Abstract
For the past two years, our journal has featured a symposium that has provided a forum for papers on important methodological issues and a place for scholars to exchange views, ideas, comments, and suggestions. I am pleased to present again in this year’s volume a symposium, this time on qualitative comparative analysis (QCA).
QCA has attracted an increasing number of users since Charles C. Ragin developed the method and published his 1987 book The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies. The original method is based on truth tables built on crisp sets. Crisp-set QCA has been extended to multivalue QCA and fuzzy-set QCA by Ragin and others to allow great flexibility. QCA has more users in Europe and a greater following in political science, but it has also seen applications in business and organizational research, education, environmental studies, health and medical research, and civil and electrical engineering, in addition to sociology.
However, as with many issues in the social sciences, QCA is not without controversy. Critics of QCA, just like its adherents, want their voices heard. The symposium in this volume features a main article by Samuel R. Lucas and Alisa Szatrowski that provides a critical assessment of QCA. The article is followed by a set of commentaries by Charles C. Ragin; Peer C. Fiss, Axel Marx, and Benoît Rihoux; Wendy Olsen; Stephen Vaisey; Jake Bowers; Jason Seawright; and David Collier, as well as a rejoinder by Lucas. In spite of the writing styles and particular wording some authors chose that may not be journal preference, the purpose of this symposium is to provide a forum for scholarly exchanges by authors holding different views. By reading the symposium, we hope that readers will be able to evaluate the merits and limitations of the QCA method and to improve our understanding of it.
Following the symposium are two sections, each containing three chapters. The three articles in the first section all deal with some aspects of modeling strategy. Adam N. Glynn and Jon Wakefield, in “Alleviating Ecological Bias in Poisson Models Using Optimal Subsampling: The Effects of Jim Crow on Black Illiteracy in the Robinson Data,” propose a solution to the famous problem of ecological fallacy that arises when making individual-level inference with group-level data. Their solution is to supplement group-level data with individual-level data by using a Poisson modeling framework. In “An Extended Cultural Consensus Theory Model to Account for Cognitive Processes in Decision Making in Social Surveys,” Zita Oravecz, Katherine Faust, and William H. Batchelder develop the hierarchical extended Condorcet model, which is an extension to the general Condorcet model (a type of cultural consensus theory model), for analyzing social surveys about knowledge, especially for dealing with “don’t know” responses. Both multilevel and spatial models are applied in neighborhood research. In the third paper of the section, “Comparing Spatial and Multilevel Regression Models for Binary Outcomes in Neighborhood Studies,” Hongwei Xu compares these two types of modeling strategies to assess their strengths and limitations for analyzing binary outcomes in neighborhood data.
The final section of the volume also contains three chapters all related to scales or scaled analysis in one way or another. Zack W. Almquist and Carter T. Butts, in “Logistic Network Regression for Scalable Analysis of Networks with Joint Edge/Vertex Dynamics,” view network dynamics as a process of change in the edge structure of a network or in its vertex sets, and they extend a logistic network regression-based approach to providing a highly scalable framework for analyzing large networks with dynamic vertex sets. In recent years, sociologists have shown increasing interest in measuring attributes of neighborhood environments. In “Creating Measures of Theoretically Relevant Neighborhood Attributes at Multiple Spatial Scales,” Michael D. M. Bader and Jennifer A. Ailshire show how to combine ecometric measures with kriging to develop city block–level estimates and how these measures can be aggregated to spatial scales. In attitudinal surveys, extreme response style and acquiescence style often present problems for researchers. In the final paper, “The Effect of Labeling and Numbering of Response Scales on the Likelihood of Response Bias,” Guy Moors, Natalia D. Kieruj, and Jeroen K. Vermunt apply a latent class factor model for diagnosing and correcting these problems and analyze the effect of labeling and numbering of response scales such as bipolar scales and agreement style scales.
We at Sociological Methodology are grateful to all the authors who submitted articles to our journal, whether or not their articles grace these pages, and to the reviewers and our board members whose devoted work guaranteed the high quality of the publications. In addition, I wish to thank my managing editor, Lisa Savage, for her highly consistent and reliable work; our editorial assistant, Andrea Wilbon Hartman, for her enthusiasm in performing the various kinds of necessary editorial assistance; copy editor Stephanie Magean and the copy editors of Sage Publications for keeping up the quality of writing; and Athena Liao, for her art design on the cover. My appreciation also goes to Jim Ballinger, Debbie Sourgen, and Sara Sarver, who coordinated the publication process at Sage, and to Janine Chiappa McKenna, Karen Edwards, and the Publications Committee of the American Sociological Association for their continuing support. Finally, we would like to acknowledge the material support from the Department of Sociology and the College of Liberal Arts and Sciences at the University of Illinois at Urbana-Champaign that houses our editorial office in the tastefully renovated and environmentally sound historic building of Lincoln Hall.