A marketing manager’s question is not only “If we raise quality, will customers increase?” More often it is “Which things should we combine to reach our target?”
Qualitative comparative analysis (QCA) was built to answer that kind of question. Statistical methods such as regression are good at showing how much one factor matters on its own, with the other conditions held constant. QCA looks at the data from a different angle.
Why QCA is attractive for decisions
1. Actions are taken in combination, not alone
Store renovation, better service, and a price review are usually carried out together. QCA takes the combination (the recipe) as its unit, for example “high quality and high service”, and asks what happens to the outcome.
2. There may be more than one road to success
The same outcome can be reached in several ways (equifinality). Under the conditions and analysis settings you choose, QCA can identify several routes to the same outcome, so a manager can choose the one that fits the budget and the company’s strengths.
3. The reasons for success are not always the mirror image of the reasons for failure
The conditions that produce loyal customers and the conditions that produce customers who leave may need separate analyses (asymmetry). QCA analyzes the high case and the low case separately.
4. It can be used with small to medium numbers of cases
With stores, brands, campaigns, or corporate customers, you may have only dozens or a few hundred cases. QCA lets you analyze them while keeping an eye on what is inside each case, It can be used for small to medium comparative case sets. When there are few cases relative to the number of conditions, however, many logical remainders arise (combinations of conditions that no case shows).
5. The result reads as an “if … then” rule
A result comes out as a short rule, such as “high quality and high service”. It is easy to explain in a meeting, and it tells people on the front line what to do.
6. It separates “necessary” from “sufficient”
You can treat “without this, success does not happen” (a necessary condition) differently from “with this, success happens” (a sufficient condition). That lets you think separately about the actions you cannot skip and the actions that win.
How it differs from regression (roughly)
| Point of view | Regression | QCA |
|---|---|---|
| The question | How much does one factor matter on its own, with the others held constant? | Which combinations are enough for the outcome? |
| Unit of analysis | Each variable | Combinations of conditions |
| Multiple roads | Interactions and nonlinearity must be built into the model | Can be identified as combinations |
| How results read | Coefficients and significance | Rules such as “A and B” |
| Typical scale | Large samples | Dozens to a few hundred cases |
Neither is better than the other. You choose according to the question.
What QCA cannot do (and what is also true of other methods)
- QCA is not the only method that cannot prove causation. Analyses of observational data, regression included, cannot fully establish causation. A sufficiency solution from QCA does not, by itself, guarantee the causal effect of carrying out an intervention. A causal reading needs support from research design, temporal order, theory, and case knowledge. The results are best used as material for thinking about candidate interventions and hypotheses to test.
- It needs theory and knowledge of the cases. The analyst decides which conditions to include and where “high” begins.
The pitfall is the threshold
That last point matters more than it first appears. In QCA, the analyst decides the threshold that separates “high” from “low”. But a decision target is exactly such a threshold: do you aim for “reasonably satisfied” or for “loved”? When the threshold changes, the answer can change too.
Threshold-sweep QCA (ThSQCA) addresses this problem. The next post continues from here.
References
- Ragin, C. C. (2008). Redesigning Social Inquiry: Fuzzy Sets and Beyond. University of Chicago Press.
- Fiss, P. C. (2011). Building better causal theories: A fuzzy set approach to typologies in organization research. Academy of Management Journal, 54(2), 393-420.
- Woodside, A. G. (2013). Moving beyond multiple regression analysis to algorithms: Calling for adoption of a paradigm shift from symmetric to asymmetric thinking in data analysis and crafting theory. Journal of Business Research, 66(4), 463-472.