Statistics & Research Tools
KAP Survey Sample Size Calculator
Work out how many people to interview for a Knowledge, Attitudes and Practices survey. Adjust for cluster sampling, finite populations and non-response in seconds.
How the KAP sample size is calculated
Most KAP and household surveys estimate a proportion, such as the share of mothers who know the danger signs of pregnancy. The standard formula for a proportion is below. It tells you the number of respondents needed to estimate that share within a chosen margin of error and confidence level.
Z is the value for your confidence level (1.96 for 95%), p is the expected proportion, and e is the margin of error. When p is unknown, use 50%, because it gives the largest and therefore safest sample.
Design effect, finite population and non-response
Three adjustments turn the basic number into a practical field target:
- Finite population correction: if your population is small, for example 2,000 households, you need fewer respondents than the base formula suggests.
- Design effect (deff): cluster sampling is cheaper but less precise than simple random sampling. A design effect of 1.5 to 2 is common in village or school based surveys, and it multiplies the sample.
- Non-response: some selected households will refuse or be absent. Divide the sample by one minus the expected non-response rate to keep your final numbers.
Worked example
A district health survey wants to estimate immunisation coverage with 95% confidence and a 5% margin of error, expecting about 50% coverage. The base sample is 385. With a design effect of 1.5 for cluster sampling and 10% non-response, the target becomes 385 × 1.5 / 0.90, or about 642 respondents.
When to use this calculator
Use it when planning baseline and endline KAP studies, household surveys, needs assessments and programme evaluations for NGOs and development organisations. If you also need to compare two groups, such as treatment and control areas, you will need a power-based calculation that this tool does not cover. Our analysts can help with that.
Frequently asked questions
What margin of error should I choose for a KAP survey?
Five percent is the most common choice. Use 3% when you need very precise estimates and the budget allows a larger sample, or up to 10% for quick rapid assessments.
Why do many KAP surveys use a proportion of 50%?
Because the formula p(1 − p) is largest at 50%. If you have no earlier estimate, 50% guarantees the sample is large enough for any true value.
How do I choose the design effect?
Use 1 for simple random sampling. For two-stage cluster samples, 1.5 to 2 is a typical planning value. If a previous survey in the same setting reported a design effect, use that figure instead.
Does this tool store my inputs?
No. The calculation runs in your browser and nothing you enter is sent to our servers.