"Validation of the Narcissistic Grandiosity Scale and creation of reduced item variants": Correction to Crowe et al. (2016).

Reports an error in "Validation of the Narcissistic Grandiosity Scale and creation of reduced item variants" by Michael Crowe, Nathan T. Carter, W. Keith Campbell and Joshua D. Miller (, 2016[Dec], Vol 28[12], 1550-1560). In the article, the reliabilities (alphas) for the NPI were reported in the "NPI: Samples 1, 2, and 3" section as follows: NPI LA: α = .73; NPI GE: α = .71; NPI EE: α = .73. These values are incorrect. The correct values are as follows: NPI LA: α = .79; NPI GE: α = .79; NPI EE: α= .54. (The following abstract of the original article appeared in record 2016-03883-001.) The Narcissistic Grandiosity Scale (NGS) is a short adjective-based measure of narcissistic grandiosity (Rosenthal, Hooley, & Steshenko, 2007). The NGS has already shown promise as a measure of grandiose narcissism, but it has never been the subject of a formal validation study. In the current study (N = 870 across 3 samples), the factor structure of NGS was examined and item response theory analyses were used to generate abbreviated versions of the scale. The NGS scales' relations to measures of grandiose and vulnerable narcissism, the five-factor model (FFM), the interpersonal circumplex, self-esteem, and the Personality Inventory of the Diagnostic and Statistical Manual for Mental Disorders-Fifth Edition (DSM-5, PID-5) were assessed. The correlation profile of the NGS was also correlated with expert ratings of prototypical cases of narcissistic personality disorder using both the FFM and PID-5 trait profiles. Overall, the NGS was found to be a unidimensional measure of narcissistic grandiosity with good convergent, discriminant, and criterion validity. The abbreviated versions of the NGS manifested strong reliability and associations entirely consistent with the full version. (PsycINFO Database Record