Abstracts

Abstract

To demonstrate the utility of content analysis for generating cultural indicators spanning long periods of time, we re-analyzed Sorokin's (1937) data on theories of truth. Based on content-analytic techniques consistent with contemporary methods, Sorokin concluded that there were no trends or cycles in concern with theories of truth, but only trendless fluctuations. Contrary to Sorokin's conclusions, our re-analysis indicates that since the Renaissance, changes in concern with theories of truth reflect: (1) increasing secularization of Western civilization, and (2) cultural problem-solving cycles. The cultural cycles are the same length as political problem-solving cycles previously observed in America and Great Britain. These results suggest that Sorokin's data have both construct and face validity.This paper provides a method for calculating the minimum and maximum possible variances from summary data for which n-tile group means are reported, but for which standard deviations are not given. The method makes no assumptions about the nature of the distribution of the data. Decision rules are derived for using the minimum and maximum possible variances, Examples and a flow-chart for calculating the minimum and maximum variances are given in Appendices.