Abstract
Some common applications of statistics in sociological inquiry are critically analyzed under two "sacred cows": (1) the treatment of assumptions as inviolate, and (2) the use of statistics as ends in themselves. The first includes the usage of assumptions, measurement scales, and dichotomies; the second includes significance tests, complex techniques, fads, continuous data analysis, and intercorrelating all variables. It is argued that sociologists should be more willing to quantify and experiment with higher-order measurement. Finally, several tentative guidelines are given for more helpful statistical usage.
Get full access to this article
View all access options for this article.
