implies direct student reporting, which is not the case.

– The outcome is labelled in various ways: "teacher-to-student victimization", "students’ victimization", "reports of violence". This can confuse readers. Suggestion: consider using consistent terminology when referring to the dependent variable across the text, tables, and figure captions.

Reply: The caption has been corrected accordingly, and now reads as follows:” Figure 1. Receiver operating characteristic curve for teachers’ reports of teacher-to-student victimization.“. In addition, the term for the dependent variable is "teacher-to-student victimization," which has been applied and used consistently throughout the manuscript (e.g., p. 6, line 287).

– The results state that the model shows excellent discrimination, but the AUC value is only shown in Figure 1. Please report the AUC value in the text and provide a brief interpretation.

Reply: The requested information has been incorporated into the manuscript, which now reads as follows: (p. 10, line 413): “Finally, the model demonstrated excellent discriminative ability, with an area under the receiver operating characteristic curve of .86, 95% CI [.81, .91], indicating a high level of accuracy in distinguishing between outcome groups (see Figure 1).”

– The regression results show that trust in the principal and job socialization were not statistically significant predictors. This should be noted and discussed briefly in the results section to provide a balanced interpretation.

Reply: The information was added to results section (p. 10, line 412: “Trust in the principal and job socialization did not emerge as statistically significant predictors.”) and into the discussion which now briefly discusses these insignificant contributions of trust in the principal and job socialization. This part reads as follows (p. 13-14, line 551): “While trust in the principal and job socialization exhibited significant associations with teacher-to-student victimization in the bivariate analyses, they did not emerge as statistically significant predictors in the multivariate analysis. It is probable that teachers’ reports of school social climate and interpersonal conflict at work already explained a substantial portion of the variance in the dependent variable, highlighting their importance in predicting teacher-to-student victimization while leaving little additional variance to be explained by these organizational climate variables.”.

6. Discussion
   – The section is titled “Multivariate Prediction of Student-to-Teacher Victimization” but the whole paper (including this section) is about teacher-to-student victimization.

Reply: Thank you for identifying this typographical error. It has been corrected accordingly.

– Line 569: The Cronbach’s alpha for the key outcome variable was only .52. This should be explicitly acknowledged as a major limitation, not just a ‘moderate’ one, as it substantially affects the reliability of findings.

Reply: I acknowledge the reviewer's comment regarding the Cronbach's alpha for the teacher-to-student victimization scale. While this value is lower than conventional thresholds, it may be attributable to the complexity and sensitivity of the subject matter, as well as the limited number of items included in the scale. In addition, the lower-than-ideal value may also be due to the heterogeneity of the scale, which assesses distinct forms of teacher-perpetrated violence rather than measuring a single, unified construct. The scale encompasses items ranging from relatively less severe actions, such as name-calling, to more severe forms of violence, including physical harm. While this breadth enhances the scale's validity, it also likely reduces its internal consistency. This limitation has been clearly noted in the discussion section of the manuscript, suggesting that future research should consider refining the measure to enhance its reliability, possibly by examining different subtypes of teacher-to-student violence or employing subscales. I have moved this limitation to the beginning of the “Limitations and Future Research Directions” section, to emphasize its importance. This part reads as follows (p. 14, line 589): “However, the internal consistency for teacher-to-student victimization was moderate, which may be attributable to the complexity and sensitivity of the subject matter, as well as to the limited number of items included in the scale. The moderate internal consistency may be also attributed to the heterogeneity of the scale, which assesses distinct forms of teacher-to-student victimization rather than measuring a single, unified construct. Future research should consider refining the measure to enhance its reliability, possibly by examining different subtypes of teacher-to-student violence or employing subscales, to enhance reliability.”.  

– Phrases like “job socialization reduces the risk…” or “trust decreases victimization” suggest causal relationships, which are not supported by the study design. Suggestion: the cross-sectional nature of the study does not allow for causal conclusions. Consider rephrasing causal language to reflect associations or correlations more cautiously.

Reply: The language has been revised throughout the manuscript to avoid implying causality, instead focusing on associations. For instance, "job socialization reduces the risk" was rephrased as " job socialization is associated with reduced risk of teacher-to-student victimization (p. 12, line 480)”. Another example appears on p. 13 (line 509): “trust in the principal is associated with decreased teacher-to-student victimization.“.  

7. Conclusion
   – There is no conclusion section. I strongly recommend adding a standalone conclusion (even if brief) that effectively distils the key messages of the paper, reinforces its contributions, and outlines practical or research implications.

Reply: A concluding paragraph has been added (p. 14, line 620): “Addressing a gap in the literature concerning the complexities of school violence, this study explored teachers' perceptions and experiences within their broader working environment to better understand and predict teacher-to-student victimization. The study sheds light on the less-examined phenomenon of teachers perpetrating violence against students by considering factors such as their own victimization, school social climate, and organizational dynamics. The current findings underscore the need for interventions that move beyond traditional approaches focused solely on individual factors and peer victimization. Emphasizing improvements to the overall school climate, organizational support structures for teachers, and the inclusion of diverse voices and experiences from all members of the school community—not just students—is critical. The study advances our understanding of how the dynamics among the entire school community influence school violence and safety. Ultimately, these steps may help foster a safer and more supportive environment for teachers, students and parents.”.

 

[1] Afifi, T. O., MacMillan, H. L., Taillieu, T., Cheung, K., Turner, S., Tonmyr, L., & Hovdestad, W. (2015). Relationship between child abuse exposure and reported contact with child protection organizations: Results from the Canadian Community Health Survey. Child Abuse & Neglect, 46, 198–206. https://doi.org/10.1016/j.chiabu.2015.05.001

[2] Tonmyr, L., Mathews, B., Shields, M. E., Hovdestad, W. E., & Afifi, T. O. (2018). Does mandatory reporting legislation increase contact with child protection? – a legal doctrinal review and an analytical examination. BMC Public Health, 18, 1021.‏ https://doi.org/10.1186/s12889-018-5864-0