Purpose The purpose of this paper is to describe the results of the construct validity by the known-groups method of a Brazilian cooking skills and healthy-eating questionnaire. Design/methodology/approach Responses obtained from university students (n=767) for Brazilian-Portuguese cooking skills and health eating questionnaire, surveyed online, were submitted to construct validity comparing two known groups. The t-test was used to compare differences between gender (male and female) and the level of cooking knowledge (high or low) in each measure of the questionnaire. Internal consistency was evaluated by obtaining the Cronbach’s coefficient. Findings Women showed significantly higher means than men in all scale measures, except in the self-efficacy for using basic cooking techniques (SECT), where no differences were found. Students classified as having high cooking knowledge and had higher score means in all scales compared to the students with low levels. Internal consistency was adequate for all scales (a>0.70), except for cooking attitude (CA) (a=0.33) and cooking behavior (CB) scales (a=0.59). Research limitations/implications SECT likely depends on cooking knowledge, independent of gender, suggesting further examination. Items and structure of CA and CB constructs also need to be examined more deeply. Practical implications A validated cooking skills and health-eating questionnaire demonstrated its ability to detect differences between groups, useful to provide data for further interventions. Originality/value No available cooking skills questionnaires were found that have been validated by the known-groups method regarding differences between gender and individuals’ level of cooking knowledge, as conducted in this study.