This study determined the internet addiction profiles of university students with latent class analysis based on their responses to Internet Addiction Test (IAT). The study group consisted of 480 university students. The participants were classified into four groups according to their total score: “normal (0-30), mild (31-49), moderate (50-79) and severe (80 and above)” level of internet addiction, respectively (Young 2010). The performance of latent classes across six factors of IAT found substantial difference among three latent classes for salience, excessive use, neglect of work and anticipation factors. Amongst these, the mean score of highest latent class (LC3) was around 60 while it was 50 and 40 for latent class 2 (LC2) and latent class 1 (LC1), respectively, in which distinction between latent classes were obvious. However, discrepancy between higher two classes (LC2 and LC3) with respect to the factors of “lack of control and the neglect of social life” were negligible low indicating the existence of only two significant classes (LC1 and LC2) for these two factors. These results suggest that the same clustering criterion cannot be applied to each factor of IAT and using same criterion for each factor might lead to inaccurate and biased classification of individuals.

这项研究根据潜在学生对互联网成瘾测试（IAT）的反应，通过潜伏类分析来确定他们的网络成瘾状况。研究小组由480名大学生组成。参与者根据他们的总得分分为四组：“网络成瘾水平”（正常（0-30），轻度（31-49），中度（50-79）和重度（80及以上））（年轻2010）。在IAT的六个因素中，潜在类别的表现在显着性，过度使用，工作疏忽和预期因素三个潜在类别之间存在实质性差异。其中，最高潜在类别（LC3）的平均得分约为60，而潜在类别2（LC2）和潜在类别1（LC1）的平均得分分别为50和40，其中潜在类别之间的区别很明显。然而，关于“缺乏控制和对社会生活的忽视”因素，较高的两个类别（LC2和LC3）之间的差异可以忽略不计，表明这两个因素仅存在两个重要的类别（LC1和LC2）。这些结果表明，不能将相同的聚类标准应用于IAT的每个因素，并且对每个因素使用相同的标准可能会导致个人分类的不准确和有偏见。