Specification of Boundaries of Constructed Types through Use of the Pattern Variable

* Dr. Marvin Bressler, of New York University, has encouraged me in worrying about the problem attacked in the following pages since 1955. His critical comments and suggestions and those of Drs. Albert K. Cohen and Sheldon Stryker, of Indiana University, have been most helpful in bringing to this paper whatever clarity it may have. Lapses still remaining result from my stubbornness and not from any lack of critical acumen on their parts. shorter version of this paper was presented at the annual meetings of the Ohio Valley Sociological Society, East Lansing, Michigan, May 4 and 5, 1962. 1 The term has been closely associated with the name of Howard Becker. See, for example, his Typology in the Social Sciences, in Harry E. Barnes, Howard Becker, and Frances B. Becker, Contemporary Social Theory (New York: Appleton-Century, 1940), pp. 17-46, or his Interpretive Sociology and Constructive Typology, in Georges Gurvitch and Wilbert E. Moore (eds.), Twentieth Century Sociology (New York: Philosophical Library, 1946), pp. 70-95. Two other particularly useful articles are Paul F. Lazarsfeld, Some Remarks on the Typological Procedures in Social Research, Zeitschrift fiir Sozialforschung, 6:119-38 (1937) and William J. Goode, A Note on the Ideal Type, American Sociological Review, 12:394-401 (1947). useful definition of the term, and one which includes most of the current types of use is that of John McKinney: the constructed type is a purposive, planned selection, abstraction, combination, and accentuation of a set of criteria that have empirical referents, and that serves as a basis for comparison of empirical cases ... . See his article Typology and Social Research, in John T. Doby et al. (eds.), An Introduction to Social Research (Harrisburg, Pa.: Stackpole, 1954), p. 147. 2 See, e.g., the section on Concepts and Indices, in Paul F. Lazarsfeld and Morris Rosenberg (eds.), The Language of Social Research (Glencoe, Ill.: Free Press,