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
COMPASSS Working Papers
(Full info in chronological order)
To view a short list in chronological order, click HERE.
To view the papers by themes, click HERE.
WP2008-53[Posted
on 06-November-2008]
Goertz ROHWER
Qualitative
Comparative Analysis: A Discussion of Interpretations
(PDF Format)
Abstract:
This paper (which is part of an ongoing project that investigates possibilities
to apply statistical (or more general: formal) notions and methods to
narrative data) tries to understand an approach proposed by Charles C.
Ragin: Qualitative Comparative Analysis (QCA).
WP2008-53[Posted
on 21-Aug-2008]
Seungyoon Sophia
LEE (University of Oxford)
A
Critique of the Fuzzy-Set Methods in Comparative Social
Policy. A Critical Introduction and Review of the Applications of Fuzzy-Set
Methods
(PDF Format)
Abstract:
This article critiques the Fuzzy-Set Qualitative Analysis (fs/QCA) methodology
by examining its applicability in three studies in the field of comparative
social policy. In each of these three test cases, I focus on the validity
of Fuzzy-Set’s claimed function – its ability to combine theoretic
discourse and evidence analysis. All three studies investigate welfare
state reform in the late twentieth century and apply fs/QCA: (1) “Welfare
Reform in the Nordic Countries in the 1990s: Using Fuzzy-Set Theory to
Assess Conformity to the Ideal Types,” (2) States of Welfare or
States of Workfare? Welfare State Restructuring in the 16 Capitalist Democracies,
1985-2002,” and (3) “The Diversity and Causality of Welfare
State Reforms Explored with Fuzzy-Sets.” This article begins by
discussing the ontology and epistemology of comparative social policy.
The Fuzzy-Set logic and set theoretic nature of social science theory
is then discussed to align ontology with Fuzzy-Set methodology. Next,
a more detailed introduction of Fuzzy-Set methods (fs/QCA) is followed.
This study suggests that fs/QCA is a unique and useful method for comparative
social policy. It advances quantitative comparative analysis by in interpreting
attributes as a configuration. By applying Fuzzy-Set logic and the principle
of calibration, it advances qualitative analysis by permitting theoretically-informed
concepts to the quantified.
WP2007-52[Posted
on 27-Nov-2007]
Maarten
VINK (University of Maastricht / University
of Lisbon) and Olaf VAN
VLIET (Leiden University)
Not quite
crisp, not yet fuzzy?... Assessing the potentials and pitfalls of multi-value
QCA.
(PDF format)
NB:
Updated version forthcoming in Field
Methods
Abstract:
This paper assesses the strengths and shortcomings of multi-value
Qualitative Comparative Analysis (mvQCA) a comparative technique for small
to medium size datasets that has been integrated in the TOSMANA software
developed by Lasse Cronqvist. Its main difference with Charles Ragin's
'crisp-set' QCA (csQCA) which only allows for conditions with 0 or B values,
is that the dataset can also contain causal conditions with three or more
categories. MvQCA thus avoids relatively crude dichotomization and arguably
better captures the richness of
information of the raw data. Unlike 'fuzzy-set' QCA (fsQCA), developed
by Ragin to go beyond the classic dichotomous approach, mvQCA is still
based on dichotomous outcomes and
applies Boolean minimization principles in a similar way to csQCA. Its
major advantage, according to its proponents, is that it deals better
with the classic QCA problem of contradictory
configurations where cases with the same explanatory characteristics display
different outcomes and in principle cannot be taken into account for logical
minimization. We discuss the logical
status of mvQCA, its impact on limited diversity, and present a re-analysis
of a recent paper to show how mvQCA uses threshold-setting to solve contradictions.
WP2007-51[Posted
on 4-Nov-2007]
Claudius
WAGEMANN (Istituto italiano di scienze umane) and Carsten
Q. SCHNEIDER (Central European University)
STANDARDS
OF GOOD PRACTICE IN QUALITATIVE COMPARATIVE ANALYSIS (QCA) AND FUZZY-SETS
(PDF format)
Abstract:
Over the last couple of years, we witness an increasing curiosity
for a methodological family, generally identified with its acronym, 'QCA'.
This stands for 'Qualitative Comparative Analysis', which was introduced
for the first time to a wider public by the American social scientist
Charles Ragin in 1987 (1987). Since then, QCA has been modified, extended
and improved several times (Ragin 2000; Ragin 2003b; Ragin 2006a Ragin
2006b; and Ragin and Sonnett 2004). These developments have contributed
to a better applicability of QCA to empirical social scientific research
questions and to its prominence within the discipline.
In this article, we will, first, present the 'state of the art' of QCA
and will introduce both its basic principles and the different variants
of this group of 'Configurational Comparative Methods' (a term coined
by Rihoux and Ragin 2007a, which might probably substitute ‘Qualitative
Comparative Analysis’ in the long run). After this, we will propose
a list of criteria for a 'good' QCA analysis. We hope that our contribution
can be a guideline for QCA users as to which aspects have to be considered
when carrying out QCA analyses in order to render them not only technically
correct, but also to make the best out of the analytically relevant information
one can generate with QCA. Furthermore, the standard of good practice
which we propose can also be a helpful instrument for readers and commentators
when they have to evaluate empirical analyses based on QCA techniques.
WP2007-50[Posted
on 7-Oct-2007]
Adrian
DUSA (University of Bucharest)
QCA
Graphical User Interface manual
(PDF format)
NB: This is an updated version of WP2006-41
Published in Journal
of Business Research, Vol.60(5), 2007, Pp.576-586
Abstract:
This manual is intended for scholars wishing to use QCA in an
R environment. It includes visualisation and factorisation functions,
along with all other basic QCA functions.
QCAGUI is a graphical user interface (GUI) for the QCA package, derived
from R Commander. Because QCA has little to do with statistics, the menus
from Rcmdr were stripped down to the very basics. In crisp sets QCA, data
is binary therefore it is fairly decent to treat it as categorical (1
- presence; 0 - absence). In order to ease the primary analysis (e.g.
tables of frequencies) and the creation of basic graphs, this package
activates some menus that are not available in Rcmdr but for factors.
Users should be aware, however, that QCAGUI is _not_ a package for statistics;
Rcmdr is better for this purpose.
Updates of the QCA packages can be followed on the R webpage:
http://cran.r-project.org/
WP2007-49[Posted
on 5-Oct-2007]
Adrian
DUSA (University
of Bucharest)
Enhancing
Quine-McCluskey
(PDF format)
Abstract:
C urrently, the only algorithm that yields an exact solution to the boolean
minimization problem is the well known Quine-McCluskey, but almost all
software solutions employ different implementations because of its two
fundamental weaknesses: it is memory hungry and slow for a large number
of causal conditions.
This paper proposes an alternative to the classical Quine-McCluskey algorithm,
one that addresses both problems, and especially the one of memory consumption.
The solutions of this new algorithm are also exact, but they are produced
not by following the cumbersome classical algorithm but using a more direct
and faster approach.
Memory restrictions limit the number of input variables (causal conditions)
at a ceiling of about 14 or 15 (because each new variable expands the
memory usage in a geometric proportion), where this alternative uses only
a very small fraction of memory and it can process about 20 input variables
with acceptable speed.
WP2007 -48[Posted
on 3-Oct-2007]
Olav
S. STOKKE (The Fridtjof Nansen Institute, Norway)
Qualitative
Comparative Analysis, Shaming, and International Regime Effectiveness
(PDF format; Link to the Fridtjof
Nansen Institute)
NB:
published in Journal of Business Research. Updated version of WP2003-5,
and of Stokke (2004).
Abstract:
The article presents and applies a set-theoretic comparative technique,
qualitative comparative analysis (QCA), to a string of case studies on
shaming as a strategy for improving the effectiveness of international
regimes for resource management. This technique is particularly attractive
when the number of cases available is greater than the researcher can
reliably handle by narrative comparison, yet too low to support statistical
procedures. QCA can capture causal conjunctions, even in small-to-intermediate-N
situations, primarily because it permits the introduction of simplifying
assumptions in a way that maintains a clear connection to the underlying
cases – thus allowing substantive evaluation of their plausibility.
A more recent fuzzy-set version lifts two limitations of the crisp-set
version of QCA examined here (i.e., that variables must be dichotomous,
and that the analysis makes no allowance for measurement error and non-modeled
causality).
WP2007-47[Posted
on 2-Aug-2007]
Maria
GJOLBERG (University of Oslo)
The Origin
of Corporate Social Responisbility: Global Forces or National Legacies?
(PDF format)
Abstract:
This article explores the relative importance of global forces and national
political-economic institutions for companies’ inclination and ability
to engage in initiatives promoting Corporate Social Responsibility (CSR).
The globalist hypothesis postulates the CSR efforts of a given company
as a function of necessities dictated by the global market place: strong
anti-globalisation and anti-corporate sentiments create a need for a positive
reputation in order to obtain a “social licence to operate”.
The institutionalist hypothesis postulates the CSR efforts of a given
company as a function of institutional factors in the national, political-economic
system: companies based in certain political economic systems have comparative
institutional advantages for success in CSR. The hypotheses are examined
quantitatively by testing an index of national CSR-performance against
a wide variety of political-economic indicators. The final analysis, based
on Qualitative Comparative Analysis (QCA), reveals causal heterogeneity
and indicates two separate roads leading to CSR success.
WP2007-46[Posted
on 8-March-2007]
Adrian
DUSA (University of Bucharest)
A mathematical
approach to the boolean minimization problem
(PDF format)
NB:
published in Quality and Quantity (2008).
Abstract:
Any minimization problem involves a computer algorithm. Many such algorithms
have been developed for the boolean minimizations, in diverse areas from
computer science to social sciences (with the famous QCA algorithm).
For a small number of entries (conditions in the QCA) any such algorithm
will find a minimal solution, especially with the aid of the modern computers.
However, for a large number of conditions a quick and complete solution
is not easy to find using an algorithmic approach, due to the extremely
large space of possible combinations to search in.
In this article I will demostrate a simple alternative solution, a mathematical
method to obtain all possible minimized prime implicants.
This method is not only easier to understand than other complex algorithms,
but it could prove to be a faster method to obtain an exact and complete
boolean solution.
WP2007-45[Posted
on 25-January-2007]
Charles
C. RAGIN and Sarah STRAND
(University of Arizona)
Using QCA to Study
Causal Order: Comment on Caren and Panofsky (2005)
(PDF format)
Abstract:
The goal of qualitative comparative
analysis (QCA) is to identify the different combinations of causally relevant
conditions linked to an outcome. The researcher typically focuses on a
qualitative outcome and seeks to identify the different conjunctural conditions
that generate it. In this way QCA allows for causal complexity--for the
possibility that no single cause may be either necessary or INUS sufficient.
Instead causes are viewed as conditions: insufficient but necessary components
of unnecessary but sufficient combinations of conditions (Mackie 1965).
Caren and Panofsky (2005) seek to advance QCA by demonstrating that it
can be used to study causal conditions that occur in sequences and introduce
a technique they call TQCA (temporal qualitative comparative analysis).
In their sequence formulation the causal conjuncture is a of conditions
or events.
While we applaud their effort, in this comment we seek to clarify aspects
of their analysis and to present a generalization of the approach that
is more amenable to truth table analysis and use of existing software,
fsQCA (Ragin 1987; 2000; Ragin, Drass, and Davies 2006). Our first task
is to correct what appear to be errors of omission in their analysis.
Specifically, they seem to have stopped the process of logical minimization
short of completion. We show that it is possible to produce a logically
simpler solution than the one they present, while still remaining true
to the principles they advocate. Our second task is to demonstrate how
to use fsQCA to implement a generalization of their procedure. This procedure
takes advantage of an under-utilized feature of fsQCA software, namely,
the facility in crisp-set analyses to code a causal condition not only
as "present" versus "absent," but also as "irrelevant."
The coding of "irrelevant" is especially important in analyses
of event sequences, where event order is relevant only if the events actually
occur. Thus, the question, "Which came first, event A or event B?”
is relevant only if both A and B are coded "present."
WP2007-44[Posted
on 25-January-2007]
Charles
C. RAGIN (University of Arizona)
Fuzzy Sets: Calibration
Versus Measurement
(PDF format)
Abstract:
This essay explores the connections between measurement and calibration
in the social sciences and addresses its long-standing neglect. My starting
point is the contrast between conventional approaches to measurement in
quantitative and qualitative social research. After sketching common measurement
practices in both types of research, I argue that a useful way for social
scientists to incorporate measurement calibration into their research
is through the use of fuzzy sets. In order to use fuzzy sets effectively,
researchers must assess the degree of membership of cases in well defined
sets (e.g., degree of membership in the set of "developed countries").
This requirement forces researchers to attend to the issue of calibration
and provides additional motivation for them to explore the conceptual
underpinnings of their measures. Fuzzy sets resonate with the measurement
concerns of qualitative researchers, where the goal often is to distinguish
between relevant and irrelevant variation--that is, to interpret it--and
with the measurement concerns of quantitative researchers, where the goal
is the precise placement of cases relative to each other.
The second half of this essay sketches a technique for calibrating conventional
interval- and ratio-scale variables according to external standards. In
the examples provided, the external standard used is a qualitative assessment
of the degree to which cases with given scores on a conventional interval-scale
measure are members of a target set. A simple estimation technique rescales
the interval-scale measure so that it conforms to these qualitative assessments.
The end product of this procedure is the calibration of the degree of
membership of cases in sets, which in turn is suitable for fuzzy-set and
other types of analysis. The examples illustrate the responsiveness of
this calibration technique to the researcher's qualitative assessments
of cases. While calibration in the social sciences is unlikely ever to
match the sophistication of calibration in the physical sciences, the
technique of qualitative calibration presented here is an important first
step.
WP2006-43[Posted
on 30-June-2006]
Axel MARX (Hogheschool Antwerp)
TOWARDS MORE ROBUST MODEL
SPECIFICATION IN QCA
RESULTS FROM A METHODOLOGICAL EXPERIMENT
(PDF format)
Abstract:
Qualitative Comparative Analysis (QCA) is a research technique which was
developed by Charles C. Ragin and has been applied in several studies
that appeared in major sociological journals. Recently, QCA has been criticized
concerning the validity of the models it generates. Lieberson has hypothesized
that QCA is unable to distinguish real from random data. In other words,
it is argued that QCA always finds a model even on the basis of random
data. The paper addresses this issue through a methodological experiment.
It uses randomly created data-matrices to show that QCA can make a distinction
between real and random data. However, it only does so under certain conditions
namely when the proportion of variables on cases goes below a certain
threshold, which differs as a function of the combination of variables
on cases. Secondly, it argues that there is an upper-limit to the number
of variables which can be used in a QCA-analysis. Both limiting conditions
are the result of the problem of uniqueness which is a consequence of
the use of Boolean algebra and have not yet been addressed in the literature.
Five implications for comparative case research-design and QCA are discussed.
WP2006-42[Posted
on 21-June-2006]
Barbara
VIS (Vrije Universiteit Amsterdam)
States of welfare
or states of workfare?
A fuzzy-set ideal type analysis of major welfare state restructuring in
sixteen advanced capitalist democracies, 1985-2002
(PDF format)
NB: forthcoming in
Policy
& Politics.
Abstract:
Did welfare states change radically from welfare towards workfare
or was such a shift absent and was welfare state change regime specific
instead? This paper assesses this question for sixteen advanced capitalist
democracies for the period 1985-2002, using an innovative method, fuzzy-set
ideal type analysis. This study shows that the mainstream welfare state
literature’s prediction of no radical and regime specific change
holds for most countries. The regulation literature’s prediction
of radical change from welfare towards workfare is supported fully only
in Ireland and moderately in Denmark. Furthermore, interesting other patterns
are revealed in six countries.
WP2006-41[Posted
on 20-June-2006][Removed 7-Oct-2007]
Adrian
DUSA (University of Bucharest)
QCA Graphical User Interface manual
(PDF format)
NB: Updated version (October 2007) available as
WP2007-50.
WP2006-40[Posted
on 29-May-2006]
Jon
KVIST (Danish National Institute of Social Research)
MEASURING THE
WELFARE STATE – CONCEPTS, IDEAL TYPES AND FUZZY SETS IN COMPARATIVE
STUDIES
(PDF format)
NB: forthcoming in Jochen Clasen and Nico Siegel
(eds.), "Welfare Reform in Adanced Societies", Edward Elgar.
Abstract:
Is the glass half-empty? Is it more empty than full? Such questions are
often linked to judgements which concern qualitative states and changes
in degree and kind. Abound in comparative studies such judgements bring
forward issues of how best to conceptualise and measure. In comparative
studies of the welfare state they prompt reflections on what constitutes
the welfare state, how to operationalise it and how to measure change
over time and space.
Comparative welfare state research has made significant progress in the
theoretical understanding of the welfare state itself, not least due to
a dialogue between qualitatively and quantitatively oriented studies (Amenta,
2003). Since 1990, when Gøsta Esping-Andersen published Three Worlds
of Welfare Capitalism, a common starting point has been the distinction
between different types of welfare state regimes: identifying a liberal,
conservative and a social democratic welfare state regime. In short, diversity
- the co-existence of similarities and differences - characterises different
welfare states.
Comparative research however has made much less progress in the measurement
of welfare state and welfare state change. A lack of consensus about how
to measure either is the main reason why scholars disagree on the direction
and magnitude of recent change in social policy, i.e. whether reforms
amount to fundamental or marginal change (Clayton and Pontusson, 1998
with Pierson, 1996, or Gilbert, 2002 with Kvist, 1999).
This chapter offers an alternative approach to measurement
and a very different strategy, that of formulating a new way of going
about measurement by using fuzzy sets and axioms in fuzzy set theory.
The aim is to advance the application of fuzzy set theory as a new method
for conceptualisation and measurement (see Ragin, 2000 for a broad introduction
to fuzzy set social science). I argue that the fuzzy set approach is particularly
useful for assessing diversity and change across a limited set of cases,
and that it can overcome some of the problems typically related to measurement
validity and precision. In other words, using fuzzy sets help to assess
whether the glass is half-full or half empty, or how, if at all, the welfare
state is retrenched or restructured.
WP2006-39[Posted
on 22-May-2006]
Carsten
Q. SCHNEIDER (Central
European University) and Bernard
Grofman (University of California at Irvine)
IT MIGHT LOOK LIKE A
REGRESSION EQUATION … BUT IT’S NOT! AN INTUITIVE APPROACH
TO THE PRESENTATION OF QCA AND FS/QCA RESULTS
(PDF format)
Abstract:
Scholars who have presented their QCA and fs/QCA results in conference
papers or journal articles will most likely have encountered the problem
that an audience not trained in these approaches tends to read the notations
and graphs displaying the results as if they stemmed from standard statistical
techniques such as linear regression or factor analysis. This leads to
gross misunderstandings, since the underlying mathematical models and
the epistemology are different, and because the notations and graphs used
in QCA und fs/QCA carry a different meaning than similar looking ones
in standard statistical approaches. Thus readers may think they know what’s
going in QCA analyses when they really don’t.
The main aim of this paper is to offer seven ways, some new to this paper,
of presenting results in QCA and fs/QCA that are designed to make the
interpretability of results from these methods clearer and more intuitive:
(1) truth tables; (2) solution formulas; (3) parameters of fit; (4) Venn
diagrams; (5) dendograms; (6) x-y plots; and (7) membership scores for
solution terms – the latter two only appropriate for fuzzy set QCA.
We show that each form tends to be confused with one or more presentational
forms commonly used in standard statistical techniques, its “false
friend(s),” and thus misinterpreted; and so we try to clarify the
implications of each of these presentational tools by pointing out what
they do not mean.
Generally speaking, the presentation of results generated with any kind
of method applied in comparative social research has multiple purposes,
not all of which can always be achieved simultaneously in one presentational
form. In grosso modo, the presentation of results aims at: (a) displaying
relations between variables; (b) highlighting descriptive or causal accounts
for specific (groups of) cases; (c) expressing the fit of the result obtained
with the data at hand. Trying to accomplish all three of these purposes
is particularly important for QCA and fs/QCA because they have been explicitly
introduced as methods for bridging the gap between qualitative (case-oriented)
and quantitative (variableoriented) approaches of social scientific research.
While the individual presentational forms serve one or more (but never
all) of the three above-mentioned purposes, using a combination of them
in a fashion that covers all three bases allows us to display the full
potential and logic of QCA and fs/QCA methods.
WP2006-38[Posted
on 15-Feb-2006]
Daniel
BOCHSLER (University of Geneva)
Electoral engineering and inclusion
of ethnic groups:
Ethnic minorities in parliaments of Central and Eastern European countries
(PDF format)
Abstract:
Inter-ethnic fire was set in many countries in Central and Eastern
Europe after the breakdown of the communist regimes. This resulted in
discrimination of groups of citizens and in violent conflicts. An important
aspect of power-sharing regimes and of the discrimination or integration
of minority groups is the way, in which parliaments are elected. Parliamentary
elections may accelerate the flames – or they may be attempts to
extinguish them. To a certain degree, the elections’ success in
fire extinguishing relies on the electoral systems.
In this article, I shall test if electoral systems are favourable to ethnic
minority integration. For that purpose, I explore three questions: 1.
Do they enable ethnic minority parties to gain representation in parliament?
2. Do they allow a plurality amongst ethnic minority parties? 3. Do they
support a policy of conciliation by giving incentives to vote for mixed-ethnic
parties? Or do they on the contrary hinder their success?
My empirical test on a database covering 106 ethnic minorities in Central
and European countries confirms the importance of electoral laws for the
representation of ethnic minority parties. More concretely, my tests using
the Qualitative Comparative Approach (QCA) show that if proportional electoral
systems are amended with high national thresholds, even medium-sized territorially
concentrated ethnic minorities are excluded from parliament. And any kind
of plurality of singlemember district systems may be poison for the representation
of not-concentrated minorities and exclude local minorities.
WP2006-37[Posted
on 30-Jan-2006]
James
MAHONEY (Northwestern University) and Gary
Goertz (Arizona University)
A tale of two cultures:
contrasting quantitative and qualitative research
(PDF format)
Abstract:
The quantitative and qualitative research traditions can be thought
of as distinct cultures marked by different practices, beliefs, and norms.
In this essay, we adopt this imagery toward the end of contrasting these
research traditions across ten areas: (1) approaches to explanation, (2)
conceptions of causation, (3) multivariate explanations, (4) equifinality,
(5) scope and causal generalization, (6) case selection, (7) weighting
observations, (8) substantively important cases, (9) lack of fit, and
(10) concepts and measurement. We suggest that an appreciation of the
alternative assumptions and goals of the traditions can help scholars
avoid misunderstandings and contribute to more productive “cross-cultural”
communication in political science.
WP2006-36[Posted
on 27-Jan-2006]
Gregory
JACKSON (King's College London, and Research Institute of Economy,
Trade and Industry (RIETI))
Employee
Representation in the Board Compared: A Fuzzy Sets Analysis of Corporate
Governance, Unionism, and Political Institutions
(PDF format)
Abstract:
Why do employees have rights to representation within corporate boards
in some countries, but not in others? Board-level codetermination is widely
considered a distinctive feature of coordinated or nonliberal models of
capitalism. Existing literature stresses three sets of explanations for
codetermination rooted in corporate governance, union strength and political
systems. The paper compares data from 22 OECD countries using the QCA
method (Qualitative Comparative Analysis) and fuzzy sets approach to explore
necessary and sufficient conditions for board-level codetermination. The
results show two central pathways toward codetermination both rooted primarily
in union coordination and consensual political systems, but with divergent
implications for corporate governance systems in Scandinavia and Germany.
WP2005-35 [Posted
on 7-Nov-2005]
Carsten Q. SCHNEIDER (CEU Budapest)
and Claudius WAGEMANN (EUI
Florence)
Reducing
Complexity in Qualitative Comparative Analysis (QCA): Remote and Proximate
Factors and the Consolidation of Democracy
(PDF format)
NB: Forthcoming in European Journal of Political
Research 2006. This is an updated version of the WP2002-1 version (now
deleted).
Abstract:
Comparative methods based on set theoretic relationships, such as ‘fuzzy
set Qualitative Comparative Analysis’ (fs/QCA) represent an useful
tool for dealing with complex causal hypotheses in terms of necessary
and sufficient conditions under the constraint of a mediumsized number
of cases. However, real world research situations might make the application
of fs/QCA difficult in two respects, namely, with regard to the complexity
of the results and the phenomenon of limited diversity. We suggest a two-step
approach as one possibility to mitigate these problems. After introducing
the difference between remote and proximate
factors, the application of a two-step fs/QCA approach is demonstrated
analysing the causes of the consolidation of democracy. We find that different
paths lead to consolidation but all of them are characterised by a fit
of the institutional mix chosen to the societal context in terms of power
dispersion. Hence, we demonstrate that the application of fs/QCA in a
twostep manner helps to formulate and test equifinal and conjunctural
hypotheses in medium-size N comparative analyses and, thus, to contribute
to an enhanced understanding of social phenomena.
WP2005-34 [Posted on
12-Oct-2005]
Svend-Erik SKAANING (University
of Aarhus)
Respect
for Civil Liberties in Post-Communist Countries:
A Multi-Methodological Test of Structural Explanations
(PDF format)
NB:
This is an update of the 6-Oct-2005
version (now deleted).
Abstract:
This paper aims at uncovering factors explaining extent of membership
in the group of liberal polities among the post-communist countries. The
empirical examination makes use of three methodological tools –
crisp-set method, fuzzy-set method, and OLS-regression – associated
with two different approaches – diversity-oriented and variable-oriented
– to test theoretically different structural conditions/variables
supposed to facilitate the development of civil liberties. Six factors
are included in the analysis: ethno-religious diversity, natural resources,
early development, and three modernization indicators, i.e., GDP/cap.,
education and agricultural employment. The results diverge considerable
between different approaches, but minimally between the QCA methods connected
to the same approach. The OLS-regression shows that early development,
ethnic diversity, and education are significantly correlated with civil
liberty; education, though, not in the theoretically expected direction.
On the other hand, the crisp-set analysis, using logical cases in the
reduction procedure, points out early developed and not early developed
as a necessary and sufficient condition for the existence and absence
of a liberal regime, respectively. Finally, the fuzzy-set analysis, very
similarly, emphasizes early development as a sufficient condition for
liberal regimes in the post-communist context, if an inclusion of all
simplifying assumptions is allowed, and three paths to a not liberal regime
are uncovered, that is, the conjunction of the necessary condition not
early developed with either not wealthy, not independent of natural resources,
or not low agricultural employment rate. Thus, early development is undoubtedly
identified as the most important factor in the explanation of the respect
for civil liberty in post-communist countries. Regarding the QCA methods,
they appear to be valuable supplements and at times even alternatives
to standard statistical tests – the fuzzy-set method probably somewhat
more than the crisp-set method – especially when the case number
is rather low and when the relationship between the phenomenon to be explained
and the explanatory factors is characterized by multiple conjunctural
causation and necessity and/or sufficiency.
WP2005-33 [Posted on
29-Aug-2005]
Frank M. HAEGE
(Leiden University)
Constructivism,
Fuzzy Sets and (Very) Small-N:
Revisiting the Conditions for Communicative Action
(PDF format)
NB: A revised version of the paper has been
published in the Journal of Business Research 60, 5: 512-521 (2007)
Abstract:
In this paper, it is argued that the fuzzy set approach can engage
in a fruitful liaison with constructivist research. There are several
important properties of fuzzy set analysis that overlap with constructivist
theorizing and research practice. To demonstrate the usefulness of the
approach, Niemann’s study on the conditions for communicative action
is replicated and re-interpreted using fuzzy sets. The result is an improvement
of the informational content, the precision and the validity of conclusions
drawn from the empirical analysis. Furthermore, the re-interpretation
points to theoretical and conceptual issues that need more consideration
in future research.
WP2005-32 [Posted on 07-July-2005]
Caty CLEMENT (Harvard
University)
The Nuts and Bolts of
State Collapse: Common Causes and Different Patterns?
A QCA Analysis of Lebanon, Somalia and the former-Yugoslavia
(PDF format)
Abstract:
This paper is the result of a research project designed to address two
questions: ‘why’ do states collapse and ‘how’
do they collapse? Rather than testing existing theories (largely non existent),
this paper suggests a new model. The first issue, the causes of state
collapse, has been the focus of in depth research over the recent years.
The bulk of the comparative work came through large N studies focussing
on long-term structural conditions and often resulting in long shopping
lists of indicators. Instead, this research develops a concise set of
four core causes (rather than indicators) based on in-depth country research
(small n) using ‘soft’ qualitative data (quantitative being
often unreliable and constraining the research).
WP2005-31 [Posted on 18-March-2005]
Gary
GOERTZ (University of Arizona) and Jack
S. LEVY (Rutgers University)
CAUSAL EXPLANATIONS, NECESSARY CONDITIONS,
AND CASE STUDIES: WORLD WAR I AND THE END OF THE COLD WAR
(PDF format)
NB: Forthcoming
monograph edited by the authors, Causal
Explanations, Necessary conditions, and Case Studies:
World War I and the end of the Cold War.
Extract:
Our focus in this anthology is on how scholars often use a certain family
of causal explanations in their analyses of historical events, such as
World War I and the end of the Cold War. The next chapter by Goertz and
Levy serves as a survey of the various ways necessary condition counterfactuals
appear in the literature on the causes of World War I and the end of the
Cold War. We do not pretend to cover exhaustively these massive debates,
but we have chosen prominent scholars whose work illustrates the various
aspects of our central theme. While the idea of
a necessary condition is simple, they show that there are extensive ramifications
for research design, theory, and causal explanations.
Although necessary condition counterfactuals are the central focus of
this
volume, not all of the contributors agree that the concept is a useful
one. In particular, Brooks and Wohlforth (chapter 9) argue that probabilistic
approaches to explanation and causation are more useful. Thompson worries
that an emphasis on necessary and sufficient condition causation will
detract from the goal of evaluating the relative causal weights of different
factors (see Goertz and Starr 2002 for a discussion of these two issues).
So while Goertz and Levy show that the necessary condition explanatory
strategy is widespread, this does not necessarily mean that it is without
problems or that other alternative strategies do not exist.
WP2005-30 [Posted on 14-January-2005]
Peer C. FISS (Queen's
School of Business)
A SET-THEORETIC APPROACH TO ORGANIZATIONAL
CONFIGURATIONS
(PDF format)
NB: Forthcoming in Academy of Management Review.
Re-worked version of WP2004-24.
Abstract:
I argue that research on organizational configurations has been limited
by a mismatch between theory and methods. While configurational theory
stresses nonlinearity, synergistic effects, and equifinality, empirical
research has largely drawn on methods that assume linearity, additive
effects, and unifinality. I introduce set-theoretic methods as a viable
alternative for overcoming this mismatch. Set-theoretic methods conceptualize
cases as combinations of attributes and use Boolean algebra to derive
simplified expressions of combinations that lead to a specific outcome.
I demonstrate the value of such methods for studying organizational configurations
and discuss their applicability for examining equifinality and limited
diversity among configurations, as well as their relevance to other research
fields such as complementarities theory, complexity theory, and the resource-based
view.
WP2004-29 [Posted on 22-December-2004]
Axel MARX and Hans
PEETERS (KULeuven)
WIN FOR
LIFE: AN EMPIRICAL EXPLORATION OF THE SOCIAL CONSEQUENCES OF INTRODUCING
A BASIC INCOME
(PDF format)
Abstract:
The aim of this paper is twofold. First of all, the paper discusses why,
how, and to what extent, natural experiments such as lotteries can contribute
to research which empirically explores possible social consequences of
the introduction of a Basic Income. The second aim is to focus on the
question of what, if anything, happens after the introduction of a Basic
Income.
The paper is structured in three parts. The first part of the paper addresses
the question of why natural experiments constitute an interesting research-strategy.
Via a comparison with a genuine experiment a theoretical case is made
to conduct lottery research, which has some distinctive strengths vis-à-vis
an experiment.
The second part of the paper discusses an ongoing pilot-project which
investigates the consequences of winning the Belgian lottery game Win
for Life (unconditional lifelong monthly allowance of 1.000 euro). It
is assessed to what extent this game represents a good proxy for a Basic
Income and what conclusions can be drawn from it. In addition, the results
of a pilot-project are discussed using Qualitative Comparative Analysis
(QCA). A specific issue of model-selection in a QCA-type of analysis is
also addressed.
In a third part, a proposal for the extension of lottery research is suggested.
WP2004-28 [Posted on 7-December-2004]
Charles RAGIN (University
of Arizona)
From Fuzzy Sets to Crisp
Truth Tables (April05 version)
(PDF format)
NB: Re-worked version of WP2004-26
Abstract:
One limitation of the truth table approach is that it is designed for
causal conditions are simple presence/absence dichotomies (i.e., Boolean
or "crisp" sets). Many of the causal conditions that interest social scientists,
however, vary by level or degree. For example, while it is clear that
some countries are democracies and some are not, there are many in-between
cases. These countries are not fully in the set of democracies, nor are
they fully excluded from this set. Fortunately, there is a well-developed
mathematical system for addressing partial membership in sets, fuzzy-set
theory. Section 2 of this paper provides a brief introduction to the fuzzy-set
approach, building on Ragin (2000). Fuzzy sets are especially powerful
because they allow researchers to calibrate partial membership in sets
using values in the interval between 0 (nonmembership) and 1 (full membership)
without abandoning core set theoretic principles, for example, the subset
relation. Ragin (2000) demonstrates that the subset relation is central
to the analysis of multiple conjunctural causation, where several different
combinations of conditions are sufficient for the same outcome.
While fuzzy sets solve the problem of trying to force-fit cases into one
of two categories (membership versus nonmembership in a set), they are
not well suited for conventional truth table analysis. With fuzzy sets,
there is no simple way to sort cases according to the combinations of
causal conditions they display because each case's array of membership
scores may be unique. Ragin (2000) circumvents this limitation by developing
an algorithm for analyzing configurations of fuzzy-set memberships that
bypasses truth tables altogether. While this algorithm remains true to
fuzzy-set theory through its use of the containment (or inclusion) rule,
it forfeits many of the analytic strengths and virtues that follow from
analyzing evidence in terms of truth tables. For example, truth tables
are very useful for investigating "limited diversity" and the consequences
of different "simplifying assumptions" that follow from the use of different
subsets of "remainders" to reduce complexity (see Ragin 1987; Ragin and
Sonnett 2004). Analyses of this type are difficult without using a truth
table as the starting point. Ragin and Sonnett (2004), for example, show
how to use QCA to aid counterfactual analysis and link the analysis of
counterfactual conditions to core practices in case-oriented research.
Truth tables are central to the analysis of counterfactuals, and the techniques
described in Ragin and Sonnett (2004) cannot be implemented without the
aid of truth tables.
Section 3 of this paper builds a bridge between fuzzy sets and truth tables,
demonstrating how to construct a conventional Boolean truth table from
fuzzy-set data. It is important to point out that this new technique takes
full advantage of the gradations in set membership central to the constitution
of fuzzy sets and is not predicated upon a dichotomization of fuzzy membership
scores. To illustrate these procedures I use data on class voting in the
advanced industrial societies, compiled by Paul Nieuwbeerta (see, e.g.,
Nieuwbeerta and de Graaf 1999; Nieuwbeerta and Ultee 1999; Nieuwbeerta,
de Graaf and Ultee 2000). It is important to point out that the approach
sketched in this paper offers a new way to conduct fuzzy-set analysis
of social data. This new analytic strategy is superior in several respects
to the one sketched in Fuzzy-Set Social Science (Ragin, 2000). While both
approaches have strengths and weaknesses, the one presented here uses
the truth table as the key analytic device. A further advantage of the
fuzzy-set truth-table approach presented in this paper is that it is more
transparent. Thus, the researcher has more direct control over the process
of data analysis. This type of control is central to the practice of case-oriented
research.
WP2004-27 [Posted on 11-October-2004]
Astrid SPREITZER
(Institute for Advanced Studies, Vienna) and Sakura
YAMASAKI (UCL, FNRS, Louvain-la-Neuve)
Beyond
methodological tenets - The worlds of QCA and SNA and their benefit to
Policy Analysis
(PDF format)
Abstract:
The aim of this paper is to present combinations of Social Network Analysis
(SNA) and Qualitative Comparative Analysis (QCA) and their benefit to
Policy Analysis. We think that QCA and SNA are particularly suited to
explain complex macro-social phenomena, just like policies. SNA gives
access to a set of actors and the relationships between them. The main
goal is to model these relationships in order to study action and structure
in their mutual dependence (Wasserman and Faust 1997). QCA on the other
hand helps to uncover regularities across cases while maintaining within-case
complexity; it offers "multiple conjunctural explanations" (Ragin 1987,
2003). First we expose our understanding of Policy Analysis and the problems
research on the topic faces. The second part of the paper focuses on SNA
and QCA as two approaches, which stand in between of the conventional
qualitative/quantitative logic of research. Therefore we will explain
the main principles of the methods but also show the communities of the
two, the underlying meta-theoretical assumptions, the opportunities they
offer to appear as supplementing to each other. Finally, it is to explore,
how the combination of SNA and QCA helps to explain policies.
WP2004-26 [Posted on 6-September-2004]
Charles RAGIN (University
of Arizona)
From Fuzzy Sets to Crisp Truth Tables
NB: Re-worked version available as WP2004-28
WP2004-25 [Posted on 26-August-2004]
Axel MARX (KU Leuven)
and Jan DOMBRECHT (Federal Government
of Belgium)
THE
ORGANISATIONAL ANTECEDENTS OF REPETITIVE STRAIN INJURIES: A Systematic
Comparative Case Analysis of Assembly, Sorting and Packaging Jobs
(WP removed in Dec. 2006. This WP will be
published soon in "Journal of Business Research".)
Abstract:
The paper presents a specific research-design – systematic comparative
case analysis - to analyse the impact of organisational characteristics
on individual level outcomes. A systematic comparative case analysis consists
of an across case and within case analysis of a limited set of comparable
cases. Across case analysis or Qualitative Comparative Analysis (QCA)
aims to identify similarities and differences between configurations of
explanatory variables. Within case analysis aims to identify the causal
mechanisms which link configurations to outcomes. Systematic comparative
case analysis is applied to a research question on the organisational
antecedents of repetitive strain injuries of the wrist in highly repetitive,
non-fragmented and simple jobs. In total, 16 cases (each consisting on
average of 15 workers) were analysed.
WP2004-24 [Posted on 3-August-2004]
Peer C. FISS (Queen's University)
Towards a Set-theoretic Approach for Studying Organizational
Configurations
(The re-worked version is posted as WP2005-30.
It is forthcoming in the Academy of Management Review.)
Abstract:
I argue that research on organizational configurations has been limited
by a mismatch between theory and methods. While configurational theory
stresses nonlinearity, synergistic effects, and equifinality, empirical
research has largely drawn on methods that assume linearity, additive
effects, and unifinality. I introduce set-theoretic methods as a viable
alternative for overcoming this mismatch. Set-theoretic methods conceptualize
cases as combinations of attributes and use Boolean algebra to derive
simplified expressions of combinations that lead to a specific outcome.
I demonstrate the value of such methods for studying organizational configurations
and discuss their applicability for examining equifinality and limited
diversity among configurations.
WP2004-23 [Posted on 10-July-2004]
Charles C. RAGIN (University
of Arizona) and John SONNETT
(University of Arizona)
Between Complexity
and Parsimony: Limited Diversity, Counterfactual Cases, and Comparative
Analysis
(PDF format)
NB: Forthcoming in: Sabine KROPP and Michael MINKENBERG (eds.),
Vergleichen in der Politikwissenschaft. Wiesbaden: VS Verlag für
Sozialwissenschaften, 2004.
Abstract:
Counterfactual analysis has a long and distinguished history in comparative
research. To some, counterfactual analysis is central to comparative inquiry
because such research typically embraces only a handful of empirical cases
(Fearon 1991). If there are only a few instances (e.g., of revolution),
then researchers, of necessity, must compare empirical cases to hypothetical
cases. The affinity between counterfactual analysis and comparative research,
however, derives not from its focus on small Ns, but from its configurational
nature. Case-oriented explanations of outcomes are often combinatorial
in nature, stressing specific configurations of causal conditions. Rather
than focus on the net effects of causal conditions, case-oriented explanations
emphasize their combined effects.
To support an argument emphasizing combinations of causal conditions,
it is necessary for researchers to compare cases that are closely matched
with each other. The ideal comparison is between pairs of cases that differ
on only one causal condition (Mill 1843). Such comparisons help researchers
establish whether or not a specific causal condition is a integral part
of the combination of conditions that generates the outcome in question.
It is very difficult to match empirical cases in this manner, however,
due to the limited diversity of empirical social phenomena.
In this paper, we discuss the impact of limited diversity on comparative
case-oriented research. We show how limited diversity is conceived in
Qualitative Comparative Analysis (QCA; see Ragin 1987, 2000), and link
QCA strategies for addressing limited diversity to counterfactual analysis.
We distinguish two kinds of counterfactual cases, "difficult" and "easy,"
and demonstrate procedures for incorporating "simplifying assumptions"
into QCA based on the analysis of "easy" counterfactual cases. We illustrate
these methods with comparative data on international fishing regimes collected
by Olav Schram Stokke (2004).
WP 2004-22 [Posted
on 02-June-2004]
Gary GOERTZ (University of
Arizona) and Jack S. LEVY
(Rutgers University)
Causal Explanations, Necessary conditions, and Case
Studies
The re-worked version of the chapter can
be found at 2005-31.
Extract:
In our discussion of causation in this chapter we make no attempt to survey
the ways a given event can be an important cause. We focus on necessary
conditions as a particular kind of important cause. Necessary conditions
are important causes because they directly imply a key counterfactual:
If X had not been present/occurred then the Cold War would not have ended.
A probabilistic version is that if X had not occurred then the end of
Cold War would have been very unlikely.
It turns out that this rather simple causal strategy has wide-spread ramifications
tions for explaining individual events. We first take a look at simple
necessary condition explanations and their intimate connection with counterfactuals.
However, necessary conditions also play an essential role in multivariate
explanations of events as well. For example, one frequently reads about
historical chains of events. If we take this metaphor seriously then each
“link” is a necessary condition factor: break one link and the chain is
broken. (...)
WP 2004-21
[Posted on 25-May-2004]
Kenneth NELSON (Swedish
Institute for Social Research)
The Last Resort. Determinants
of Generosity of Means-Tested Minimum Income Protection in Welfare Democracies
(PDF format)
Abstract:
This study evaluates institutional linkages between different types of
social security programs in eighteen welfare states in the early 1990s.
The purpose is to analyze the determinants of cross-national variations
in the level of minimum income protection. Three hypotheses of an institutional
relationship between social insurance and the generosity of minimum income
protection are tested by means of OLS-regression, qualitative comparative
analysis and fuzzy set analysis. From an economic point of view it is
hypothesized that the impact of social insurance on the generosity of
minimum income protection is mediated through its effects on the costs
for means-tested benefits. From a political perspective, the hypothesis
is that this impact derives from the degree to which social policies promote
cross-class interests in support for the welfare state. Finally, from
a strictly institutional perspective, the hypothesis is that social insurance
sets certain upper limits to the level of means-tested benefits, which
determine the possibilities of raising the value of minimum income protection.
The empirical analyses show that not all aspects of social insurance are
of equal importance in explaining cross-national variations in the level
of minimum income protection. The most important aspect seems to be the
degree to which social insurance provides income security, which supports
the middle-class inclusion hypothesis on institutional dependencies between
different tiers of the social security system.
WP 2004-20 [Posted
on 15-Apr-2004]
Lasse CRONQVIST (Marburg
University)
Presentation
of TOSMANA. Adding Multi-Value Variables and Visual Aids to QCA
(PDF format)
N.B.: This is a new version of a paper posted earlier (see WP 2003-14)
written in September 2003.
Abstract:
In this presentation the TOSMANA (Tool for Small-N Analysis) software
is described. TOSMANA is a tool for case-based comparative analysis, implementing
existing techniques as Qualitative Comparative Analysis (QCA) as well
as new approaches for macro-qualitative comparative analysis. The different
parts of TOSMANA are explained and some ideas on further development are
introduced.
WP 2004-19 [Posted on 30-Jan-2004]
Gary GOERTZ (University of
Arizona) and James MAHONEY
(Brown University)
Two-Level
Theories and Fuzzy Sets
(PDF format)
N.B.: This is a new version of a paper posted earlier (see WP 2003-6)
written in April 2003.
Abstract:
Two-level theories explain outcomes with causal variables at two levels
of analysis that are systematically related to one another. Although many
prominent scholars in the field of comparative analysis have developed
two-level theories, the empirical and methodological issues that these
theories raise have yet to be investigated. In this article, we explore
different structures of two-level theories and consider the issues involved
in testing these theories with fuzzy-set methods. We show that grasping
the overall structure of two-level theories requires both specifying the
particular type of relationship (i.e., causal, ontological, or substitutable)
that exists between and within levels of analysis and specifying the logical
linkages between levels in terms of necessary and sufficient conditions.
We argue that for the purposes of testing these theories fuzzy-set analysis
provides a powerful set of tools. However, to realize this potential,
investigators using fuzzy-set methods must be clear about the two-level
structure of their theories from the onset. We illustrate these points
through an empirical, fuzzy-set test of Skocpol’s States and Social Revolutions.
WP 2004-18 [Posted on 30-Jan-2004]
James MAHONEY (Brown University)
and Gary GOERTZ (University
of Arizona)
The
Possibility Principle: choosing negative cases in comparative research
(PDF format)
N.B.: This is a new version of a paper posted earlier (see WP 2003-8)
written in July 2003.
Abstract:
A central challenge in qualitative research involves selecting the “negative”
cases (e.g., nonrevolutions, nonwars) to be included in analyses that
seek to explain positive outcomes of interest (e.g., revolutions, wars).
Although it is widely recognized that the selection of negative cases
is consequential for theory testing, methodologists have yet to formulate
specific rules to inform this selection process. In this paper, we propose
a principle – the Possibility Principle – that provides explicit, rigorous,
and theoretically-informed guidelines for choosing a set of negative cases.
The Possibility Principle advises researchers to select only negative
cases where the outcome of interest is possible. Our discussion elaborates
this principle and its implications for current debates about case selection
and strategies of theory testing. Major points are illustrated with substantive
examples from studies of revolution, economic growth, welfare states,
and war.
WP 2003-17 [Posted on 29-Sept-2003]
Alain GOTTCHEINER (Université
Libre de Bruxelles)
Contradictions
and their use in falsification : the case of comparative linguistics and
QCA's contribution
(PDF format)
Abstract:
Linguists searching about laws of phonetic changes make use of the entire
corpus at their disposal. By so doing, they find laws that correctly describe
observed changes, especially " splits ", but can't be checked. Such a
law may always be found if using enough parameters, but doesn't guarantee
a fair description. In a Popperian perspective, we'd like to suggest working
on a partial corpus, trying to establish laws that correctly account for
all matching multiplets considered, then applying these assumed laws to
the rest of the corpus ; if no counterexample is found, the set of laws
gains in credibility.
In this approach, QCA may be very useful, because it allows us to : 1)
consider all possible influences (position in the word, preceding and
following phoneme, umlaut/ablaut, position relative to stress, …) as conditions
; 2) use contradictions as guides to the detection of influences we forgot
to use ; 3) modify the corpus and set of conditions at will ; 4) produce
several laws, among which we may choose the most plausible ; 5) find implications
that aren't seen at first glance.
WP 2003-16 [Posted on 24-Sept-2003]
David LEVI-FAUR (University
of Haifa and Australian National University)
Comparative Research
Designs in the Study of Regulation: How to Increase the Number of Cases
without Compromising the Strengths of Case-Oriented Analysis
(PDF format)
Abstract:
The aim of this chapter is to explore the role of variations and similarities
in Medium-N comparative analysis and to suggest a technique that could
maximize their explanatory power in designs that combine two or more comparative
approaches to the study of regulatory change. The chapter identifies four
popular comparative approaches to the study of politics and policy in
general and regulation in particular. These four might best be titled
the National Patterns Approach (NPA), the Policy Sector Approach (PSA),
the International Regime Approach (IRA), and the Temporal Patterns Approach
(TPA). While these approaches are not necessarily contradictory they represent
different assumptions as to the determinants of political and regulatory
change. Each of these approaches omits some important sources of variations
and similarities in the regulation of the economy and society. To overcome
these omissions it is suggested that combinations of these approaches
- through complex research designs - might prove a sounder and more effective
method for the study of regulation.
WP 2003-15 [Posted on 24-Sept-2003]
Jon KVIST (Danish National Institute of
Social Research)
CONCEPTUALISATION, CONFIGURATION,
AND CATEGORISATION - DIVERSITY, IDEAL TYPES AND FUZZY SETS IN COMPARATIVE
WELFARE STATE RESEARCH
(PDF format)
Abstract:
This paper advances a new method for studying ideal types, fuzzy-set theory,
which is a framework that allows a precise operationalisation of theoretical
concepts, the configuration of concepts into analytical constructs, and
the categorisation of cases. In a Weberian sense ideal types are analytical
constructs used as yardsticks to measure the similarity and difference
between concrete phenomena. Ideal type analysis involves differentiation
of categories and degrees of membership of such categories. In social
science jargon, this means analysis involving the evaluation of qualitative
and quantitative differences or, in brief, of diversity. Fuzzy set theory
provides a calculus of compatibility. It can measure and compute theoretical
concepts and analytical constructs in a manner that is true to their formulation
and meaning. This paper sets out elements and principles of fuzzy set
theory that are useful for ideal type analysis and presents two illustrative
examples of how it can be used in comparative studies. The examples concern
changing Nordic welfare policies in the 1990s, unemployment and child
family policies, and relate to their conformity to predefined ideal typical
models.
WP 2003-14 [Posted on 23-Sept-2003]
Lasse CRONQVIST (Marburg
University)
Presentation of TOSMANA. Adding Multi-Value Variables
and Visual Aids to QCA
The re-worked version is posted as WP2005-20.
Abstract:
In this presentation the outlines of the TOSMANA software development
are described. TOSMANA is intended to be a tool for case-based comparative
analysis, implementing existing techniques as Qualitative Comparative
Analysis (QCA) as well as new approaches towards macro-qualitative comparative
analysis. The different parts of TOSMANA are explained and some ideas
on further development are introduced.
WP 2003-13 [Posted on 23-Sept-2003]
Tsutomu WATANABE (Shinshu
University)
Where Theory
and Reality Meet: Using the Full Potential of QCA by Exploiting the Intersection
Function of the QCA Software. International Comparison Analysis about
the Occurrence of Social Movement
(PDF format)
WP 2003-12 [Posted on 23-Sept-2003]
Catherine MOURY (University
of Siena and Université catholique de Louvain)
Use of Fuzzy Set
in an Explanatory Research: a Study on the Characteristics of Coalition
Agreement
(PDF format)
Abstract:
This research aims to understand why the complete and precise character
of the coalition agreement varies. I assume that the fuzzy set is the
appropriate method in such an explanatory phase of the research, because
it allows the “necessary dialogue between ideas and evidence”. The results
suggest to consider the redaction of the agreement as a two level game,
between the parties and between the government and the parties which support
it.
WP 2003-11 [Posted on 23-Sept-2003]
Patrick DUMONT (Université
catholique de Louvain) and Hanna
BÄCK(Uppsala University)
Why
So Few, and So Late? Green Parties and the Question of Governmental Participation
(PDF format)
Abstract:
Green parties have been represented in the parliaments of European Union
countries since 1981 but only a few have entered national governments,
and this has not happened until recently. Using a data-set comprising
44 government formation opportunities where Greens were represented in
parliament, we show that parties involved in these bargaining situations
were more office-oriented than earlier studies argued. As Green parties
are less office-seeking than other parties, this general tendency for
office-seeking behaviour in government formation may partly account for
the scarcity of Greens in government. Further, we test a number of hypotheses
derived from theories that account for the specific nature of Green parties
in terms of their office-, policy- and vote-seeking orientations. We find
that Greens participate in government when they have lost at least one
election, when they are identified as a clear electoral threat by the
main party of the left and when the policy-distance between
the Greens and the main left party is small. As these simultaneous conditions
only materialized recently, and in a few countries, we argue that our
analysis, which is the first comparative and multivariate test on this
question, helps explaining the scarcity and the delay of Green governmental
participation.
WP 2003-10 [Posted on 10-Sept-2003]
Charles RAGIN (University
of Arizona)
Making
Comparative Analysis Count
(PDF format)
WP 2003-9 [Posted on 21-Aug-2003]
Charles RAGIN (University
of Arizona)
Recent advances
in fuzzy-set methods and their application to policy questions
(PDF format)
Abstract:
Fuzzy sets have many potential applications in the social sciences. The
ideas and suggestions presented in Fuzzy-Set Social Science scratch only
the bare surface of their potential uses, for there are many ways to integrate
fuzzy sets (and set-theoretic thinking more generally) into social research.
In this paper, I sketch several recent advances in fuzzy-set methods,
illustrating them with examples drawn from policy research. While these
new fuzzy-set methods build on arguments presented in Fuzzy-Set Social
Science, they also forge a strong link to some of the crisp-set principles
presented in The Comparative Method, especially those concerning the issue
of limited diversity. Thus, the techniques presented in this paper are
generally relevant to both crisp-set and fuzzy-set analysis.
The first advance I present is the elaboration and refinement of the concepts
of "consistency" and "coverage" in set-theoretic analysis. Specifically,
I show how to assess the consistency and coverage of combinations of causal
conditions. Consistency concerns the degree to which a combination of
causal conditions is consistent with an argument of sufficiency; coverage
concerns the relative importance of combinations of sufficient conditions
in the effort to explain or "cover" instances of the outcome. In this
discussion I emphasize the fuzzy-set analysis of sufficient combinations
of causal conditions, but the two principles apply just as well to the
fuzzy-set analysis of necessary conditions and to the analysis of necessity
and sufficiency as set-theoretic relations using crisp sets.
The second advance I discuss is a new algorithm for the incorporation
of "simplifying assumptions" into the results of applications of QCA and
fs/QCA. This new algorithm allows the direct incorporation of theoretical
and substantive knowledge into the evaluation of simplifying assumptions
in situations of "limited diversity" (which is the rule in the study of
naturally occurring social phenomena). I illustrate this algorithm with
crisp sets, and then extend it to fuzzy sets. Along the way, I also introduce
a new algorithm for the fuzzy-set analysis of social data. The new algorithm
is more amenable to the analysis of limited diversity than the one presented
in Fuzzy-Set Social Science.
WP 2003-8 [Posted on 4-Aug-2003]
James MAHONEY (Brown University)
and Gary GOERTZ (University
of Arizona)
The
Possibility Principle and case selection: choosing negative cases in comparative
analysis
(PDF format)
Abstract:
A central challenge in qualitative research involves selecting the ``negative''
cases (e.g., nonrevolutions, nonwars) to be included in analyses that
seek to explain positive outcomes of interest (e.g., revolutions, wars).
Although it is widely recognized that the selection of negative cases
is highly consequential for theory testing, methodologists have yet to
formulate specific rules to inform this selection process. In this paper,
we propose a principle -- the Possibility Principle -- that provides explicit,
rigorous, and theoretically-informed guidelines for choosing a set of
negative cases. The Possibility Principle advises researchers to select
only negative cases where the outcome of interest was possible. An outcome
is considered possible if one or a small number of independent variables
predict its occurrence. Our discussion elaborates this principle and its
implications for both theory formulation and theory testing. Major points
are illustrated with substantive examples from studies of revolution,
economic growth, democracy, and interstate war.
WP 2003-7 [Posted on 11-June-2003]
Gary GOERTZ (University
of Arizona)
Assessing the
importance of necessary or sufficient conditions in fuzzy-set social science
(PDF format)
Abstract:
Assessing the importance of necessary or sufficient conditions in fuzzy-set
social science
Political scientists of all stripes have proposed numerous necessary or
sufficient condition hypotheses. For methodologists a question is then
how can we assess the ``importance'' of these necessary conditions. This
paper addresses two central questions about the importance of necessary
or sufficient conditions. The first regards their ``absolute'' importance
which is addressed via the concept of the trivialness of necessary or
sufficient conditions. The second importance question deals with the relative
importance of necessary or sufficient conditions: for example, if X_1
and X_2 are necessary or sufficient conditions, is one more important
than the other? The paper develops measures to assess the importance of
necessary or sufficient conditions in three related contexts: (1) Venn
diagrams, (2) 2x2 tables, and (3) fuzzy logic, with an emphasis on fuzzy
logic methods. The empirical analysis uses the measures of absolute and
relative importance to extend Ragin's (2000) discussion of the causes
of IMF riots.
WP 2003-6 [Posted on 13-May-2003]
Gary GOERTZ (University
of Arizona) and James MAHONEY
(Brown University)
Two-level
theories and fuzzy logic
(PDF format)
Abstract:
Two-level theories and fuzzy-set analysis
Two-level theories explain outcomes with causal variables at two levels
of analysis that are systematically related to one another. Although many
prominent scholars in the field of comparative analysis have developed
two-level theories, the empirical and methodological issues that these
theories raise have yet to be s investigated. In this article, we explore
different structures of two-level theories and consider the issues involved
in testing these theories with fuzzy-set methods. We show that grasping
the overall structure of two-level theories requires both specifying the
particular type of relationship (i.e., causal, ontological, or substitutable)
that exists between levels of analysis and specifying the logical linkages
between levels in terms of necessary and sufficient conditions. For the
purposes of testing these theories, we illustrate using Skocpol's States
and Social Revolutions that fuzzy-set analysis provides a powerful set
of tools. However, as the Skocpol example shows, fuzzy-set methods are
not effective if the investigator fails to be clear about the two-level
structure of these theories from the onset.
WP 2003-5 [Posted
on 12-Jan-2003] [Removed Nov. 2004][Updated version as WP2007-48]
Olav Schram STOKKE (The Fridtjof
Nansen Institute)
Boolean analysis, mechanisms, and the study of regime
effectiveness
NB: An
updated and published version of this WP is available as
WP2007-48. See also:
Stokke, Olav Schram (2004). "Boolean Analysis, Mechanisms, and the
Study of Regime Effectiveness". Arild Underdal and Oran R. Young (eds.),
Regime Consequences: Methodological Challenges and Research Strategies.
Dordrecht: Kluwer Academic. pp. 87-119.
WP 2002-4 [Posted on 19-Nov-2002]
Corentin CURCHOD (Ecole
Polytechnique)
Diversity-oriented
research. Between complexity and generality
(PDF format)
Warning
This paper is NOT a working paper. This is a few notes I took while reading
Charles Ragin's book on Fuzzy-Set (Ragin [2000]. "Fuzzy-Set Social Science",
Chicago: The University of Chicago Press). I also selected a few quotations
from the book, which seem useful to me. They appear in the text with a
left borderline. Please do not quote this document, and check the original
book before reusing quotations.
Avertissement
Ce papier N'EST PAS un working paper. Il s'agit de quelques notes prises
pendant la lecture de l'ouvrage de Charles Ragin concernant les ensembles
flous (Ragin [2000]. "Fuzzy-Set Social Science", Chicago: The University
of Chicago Press). J'ai également sélectionné dans l'ouvrage quelques
citations qui me semblaient utiles. Elles apparaissent dans le texte qui
suit avec une bordure à gauche. Ne pas citer ce document, et vérifier
l'ouvrage original avant de réutiliser les citations.
WP 2002-3 [Posted on 8-Nov-2002]
Corentin CURCHOD (Ecole
Polytechnique)
La méthode comparative
en sciences de gestion : Vers une approche quali-quantitative de la réalité
managériale
(PDF format)
WP 2002-2 [Posted
on 5-Nov-2002]
Dirk BERG-SCHLOSSER
(Philipps-University, Institute of Political Science)
Macro-quantitative
vs macro-qualitative methods in the social sciences - testing empirical
theories of democracy
(PDF format)
WP 2002-1 [Posted
on 31-Oct-2002]
Carsten Q. SCHNEIDER,
Claudius WAGEMANN (European
University Institute, Dpt of Political and Social Sciences)
How to draw causal inference (despite) using QCA: the
'two-step, multi-equation FS/QCA approach'
NB: Updated version available as
WP2005-35
Last modified: 11-Nov-2008
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