Menstrual cycle as focus of study and self-reports of moods and behaviors

Retrospective questionnaires show cyclical variations in moods and behaviors across the menstrual cycle. However, results obtained from daily mood questionnaires are inconsistent. In the present study, which ran for 11 weeks, self-report measures of menstrual symptomatology, using the 8-factor Moos (1968, 1969 a,c) Menstrual Distress Questionnaire (MDQ), were investigated. The MDQ was administered under conditions that made the menstrual cycle a salient (retrospective questionnaire) or not a salient (daily questionnaire) part of the study. The study included women who were taking and not taking oral contraceptives. A 2×3 analysis of variance (with the two groups of women and three menstrual cycle phases as independent variables) yielded broad cyclical variations only in the menstrual cycle salient condition. When the menstrual cycle was not a salient part of the study, only the pain factor reached significant cyclical variation (p<.01).A 2×3 analysis of variance (with the two types of questionnaires and three menstrual cycle phases as independent variables) indicated that for women not taking oral contraceptives the two questionnaires differed on pain (p<.01), concentration (p<.01), autonomic reaction (p<.05), and water retention (p<.01). Phase effects were significant on all factors, with largest cyclical variations in the menstrual cycle salient condition. A similar analysis for women taking oral contraceptives yielded fewer significant differences. These results suggest that questionnaires that make the menstrual cycle a salient part of the study may exaggerate possible cyclical variations in moods and behaviors.The authors are grateful to Dawn Christie Burns and Ross Hodges for their help in data reduction.