Etic psychometric tools work less well in non-Western than in Western cultures, whereas data collected online in the former societies tend to be of superior quality to those from face-to-face interviews. This represents a challenge to the study of the universality of models of personality and other constructs. If one wishes to uncover the true structure of personality in a non-Western nation, should one study only highly educated, cognitively sophisticated Internet users, and exclude the rest? We used a different approach. (...) We adapted a short Big Five tool, previously tested successfully in 19 countries on all continents, to Mongolian culture. EFA and CFA analyses across a nationally representative sample of 1,500 adult Mongolians recovered the Big Five satisfactorily. A Big Two model was also recovered reasonably well. Correlations between personality traits and age, as well as gender differences, were not different from those reported for Western samples. Respondents with higher education, or higher-than-average socioeconomic status, or urban dwellers, or Internet users, did not yield a clearer Big Five than the whole sample. Our method may be preferable to exclusion of specific demographic groups in Big Five studies of non-Western populations. (shrink)