size of sample calculator

What is sample size, and why does it matter?

Sample size is the number of observations, people, or records you need to include in a study to get results you can trust. If your sample is too small, your results may swing wildly and fail to represent the true population. If your sample is too large, you may waste time, money, and effort collecting unnecessary data.

In survey design, market research, product testing, and academic work, the right sample size helps you strike the balance: enough data for reliable decisions, but not more data than needed.

How this sample size calculator works

This calculator uses the standard sample-size formula for estimating a population proportion:

n₀ = (Z² × p × (1 − p)) / e²

• Z = z-score based on confidence level (for example, 1.96 for 95%).
• p = expected proportion as a decimal (50% becomes 0.50).
• e = margin of error as a decimal (5% becomes 0.05).

If you provide a finite population size, the calculator applies finite population correction:

n = n₀ / (1 + (n₀ − 1) / N)

where N is your total population size.

How to choose each input

1) Confidence level

Confidence level tells you how certain you want to be that your interval includes the true population value. 95% is the most common choice in business and social science.

• 90%: smaller sample, lower certainty
• 95%: common default
• 99%: higher certainty, larger sample

2) Margin of error

Margin of error controls precision. Smaller margins require larger samples.

• ±5% is often acceptable for general surveys
• ±3% is preferred for higher-stakes decisions
• ±2% can require very large sample sizes

3) Estimated proportion

If you do not know the expected proportion, enter 50%. This is conservative and usually produces the largest required sample size.

4) Population size (optional)

If your population is small (for example, a class of 280 students or a customer list of 1,200 accounts), include population size so the result reflects that finite universe.

Quick reference (95% confidence, p = 50%, large population)

Margin of Error	Required Sample Size
±10%	97
±5%	385
±4%	601
±3%	1,068
±2%	2,401

Worked example

Suppose you want to estimate customer satisfaction in a region with a known population of 12,000 users. You choose:

• Confidence level: 95%
• Margin of error: 4%
• Estimated proportion: 50%
• Population: 12,000

Without correction, the required sample is about 601. With finite population correction, this falls slightly. The calculator handles this automatically and rounds up to the next whole person.

Practical tips for collecting a valid sample

• Randomize selection whenever possible to avoid selection bias.
• Plan for non-response. If you need 400 completes and expect a 40% response rate, invite around 1,000 people.
• Use clear screening criteria so only the target population is included.
• Avoid convenience-only samples for high-impact decisions.
• Track subgroup counts if you plan to compare age groups, regions, or segments.

Common mistakes to avoid

• Using an unrealistically low margin of error without budget/time support.
• Ignoring population size when the total group is small.
• Assuming a sample is representative just because it is large.
• Forgetting to account for drop-off and incomplete responses.

FAQ

Should I always use 50% for estimated proportion?

Use 50% when you are uncertain. If prior data suggests a different value (for example 20% conversion), using that estimate can reduce the sample size requirement.

Why is my sample size so large at 99% confidence?

Higher confidence requires a wider safety buffer, which increases required sample size substantially. This is expected and not an error.

Can this calculator be used for A/B tests?

This tool is for estimating a single proportion. A/B testing typically uses power analysis for two groups. For experiments, use a dedicated test power calculator.

Bottom line

A well-chosen sample size improves decision quality and keeps research efficient. Start with 95% confidence, 5% margin of error, and 50% estimated proportion if you do not have prior data. Then adjust according to your budget, risk tolerance, and analytical goals.