Misunderstanding the Adjective Generation Technique (AGT): Comments on Bem's rejoinder 1

We are pleased to note that Bem (1977) does not question our basic results (AUen & Potkay, 1977). Once again we have demonstrated that individuals' self-descriptions vary markedly on a day-to-day basis (AUen & Potkay, 1973). In addition, this time we have shown that the intraindividual variance is controlled in large part by "significant events." Accepting our results, Bem's primary objection to the research is that the Adjective Generation Technique (AGT) is a nonidiographic measvtre. Contrary to this opinion we believe the AGT to be a uniquely idiographic person-centered instrument. It relies entirely upon the individual's own phenomenology and construct system for adjective descriptions. It allows each person fuU opportunity to convey his or her own central set of predispositions, at the discretion of each person alone. The AGT is not "quasi-projective," it is fully projective. Subjects simply are asked to think of themselves at the end of a given day and to write down adjectives which best describe themselves. The openendedness of this instruction is clear and obvious. No cues are provided to subjects whatsoever. Bem has argued that the basic "deficiency" of the AGT as an idiographic method is that we have had people other than our subjects rate the favorability of the words that our subjects generated. But does having someone other than our subjects rate adjectives render our technique only quasi-idiographic? Not necessarily. It may be that others' favorability ratings can be reasonably substituted for the subjects' own ratings just as persons' inferences