Search

Symbolic interaction theory as derived mainly from the work of George Herbert Mead views the self-concept as a structure of attitudes which arises in social experience and organizes the individual's perception of his social milieu. Recently, tests based upon this symbolic interaction orientation have been developed for the elicitation of self attitudes. One such test frequently used is the Twenty Statements Test (TST) constructed by Manford Kuhn and Thomas McPartland. The TST attempts to ascertain the specific statements respondents use to identify themselves to themselves. The applicability of the TST to problems of self theory is indicated by the rather wide use which has been made of the instrument. Investigations using the TST have focused upon the following: self-concept and social position; a self-concept and behavior; and the self-concept and attitudes toward other social objects. While TST results have been of substantive interest, little attention has been given to the methodological issues involved in the instrument and its use. For example, little is known concerning the validity and reliability of the test and its scoring procedures. If substantive assertions based upon TST results are to be considered as sound, judicious concern should be directed toward basic methodological issues. It is the purpose of this paper to consider empirically one methodological aspect of the TST. The focus is upon the test format itself. The question is posed: Do variations in the TST format produce variations in individuals' response patterns? Specifically, does the length of the test influence the number and the nature of statements made by respondents? This paper is directed toward answering this question. [ABSTRACT FROM AUTHOR]