This study examines the factorial validity, factorial invariance across gender, and construct validity of a Swedish version of the Self-Presentation in Exercise Questionnaire (SPEQ; Conroy, Motl, & Hall, 2000). The a priori two-factor 14-item, 11-item, and 9-item models fail to reach acceptable levels of fit in a calibration sample. A modified 8-item model demonstrates acceptable fit and loose and partial cross-validity on a second independent sample. The factorial structure is invariant across gender. Men depict higher latent mean scores on both SPEQ factors of Impression Motivation and Impression Construction. Different patterns for men and women in the relation between SPEQ factors and exercise identity, social physique anxiety, and exercise participation are detected. Psychometric weaknesses in reliability and content validity from a cultural equivalence perspective are discussed. (Contains 6 tables and 4 footnotes.)