The Dimensionality and Structure of Trait Narcissism

Although trait narcissism has been studied empirically since 1980, there is still much debate and uncertainty about its conceptualization and assessment. Notably, there are currently two theorized forms or presentations of narcissism, “grandiose” and “vulnerable,” whose definitions and labels vary across the literature, and whose scales are often psychometrically problematic. As such, observed correlations between measures of the two purported forms/ presentations range from near-zero to large, and the two forms have distinct nomological nets. It is, therefore, unclear whether the two forms represent two distinct constructs or two expressions of a common, underlying narcissism factor. This issue is complicated by debates among researchers over whether “pathological” narcissism is an extreme form of “normal” narcissism, or the two are entirely separate traits. The current study examined the relation between “grandiose” and “vulnerable” narcissism, and between “normal” and “pathological” narcissism in order to resolve these debates by clarifying narcissism’s underlying structure. Three samples were combined for this study, resulting in a total included sample of 866 psychiatric outpatients, high-risk community members, and undergraduate students who completed batteries of multiple self-report scales assessing personality traits, functioning, and psychopathology. Correlations and regressions were used to examine the relations between several narcissism scales and scales assessing other constructs. Additionally, several exploratory and confirmatory factor analyses were conducted to further clarify the relation between the various “types” of narcissism and identify/interpret the common variance connecting them. Results indicate that “normal” and “pathological” narcissism share a “core” of antagonism-related variance that varies with respect to the presence of general personality pathology. Furthermore, “vulnerable” narcissism measures assess general personality pathology without any narcissism-specific variance, indicating that “grandiose” and “vulnerable” narcissism are distinct traits but for their shared relations with pathology and dysfunction. Therefore, the current study concludes that trait narcissism itself is an antagonism-based construct, and that its noted manifestations and “forms” vary based on the amount of general personality pathology present. It is hoped that the current study will help narcissism research continue with more clarity, as well as contribute to the overall quality of personality research. Implications, recommendations, and future directions are discussed.