How much sample size is enough for research?
It depends on your design, population, expected effect size and analysis. As rough guides: survey studies often target 200–400 respondents, PLS-SEM needs about 10 cases per path pointing at a construct, and covariance-based SEM usually wants 200+. A power analysis gives the precise number.
What actually determines sample size
Four things drive it: the size and variability of your population, the confidence level and margin of error you accept, the effect size you expect to detect, and the statistical power you want (commonly 0.80). Your analysis method then sets minimums.
Useful rules of thumb
For descriptive surveys, Cochran's formula gives a starting number (often ~385 for a large population at 95% confidence, 5% margin). For PLS-SEM, a common guide is 10 times the largest number of paths pointing at any one construct. For covariance-based SEM (AMOS), aim for 200+ and ideally 5–10 cases per estimated parameter. For qualitative research, you sample to thematic saturation rather than a fixed number.
Calculate it properly
Rules of thumb are a starting point; a power analysis (for example in G*Power) gives the defensible number for your specific test and effect size. Our free Sample Size Calculator gives a quick survey estimate you can refine with a mentor.
Key takeaways
- There is no single 'right' number — it depends on design and analysis.
- Surveys often target 200–400; SEM needs careful minimums.
- Use a power analysis for a defensible figure.
- Qualitative studies sample to saturation, not a fixed count.
People also ask
What sample size do I need for PLS-SEM?+
A common guideline is at least 10 times the largest number of structural paths directed at any single construct, but confirm with a power analysis for your effect size.
What is a good sample size for a PhD survey?+
Many PhD survey studies target 200–400 valid responses, but the right number depends on your population, expected effect size and planned analysis.
Ethical, compliant guidance: We provide academic support, mentoring, analysis, editing and structuring — not authorship. Your work stays compliant with university policies.
Not sure where to start?
Book a free 15-minute consult. We'll map your next three steps — no obligation.