QCA Software Review
This is a review of the most useful software to process QCA. If you have any suggestion or comment concerning this page, please send a message to Hisako Nomura
FSQCA
FSQCA is capable of computing logical truth tables for both fuzzy set and dichotomous (crisp) set data. The programme includes both "fuzzy sets" procedures (with multi-value variables; see Ragin, Fuzzy-set Social Science, 2000) and QCA procedures (crisp sets, i.e. dichotomous variables). It also includes calculations of consistency and coverage measures for both crisp and fuzzy set analyses, as described in a paper by Ragin recently published in the journal Political Analysis. Additionally, coverage can be partitioned to show the relative empirical weight of the combinations of conditions shown in a truth table solution. Finally, a new procedure has been implemented which allows the derivation of three solutions for each analysis: the complex solution, the most parsimonious solution, and the "intermediate" (i.e., theory - and knowledge-informed) solution, as described in a paper by Ragin and John Sonnett on counterfactual analysis (QCA Working Paper WP2004-23).
TOSMANA
Tosmana is a Tool for Small-n Analysis, used to perform social science research on data sets with a small number of cases.
Tosmana also implements classical Boolean algebra, but it seeks to tackle one of the main limitations of QCA - its restriction to Boolean sets: Every element of a data set has to be T (true) or F (false), but often we want to use more variable values like {low, medium, high}. Therefore Tosmana introduces Multi-Value Minimization as an additional feature of Boolean Minimization, but it also can be used for Boolean Minimization.
STATA
Fuzzy command is now available in STATA. It is capable of creating, testing, and performing logical reductions on both fuzzy and dichotomous (crisp) set-theoretic data. It has a capacity to test sets of configurations for logical necessity and sufficiency probabilistically. The user has to specify which variate is the outcome. Fuzzy also has the ability to perform other useful statistical tests. Certain features are not yet available such as a way to view and code the truth table spreadsheet directly. Fuzzy does not derive an intermediate solution, either, as needed for counterfactual analysis. Yet, fuzzy in STATA gives an interface between csQCA/fsQCA and statistical analysis as well as potentially a faster calibration process. It will enable sensitivity analyses via programming. This is a strong signal of the growing acceptance of non-universalist, non-frequentist methods in social science. For new users to the fuzzy command in STATA, simply type: find it fuzzy in the command box.
THE QCA PACKAGE IN R
The QCA package in the R programming environment is the only open-source
program that performs Qualitative Comparative Analysis in the social sciences.
Starting with version 0.4-0 it offers one of the most solid implementations of
the exact Quine-McCluskey minimization procedure, one that is both exact and
extremely fast. From version 0.5-0, the package gained an enhanced algorithm
("eqmcc") that is not only faster than the previous one, but it also consumes
significantly less memory. This algorithm is also capable of performing mvQCA
with exact solutions, start ing with version 0.6-0. Future developments aim to
implement fuzzy sets operations.
With the current version, users can perform QCA with up to 17 causal
conditions at once in a reasonable time. Due to its rapid and exact nature,
the algorithm can also be adapted for temporal QCA, where the sequence of the
causal conditions does matter in the appearance of a phenomenon.
The algorithm behind the code is presented in the COMPASSS Working Paper
"Enhancing Quine-McCluskey" (WP2007-46), now extended for multi-valued causal
conditions.
The software can be used either in command line mode or using the
companion package QCAGUI that provides a user interface.