期刊
SOCIOLOGICAL METHODS & RESEARCH
卷 51, 期 2, 页码 541-565出版社
SAGE PUBLICATIONS INC
DOI: 10.1177/0049124119882456
关键词
qualitative comparative analysis; Boolean algebra; material implication; method evaluation; causal inference; set theoretic methods; robust sufficiency
资金
- ICUB (Research Institute of the University of Bucharest)
The objective of qualitative comparative analysis is to find solutions that meet the conditions for a desired outcome, while being as simple as possible. Different search strategies lead to different types of solutions, and there is an ongoing debate on which type is closest to the true causal structure. This article introduces the logics behind each simplification system, explores the concept of robust sufficiency, and provides improved procedures to measure correctness.
The main objective of the qualitative comparative analysis is to find solutions that display sufficient configurations of causal conditions leading to the presence of an outcome. These solutions should be less complex than the original observed configurations, as parsimonious as possible, without sacrificing the sufficiency requirement. Sufficiency and parsimony are two requirements that act in opposition, and an optimal solution is one that accommodates both. There are different search strategies that lead to different types of solutions, with an ongoing debate about which solution type is closest to the true, underlying causal structure. This article presents the different logics behind each simplification system in order to explain how and why they lead to different results and introduces the concept of robust sufficiency to clear the debate. It analyses the correctness ratios for the different solution type and provides an improved set of procedures to measure correctness that captures the best features from each system. Out of the competition between the conservative and the parsimonious search strategies, the intermediate solution emerges as the best hybrid that is suitable for causal analysis, outperforming the parsimonious solution in recovering a known (even parsimonious) causal structure.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据