Skip to content
phdguide

Sample Size Calculator

Find the sample size your survey needs.

Recommended sample size
385
respondents (at 95% confidence, ±5% margin)

Uses Cochran's formula with finite-population correction when a population size is provided: n₀ = z²·p·(1−p) / e². Plan for non-response by inflating the target (e.g. ÷ expected response rate).

Ethical, compliant guidance: We provide academic support, mentoring, analysis, editing and structuring — not authorship. Your work stays compliant with university policies.

About this tool

‘How many respondents do I need?’ is the most common — and most consequential — methodology question. Too few and your tests lack power; too many and you waste months of fieldwork.

This calculator gives you a defensible sample size from your population, confidence level and margin of error, following the standard Cochran/Krejcie–Morgan logic committees recognise.

How to use it

  1. 1Choose the confidence level — 95% is the research standard.
  2. 2Set the margin of error (5% is the common choice) and leave response distribution at 50% for the most conservative estimate.
  3. 3Enter your population size if it's known and finite; leave it blank for large or unknown populations.
  4. 4Read the recommended sample size and note all four parameters — you'll report them in your methodology chapter.

Frequently asked

Is 100 respondents enough for my study?+

It depends on your population and analysis. Simple descriptive surveys may manage; regression and SEM need power-based numbers — often 200–400 for typical models. Justify with a calculation, not a convention.

What sample size does SEM need?+

CB-SEM (AMOS) usually wants 200+ with reasonably normal data; PLS-SEM (SmartPLS) tolerates less but still requires power-based justification — the '10-times rule' is only a floor.

Not sure where to start?

Book a free 15-minute consult. We'll map your next three steps — no obligation.

Ethical, compliant guidance

We provide academic support, mentoring, analysis, editing and structuring — not authorship. Your work stays compliant with university policies.