If you have searched for survey sample sizes, you have probably met the number 385. It appears so often that many students treat it as a rule. It is not a rule — it is the answer to one specific calculation.
Where 385 comes from
For a large population, the standard formula for estimating a proportion is:
n = z² × p(1 − p) ÷ e²
With 95% confidence (z = 1.96), the most cautious proportion (p = 0.5) and a 5% margin of error (e = 0.05):
1.96² × 0.25 ÷ 0.0025 = 384.16, rounded up to 385.
Change any input and the answer changes. A 3% margin of error needs about 1,068 responses; 90% confidence with a 5% margin needs about 271.
When your population is small
If you are surveying a defined group — say 400 employees of one company — apply the finite population correction:
n_adjusted = n ÷ (1 + (n − 1) ÷ N)
For N = 400, the 385 above drops to about 197. Many student projects over-survey because they skip this step.
Three things the formula does not fix
- Response rate. If you expect 40% of invitations to be answered, invite 2.5 times your target.
- Sub-group analysis. Comparing men and women, or three departments? Each group needs enough responses on its own.
- Your analysis method. Regression, SEM and factor analysis have their own sample size guidance — PLS-SEM, for example, is often planned with power analysis rather than this formula.
Do it in seconds
Our free Sample Size Calculator runs the formula, applies the finite population correction and shows the working, so you can copy it straight into your methodology chapter.
Not sure which approach fits your design? Bring your research questions to a free consultation.
Reach students and researchers with a clearly labelled placement.
Questions & discussion