Purpose: The purpose of this study was to develop a Korean version of Physical Activity Scale for the Elderly (K-PASE) from PASE and to evaluate the validity and reliability of the K-PASE. This scale was originally developed by Washburn et al. (1993). A valid and reliable physical activity scale for the elderly is needed to assess accurately physical activity of the elderly, because there has been no valid and reliable physical activity scale for the elderly in Korea. Methods: The K-PASE was developed from PASE through linguistic validation and cultural adaptation for use with this population. Convenient sampling was used to recruit participants. Reliability was evaluated by conducting the test-retest and convergent validity was evaluated by Pearson correlation. Results: Test-retest reliability, assessed over a 2 week interval, was r=.94. Convergent validity was established by correlating the KPASE scores between related variables which were the Tokyo Metropolitan Institute of Gerontology Index of Competence (TMIG-IC) and hand-grip strength. Scores of the K-PASE had a low significant correlation with TMIG-IC (r=.246, p<.001), and hand-grip strength (r=.251, p<.001). The mean score of physical activity of Korean elderly assessed by the K-PASE was 96.5. Conclusion: We conclude that the K-PASE is a reliable and valid instrument to assess physical activity for Korean elderly.