Further Considerations of the Impucative Meaning Procedure: a Replication and Extension

This study was a replication and extension of an earlier one (Kilty, 1967), which compared the unidimensional (affect only) and the multidimensional (affect, cognition, and behavioral intentions) models of attitude structure, by using the implicative meaning (IM) procedure. According to the multidimensional model, this instrument is a measure of attitudinal cognition. The unidimensional model maintains the IM procedure is an indirect measure of affect. Thus, the unidimensional model predicts high correlations between IM scores and scores from an independent measure of affect, while the multidimensional model predicts relatively low correlations. The results, though, showed that by varying the type of concept, the type of belief, the number of beliefs, and the type of belief statement, significantly different levels of correlation could be obtained. The type of concept (e.g., abstract or concrete) appeared to be of most importance. In addition, widely different levels of correlation were found on the basis of method of correlation. The level was considerably higher when correlations were derived by subjects over concepts than by concepts over subjects. Further Considerations of the Implicative Meaning Procedure: A Replication and Extension 2 Keith M. Kilty University of Illinois For the past several decades, much of the research concerning attitude structure has stemmed from one of two theoretical orientations. From the first, attitude has been considered as a unidimensional construct, consisting solely of affect (e.g., Edwards, 1967; Fishbein, 1963, 1965; Osgood, Suci, 5 Tannenbaum, 1957; Thurstone, 1931). From the second, the multidimensional approach, attitude has been viewed as a three-component construct, consisting of affect, cognition, and behavioral intentions (e.g., Allport, 1935; Davis 6 Triandis, 1965; Katz G Stotland, 1959; Krcch, Crutchfield, 5 Ballachey, 1962; Smith, Bruner, 5 White, 1956; Triandis, 1964, 1967). The present study was a replication and extension of an earlier one (Kilty, 1967) which compared these two models by means of an attitude instrument, the implicative meaning (IM) procedure. This instrument was originally developed to measure the cognitive component of attitude (Davis f, Triandis, 1967; Kilty, 1967; Triandis, 1967). Subjects were presented »•fences in the form of "if, then" clauses, with the artitude objects in the "if" parts and related objects in the "then" parts (e.g.. If one has FEAR, This research was supported by the Advance^ Research Projects Agency, ARPA Order No. 454, under Office of Naval Research Contract NR 177-472, Nonr 1834(36) (F. E. Fiedler and H. C. Triandis, Principal Investigators). Gratitude is expressed to M. R. "ehling for assistance with the computers and to E. E. Taylor for assistance with data collection and analysis. H. C. Triandis gave valuable advice and guidance throuphout the study, and his comments improved an earlier draft of this paper. 2 Now at Yale University. 2 then one has flight.). Probability ratings of the statements and evaluations of the related objects were obtained, and IM scores were derived from these ratings.