Cognitive Style Research: A Perspective for Integration

Cognitive style is a continuing area of interest in MIS research. The work is often criticized for its fragmentation and lack of validity. This paper proposes the uses of a single instrument, the MyersBriggs Type Indicator (MBTI). It reviews the overall issue of validity and identifies the four steps cognitive style research must accornplish to achieve coherence. It assesses existing research related to those steps, focusing on findings from studies that use the MBTI. ;t presents data on cognitive style differences among occupational specialties. ISSUES OF COGNITIVE STYLE IN style research in MIS. The work is fragMIS RESEARCH mented and uses a variety of overlapping constructs and measures. The empirical The link between cognitive style and the results are generally equivocal and inconimplementation and use of information sistent and, all in all, the research has not systems and models is a recurring theme in generated convincing evidence to support MIS/MS research. Studies of cognitive the hypotheses implicit in the two assumpstyle reflect two central assumptions: tions listed above (Taylor and Benbasat, 1980; Wade, 1981). 1. There are systematic di fferences among individuals in terms of perception, thinking, and judgment That said, the cognitive style theme is of that significantly influence their persistent interest and influence in MIS. choice of and response to informaThe work of Churchman ( 1964), Churchman tion. and Schainblatt ( 1965), and Mason and Mitroff (1973) constitutes an unfinished 2. The difference between managers' program for research on the dynamic interand analysts' cognitive styles is a action between information and personalmajor explanation of difficulties in ity. Taylor and Benbasat's critique of implementation. previous studies ( 1980) points to the high potential payoff from "sound research into Table I traces the evolution of cognitive the psychological characteristics of informotion system users," though it justifiably highlights "inadequately formulated *We, the authors, wish to thank Roberta theory, '1 11 use of a great many inadequately Fallon for her time, patience, and contrivalidated measuring instruments," and butions to this paper. "faulty research designs."