The construct of intuitive eating is most often measured using the 23-item Intuitive Eating Scale-2 (IES-2). Here, we examined the psychometrics of a Romanian translation of the IES-2. Eight-hundred-and-thirty Romanian adults completed the IES-2 along with measures of positive body image, symptoms of disordered eating, and psychological well-being. Exploratory factor analyses (EFAs) with a split-half subsample (n = 420) indicated that IES-2 scores reduced to three factors in women and four in men, both of which diverged from the parent model. Confirmatory factor analysis with a second split-half sample (n = 410) indicated that the parent model had poor fit indices, whereas fit of the EFA-derived models were acceptable but not uniformly ideal. Scores on the 3-factor model - which had comparatively better fit of the models tested - had adequate internal consistency and evidenced scalar invariance across gender. However, evidence of test-retest reliability after four weeks (n = 205) was poor and evidence of construct validity, assessed through correlations with additional measures included in the survey, was weak at best. Based on these results, we question the degree to which the construct of intuitive eating can be applied to nations undergoing nutrition transitions.