This study compares condom use reported two ways. 251 heterosexual participants (63% female) reported condom use on a prospective daily diary and on a retrospective questionnaire. Proportion of condom use with vaginal sex was calculated from the diary data and contrasted with retrospective categories. Responses were consistent for some participants, especially those who used condoms never or always, but responses from others showed considerable variability. Participants with few sexual encounters were more likely than those with more encounters to use "never" and "every time" endpoints. The results call into question the way participants interpret the meaning of retrospective categories.