Critical Value Calculator

In statistics, a critical value defines the threshold of the rejection region. If your test statistic falls beyond this threshold, you reject the null hypothesis. If it falls inside the threshold, you fail to reject the null hypothesis.

Think of it like a finish line for a race. If your test data crosses that line, it proves that your results are real and not just random luck. This concept is the foundation of statistical significance.

The critical value is not your test result. It is the cutoff you compare your test statistic against. Researchers pick an alpha level first, usually 0.05, and the critical value is the number on the distribution that matches that cutoff. A result past that number is considered unlikely if the null hypothesis is true.

Finding this value by hand requires looking up numbers on a Z-table or a t-distribution table. Our tool does this for you. If you want to understand the steps, here is how the process works.

Step 1: Choose the Test Type

You must know which statistical test you are running. The most common types are the Z-test, the T-test, the Chi-Square test, and the F-test. Use a Z-test when you know the population standard deviation and have a large sample. Use a T-test when the sample is smaller or you must estimate the standard deviation from the data. Use a Chi-Square test for categorical counts, and an F-test when you are comparing variances or running ANOVA.

Step 2: Find the Alpha Level

The alpha level is your chance of being wrong. It is also called the significance level. The most common alpha level is 0.05, which means there is a 5% risk of concluding that a difference exists when there is no actual difference.

Step 3: Calculate Degrees of Freedom

You only need degrees of freedom for a T-test, Chi-Square test, or F-test. You do not need it for a standard Z-test. Degrees of freedom are usually based on your sample size.

Step 4: Choose One-Tailed or Two-Tailed

You must decide if your test is a one-tailed test or a two-tailed test. Use a one-tailed test if you are only checking if a value is greater than or less than a certain number. Use a two-tailed test if you are checking if a value is simply different from a certain number, in either direction.

There is no single algebraic formula to calculate these values. Instead, statisticians use inverse cumulative distribution functions. These are complex math equations built into our tool. Here is how the logic works for each test type.

Z-Critical Value

The Z-test uses the standard normal distribution. For a two-tailed test with an alpha of 0.05, the Z-critical value is exactly 1.96. This means if your test statistic is greater than 1.96 or less than negative 1.96, your result is significant.

T-Critical Value

The T-test uses a t-distribution. This distribution changes shape based on your sample size. A small sample size creates a flatter curve with larger critical values. A very large sample size makes the t-distribution look almost exactly like the Z-distribution.

Chi-Square Critical Value

The Chi-Square test checks for independence or goodness of fit. It uses only the right tail of the distribution. The curve is heavily skewed to the right, and the critical value depends entirely on your alpha level and degrees of freedom. You can read more about this in our chi-square calculator guide.

F-Critical Value

The F-test compares two different sample variances. It is most often used in an ANOVA calculator. The F-distribution is also skewed to the right and requires two separate degrees of freedom. One degree of freedom belongs to the numerator, and the other belongs to the denominator.

Understanding hypothesis testing is easier when you visualize a graph. Picture a bell curve. The critical value marks the cut-off point at the extreme edges of the curve. The area beyond this point is the rejection region.

If your test statistic lands in the rejection region, you reject the null hypothesis. If your test statistic lands in the middle of the curve, you do not have enough evidence to reject the null hypothesis. Always compare your calculated test statistic to the critical value before making a final conclusion.

Worked example

A clinic compares mean recovery days against a published benchmark with a two-tailed Z test at α = 0.05. The critical values are −1.96 and 1.96. The observed Z statistic is 2.21. Because 2.21 is past 1.96, the result is significant and H₀ is rejected.

A second study uses a two-tailed t test with n = 18, so df = 17, still at α = 0.05. The t critical value is about 2.110. The observed t statistic is 1.87. Because 1.87 is inside ±2.110, fail to reject H₀. Set the calculator to Z, two-tailed, α = 0.05, then to t with 17 degrees of freedom, to reproduce both cutoffs.