An Examination of the Blau -- Scott and Etizioni Typologies: A Critique

The conclusions reached by R. H. Hall, J. E. Haas & N. J. Johnson (see SA 0624/C9026) in their comparative evaluation of the Blau -- Scott & A. Etzioni typologies (no reference given) fail to reflect their own empirical & statistical findings. In their article indicators of such concepts as complexity, formalization, goal specificity & org'al activities were correlated with the typologies, while the chi-square statistic was used to determine the nature of -- he relationships. No cutting points were introduced to indicate the level of signif, nor were the levels themselves published; only the chi-square value appeared in the tables. Analyzing the df presented in the tables, the reader can determine the level of signif of the X2 values. The results of this simple exercise raise serious doubts about the conclusions which Hall et al drew from their findings. Their approach shows little concern for conceptual definition & operationalization, esp in the issue of conceptual dimensionality. Because of these methodological shortcomings, they overstated their negative evaluation of the typologies. REPLY TO WELDON by Richard H. Hall, J. Eugene Haas & Norman J. Johnson (U of Minnesota, Minneapolis, U of Colorado, Boulder, & Carnegie -- Mellon U, Pittsburgh, Pa), states that the purpose of a typology is to allow us to combine a number of variables into a single construct & thus allow us to deal with extremely complex phenomena in a relatively simple fashion. The argument is that the 2 typologies do not do this, despite Weldon's often correct evaluation of levels of signif & their usage. The purpose was to point out the pitfalls of typologies based on a limited number of variables. E. Weiman.