Abstract
Eysenck's PEN model is one of the most relevant and fruitful models with empirical support, and continues eliciting a large research corpus. Neverthe less, the systematic limitations regarding the psychoticism dimension and questionable inclusion of social desirability as a personality dimension have limited the model. The current research aimed to estimate an alternative PEN model including social desirability as a control and test its validity and reliability. This sample consists of 2969 Spanish young adults. Confirmatory factor analysis was carried out to test the fitting of four different models to the data. Once the best-fitting model was obtained, multiple-group analyses were carried out to assess the configural, metric, and scalar invariance of the model across sexes. The results showed that the three-dimension PEN model and two-dimension EN model controlling social desirability best fit the data and were invariant across sexes. Despite the apparent appropriateness of both models, the EN model controlling for social desirability is more appropriate due to the weakness of the P dimension.