PhDGujarat
Statistics laboratory · free forever

Sample size calculator

Cochran's formula with finite population correction. It also gives you the number of people you must approach — the figure most calculators leave out and most scholars discover too late.

Your study

Change anything and the answer updates as you type.

Required sample

373

completed responses needed for your stated precision

Approaches to make

622

at your expected response rate — this is the number people forget

How to report it

What this assumes

Every sample size figure rests on assumptions, and an examiner may ask about them.

  • p = .50. Maximum variability, the most conservative assumption. If you have a prior estimate of the proportion, the required sample falls.
  • Simple random sampling. Cluster or stratified designs need a design effect adjustment, usually raising the figure.
  • Finite population correction applied. This is why a population of 1,000 needs fewer responses than one of 100,000.
  • Not for SEM. Structural equation modelling uses different rules — commonly 10 cases per estimated parameter, or a minimum in the 200 range.

Not sure this is the right calculation for your design? Tell us your design and sampling method and we will tell you which rule applies — at no charge.

Ask about your design