All free tools
Free · no signup

Sample size table (Krejcie & Morgan)

The table every survey methods course cites: how many respondents you need for a given population, at 95% confidence and a ±5% margin of error. Need other settings? Use the sample size calculator.

Let the AI do this — sign up freeUpload your data, get the write-up back.
Required sample size by population size (95% confidence, ±5%)
Population (N)Sample (n)Population (N)Sample (n)Population (N)Sample (n)
10102301441,400302
15142401481,500306
20192501521,600310
25242601551,700313
30282701591,800317
35322801621,900320
40362901652,000322
45403001692,200327
50443201752,400331
55483401812,600335
60523601862,800338
65563801913,000341
70594001963,500346
75634202014,000351
80664402054,500354
85704602105,000357
90734802146,000361
95765002177,000364
100805502268,000367
110866002349,000368
1209265024210,000370
1309770024815,000375
14010375025420,000377
15010880026030,000379
16011385026540,000381
17011890026950,000381
18012395027475,000382
1901271,000278100,000383
2001321,100285250,000384
2101361,2002911,000,000+384
2201401,300297

Computed from the Krejcie & Morgan (1970) formula: χ² = 3.841, P = .50, d = .05. Note how n saturates: beyond N ≈ 100,000 the answer is effectively 384 no matter how large the population.

How to use the table

Find the population closest to yours (round UP for safety) and read off the required sample. A school of 320 students needs n = 175 completed responses; a company of 5,000 employees needs 357; any large national population needs 384. These are completed questionnaires — divide by your expected response rate to get the number of invitations.

The table hard-codes the most common survey scenario: estimating a proportion at 95% confidence with a ±5% margin of error and the conservative P = .50. If you need ±3%, 99% confidence, or you expect the true proportion to be far from 50%, the numbers change substantially — use the interactive sample size calculator on this site instead of scaling the table.

Krejcie & Morgan’s enduring lesson is the saturation effect: sample size is driven by desired precision, not by population size. Once the population exceeds ~100,000, "how big is the country?" stops mattering entirely.

Frequently asked questions

Why does the table stop growing around 384?
The margin of error depends on the sample’s absolute size, not its share of the population (for large populations). At 95%/±5% with P = .50, the infinite-population answer is 1.96² × .25 / .05² = 384.16 → 385; the finite population correction only reduces it when N is small.
Is this table valid for experiments or group comparisons?
No. It sizes a descriptive estimate of a single proportion. Detecting a difference between groups is a power problem, driven by the expected effect size — use the two-group sample size calculator for that.
The table says my population of 60 needs 52 respondents — really?
Yes — for small populations the finite population correction helps less than intuition suggests, and near-census coverage is needed for ±5% precision. In practice, for a population under ~100 it is usually easier to just survey everyone.

Related free tools

Sample size calculatorSample size for two groupsEffect size (Cohen's d)Descriptive statisticsOne-sample t-testIndependent t-testPaired t-testz-table (standard normal)t-table (critical values)Chi-square table

Have a question the table can't answer?

The AI inside Kahubi works with your actual data and library — it runs analyses, checks assumptions, and writes the results up in your style.

Kahubi AI
Hi! Ask me anything you’d normally google mid-study — study design, statistics, literature, writing.

Free accounts include AI chat, literature search, and every tool on this site.

Stop copying numbers between tools

Inside Kahubi, the AI agent runs this analysis directly on your uploaded dataset — then writes the results section in your own writing style, with the statistics reported correctly.

Try Kahubi freeMore free tools