Saved

Press the star next to a tool's name to save it.

All my tools

Critical Value Calculator

Use this critical value calculator to find the z, t, chi-square or F critical value for any significance level, degrees of freedom and one- or two-tailed test.

Restored your last values ·

Choose the distribution, the kind of test and the significance level. For t and chi-square, enter the degrees of freedom, and for F both of them. The calculator gives the critical value or values that bound the rejection region.

Everything is worked out in your browser. Nothing you type is sent anywhere.

–

–

Choose the distribution and the test, enter the significance level, then press Calculate.

See full result

A critical value calculator finds the cut-off that a test statistic must pass for you to reject the null hypothesis. Choose the normal, t, chi-square or F distribution, enter the significance level and the degrees of freedom, and pick a two-tailed, right-tailed or left-tailed test. You get the critical value or values and the matching confidence level. It replaces looking up a table, and works for any significance level. It runs in your browser.

How to use the critical value calculator

  1. Choose the distribution and the test.
  2. Enter the significance level, and the degrees of freedom for t, chi-square and F.
  3. Press Calculate.

What a critical value is

A critical value marks the edge of the rejection region. If the test statistic falls beyond it, the result is significant at your level. The area of the rejection region equals the significance level. For a right-tailed test, all of it is in the right tail. For a left-tailed test, it is in the left tail, and for a two-tailed test it is split between both tails, half in each. The calculator shows this split.

The normal distribution

The Wikipedia article on the Z-test says that the test has one critical value for each significance level, for example 1.96 for a 5 percent two-tailed test, and that the same value, written z with the subscript α ÷ 2, is used for a confidence interval, with 1.96 for a 95 percent confidence level.[1] The calculator gives ±1.96 for this case. For a one-tailed test at 5 percent the value is 1.6449, and for 1 percent two-tailed it is ±2.5758. For the left-tailed test, the sign changes.

The t distribution

The t distribution has fatter tails than the normal one for small samples, so its critical values are larger, and they shrink toward the normal values as the degrees of freedom grow. With 10 degrees of freedom the two-tailed 5 percent value is ±2.2281, with 30 it is ±2.0423, and with 1 degree of freedom the right-tailed 5 percent value is 6.3138. The calculator accepts any degrees of freedom above zero, also those that are not whole numbers, as in Welch's test.

The chi-square distribution

The chi-square distribution is not symmetric and starts at zero, so its critical values are positive. For a right-tailed test at 5 percent, the critical value is 3.8415 with 1 degree of freedom, 5.9915 with 2, 11.07 with 5 and 18.307 with 10. At 1 percent with 1 degree of freedom it is 6.6349. For a two-tailed test, the calculator gives the lower and the upper values: with 10 degrees of freedom, 3.247 and 20.483.

The F distribution

The F distribution has two degrees of freedom, one for the numerator and one for the denominator. The one-way ANOVA article gives the critical value for 2 and 15 degrees of freedom at the 5 percent level as 3.68.[2] The calculator gives 3.6823 for a right-tailed test. Because the F statistic is a ratio of variances, ANOVA tests use the right tail only, and a two-tailed test is used mainly to compare two variances.

Using the result

Work out your test statistic, then compare it with the critical value. For a right-tailed test, reject if the statistic is larger than the value. For a two-tailed test, reject if its absolute value is larger, or for chi-square and F if it is outside the interval. The decision is the same as with the p-value: a statistic beyond the critical value has a p-value below your significance level. Confidence intervals use the same values, so a 95 percent interval uses the critical value for α = 0.05, two-tailed.

Worked examples

A two-tailed t-test with 14 degrees of freedom at the 5 percent level uses 2.1448, so a statistic of 2.5 is significant and one of 1.9 is not. A chi-square test with 4 degrees of freedom at 5 percent has a critical value of 9.4877, so a statistic of 11.2 rejects the null hypothesis. A right-tailed z-test at 1 percent has a critical value of 2.3263. For an ANOVA with 3 and 20 degrees of freedom the 5 percent critical F is 3.0984. You can enter each of these to check the calculator against a table.

Critical values and confidence intervals

A confidence interval for a mean is the sample mean plus or minus the critical value times the standard error. For a 95 percent interval, use the two-tailed value for α = 0.05. With a sample of 25, the t value has 24 degrees of freedom and is 2.0639, a little larger than the 1.96 of the normal distribution, which makes the interval wider to allow for the estimated standard deviation. The calculator gives the matching confidence level beside each value.

Why not use a table

Tables list only a few significance levels and degrees of freedom, and you have to interpolate between them. The calculator works for any value, so you can use a level such as 0.025 or 0.003, or 37 degrees of freedom, directly. It also gives both critical values for chi-square and F in a two-tailed test, which many tables do not show.

Common mistakes

  • Using the wrong tail. A two-tailed test uses α ÷ 2 in each tail.
  • Using the z value for a small sample. Use the t distribution.
  • Forgetting the degrees of freedom. They change the value a great deal when small.
  • Mixing up confidence and significance. 95 percent confidence matches α = 0.05.

What next

Use this often? Add Critical Val to your Home screen.

Frequently asked questions

What is a critical value calculator?

A tool that finds the cut-off value of a test statistic for a given significance level, distribution and tail.

What is the critical value of z at 5 percent?

±1.96 for a two-tailed test and 1.645 for a one-tailed test.

How do I find a critical t value?

Enter the degrees of freedom and the significance level. For 10 degrees of freedom the two-tailed 5 percent value is ±2.228.

Why does the chi-square test have two critical values for a two-tailed test?

The distribution is not symmetric, so the lower and upper cut-offs differ.

What is the F critical value for 2 and 15 degrees of freedom?

3.68 at the 5 percent level, as in the ANOVA example.

What went wrong?

We only receive the reason you pick, never what you typed into the tool.

Thanks for your feedback.

More Math Tools