大學生自戀性格、同理心、道德脫離與網路霸凌關係之研究

Abstract

本研究旨在探究大學生網路霸凌之現況，同時瞭解不同性別及有/無網路霸凌傾向之大學生在自戀、同理心、道德脫離與網路霸凌之差異，以及大學生之自戀、同理心、道德脫離對網路霸凌之預測性。本研究採用問卷調查法，翻譯自國外學者所編制之「自戀性格量表」與「道德脫離量表」，採用林苡彤（2009）的「同理心量表」，並自編「網路霸凌量表」以進行施測。蒐集臺灣地區各大專院校共473名有效樣本為研究對象，分別以描述性統計、ANOVA檢定、MANOVA檢定、相關性檢定及SEM結構方程分析等進行統計處理。主要研究發現如下：（一）大學生具有網路霸凌傾向之人數為366人，比例高達七成七，其中男性、大二生及填寫網路問卷者的網路霸凌傾向之比例較高；（二）男、女性網路霸凌傾向者雖然在同理心上沒有顯著差異，但男性網路霸凌傾向者的自戀、道德脫離與網路霸凌程度皆顯著高於女性；（三）不同問卷形式之網路霸凌傾向者雖然在同理心上沒有顯著差異，但問卷來自網路之網路霸凌傾向者的自戀、道德脫離與網路霸凌程度皆顯著高於紙本問卷的網路霸凌傾向者；（四）大學生網路霸凌傾向者的網路霸凌假設模型成立，模型結果顯示大學生網路霸凌傾向者的「自戀性格」會正向預測「同理心」，透過負向預測「道德脫離」，從而正向預測「網路霸凌」，且自戀性格與道德脫離也能正向預測網路霸凌，而同理心則無；（五）男、女性網路霸凌傾向者在自戀、同理心、道德脫離對網路霸凌的假設模型上之差異未達顯著水準，顯示男、女性網路霸凌傾向者之假設模型相同，然而，假設模型卻不成立，僅自戀性格與道德脫離能正向預測網路霸凌；（六）網路/紙本問卷網路霸凌傾向者在自戀、同理心、道德脫離對網路霸凌的假設模型上之差異未達顯著水準，顯示網路/紙本問卷網路霸凌傾向者之假設模型相同，且假設模型成立。研究結果顯示兩種問卷形式的網路霸凌傾向者之「自戀性格」能正向預測「同理心」，透過負向預測「道德脫離」，從而正向預測「網路霸凌」，且自戀性格與道德脫離也能正向預測網路霸凌，而同理心則無。綜合上述，大學生具有網路霸凌傾向之現象已相當嚴重，尤其自戀性格與道德脫離是造成此現象的主要因素，且不管問卷來自於網路或紙本，自戀性格都會因為同理心的介入，而降低道德脫離，進而抑制網路霸凌的參與。最後，根據研究發現，自戀性格、同理心與道德脫離對網路霸凌並非只有直接的影響效果，四者之中尚存有中介模式。另外，對於大學生網路霸凌之現象，則可針對自戀性格與道德脫離進行介入與輔導，也可透過提升同理心之策略，減緩自戀大學生在道德認知上的不一致，進而降低其參與網路霸凌之機會。

The main purposes of this study were to: 1) explore the phenomena of cyberbullying among college students; 2) analyze the differences on narcissism, empathy, moral disengagement and cyberbullying between male and female students who displayed the tendency of cyberbullying, and between students who had never cyberbullied and who had cyberbullied; and 3) investigate the path model of the relationship among narcissism, empathy, moral disengagement, and cyberbullying for students who had a cyberbullying tendency, and examine whether this path model might vary due to gender and methods of data collection (online or paper-based questionnaires). Participants were 473 students from different colleges in Taiwan. The instruments used in this study include 16-item Narcissism Personality Instrument (NPI-16), Empathy Scale, Moral Disengagement Scale and Cyberbullying Scale. Univariate Analysis of Variance (ANOVA), Multivariate Analysis of Variance (MANOVA), Pearson product-moment correlation and Structural Equation Modeling were used to analyze the data. The main findings of the study were as follows: 1. Three hundreds and sixty-six college students had participated in cyberbullying at least one time, and the proportion was up to 77%. Among these students, male, sophomore, and students who filled out questionnaires online occupied the highest proportion. 2. Male who had a cyberbullying tendency scored higher in narcissism, moral disengagement, and cyberbullying than female, but there was no difference in empathy. 3. Students who had a cyberbullying tendency and who filled out online questionnaires scored higher in narcissism, moral disengagement, and cyberbullying than ones who filled out paper-questionnaires, but they did not differ in empathy. 4. The proposed mediation model was supported. There was a direct effect from narcissism to cyberbullying, and an indirect effect from narcissism to cyberbullying through empathy and moral disengagement for students who had a cyberbullying tendency. 5. The path model did not vary according to gender, but the proposed mediation model was not supported. Narcissism and moral disengagement positively predicted cyberbullying, but empathy did not predict moral disengagement as well as cyberbullying. 6. The path model did not vary according to methods of data collection. Narcissism and moral disengagement positively predicted cyberbullying. but empathy did not predict cyberbullying. There was a direct effect from narcissism to cyberbullying, and an indirect effect from narcissism to cyberbullying through empathy and moral disengagement for students who had a cyberbullying tendency.