Corry Nida, Merritt Rebecca Davis, Mrug Sylvie, Pamp Barbara
Department of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, Baltimore, MD 21224, USA.
J Pers Assess. 2008 Nov;90(6):593-600. doi: 10.1080/00223890802388590.
In the first study, we administered the 40-item Narcissistic Personality Inventory (NPI; Raskin & Terry, 1988) to 843 female and 843 male college students, most of whom were Euro-American, to comprehensively assess the NPI factor structure using confirmatory factor analysis. Initial exploratory common factor analyses (N = 724) revealed a 2-factor model (Leadership/Authority and Exhibitionism/Entitlement). Subsequently, we used confirmatory factor analysis in a separate sample (N = 724) to evaluate the Emmons (1987) 4-factor model, the Raskin and Terry (1988) 7-factor model, the Kubarych, Deary, and Austin (2004) 2- and 3-factor models, and our 2-factor model. Finally, we assessed construct validity by correlating the scale scores with the Five-factor model of personality in an independent sample (N = 238). The 2-factor models for the NPI we obtained in this study and by Kubarych et al. (2004) appeared to be the most parsimonious models, with both a good fit to the data and satisfactory internal consistency values; so they are recommended for use. However, additional NPI research is needed to rescale, modify, or omit several NPI items and develop gender-equivalent items.
在第一项研究中，我们对843名女大学生和843名男大学生施测了包含40个条目的自恋人格问卷（NPI；Raskin & Terry，1988），其中大多数是欧美裔，通过验证性因素分析全面评估NPI的因素结构。最初的探索性共同因素分析（N = 724）揭示了一个双因素模型（领导力/权威和表现欲/特权感）。随后，我们在另一个样本（N = 724）中使用验证性因素分析来评估埃蒙斯（1987）的四因素模型、拉斯金和特里（1988）的七因素模型、库巴里奇、迪尔里和奥斯汀（2004）的双因素和三因素模型，以及我们的双因素模型。最后，我们在一个独立样本（N = 238）中通过将量表得分与五因素人格模型进行相关分析来评估结构效度。我们在本研究中以及库巴里奇等人（2004）所获得的NPI双因素模型似乎是最简约的模型，它们对数据的拟合度良好且内部一致性值令人满意；因此推荐使用。然而，需要进行更多的NPI研究来重新调整、修改或删除几个NPI条目，并开发性别等效的条目。
