Patient satisfaction is a variable of increasing interest to researchers, clinicians, and medical educators. Of several studies reviewed, only a few have shown evidence of careful methodology. Most surveys have focused on general evaluations of doctors and/or health care services or of a particular facility. The present article reports the development of a scale to measure patient satisfaction with an encounter with a physician or other primary care provider. Methods of item generation and pretesting are detailed. The overall reliability of the scale (Cronbach's coefficient alpha) is 0.93. The distribution of satisfaction scores is broader than that reported for other scales and approaches the normal in shape. Clinical and research applications of the scale are suggested.