Exploring the robustness of set theoretic findings from a large n fsQCA: an illustration from the sociology of education
Exploring the robustness of set theoretic findings from a large n fsQCA: an illustration from the sociology of education
复制标题
DOI：
10.1080/13645579.2015.1033799
复制
发表时间：
2016-01-01
影响因子：
3.3
通讯作者：
Glaesser, Judith
中科院分区：
文献类型：
--
作者：
Cooper, Barry;Glaesser, Judith
Ragin的定性比较分析(QCA)通常用于研究人员具有良好案例知识的中小型样本。在没有深入案例知识的情况下，利用它来分析大型调查数据集，提出了新的挑战。我们提出了应对这些挑战的方法。我们首先报告了英国国家儿童发展研究数据集(最高学历作为社会阶层、性别和能力的一组理论函数)的结构分析得出的单一QCA结果。然后，我们通过使用Dusa和Thiem的R QCA软件包来讨论我们分析的稳健性，以探索(I)改变模糊集理论的能力校准，(Ii)模拟能力测量中的错误，以及(Iii)改变评估教育成就的因果配置的准充分性的阈值的结果。我们还考虑了使用Bootstrapping在模拟重采样下的分析行为。本文为其他希望使用大n数据的QCA的人提供了建议的方法。
Ragin's Qualitative Comparative Analysis (QCA) is often used with small to medium samples where the researcher has good case knowledge. Employing it to analyse large survey datasets, without in-depth case knowledge, raises new challenges. We present ways of addressing these challenges. We first report a single QCA result from a configurational analysis of the British National Child Development Study dataset (highest educational qualification as a set theoretic function of social class, sex and ability). We then address the robustness of our analysis by employing Dusa and Thiem's R QCA package to explore the consequences of (i) changing fuzzy set theoretic calibrations of ability, (ii) simulating errors in measuring ability and (iii) changing thresholds for assessing the quasi-sufficiency of causal configurations for educational achievement. We also consider how the analysis behaves under simulated re-sampling, using bootstrapping. The paper offers suggested methods to others wishing to use QCA with large n data.