COMParative methods for the Advancement of Systematic cross-case analysis and Small-n Studies
International ressource site
NEWS: training, meetings, publications
Didactics: Boolean algebra and QCA logic
CCM Textbook
Software : QCA, fs/QCA and others
If you have any suggestion or comments concerning this page, please send an e-mail to Gisèle DE MEUR.
-
QCA softwares
DOWNLOAD QCA 2.0
(zip file)
GO
TO DOWNLOAD PAGE OF QCA 3.0
-
FS/QCA software (version
2.0)
GO
TO DOWNLOAD PAGE OF FS/QCA
-
TOSMANA software (updated
version, July 2006)
GO TO DOWNLOAD PAGE OF
TOSMANA
- QCA in R (updated version,
March 2007)
QCA softwares
DOWNLOAD QCA 2.0
(zip file)
GO
TO DOWNLOAD PAGE OF QCA 3.0
FS/QCA software (version
2.0)
GO
TO DOWNLOAD PAGE OF FS/QCA
TOSMANA software (updated
version, July 2006)
GO TO DOWNLOAD PAGE OF
TOSMANA
QCA software
Charles Ragin's website, hosted at the University of Arizona, allows to download the QCA software and manual.
Qualitative Comparative Analysis (QCA) is an approach and a technique that uses Boolean algebra to implement principles of comparison used by scholars engaged in the qualitative study of social phenomena. These phenomena be may be macro-, meso- or microsociological. Typically, qualitatively oriented scholars examine only a few cases at a time, but their analyses are both intensive -addressing many aspects of cases- and integrative -examining how the different parts of a case fit together, both contextually and historically.
By formalizing the logic of qualitative analysis, QCA makes it possible to bring the logic and empirical intensity of qualitative approaches to studies that embrace more than a handful of cases--research situations that normally call for the use of variable-oriented, quantitative methods. Boolean methods of logical comparison represent each case as a combination of causal and outcome conditions. These combinations can be compared with each other and then logically simplified through a bottom-up process of paired comparison.
Computer algorithms developed by electrical engineers in the 1950s provide techniques for simplifying this type of data. The data matrix is reformulated as a "truth table" and reduced in a way that parallels the minimization of switching circuits (see Charles Ragin, The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies, 1987; see also the textbook by **, Switching Circuits, Schaum, **, **). These minimization procedures mimic case-oriented comparative methods but accomplish the most cognitively demanding task -making multiple comparisons of configurations- through computer algorithms. The goal of the logical minimization is to represent -in a logically shorthand manner- the information in the truth table regarding the different combinations of conditions that produce a specific outcome.
The successive versions of the QCA (Qualitative Comparative Analysis) software have been developed by Kriss Drass and Charles C. Ragin.
QCA 2.0
The first operational and downloadable version
of QCA was QCA 2.0, released in 1986**. This software operates
under a DOS-environment.
DOWNLOAD QCA 2.0
QCA 3.0
A more advanced version, more user-friendly (but still
DOS-environment) was launched from 1992** onwards : QCA
3.0.
GO TO DOWNLOAD PAGE OF
QCA 3.0
FS/QCA software
The
fs/qca page (www.fsqca.com) provides more detailed information on
the software.
The development (currently under way) of the fs/QCA software corresponds to a major qualitative leap as compared to the previous versions :
- it includes both "fuzzy sets" procedures (with multivalue
variables ; see Ragin, Fuzzy-set Social Science, 2000) and QCA
procedures (crisp sets, i.e. dichotomous variables);
- it allows one to use a broader range of algorithms (in the QCA procedure);
- it operates much faster;
- it allows one to include a larger number of conditions (i.e. independent
variables);
- it is much more user-friendly (windows environment).
The most recent has complete routines for truth table analysis
of fuzzy
set data, which is now the preferred procedure. fsQCA 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 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.
The fsQCA download page
http://www.u.arizona.edu/~cragin/fsQCA/software.shtml
has links to four recent papers that should help users understand all
the new procedures, including some instructions for transforming interval-scale
variables into fuzzy sets.
The latest version 2.0 (July 2006) and a manual can be downloaded here.
Tosmana software TOSMANA Version 1.255, July 06
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.
The TOSMANA software, developped by Lasse Cronqvist, and its manual can be downloaded at the tosmana homepage.
QCA in R
Visit also R's
webpage.
Dear friends,
Starting with version 0.4-0, the QCA package in R has one of the most
solid implementations of the exact Quine-McCluskey minimization procedure.
It employs a novel method of determining the prime implicants, one that
is both exact and fast.
Usually, an exact procedure is resource consuming and most importantly
time consuming, but this new algorithm shows a significant increase
in speed.
Timings may vary between machines (the newer the machine, the quickest
the solution). For a machine with 1GB of RAM (physical memory is important)
and a Core2Duo processor at 2GHz, here are some timings for large numbers
of conditions:
- less than 10 conditions -- almost instantly
- 10 conditions -- 0.9 seconds
- 11 conditions -- 2.5 seconds
- 12 conditions -- 9 seconds
- 13 conditions -- 37 seconds
- 14 conditions -- 130 seconds
- 15 conditions -- 600 seconds
I believe this is fast enough for anyone, given that an exact
solution is presented.
I present the new algorithm in a COMPASSS working paper
(WP2007-46).
The users' manual can be downloaded HERE.
The source code and the binary version for Windows can be found on section
"Packages" on CRAN:
http://cran.r-project.org/
Comments and suggestions are welcome, as well as code development.
Adrian DUSA
adi@roda.ro
Romanian Social Data Archive
1, Schitu Magureanu Bd
050025 Bucharest sector 5
Romania
Tel./Fax: +40 21 3126618 \
+40 21 3120210 / int.101
A tribute to Kriss Drass (message from Charles Ragin, September 2001)
It is with great sadness that I report the death of Kriss Drass, the multi-talented programming genius behind both QCA and fs/QCA. Kriss was a great friend and a fine human being who touched all those around him with his kindness and generosity. He developed both QCA and fs/QCA out of his love of analytic challenges, in the midst of his demanding career as a professor of Criminal Justice at University of Nevada, Las Vegas. Work on fs/QCA will continue; the program was very close to completion before his sudden and unexpected death.
Last modified: 08-Mar-2007
Benoît RIHOUX, Centre de Politique Comparée
Gisèle
DE MEUR, Lab.
de recherche en MAThématiques et sciences humaines
Afdeling
Arbeids- en Organisatiesociologie
Peter BURSENS,
Onderzoeksgroep
Internationale Politiek