Annual Review of Sociology
I Abstract The last decade featured the emergence of a significant and growing literature concerning comparative-historical methods. This literature offers methodological tools for causal and descriptive inference that go beyond the techniques currently available in mainstream statistical analysis. In terms of causal inference, new procedures exist for testing hypotheses about necessary and sufficient causes, and these procedures address the skepticism that mainstream methodologists may hold
... logists may hold about necessary and sufficient causation. Likewise, new techniques are available for analyzing hypotheses that refer to complex temporal processes, including path-dependent sequences. In the area of descriptive inference, the comparative-historical literature offers important tools for concept analysis and for achieving measurement validity. Given these contributions, comparative-historical methods merit a central place within the general field of social science methodology. 2 Statistical methods dominate the methodology section of the American Sociological Association, the required courses on methodology in leading graduate programs, and the leading methodological journals in the social sciences. Downloaded from arjournals.annualreviews.org by NORTHWESTERN UNIVERSITY -Evanston Campus on 11/03/09. For personal use only. 4 The formulation of this sentence assumes that variables are measured dichotomously. However, one need not make this assumption. For example, with continuous measurement, one can hypothesize that a particular value (or range of values) on an independent variable is necessary or sufficient for a particular value (or range of values) on a dependent variable. Annu. Rev. Sociol. 2004.30:81-101. Downloaded from arjournals.annualreviews.org by NORTHWESTERN UNIVERSITY -Evanston Campus on 11/03/09. For personal use only.