Calculate the right sample size for your research. Supports survey, experimental, correlational, and qualitative designs. Get a ready-to-use justification paragraph for your methodology chapter.
Enter 0 if population is unknown or very large
Sample size determination is one of the most critical decisions in research methodology. Too small a sample leads to insufficient statistical power (Type II error), while too large a sample wastes resources without meaningful gains in precision. The correct approach depends entirely on your study design.
For survey research, Cochran's (1977) formula is the gold standard. It accounts for your desired confidence level, acceptable margin of error, and population variability. When your population is finite (known), a correction factor reduces the required sample size.
For experimental designs, Cohen's (1992) power analysis determines sample size based on the effect size you expect to detect. This ensures your experiment has sufficient power to find real differences between groups.
For qualitative research, sample size is not determined by formulas but by the principle of theoretical saturation — collecting data until no new themes emerge. Established guidelines exist for each qualitative methodology (phenomenology, grounded theory, case study, etc.).
Quick reference for survey research at 95% confidence, ±5% margin of error:
| Population (N) | Sample (n) | Population (N) | Sample (n) |
|---|---|---|---|
| 100 | 80 | 1,000 | 278 |
| 200 | 132 | 2,000 | 322 |
| 300 | 169 | 5,000 | 357 |
| 500 | 217 | 10,000 | 370 |
| 700 | 248 | 100,000 | 384 |
Source: Krejcie, R.V. & Morgan, D.W. (1970). Determining sample size for research activities. Educational and Psychological Measurement, 30, 607-610.
For surveys with a defined population → Cochran's formula. For experiments comparing groups → Cohen's power analysis. For correlational studies → Fisher's Z method. For qualitative → saturation guidelines. Our calculator automatically uses the correct formula based on your study type.
Enter 0 or leave it blank. The calculator will use the infinite population formula (no correction). For most large populations (>10,000), the correction makes minimal difference anyway.
There's no universal minimum, but general rules: surveys need at least 30 for basic statistics. Experiments need at least 15-20 per group. Correlational studies need at least 30. Qualitative studies vary by methodology (5-60).
Yes. Always add 10-20% extra to account for non-response, incomplete data, or dropouts. For longitudinal studies, add 20-30% for attrition.
Cite the underlying formula/reference (Cochran 1977, Cohen 1992, or Krejcie & Morgan 1970) — not the calculator itself. We provide the correct citation in the results.
Effect size is the magnitude of the difference you expect to find. Small (d=0.2): subtle, hard to notice. Medium (d=0.5): noticeable, typical in social sciences. Large (d=0.8): obvious, strong effect. Check prior studies in your field, or use medium if unknown.