Effects of anonymity and experimenter demand on client satisfaction with mental health services

Among the methodological issues associated with the measurement of client satisfaction are those concerning anonymity of response, experimenter demand effects, and reliability of instruments. In this study, 100 inpatients were randomly assigned in a 2 X 2 factorial design; subjects were asked or not asked to sign their names to the CSQ-8 questionnaire and were asked or not asked to "tell it like it is." The reported satisfaction of those given anonymity was significantly lower but there was no effect due to demand. Furthermore, subjects in the anonymity group had lower scores on all 8 items of the questionnaire. A factor analysis of the CSQ-8 revealed one major factor; the internal consistency of the scale was .93. The CSQ-8 was found to be a useful measure of overall client satisfaction but, because ceiling effects are a pervasive problem in client satisfaction research, anonymity should be assured whenever possible.