Calculate required sample size for surveys and experiments.
Sample size determines how many respondents you need to get statistically valid survey results. A larger sample gives more accurate results but costs more time and money. The key factors are: population size, confidence level (usually 95%), margin of error (usually ±5%), and expected response distribution.
| Confidence Level | Z-Score | Margin of Error | Min Sample (large pop) |
|---|---|---|---|
| 90% | 1.645 | ±5% | 271 |
| 95% | 1.960 | ±5% | 385 |
| 95% | 1.960 | ±3% | 1,068 |
| 99% | 2.576 | ±5% | 664 |
| 99% | 2.576 | ±1% | 16,590 |
n = Z² × p(1−p) ÷ e², where Z = z-score for confidence level, p = expected proportion (use 0.5 for maximum sample), e = margin of error as decimal. For small populations, apply finite population correction: n_final = n ÷ (1 + (n−1)/N).