Part 2 · Why not just use regression?
Module 4 · Why we need something different
"Can't I just run a regression?"
- Explain, in plain language, the question regression is built to answer.
- Explain why that question misses how immunization programs actually work.
- Say when CNA is the better tool, and when it is not.
It is a fair question, and one every analyst should ask. Logistic regression is an excellent, well-tested tool, and if your question is "which single factor moves the outcome most, holding the others constant," you should reach for it, not CNA. The two methods are not rivals so much as instruments tuned to different questions.
The catch is that regression, by design, estimates the average independent contribution of each factor. It answers "how much does supervision matter on its own?" But immunization coverage rarely fails because of one factor acting on its own. It fails when a broken cold chain meets missing supervision in the same place, at the same time. That is a question about combinations, and it is a different question than the one regression is built to answer.
| The question you're asking | Regression | CNA |
|---|---|---|
| Which single factor matters most? | ✓ | – |
| Which combinations of conditions matter? | – | ✓ |
| Are there multiple pathways to the same result? | Limited | ✓ |
| Do conditions form causal chains? | Limited | ✓ |
Different tools for different questions. CNA does not replace regression; it answers a question regression was never designed to ask.
Regression ranks ingredients. CNA reads recipes.
Regression is brilliant at telling you which ingredient, on average, has the biggest effect. CNA is built to tell you which combinations of ingredients produce the dish, including the fact that there may be more than one recipe that works. When a program fails because several things went wrong together, you need a method that thinks in recipes.
This is also why CNA does not report coefficients, p-values, or "the effect of supervision." Those numbers answer the ranking question. CNA answers the recipe question, and it speaks in a different language: combinations, pathways, and the conditions an outcome depends on. Keeping the two questions straight now will save you a great deal of confusion later.
Think about a district you know that struggles with coverage.
- If you had to name the one factor behind its struggle, could you? Or does it really take two or three things together to explain it?
- Might a neighbouring district be struggling for entirely different reasons, a different recipe for the same poor result?