The purpose of this study was to compare two methods of measuring menstrual distress. Seventy-three women, 18 to 35 years of age, selected from lower-middle to upper-middle income neighborhoods kept a daily diary for a two-month period. Following completion of the diary, the women responded to the Moos Menstrual Distress Questionnaire (MDQ) with reference to their last menstrual period. Estimates for all symptoms on the MDQ exceeded those in the diary. The greatest discrepancies between the two methods were found for water retention and negative affect symptoms. Concordance of perimenstrual symptom reporting across the two measures was statistically significant only for menstrual cramps, κ = .248, p < .021, and premenstrual backache, κ = .203, p < .036). Although these results suggest that estimates of dysmenorrhea symptoms are least affected by the data collection method, they are consistent with the hypothesized joint effects of menstrual stereotypes and recall bias on retrospective symptom reports; these sources of bias should be considered in future menstrual distress research.