How to Use the Slovin's Formula Calculator

Enter a population size and margin of error to compute the required sample size, review the sampling fraction and comparison tables, and interpret the result with the correct assumptions.

  1. 1. Choose what to solve: Pick Sample size if you have a population and margin of error and want the required sample. Pick Population if you have a sample and margin of error and want the implied population. Pick Margin of error if you have a population and sample and want the implied error parameter. Each mode uses the same Slovin formula in a different direction.
  2. 2. Enter the population size: Enter the total population N as a whole number greater than zero. Decimals are not accepted because a population is a count of people or items. For a study of 1,000 employees, enter 1000.
  3. 3. Enter the margin of error: Enter the margin of error as a percentage — 5 for 5%. The calculator converts it to a decimal (0.05) internally before applying the formula. Common choices are 1%, 3%, 5%, and 10%. Smaller error tolerances require a larger sample.
  4. 4. Review the exact and rounded sample size: The results panel shows the exact mathematical result and the required sample rounded up to the next whole respondent. A fractional sample is meaningless — always round up. The hero figure shows the rounded sample size directly.
  5. 5. Check sampling fraction and response rate: The sampling fraction shows what percentage of the population the sample represents. If you plan to send invitations and expect nonresponse, open the response-rate planning panel and enter the response rate. The calculator returns the invitations needed to end up with the required number of completed responses.
  6. 6. Compare across margins and populations: The margin-of-error table shows the required sample at every level from 1% to 10% for the current population. The population table shows the required sample across a range of population sizes at the current margin. Use these to understand how sensitive the result is to each input.
  7. 7. Check the assumptions before reporting: Slovin's formula does not model confidence level, proportion, power, or effect size. If your study needs any of those, use a dedicated sample-size method. Before reporting the result, confirm that a simple finite-population estimate is appropriate for your research design.

What Slovin's Formula Does

Slovin's formula estimates how many people you need to survey when you already know the size of the group you are studying. It takes two inputs — population size and a chosen margin of error — and returns a required sample size. It is widely used in undergraduate and graduate research, market surveys, and quick planning studies because it is easy to explain and defend.

What it does not do matters just as much. Slovin's formula does not take a confidence level as input, does not use an expected proportion, does not model statistical power, and does not account for complex sampling designs. It is a screening-level estimate, not a full study design.

The Formula

n = N / (1 + N × e²)

  • N — total population size, a whole number greater than zero.
  • e — margin of error as a decimal. A 5% margin is 0.05, not 5.
  • n — the required sample size. Round up to the next whole respondent.

As N grows large, n approaches a ceiling determined by e alone. At a 5% margin of error, the required sample caps out around 400 no matter how large the population becomes.

Worked Example

Population N = 1,000 and margin of error e = 5% (0.05):

  1. e² = 0.0025
  2. N × e² = 1,000 × 0.0025 = 2.5
  3. 1 + N × e² = 3.5
  4. n = 1,000 ÷ 3.5 = 285.714
  5. Required sample = 286 respondents

That is the number of completed responses needed, not the number of invitations to send. At an 80% response rate, you would invite roughly 358 people to end up with 286 completed responses.

Reverse Slovin: Population and Margin of Error

The formula rearranges in two directions, and both are supported by this calculator.

Solve for population: N = n / (1 − n × e²). This is only valid when 1 − n × e² is greater than zero. If the denominator is zero or negative, the sample and margin of error you entered cannot come from a valid Slovin calculation.

Solve for margin of error: e = √((N − n) / (n × N)). This is the error parameter Slovin's formula is producing for a given population and sample. It is the algebraically implied parameter, not a confidence interval computed from survey data.

Sampling Fraction and Response-Rate Planning

The sampling fraction is n / N, usually reported as a percentage. A sample of 286 from a population of 1,000 gives a sampling fraction of 28.6%. The unsampled population is N − n.

Response-rate planning is separate from Slovin's formula. The formula returns completed responses needed. If your expected response rate is 80%, invitations required = 286 ÷ 0.80 = 357.5, which rounds up to 358. This step does not change the statistical sample — it only estimates how many people you need to contact to end up with that many completed responses.

Comparison Tables

The results panel shows two comparison tables. The margin-of-error table gives the required sample at every level from 1% to 10% for the current population, so you can see how sensitive the result is to your chosen tolerance. The population table gives the required sample across a range of population sizes at the current margin, so you can see how little the answer changes once the population is large. Both tables use the same engine as the main calculation, so they always agree with it.

Assumptions and Limitations

Slovin's formula is often presented as a universal sample-size method. It is not. Five things it does not model:

  • Confidence level — you cannot set one explicitly. Under the common assumptions of proportion 0.5 and 95% confidence, the standard sample-size formula simplifies to roughly 1 / e² for large populations, and Slovin's formula is a finite-population variant of that. The 95% figure is a side effect, not a design choice.
  • Expected proportion — the formula assumes the worst-case proportion of 0.5, which maximizes the required sample. If you have a reliable prior estimate, a proportion-based method will give a smaller and still valid sample.
  • Statistical power — Slovin's formula does not model power. If your study tests a hypothesis or compares groups, a power-based method is required.
  • Effect size — no effect size is specified, so the result is not tied to what the study is trying to detect.
  • Complex sampling designs — cluster, stratified, and multi-stage designs all require adjustments Slovin's formula does not include.

Slovin vs. Cochran and Yamane

Yamane's formula (1967) is usually written the same way — n = N / (1 + Ne²) — and gives the same number for the same inputs. The difference is that Yamane's is derived assuming 95% confidence and proportion 0.5, making its assumptions explicit, while Slovin's is often presented without derivation.

Cochran's formula takes a confidence level, a population proportion, and a margin of error as inputs, and does not include a finite-population correction by default. It is more flexible when you have a reliable prior estimate of the proportion. Each method answers a different question — pick the one that matches your design, not the one that gives the smallest number.

Common Mistakes

  • Entering 5 instead of 0.05 for the margin of error.
  • Forgetting to square e — the formula uses e².
  • Rounding intermediate values instead of only the final result.
  • Treating Slovin's formula as if it takes a confidence level as input.
  • Confusing N (population) with n (sample).
  • Reporting the required completed responses as if it were the number of invitations to send.
  • Using Slovin's formula for a study that needs power, effect size, or a specified proportion.

Reporting the Result

A complete methodology sentence usually states the population, the margin of error, the formula, and the resulting sample size. For example: Using Slovin's formula with a population of 1,000 and a 5% margin of error, the calculated sample size was 285.714, resulting in a required sample of 286 respondents. Add the sampling fraction if useful, and note that Slovin's formula does not model power or proportion. Treating the result as a floor for a well-designed study — not as a ceiling — keeps the reporting honest.

Frequently Asked Questions

Q: Why is the margin of error squared in Slovin's formula?

The squared margin of error reflects the fact that sample-size requirements grow with the inverse square of the error tolerance. Reducing your margin from 5% to 2.5% does not just double the required sample — it multiplies it by roughly four. The e² term produces that quadratic sensitivity.

Q: Should Slovin's sample size be rounded up?

Yes. A fractional sample is meaningless — you cannot survey part of a person. Always round up to the next whole respondent. The calculator shows both the exact result and the rounded result so you can see the difference.

Q: Can Slovin's formula calculate population size?

Yes, as a reverse calculation. Given a sample size n and a margin of error e, the implied population is N = n / (1 − n × e²). This is only valid when the denominator is greater than zero — if n × e² is at least 1, the reverse population is undefined because the sample and margin cannot come from a valid Slovin calculation.

Q: Can Slovin's formula calculate margin of error?

Yes, as a reverse calculation. Given a population N and a sample n, the implied margin of error is e = √((N − n) / (n × N)). This is the parameter Slovin's formula is producing for those two numbers. It is not a confidence interval computed from survey data.

Q: Can Slovin's formula be used for thesis research?

It is widely used in undergraduate and graduate theses because it is easy to explain and produces a defensible number for a methodology section. It is appropriate when your study is a simple finite-population survey and a margin of error is the only precision specification you need. If your study tests a hypothesis or compares groups, a power-based method is more appropriate.

Q: Does Slovin's formula give a 95% confidence level?

Not directly. Slovin's formula does not take a confidence level as input. Under the common assumptions of population proportion 0.5 and 95% confidence, the standard sample-size formula simplifies to roughly 1 / e² for large populations, and Slovin's formula is a finite-population variant of that. But the 95% is a side effect of the standard assumptions, not a design choice you can set.

Q: When should I not use Slovin's formula?

When your study needs a specific statistical power, when you must specify a confidence level explicitly, when you have a reliable prior estimate of the proportion you are measuring, when you are running a controlled experiment where effect size matters, or when your sampling design involves clustering or stratification. Those cases call for a dedicated sample-size method.

Q: What is the difference between Slovin's formula and Cochran's formula?

Slovin's needs a population size and a margin of error and assumes the worst-case proportion of 0.5. Cochran's takes a confidence level, a population proportion, and a margin of error as inputs, and does not include a finite-population correction by default. Cochran's is more flexible when you have a reliable prior estimate of the proportion; Slovin's is simpler and does not require one.

Q: Is Slovin's formula the same as Yamane's formula?

For the same inputs, they give the same number. Yamane's formula is usually written as n = N / (1 + N × e²), identical to Slovin's. The difference is that Yamane's is derived assuming 95% confidence and population proportion 0.5, making its assumptions explicit, while Slovin's is often presented without derivation.

Q: Does the calculator handle non-whole-number populations?

No. Slovin's formula requires population size to be a whole number, because a population is a count of people or items. If you enter a decimal, the calculator returns an error asking you to round to the nearest whole number. Sample size can be fractional internally and is rounded up at the end.

Q: Is response-rate planning part of Slovin's formula?

No. Response-rate planning is a separate step. The formula returns the number of completed responses needed. Dividing that by the expected response rate gives the number of invitations you need to send to end up with that many completions. The calculator keeps the two quantities distinct so you never confuse the statistical sample with the recruitment target.

Loaded shared calculation