Societal attitudes to menstruation are largely negative, and so serve to reinforce the inferior social status of women. These attitudes have extended into medical practice and research on menstruation. As a result, much research on 'premenstrual syndrome' has been biased both in methodology and interpretation. Correlations between hormone levels, mood changes and other behavioural measures, which cycle together, are often incorrectly assumed to indicate causal relationships. These problems are discussed along with the report of a pilot study which used a new, and more comprehensive method of studying menstruation. This study has demonstrated numerous correlations between hormone levels, mood and psycho-physical performance. A relationship was found between task performance and early education about menstruation. Studied of the kind reported here provide a basis on which to formulate new questions about causal influences, which can only be established by entirely different methods of study.