BEYOND MIXED METHODS USING QUALITATIVE COMPARATIVE ANALYSIS QCA

BEYOND MIXED METHODS: USING QUALITATIVE COMPARATIVE ANALYSIS (QCA) TO INTEGRATE CROSS-CASE AND WITHIN-CASE ANALYSES © BARRY COOPER, JUDITH GLAESSER, LOUIS COHEN, LAWRENCE MANION AND KEITH MORRISON © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

STRUCTURE OF THE CHAPTER • • • What is Qualitative Comparative Analysis Starting from a ‘quantitative’ stance Starting from a ‘qualitative’ stance Qualitative Comparative Analysis (QCA) QCA: sufficiency © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

BEYOND MIXED METHODS TO INTEGRATED METHODS • Qualitative Comparative Analysis (QCA) from Ragin. • QCA is a Boolean analytic approach drawing on logic and set theory. • It is an appropriate approach for those willing to mix methods. • It can bridge the ‘methodological gulf between intensive case-oriented research and extensive variable-oriented research’ (Ragin, 1994, p. 304). • This bridging requires ‘tools that preserve the intensity of the case-oriented approach, especially its attention to combinations and configurations of causes and conditions, when examining many cases’ © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors (ibid. , p. 304).

BEYOND MIXED METHODS TO INTEGRATED METHODS • Instead of focusing on assessing the relative importance of independent variables in predicting some outcome, QCA employs a model of conjunctural causation and draws on the concepts of necessary and sufficient conditions. • QCA strives to provide causal explanation in addition to descriptive knowledge, to explain ‘why’. • An understanding of what are more basic and what are more superficial causes requires more than just the analysis of regularities that is usually delivered by quantitative research methods • QCA combines cross-case and within-case analysis. © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

BEYOND MIXED METHODS TO INTEGRATED METHODS • We might hypothesize, for example, that, for individuals from higher-class origins, high ability will tend to be sufficient but not necessary for later high educational achievement while, for lower-class respondents, high ability will tend to be necessary but not sufficient. Such an approach can be developed in more complex ways. It might be, for example, that the conjunction ‘being male, from a higher class origin, and of a certain level of ability, but not from ethnic background X’ is sufficient for some outcome. QCA facilitates such cross-case analyses. © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

STARTING FROM A ‘QUANTITATIVE’ STANCE Qualitative analysis can add (causal) explanation to quantitative data on regularities, correlation, regression and constant conjunction © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

STARTING FROM A ‘QUALITATIVE’ STANCE Qualitative work can be used in an exploratory, hypothesis-generating way (e. g. using case studies) and any initial findings can then be confirmed and elaborated by conducting a larger scale study. © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

QUALITATIVE COMPARATIVE ANALYSIS (QCA) • QCA can be used with ‘crisp sets’ representing binary conditions such as male/female, with multivalued conditions such as high/moderate/low social class, or by employing ‘fuzzy sets’ with fully continuous conditions, e. g. measured ability. • QCA employs set theory as the basis for the Boolean analysis of conjunctural causation. • In QCA, cases are seen as members of various sets defined by conditions such as ‘being male’, ‘having high ability’, ‘achieving a degree’. © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

QUALITATIVE COMPARATIVE ANALYSIS (QCA) • First, the condition can be more complex, e. g. a conjunction of factors such as MALE*HIGH ABILITY*HIGH CLASS where * denotes set intersection (logical AND). Here a case must be a member of all three sets to belong to the configuration. • Second, because the empirical social world is not always tidy, the researcher might relax the criterion for sufficiency, e. g. where, say, 80 per cent of those with a simple or complex condition achieve the outcome. • Third, QCA uses Boolean minimization to simplify its analyses. • Fourth, it may be the case, especially when the dataset is small, that there may be no cases for some conjunctions. © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

QCA SUFFICIENCY: A WORKED EXAMPLE • A research student is investigating what is sufficient for achieving admission to university, believing that class and ability are two of the key factors. She begins by undertaking a simple Boolean cross-case analysis of the relation between entering university (U) and high class origins (HC) and high ability (HA), constructing a ‘truth table’ (a complex crosstabulation) of numbers of students in each set. • She plans to use in-depth interviews with a small number of cases in order to explore, via process-tracing, what it is about class and ability that explain the patterns in the ‘truth table’. © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

QCA SUFFICIENCY: A WORKED EXAMPLE • As a result of the interviewing of selected cases, alongside establishing what explains the typical cases (who achieves the outcome), she begins to suspect that two other factors – sex and ethnicity – interact with class and ability in determining whether the outcome is achieved. • She returns to the data set to undertake another round of cross-class analysis, but this time incorporating the additional factors of sex and ethnicity. • She conducts case studies to explore ‘what is it about being a female from a minority ethnic background that explains this low percentage, given the positive class and ability factors? ’. © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

QCA AND INTEGRATED METHODS QCA, using set theory and Boolean processes, is able to integrate quantitative and qualitative methods iteratively to provide explanations and causal explanations, combining cross-case and within-case analysis. © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors

SOME REFERENCES FOR RAGIN’S WORK ON QCA • Ragin, C. C. (1987) The Comparative Method: Moving beyond Qualitative and Quantitative Strategies. Berkeley, Los Angeles and London: University of California Press. • Ragin, C. C. (2000) Fuzzy-Set Social Science. Chicago, IL: University of Chicago Press. • Ragin, C. C. (2006) Set relations in social research: evaluating their consistency and coverage. Political Analysis, 14 (3), pp. 291– 310. • Ragin, C. C. (2008) Redesigning Social Inquiry: Fuzzy Sets and Beyond. Chicago, IL: University of Chicago Press. • Ragin, C. C. and Davey, S. (2014) fs/QCA [Computer Program], Version 2. 5. Irvine, CA: University of California. Available from: www. socsci. uci. edu/~cragin/fs. QCA/software. shtml [Accessed August 2016]. • Ragin, C. C. and Schneider, G. A. (2011) Case-oriented theory building and theory testing. In M. Williams and W. P. Vogt (eds) The SAGE Handbook of Innovation in Social Research Methods. London: Sage, pp. 150– 166. © 2018 Louis Cohen, Lawrence Manion and Keith Morrison; individual chapters, the contributors