"Reals", "roles", and demand features: A critical look at interview versus questionnaire measures of subjective states.

Abstract

AbstractThis study tests the hypothesis that questionnaire self - report measures of altered state experience are, by their very nature, affected by demand features (Orne, 1970), while interview self - report measures of experience are more resistant to these vulnerabilities. It was specifically hypothesized that an interview measure would distinguish the experiences of actual meditators (reals) from their role - playing counterparts (roles), while a questionnaire measure, the PCI (Pekala, 1991), would fail to so distinguish. It was further hypothesized that the order in which subjects were to complete the measures would influence their PCI intensity scores, but not the content of their interviews. These hypotheses were generally confirmed. Together with additional differences between the reals and roles groups, these findings support the greater validity of more subjective interview methods in research on qualitative states.States of consciousness in another cannot be directly observed. One cannot know, in any detail, what another person is experiencing without that person providing some particular indication or report. Researchers investigating altered states of consciousness (e.g., meditation and hypnosis) must choose between two self - report methods: the directed interview, generating descriptions later open to empirical content analysis of various kinds, and the more widely used (Fear, 1978) questionnaire. The latter is usually favoured by virtue of its efficiency and supposed greater objectivity. The present study is concerned with the question of whether there are principles of psychological research on states of consciousness in general which would favour one or the other method.There are initial advantages and disadvantages to both questionnaire and interview techniques. In general, the advantages of the questionnaire center on convenience. The questionnaire can be administered to a larger number of people inexpensively (Rosenthal & Rosnow, 1984). In contradistinction to what is customarily believed about interviews, the behavior of the researcher, who need not even be present, will not directly influence the subject's response to any one question. Nonetheless, the necessary specificity of questionnaire items must create their own demand featuresand subject expectations.The interview allows the researcher to ask questions that are more open - ended and to clarify the subject's understanding of a given question. The subject is allowed to respond in his/her own words, and in greater detail. This can result in much more information than with the structured, fixed - response questionnaire. The danger exists, of course, that the interviewer might unwittingly lead the subject's answers. While this may be true in principle, very little formal research has been done on this problem within the states of consciousness field. Nor has there been any attempt to compare interview and questionnaire measures in terms of the demand characteristics and other unintended influences on the subject.Typically in consciousness research what begins as an open - ended interview ends up as a standardized questionnaire. Pekala's (1991) Phenomenology of Consciousness Inventory (PCI) epitomizes this development. The PCI condenses a wide range of research into consciousness (from William James to Tellegen) into a single questionnaire. Given this extensive survey of the literature, it is reasonable to expect that the PCI should capture multiple dimensions of experience.In the present research the PCI will be compared with an interview and content analysis format developed by Hunt (1971; Hunt & Chefurka, 1976; Hunt & Popham, 1987) for the assessment of experience in meditation and systematic introspection. Because the PCI was able to successfully distinguish groups of subjects who had undergone hypnotic induction from a control group, simply sitting silent and still with eyes closed, it can be considered a valid measure of altered state induction (Pekala, 1991). …