Bullying is a global public health event that frequently occurs in primary and secondary schools.It can cause severe short-term and long-term negative impacts on both the bully and the bullied.There is no bullying without bullies,so identifying bullies is important.Although many studies have described the characteristics of bullies,these characteristics are scattered in various studies,there is a lack of generalization of various characteristics of bullies in the same study,and there are few theoretically guided bullying qualities.Sexuality studies have not been able to form an identifiable indicator system for bullies.This study is based on the evaluation-potency-accuracy(EPA model)theory of class attribute thinking,and obtains the generalization indicators and thematic characteristics of bullies;after multi-group questionnaire evaluation and teacher nomination research,the final develop an indicator system for identifying potential bullies for reference by bullying intervention practitioners.Study 1 explored characteristics,valence,and gender differences of bullies.A semi-open questionnaire was constructed based on the EPA model,and 200 primary and secondary school students in southwest China were recruited to fill in the characteristic vocabulary of the bully’s personality,behavior,physiology,learning and family.Transcribe feature word texts into NVIVO11.0,use topic analysis to form generalized indicators of bullies and refine topics,and use chi-square calculation tools to analyze gender differences in generalized indicators.As a result,2835 initial codes were obtained,17 main categories and 48 subcategories were extracted from the 2737 codes,and five themes were formed.Most of the characteristics of bullies are negative,however bullies also display some positive characteristics.The results also found significant gender differences in nine bully characteristics.Study 2 explored the accuracy and cross-group differences of bully feature words and dimensions.The characteristic words formed in the first study were extracted,and38 personality characteristic words,25 behavior characteristic words,3 physiological characteristic words,2 learning characteristic words and 6 family characteristic words were obtained.720 middle school students were recruited in schools in Southwest and Shandong.Students were asked to judge the degree of accuracy of feature words in evaluating bullies,and in order to further confirm the accuracy of the dimension,they answered the student group personality part according to the pre-constructed theoretical dimension.SPSS 25.0 was used for exploratory and confirmatory factor analysis,because the vocabulary of the personality part was the most complex and the vocabulary of the other parts was more direct and easy to understand.The results show that 98.6%of the feature word scores above 3 tend to be stable and have high accuracy;the dimension obtained from the exploration is stable,and the model fitting index obtained from the verification is good,and the dimension is stable and reflects the overall characteristics of the bully.Accuracy,but 5 feature words with specific support dimensions are deleted.Study 2 further recruited 300 teachers and 300 parents through the Internet to conduct the same procedure,and analyzed whether the overall judgments of the adult and student groups were consistent according to the scores.The results show that the Fleiss kappa value of the teacher and parent group and the student group is 0.267,and the consistency is low.Study 3 activated the application for the characteristics of bullies,and further checked the accuracy of the characteristics of bullies.Based on the final bully categorization characteristic index formed by the student group of the second middle school as the standard,200 middle school students and 5 class teachers in southwest China were recruited to evaluate the bully detection rate by the bully categorization index.Before the questionnaire is distributed,the seats are numbered,and students complete the anonymous bullying and bullying questionnaires to obtain the bully selfassessment list.The two lists were compared to determine the accuracy of the indicators of characteristics of bullies,and the bullies nominated by the head teacher had a 50%detection rate.Based on the guidance of EPA theory,three studies conducted qualitative and quantitative research on the characteristics of bullies in primary and secondary schools,and formed the typical generalized characteristics of bullies in five aspects: personality,behavior,physiology,learning and family.Most of the characteristics of bullies are negative,however bullies may also display some positive characteristics.From the perspective of accuracy,the dimension of the bully’s characteristic index and the characteristic vocabulary are highly accurate,which has a certain effect on improving the identification of bullies.There are some differences in the descriptions of the characteristics of bullies among students of different genders;there is a large difference between the descriptions of the characteristics of bullies by teachers and parents and the descriptions of students.The indicators of bullying characteristics are obtained by various methods and have high reliability.However,due to the influence of time and the epidemic,the research on the identification of bullies in multiple schools and multiple classes has not been carried out,and the self-evaluation of bullies may be underestimated,so it is very likely that there are other unexposed bullies in the nomination list of the class teacher.However,for these reasons,the detection rate of bullies is not high.Future research can continue to explore the characteristics of bullies in depth by distinguishing school age,and further expand the research on the nomination of students and class teachers to identify bullies,so as to provide more accurate standards for reducing the occurrence of bullying. |