Build random samples from ranges or custom lists. Compare required size, confidence, and expected variability. Download polished outputs for audits, classrooms, surveys, and research.
Use a numeric range or a custom list. The page calculates recommended sample size and also draws the requested random sample.
This example shows how the calculator can be used for a simple survey frame numbered from 1 to 500.
| Scenario | Population | Confidence | Margin of Error | Estimated Proportion | Finite Corrected Size | Adjusted Target |
|---|---|---|---|---|---|---|
| Customer satisfaction survey | 500 | 95% | 5% | 50% | 217.49 | 218 |
| Classroom audit checklist | 120 | 90% | 7% | 50% | 67.44 | 67 |
| Inventory quality review | 2,000 | 95% | 4% | 40% | 414.53 | 415 |
The calculator first estimates an infinite-population sample size with
n₀ = (Z² × p × (1 − p)) / e².
Here, Z is the confidence z-score, p is the estimated proportion, and e is the margin of error.
It then applies finite population correction using
n = n₀ / (1 + ((n₀ − 1) / N)),
where N is the frame size. This reduces the required sample when the population is not very large.
Finally, it inflates the target for design effect and expected nonresponse using
Adjusted Target = ceil((n × DEFF) / (1 − r)).
In that expression, DEFF is design effect and r is the nonresponse rate as a decimal.
The random sample itself is drawn from your frame. Without replacement, each unit can appear once. With replacement, repeated draws are allowed.
Choose Numeric range when your population can be identified by numbers, such as records 1 to 1000.
Choose Custom list when you already have names, codes, or item labels. Enter one per line or separate them with commas.
Enter the requested sample size to draw an actual sample. Then set confidence, margin of error, estimated proportion, design effect, and expected nonresponse.
Use an optional random seed when you need repeatable results for classroom demonstrations, audits, or internal reviews.
Click Calculate Random Sample. The page shows the recommended sample size, summary statistics, a sample table, and a chart above the form.
Use the Download CSV and Download PDF buttons to export the summary and sampled records for sharing or documentation.
It estimates an appropriate sample size and also generates a random sample from a numeric range or a custom population list.
A 50% proportion creates the most conservative sample size for many proportion studies, so it is commonly used when no better estimate exists.
Finite population correction lowers the required sample when the total population is limited, preventing unnecessary oversampling in smaller frames.
Use replacement when repeated selections are acceptable or needed in simulation exercises. Survey samples usually work without replacement.
Design effect increases the target sample when the sampling plan is less efficient than simple random sampling, such as clustered designs.
A seed makes the random draw repeatable. That helps with validation, training examples, audit trails, and collaborative reviews.
Yes. The custom list mode accepts names, codes, or labels. The graph will switch to category frequencies for nonnumeric samples.
No. It is excellent for planning and simple random sampling, but complex weighting, stratification, and multistage designs need deeper analysis.
Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.