Traditional methods indicate that the scale is highly reliable in its English and Chinese language versions. On the other hand, IRT analysis identifies anomalies with individual scale items. In particular, items tend not to discriminate well between individuals who score highly on self-esteem and ...