The Risser Patient Satisfaction Scale was used to evaluate the effectiveness of implementing the key concepts of primary nursing. An experimental control design was used. Forty-six subjects from two units completed the questionnaire. Estimates of reliability and homogeneity are reported. There is reason to question the discriminant validity of the subscales. No significant differences were obtained between the units on any of the subscales or total scale score. A number of explanations are offered for interpreting the measuring of the nonsignificance of the differences. Work needs to continue on revising existing patient satisfaction measures or developing new ones. Other criteria with theoretical importance should be used in conjunction with satisfaction measures in assessing the effects of primary nursing.