Background : Constipation is a very common medical issue among general humans worldwide. The association between physical activity (PA) and constipation is still inconsistent. Additionally, no evidence regarding this issue in young population. Aims : This study aimed to evaluate the association between self-reported physical activity and constipation in Japanese young population, taking the presence or absence of an exercise partner as an additional variable. Methods : The study subjects consisted of 12497 Japanese university students. Information on constipation, exercise frequency, exercise intensity, and exercise partners was obtained from a self-administered questionnaire. Age, sex, body mass index, drinking, smoking, anemia, and sport injury were selected as potential confounding factors. Results: The prevalence of constipation was 6.5% in this cohort. Low, moderate, and high frequency of exercise was independently inversely associated with constipation (adjusted odds ratios [a OR] were low: a OR 0.77 [95% confdence interval (CI) 0.63–0.94], moderate: a OR 0.75 [95% CI 0.60–0.94] and high: a OR 0.70 [95% CI 0.53–0.91], p for trend p=0.002). Low, moderate, and high intensity of exercise was independently inversely associated with constipation (low: a OR 0.77 [95% CI 0.62–0.97], moderate: a OR 0.77 [95% CI 0.63–0.95] and high: a OR 0.70 [95% CI 0.53–0.87], p for trend p=0.001). Exercise with groups and with friends was independently inversely associated with constipation whereas the association between exercising alone and constipation was not signifcant (groups: a OR 0.70 [95% CI 0.53–0.90] and friends: a OR 0.56 [95% CI 0.42–0.74]) Conclusion: In the young Japanese population, frequency and intensity of exercise and presence of exercise partner might be independently inversely associated with constipation.