This study was conceived to assess the agreement between self- and other reports of egoistic (E) and moralistic (M) self-enhancement (SE). A self-report scale assessing E-SE and M-SE was filled out by 304 participants (mean age = 39.11, SD = 10.12, 54% female). The scale, with items worded in third person, was also completed by 304 informants (one for each target participant). Data were subjected to a correlated-traits correlated-methods model. Confirmatory factor analysis yielded two distinct factors for both self- and other ratings of E-SE and M-SE. A significant correlation across raters was observed for each factor. Proportion of trait variance was .58 for E-SE and .35 for M-SE. Proportion of method variance was .24 and .41, respectively. E-SE and M-SE were substantially correlated within each data source. However, the two measures became orthogonal after method variance was partialed out. Measures of E-SE and M-SE include both trait and artifactual components of variance, which are likely to reflect individual differences in the style of self-presentation and response bias (i.e., tendencies to exaggerate agentic and communal qualities). The trait components of E-SE and M-SE represent two unrelated tendencies that can be captured and distinguished not only through self-judgments, but also by an external observer.