QCA answers a question regression is poorly suited to: which combinations of conditions produce an outcome, when several different combinations can each produce it on their own. It treats cases as configurations rather than as bundles of independent variables, and it works in the awkward middle range — roughly 10 to 60 cases — where there are too many for close comparative reading and too few for statistical modelling. That combination of properties has made it one of the few formal methods to gain real traction in evaluation, and also one of the most frequently misapplied.
Three ideas that make QCA different
Equifinality. Several distinct causal paths can lead to the same outcome. A regression coefficient assumes one additive pathway; QCA expects and reports multiple, and a result showing three different routes to success is a substantive finding rather than a modelling failure.
Conjunctural causation. Conditions work in combination, not in isolation. A condition may be irrelevant on its own and decisive in the presence of two others. QCA is built to express that; interaction terms in a regression express it badly beyond two or three variables.
Asymmetry. The conditions that produce an outcome are not simply the negation of those that prevent it. QCA analyses presence and absence separately, and the two analyses regularly identify different conditions. Treating them as mirror images is one of the commonest errors.
Underneath all three sits set theory: a case is a member of the set of "high-performing sites" to some degree, and analysis asks about necessity (the outcome only occurs where this condition is present) and sufficiency (wherever this combination is present, the outcome follows).
Origins and variants
QCA was introduced by Charles Ragin in The Comparative Method (1987) and extended to fuzzy sets in Fuzzy-Set Social Science (2000). The practical reference for both variants is Rihoux and Ragin's edited Configurational Comparative Methods — see Crisp-Set Qualitative Comparative Analysis and Qualitative Comparative Analysis using Fuzzy Sets.
The standard technical treatment, and the source for the diagnostic thresholds most reviewers apply, is Schneider and Wagemann's Set-Theoretic Methods for the Social Sciences.
Its evaluation use was argued most influentially by Blackman and colleagues in Using Qualitative Comparative Analysis to understand complex policy problems. The variants in general use are csQCA (crisp sets: conditions are present or absent), fsQCA (fuzzy sets: graded membership between 0 and 1) and mvQCA (multi-value). fsQCA is now the default for evaluation work because most real conditions are matters of degree.
How a QCA is carried out
Select cases and conditions
Cases must be comparable enough that the same conditions mean the same thing in each. Conditions are chosen from theory and case knowledge, not from what happens to be in the dataset.
Keep the number small. With four conditions there are 16 logically possible configurations; with eight there are 256, and with 40 cases most of those rows will be empty — the "limited diversity" problem, and the fastest way to an unpublishable analysis. Four to seven conditions is the normal working range.
Calibrate
Calibration converts raw data into set membership, and it is the step that determines everything downstream. For each condition and for the outcome, define three anchors: full membership, full non-membership, and the crossover point of maximum ambiguity.
Anchors must be justified substantively — from theory, from an external standard, from known breaks in the cases — never from the sample distribution. Calibrating at the median is not calibration; it is a relabelling of quantiles, and it is the single most common reason a QCA is rejected in review.
Test necessity
Before analysing sufficiency, check each condition individually for necessity: does the outcome occur only where this condition is present? Consistency above about 0.9 with meaningful relevance is the usual bar.
Necessity claims are strong and comparatively rare. Report them separately; folding them into the sufficiency solution obscures both.
Build the truth table and set thresholds
Each logically possible configuration becomes a row, with the cases that display it and the proportion showing the outcome. Two thresholds are then set: a frequency threshold (how many cases a row needs to be analysed at all) and a consistency threshold (how consistently a row must show the outcome to count as sufficient).
Both are judgements and both must be reported. A common and defensible practice is to look for a natural gap in the consistency distribution rather than applying 0.8 mechanically.
Minimise, and choose a solution to report
Boolean minimisation reduces the sufficient configurations to their simplest expression. It produces three solutions differing in how they treat logical remainders — configurations no observed case displays.
The conservative solution assumes nothing about remainders and is the safest. The parsimonious solution allows any simplifying assumption and is the most compact, sometimes at the cost of implausibility. The intermediate solution incorporates only assumptions consistent with stated directional expectations, and is what most published work reports — with the expectations stated explicitly.
Interpret against the cases
Return to the cases. Each solution path should be interpretable as a recognisable causal story, and the deviant cases — those consistent in kind but not covered, or covered but without the outcome — are usually the most informative part of the analysis.
A QCA that never goes back to the cases has abandoned the qualitative half of its name.
Reading consistency and coverage
| Measure | What it asks | Rough guide | What a weak value means |
|---|---|---|---|
| Consistency (sufficiency) | How often does this configuration actually produce the outcome? | ≥ 0.80, ideally ≥ 0.85 | Cases with the configuration but no outcome — the claim is unreliable |
| Raw coverage | What share of the outcome does this path account for? | No fixed bar; report it | A real but narrow path — genuine, and rare |
| Unique coverage | What share does only this path explain? | Report per path | Near zero means the path is redundant with another |
| Solution coverage | What share does the whole solution account for? | ≥ 0.60 is comfortable | Most instances of the outcome are unexplained |
| Consistency (necessity) | Does the outcome occur only with this condition? | ≥ 0.90 | Counter-examples exist; it is not necessary |
| Relevance of necessity | Is the condition present nearly everywhere anyway? | Low value = trivial | A "necessary" condition that is ubiquitous explains nothing |
Worked example: what made the pilot sites work
A foundation funded a community health-worker programme in 24 districts. Twelve reached the retention target; twelve did not. The evaluation question was what distinguished them.
Five conditions were selected from prior work and site knowledge: supervisory visit frequency, payment reliability, whether workers were selected by the community, distance to the referral facility, and whether a district health officer had championed the programme. Each was calibrated with substantive anchors — payment reliability, for instance, at full membership where no payment had been more than two weeks late in a year, full non-membership where delays exceeded six weeks more than twice, crossover at a single month-long delay.
Necessity analysis found payment reliability necessary at consistency 0.94 — no district hit the retention target without it — and relevance was acceptable, since a third of districts lacked it.
The truth table with a frequency threshold of 1 and a consistency threshold of 0.83 produced two sufficient paths: reliable payment and community selection and frequent supervision; or reliable payment and an official champion and short referral distance. Solution consistency 0.88, solution coverage 0.83.
The reading was that reliable payment was the floor, and above it two different arrangements worked: strong community ownership with close supervision, or strong institutional backing where the health system was physically accessible. Distance mattered in the second path and not the first. One deviant case — reliable payment, community selection, frequent supervision, and no outcome — turned out on return to the field to have lost its entire cohort to a mining-sector recruitment drive, which was reported as a boundary condition rather than folded into the model.
Mistakes that discredit a QCA
- Calibrating from the sample distribution. Anchors set at the median or at quartiles encode nothing substantive and make the results an artefact of the sample.
- Too many conditions. More than seven with a moderate number of cases guarantees limited diversity, and the solution then rests mostly on assumptions about configurations never observed.
- Reporting the parsimonious solution without saying so. It is the most compact and the most assumption-laden. Silence about which solution is being shown is a serious reporting failure.
- Assuming symmetry. The analysis of the negated outcome must actually be run. It frequently produces different conditions.
- Reading paths as variables. A condition appearing in one path is not "a driver of the outcome" in general; it operates in that combination.
- Never returning to the cases. Without case-level interpretation, QCA is Boolean algebra over a spreadsheet.
- Hiding the thresholds. Frequency and consistency cut-offs change the solution. Undeclared, they invite the suspicion that they were tuned.
Limitations
QCA is sensitive to case selection and to calibration in a way that is uncomfortable for a formal method: small, defensible changes to anchors can alter the solution, which is why robustness checks across plausible alternative calibrations and thresholds are now expected rather than optional.
It does not handle measurement error gracefully — a single miscoded case can eliminate a necessity claim — and it says nothing about temporal order unless conditions are constructed to encode it.
It also does not establish mechanism. A solution path states that a combination is regularly associated with the outcome; why it produces it is a separate question, which is why Schneider and Rohlfing's Combining QCA and Process Tracing has become a standard follow-on design: QCA to find the configurations, process tracing within selected cases to establish what actually happened.
Where software helps
Minimisation itself needs dedicated software — the R packages QCA and SetMethods, or the fsQCA application — and no general qualitative tool substitutes for them.
The work a qualitative platform does sit upstream and downstream. Upstream: the condition scores usually have to be derived from case documents, interviews and site reports, which is a systematic document-analysis job across dozens of sources, and calibration is only defensible if each score traces back to the evidence behind it. Downstream: interpreting solution paths and investigating deviant cases means returning to that same material with a specific question. Evidano supports QCA as a named methodology on the case-material side; the set-theoretic analysis belongs in the dedicated packages.
Topics
- qualitative comparative analysis
- QCA
- truth table
- set-theoretic methods
- evaluation
- mixed methods
- configurational analysis
Other methods in outcome and impact assessment methods
Written guides are linked directly; the rest have a reference entry in the methodology directory.
Published research using these methods
Studies and evaluations where this family of method was applied with Evidano — the work, not the claim.
- Published research2023
AI-assisted outcome harvesting in a published evaluation
A monitoring and evaluation team analysed 12 years of reports with researchers and Evidano. Here is the workflow.
12 yrs of reports (2011 to 2022) analysed with Evidano
JET Education Services Annual Report 2023, "Working Towards Impact"
Keep reading
- Research MethodsOutcome Harvesting: a practical guideHow to run an outcome harvest: the six steps, what counts as an outcome, how substantiation works, and when harvesting beats a conventional results framework.
- Research MethodsOutcome Mapping: measuring the change you can actually influenceOutcome mapping tracks behaviour change in the partners a programme works through, not the impact it cannot control. The three stages, progress markers, and real limits.
- Research MethodsRipple Effects Mapping: capturing the effects nobody plannedA participatory group method for surfacing the secondary and unintended effects of a programme. How a REM session runs, how the map is coded, and what it cannot show.