The present study applied item response theory (IRT) to the NEO five factor inventory (NEO-FFI) completed by a community based sample of adolescents. The results revealed that many of these personality items may not be discriminating well, with some traits demonstrating greater reliability than others. Furthermore, the threshold values highlighted that the majority of the items had skewed responses, suggesting a limited utility of some response categories. Generally, removing poorly discriminating items does not harm external validity, suggesting IRT reduces measurement error and increases reliability without compromising validity.