科技大學學生網路霸凌受害經驗量表之發展

The Development of Cyberbullying Victimization Scale for University of Science and Technology Students

本研究旨在發展科技大學學生網路霸凌受害經驗量表，以臺南地區六所科技大學四技部大學生為調查對象，採分層立意取樣方式發放問卷，分二階段進行施測，第一階段是進行預試調查，共回收有效問卷190份，回收率86.63％，主要目的是為進行量表之項目分析、探索性因素分析及Cronbach's α信度分析等；第二階段是正式調查，計回收可用問卷1,921份，有效回收率為77.30％，主要目的是為驗證量表的指標信度、建構信度、聚歛效度、區別效度、及複核效度等。本研究使用SPSS18.0與Amos18.0統計軟體，進行結構方程模式（SEM）分析資料，結果發現：1、網路霸凌受害經驗量表經預試調查的項目分析及探索性因素分析共保留21題計三個因素，分別為「偽裝&詐騙」、「詆毀&排擠」、「騷擾&猥褻」，解釋變異量達57.81％，信度分析顯示Cronbach's α係數達0.93。2、整體量表經正式調查的驗證性因素分析，顯示二階模式具有最佳的適配度，其建構信度達0.91，平均變異萃取量達77.60％，且具有良好的聚歛效度、區別效度以及複核效度，顯示模式具有預測力與穩定性。

The purpose of this study was to explore the cyberbullying victimizations of undergraduate students in university of science and technology. The samples of this study were collected from 6 universities in Tainan district by stratified purposive sampling method. The results of this study were listed as following: First, The cyberbullying victimizations scale had 21 items and three factors include “masquerade & trickery”, “denigration & exclusion”, “harassment & obscene”. By exploratory factor analysis showed that total explained variance amounted to 57.81%. By reliability analysis showed that Cronbach's was up to 0.930. Second, by confirmatory factor analysis showed that construct reliability was up to 0.912 and average variance extracted amounted to 77.60%. Finally, this scale had good convergent validity, discriminant validity, and cross-validation. The model’s stability and predictive power were confirmed.