This paper considers different approaches to measuring gender. It critically reviews gender role theorising and describes how this has informed two approaches to measuring gender as an individual phenomenon: gender orientation (the assessment of individual traits) and gender ideology (assessing individual endorsement, and internalisation, of social norms). It is argued here that social constructionist perspectives offer a viable alternative to gender role theory and that these inform an alternative approach to measuring gender as a social phenomenon: gender (re)presentation. This approach assesses group level endorsement of dominant gender representations. Endorsement is not seen to reflect individual traits or internalised social norms.Rather, it is understood as a social practice, made meaningful through shared understanding of dominant gender representation. This approach is introduced through critique of the traditional concept of attitudes and a reformulation thereof. The practical measurement implications and benefit of this reformulation are outlined.