Abstract
PURPOSE: The discourse around menstrual cycles is often pathologized, potentially reinforcing negative perceptions of menstruation. The extent to which individuals have internalized the idea that bodily and social experiences before menstruation are the manifestation of ill-health, thereby biasing reports of premenstrual experiences towards negative emotions, remains unclear. METHODS: Using an online experimental design, we investigate whether phrasing the premenstrual experience as having both negative and positive dimensions would enable individuals to report more diverse and positive experiences than are reported in the absence of specific emotional prompts. Participants were recruited using a period tracker app and randomly allocated to one of three conditions: control (describe your premenstrual experience); treatment 1 (describe your negative and positive premenstrual experience); treatment 2 (describe your posititive and negative premenstrual experience). Sentiment analysis was used to derive polarity scores, and a two-part Bayesian model assessed the impact of phrasing order. RESULTS: Among 2,637 participants, responses skewed negatively (mean -0.25). Compared to the control, treatment conditions 1 and 2 reported premenstrual experiences 64% and 62% less negative, respectively. Positive themes, notably 'sex, libido, and energy'emerged. The absence of positive prompts in questioning led to more negative and less diverse reports. CONCLUSIONS: These findings support existing literature on the predominance of negative premenstrual phases and underline the need to broaden measurements to encompass positive symptoms. The study also pioneers the use of text analysis for investigating premenstrual symptoms.