由於大部分的人格測驗是自陳量表，因此應徵者對於人格測驗是否會有意圖地作假，一直是個 很重要的研究議題。本研究綜合運用試題反應理論（IRT）的差異試題功能（DIF）分析方法和多群 組驗證性因素分析（MCFA），從試題層次、試題組合層次、和量表層次，探討比較應徵者和在職 者在社會期許量表和人格測驗中作答反應差異，進而剖析作假的本質。研究發現：在試題層次，應 徵者的得分並非全然比在職者高，且DIF 分析結果顯示不論是人格測驗或社會期許量表中，皆有一 些題目為應徵者傾向於高估，而另有一些題目為在職者傾向於高估，此現象使得在試題組合層次， 兩組各自形成其獨特的作答反應組型；在量表層次，跨群組的因素結構恆等性檢定結果顯示不同樣 本群組的因素結構皆有差異。最後針對研究結果，提出實務應用和研究方法的建議。

The question of whether applicants respond to self-report personality measures differently when responding for selection purposes has been a crucial concern for decades. This study conducted an item response theory (IRT) based differential item functioning (DIF) procedure and multi-group confirmatory factor analysis (MCFA) to investigate item-level, item-composite level and scale-level response variations on polytomous Likert-type personality scales between applicants and incumbents, and thereby analyzed the essence of applicant faking. The results indicated that: (a) in the item level, applicants did not constantly score higher than incumbents; (b) several items exhibited differential item functioning, however, because DIF items did not systematically function with bias toward a particular group, therefore lead to specific response patterns for the two groups in the item-composite level; (c) results of MCFA showed that the personality factor structures are different across groups in the scale level. Implications based on the study findings are also discussed.