Survey sample size calculator

How the sample size is worked out

The calculator uses the standard formula for a proportion: n = z² × p(1−p) ÷ e², where z comes from the confidence level, p is the expected answer split and e is the margin of error. When you give a population size, it applies the finite population correction: n′ = n ÷ (1 + (n−1)/N).

Two things that trip up fieldwork plans:

• Sub-groups multiply the cost. Precision applies to each group you report separately, so four regions at ±5 points needs four times the completes, not one bigger sample.
• Margin of error scales with the square root. Halving it, from ±4 to ±2 points, needs four times the completes. That's why a small tightening of precision blows a fieldwork budget.

Why the invitation number matters more than the sample size

Sample size is arithmetic. Getting those completes is the hard part. At a 4% response rate, 1,068 completes needs about 26,700 invitations to land, be read and be acted on. If a chunk of those invitations never reaches an inbox, you don't find out from the survey platform. You find out from a field close date you can't hit and a top-up sample order you didn't budget for.

Sterling Clicks works on that gap only: getting more of the people already on your lists to start and complete your surveys, without changing your questionnaire, your platform or your incentive.