教師關懷、社團影響、學校活動與青少年偏差行為之研究

Teacher Caring, Influence of Student Societies, School Activities and Delinquency

本研究旨在探討教師關懷、社團影響及學校活動，與青少年偏差行為的發生與否及發生頻率間之關聯性。研究數據取自臺灣教育長期追蹤資料庫2007年針對臺灣高中職、五專學生所進行的第四波調查資料，有效樣本為4,056名學生；使用零膨脹負二項迴歸模型進行分析，以了解教師關懷等因素是否影響青少年偏差行為的發生及發生頻率。研究結果顯示：教師關懷、社團影響及學校活動雖非偏差行為發生與否的影響因素，但卻能降低偏差行為的發生頻率。換言之，教師關懷等因素均非青少年偏差行為發生與否的觸發因素；但感受較多的教師關懷、社團正向影響或學校活動投入，將使青少年偏差行為發生頻率隨之降低。背景變項中，性別同時影響偏差行為發生與否及發生頻率，即男生出現偏差行為的機率及頻率高於女生；相較於都市，城鎮地區青少年發生偏差行為機率較低；就讀綜合學程的青少年，偏差行為發生頻率比普通學程更高。最後，依據研究結果提出建議，以供學校、教師作為實際輔導工作及未來研究之參考。

Hirschi’s social control theory implies that when students lack a sense of attachment or identity, or dislike participating in school activi­ties, deviations or criminal behavior might occur. We hypothesize that the effects of school factors should play important roles in the expla­nation of youth delinquency. If teenagers have negative relationships with teachers, or dislike going to school, these are likely to result in the occurrence of deviant behavior. Few empirical studies have analyzed the influencing factors that trigger the “occurrence” of delinquency, and whether relevant factors have an effect on the “frequency” of the occurrence of deviant behavior. Therefore, this study attempted to examine the impact and influence which teacher caring, student societies and school activities have on the occurrence and frequency of delinquent behavior by using the zero-inflated negative binomial regression model. Adopting the dynamic change development and life course interpretation perspective allowed us to explore the causes of various developmental aspects of adoles­cents’ deviant behavior.The study utilized data collected from 4,056 high school, technical high school and five-year junior college program students in Taiwan under the Taiwan Education Panel Survey (TEPS) published in 2007. The database was planned and funded by Academia Sinica and the Ministry of Education in Taiwan. TEPS was the first database in Taiwan to sample and collect long-term data from junior high school, high school, vocational high school and five-year junior college students across the country. Deviant behavior usually occurred beyond normality and routine. The actual situation was much more complex and unlikely to present itself in a normal statistical distribution. The occurrence of deviant behavior in a group is relatively rare. Consequently, it tends to be unique and occur in complex situations, rather than being normal or routine; as a result, it is difficult to identify in normal statistical distri­butions. Therefore, the use of behavioral types or continuous values cannot reflect the true situation of deviation behavior. In order to overcome the excessive dispersion and heterogeneity caused by a large number of “zeros” in the data, this study utilized the zero-inflated negative binomial regression model (ZINB) in statistical analysis. Through ZINB analysis, first, based on the number of occurrences of deviant behavior among adolescents, students were divided into sus­ceptible groups and non-susceptible groups. Then, for the susceptible groups, a negative binomial regression model was further used to test the impact of independent variables on deviant behavior.The major findings of this study were as follows: (1) The influence which teacher caring, student societies and school activities had on students was not significantly associated with the occurrence of delin­ quency. However, students who experienced a higher level of teacher caring and more positive experiences in student societies and school activities appeared to have a lower frequency of delinquency. (2) Gen­der was significantly associated with the occurrence and frequency of delinquency, such that male students showed higher levels of occur­rence and frequency of delinquency than female students. (3) Students from townships showed a lower level of occurrence of delinquent behavior than those from cities. (4) Students from comprehensive high schools showed a higher level of frequency of delinquent behavior than those from normal high schools. Based on the research results, this paper concludes with suggestions of applied counseling work references for schools and teachers and possible directions that future research may take.