Glossary
Short definitions of the terms used on this blog.
Short definitions, in plain language. Each is a first step, not a full treatment.
- QCA (qualitative comparative analysis)
- A method that looks for combinations of conditions that go with an outcome, using set relations rather than averages.
- Condition and outcome
- A condition is a possible cause (for example, high quality). The outcome is what you want to explain (for example, loyalty).
- Crisp set and fuzzy set
- In a crisp set, each case is in or out (0 or 1). In a fuzzy set, each case has a degree of membership between 0 and 1.
- Calibration
- Turning raw scores into set membership scores, using chosen thresholds for “fully in”, “fully out”, and the crossover point.
- Truth table
- A table with one row for each combination of conditions, showing how many cases fall in that row and how consistently they show the outcome.
- Necessity and sufficiency
- A condition is necessary if the outcome is not found without it. A condition (or combination) is sufficient if the outcome is found whenever it is present.
- Consistency (inclS) and coverage (covS)
- Consistency shows how well the cases follow a sufficiency relation. Coverage shows how much of the outcome the relation accounts for.
- PRI
- A measure that helps detect a condition combination that looks sufficient for both the outcome and its absence.
- Configuration (recipe)
- A combination of conditions that together are linked to the outcome.
- Equifinality
- More than one configuration can lead to the same outcome.
- Asymmetry
- The causes of the outcome’s presence need not be the mirror image of the causes of its absence.
- Minimization and solution
- The step that simplifies the truth table into shorter expressions. The result is called the solution.
- Threshold sweep
- Rerunning the same analysis across a range of thresholds, under identical settings, and recording how the solutions change.