What is a F Critical Value Calculator?
An F critical value calculator finds the value of the F distribution with numerator df₁ and denominator df₂ that leaves an area α in the tail. ANOVA and regression F-tests use the right tail: for df = (3, 36) at α = 0.05 the critical F is 2.8663, so you reject H₀ when F ≥ 2.8663. A two-tailed option returns both cut-offs for comparing two variances, using F_L = 1 / F_U with the df swapped.
What the F Critical Value Calculator does
Enter df₁, df₂ and α and choose the tail. The calculator returns the critical F (or the lower and upper values), shades the rejection region on the F curve, and lists right-tailed, left-tailed and two-tailed values at six common α levels for your df pair. Because it works for any positive df, you do not have to interpolate between the widely spaced columns of a printed F-table. In effect it is an F critical value table (a critical F value chart for every df pair and α) and an F distribution calculator for quantiles, with the working shown step by step.
Compare an F statistic with the right-tailed value in a one-way or factorial ANOVA, an overall regression F-test, or a test comparing nested models. Use the two-tailed values for the F-test of two variances when the alternative is σ₁² ≠ σ₂². To calculate the p-value for an F test instead of the cut-off, use the P-Value from F Calculator.
How to use the F Critical Value Calculator
- Enter the Numerator df (df₁), the degrees of freedom of the mean square on top of the ratio (k − 1 in ANOVA).
- Enter the Denominator df (df₂), the degrees of freedom of the mean square on the bottom (N − k in ANOVA).
- Enter the Significance level (α) and choose the Tail. Leave it on Right unless you are comparing two variances.
- Read the critical F and the rejection rule. The steps show how a lower value comes from the reciprocal rule.
How to find the F critical value
To find the critical value of F by hand you need four things: the numerator df, the denominator df, the significance level α and the tail of the test. Then read the value from an F table or compute the F quantile.
- Work out df₁ (numerator). One-way ANOVA: k − 1 for k groups. Regression: the number of predictors k. Two variances: n₁ − 1 for the sample whose variance is on top.
- Work out df₂ (denominator). One-way ANOVA: N − k. Regression: n − k − 1. Two variances: n₂ − 1. The Degrees of Freedom Calculator handles each case.
- Choose α. 0.05 is the usual default; 0.01 is stricter.
- Pick the tail. ANOVA and regression F-tests are always right-tailed. A two-sided test of two variances is two-tailed.
- Look up the value. In the F critical value table below, go to the column for df₁ and the row for df₂. Printed tables give right-tail values only; for a two-tailed test use the α/2 table, or let the calculator return both cut-offs.
How to calculate the F value
The F value (F statistic) is a ratio of two variance estimates. The formula for the F value depends on the test, but it is always "variance explained" over "variance left over", each divided by its df. The same df₁ and df₂ are what you enter above to get the critical value. To calculate the F value from raw data, use the ANOVA Calculator for group means or the F-Test Calculator for two variances.
| Test | F value formula | df₁ | df₂ |
|---|---|---|---|
| One-way ANOVA | F = MS_between / MS_within = [SS_between / (k − 1)] / [SS_within / (N − k)] | k − 1 | N − k |
| Multiple regression (overall F-test) | F = MSR / MSE = [SSR / k] / [SSE / (n − k − 1)] | k | n − k − 1 |
| Two variances | F = s₁² / s₂² | n₁ − 1 | n₂ − 1 |
Comparing the F value with the F critical value
Compare the F statistic from your test with the critical F for the same df and α:
- If F is greater than (or equal to) the F critical value, it lands in the rejection region. Reject H₀: the group means (or the regression coefficients) are not all equal at that α.
- If F is less than the F critical value, do not reject H₀. This is not proof that H₀ is true; the data are simply consistent with it.
- The decision always agrees with the p-value: F ≥ F critical exactly when p ≤ α.
- Example: three teaching methods with 8 students each (N = 24) give SS_between = 84 and SS_within = 176.4. Then MS_between = 84 / 2 = 42, MS_within = 176.4 / 21 = 8.4 and F = 42 / 8.4 = 5.00. The F critical value for df = (2, 21) at α = 0.05 is 3.4668. Because 5.00 exceeds 3.4668, reject H₀ that the three method means are equal (p ≈ 0.0167).
F critical value formula
Critical values are quantiles of the F(df₁, df₂) distribution, F⁻¹(p; df₁, df₂).
| Symbol | Meaning |
|---|---|
| df₁ | numerator degrees of freedom |
| df₂ | denominator degrees of freedom |
| α | significance level |
| F⁻¹(p; a, b) | value with cumulative probability p under F(a, b) |
The reciprocal rule holds because if X has an F(df₁, df₂) distribution, 1/X has an F(df₂, df₁) distribution. It is why F-tables only need to print upper-tail values.
The F distribution is named after Ronald Fisher, who developed analysis of variance in the 1920s; George Snedecor tabulated it and named it F in 1934. Its inverse CDF has no closed form, so the calculator finds it numerically. Values here agree with the NIST/SEMATECH e-Handbook table of F critical values.
Worked example: two-tailed F critical values for two variances
Comparing two machine variances: n₁ = 10, n₂ = 15
Problem. Machine A gives n₁ = 10 parts and machine B n₂ = 15 parts. Find the critical values for a two-tailed F-test of equal variances at α = 0.05, with F = s²_A / s²_B.
- 01df₁ = 10 − 1 = 9 and df₂ = 15 − 1 = 14; α/2 = 0.025 goes in each tail.
- 02Upper critical value: F⁻¹(0.975; 9, 14) = 3.2093.
- 03Lower critical value via the reciprocal rule: F⁻¹(0.975; 14, 9) = 3.7980, so F_L = 1 / 3.7980 = 0.2633.
- 04Rejection rule: reject H₀ if F ≤ 0.2633 or F ≥ 3.2093.
Answer. The variances differ significantly at the 5% level only if s²_A / s²_B is below 0.2633 or above 3.2093. An observed ratio of 3.5, for example, would lead to rejecting H₀.
F critical value table (α = 0.05)
This F distribution table gives right-tail critical values: each entry leaves 5% of F(df₁, df₂) to its right. Columns are the numerator df₁ and rows the denominator df₂. It doubles as an ANOVA critical value table: these are the F test critical values for ANOVA and regression at α = 0.05; for example df = (3, 20) gives 3.098. For other α levels, a two-tailed or left-tailed value, or any df not listed, use the calculator above or the full F-table. The last row (df₂ = ∞) equals χ²(df₁) / df₁.
| df₂ \ df₁ | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 161.448 | 199.500 | 215.707 | 224.583 | 230.162 | 233.986 | 236.768 | 238.883 | 240.543 | 241.882 |
| 2 | 18.513 | 19.000 | 19.164 | 19.247 | 19.296 | 19.330 | 19.353 | 19.371 | 19.385 | 19.396 |
| 3 | 10.128 | 9.552 | 9.277 | 9.117 | 9.013 | 8.941 | 8.887 | 8.845 | 8.812 | 8.786 |
| 4 | 7.709 | 6.944 | 6.591 | 6.388 | 6.256 | 6.163 | 6.094 | 6.041 | 5.999 | 5.964 |
| 5 | 6.608 | 5.786 | 5.409 | 5.192 | 5.050 | 4.950 | 4.876 | 4.818 | 4.772 | 4.735 |
| 6 | 5.987 | 5.143 | 4.757 | 4.534 | 4.387 | 4.284 | 4.207 | 4.147 | 4.099 | 4.060 |
| 7 | 5.591 | 4.737 | 4.347 | 4.120 | 3.972 | 3.866 | 3.787 | 3.726 | 3.677 | 3.637 |
| 8 | 5.318 | 4.459 | 4.066 | 3.838 | 3.687 | 3.581 | 3.500 | 3.438 | 3.388 | 3.347 |
| 9 | 5.117 | 4.256 | 3.863 | 3.633 | 3.482 | 3.374 | 3.293 | 3.230 | 3.179 | 3.137 |
| 10 | 4.965 | 4.103 | 3.708 | 3.478 | 3.326 | 3.217 | 3.135 | 3.072 | 3.020 | 2.978 |
| 12 | 4.747 | 3.885 | 3.490 | 3.259 | 3.106 | 2.996 | 2.913 | 2.849 | 2.796 | 2.753 |
| 15 | 4.543 | 3.682 | 3.287 | 3.056 | 2.901 | 2.790 | 2.707 | 2.641 | 2.588 | 2.544 |
| 20 | 4.351 | 3.493 | 3.098 | 2.866 | 2.711 | 2.599 | 2.514 | 2.447 | 2.393 | 2.348 |
| 24 | 4.260 | 3.403 | 3.009 | 2.776 | 2.621 | 2.508 | 2.423 | 2.355 | 2.300 | 2.255 |
| 30 | 4.171 | 3.316 | 2.922 | 2.690 | 2.534 | 2.421 | 2.334 | 2.266 | 2.211 | 2.165 |
| 40 | 4.085 | 3.232 | 2.839 | 2.606 | 2.449 | 2.336 | 2.249 | 2.180 | 2.124 | 2.077 |
| 60 | 4.001 | 3.150 | 2.758 | 2.525 | 2.368 | 2.254 | 2.167 | 2.097 | 2.040 | 1.993 |
| 120 | 3.920 | 3.072 | 2.680 | 2.447 | 2.290 | 2.175 | 2.087 | 2.016 | 1.959 | 1.910 |
| ∞ | 3.841 | 2.996 | 2.605 | 2.372 | 2.214 | 2.099 | 2.010 | 1.938 | 1.880 | 1.831 |
F critical value table (α = 0.01)
The same layout at the stricter 1% level. Each entry leaves 1% of F(df₁, df₂) in the right tail.
| df₂ \ df₁ | 1 | 2 | 3 | 4 | 5 | 6 | 8 | 10 |
|---|---|---|---|---|---|---|---|---|
| 5 | 16.258 | 13.274 | 12.060 | 11.392 | 10.967 | 10.672 | 10.289 | 10.051 |
| 10 | 10.044 | 7.559 | 6.552 | 5.994 | 5.636 | 5.386 | 5.057 | 4.849 |
| 12 | 9.330 | 6.927 | 5.953 | 5.412 | 5.064 | 4.821 | 4.499 | 4.296 |
| 15 | 8.683 | 6.359 | 5.417 | 4.893 | 4.556 | 4.318 | 4.004 | 3.805 |
| 20 | 8.096 | 5.849 | 4.938 | 4.431 | 4.103 | 3.871 | 3.564 | 3.368 |
| 30 | 7.562 | 5.390 | 4.510 | 4.018 | 3.699 | 3.473 | 3.173 | 2.979 |
| 40 | 7.314 | 5.179 | 4.313 | 3.828 | 3.514 | 3.291 | 2.993 | 2.801 |
| 60 | 7.077 | 4.977 | 4.126 | 3.649 | 3.339 | 3.119 | 2.823 | 2.632 |
| 120 | 6.851 | 4.787 | 3.949 | 3.480 | 3.174 | 2.956 | 2.663 | 2.472 |
| ∞ | 6.635 | 4.605 | 3.782 | 3.319 | 3.017 | 2.802 | 2.511 | 2.321 |
F critical values in Excel, R, Python and on a TI-84
Every package returns the same quantile; they differ only in whether they take the right-tail area α or the cumulative area 1 − α. The examples use df = (3, 36) and α = 0.05, which all return 2.8663.
| Tool | Right-tailed critical value | Left-tailed critical value |
|---|---|---|
| Excel / Google Sheets | =F.INV.RT(0.05, 3, 36) | =F.INV(0.05, 3, 36) |
| R | qf(0.05, 3, 36, lower.tail = FALSE) | qf(0.05, 3, 36) |
| Python (SciPy) | f.isf(0.05, 3, 36) | f.ppf(0.05, 3, 36) |
| TI-83 / TI-84 | No inverse F function: solve Fcdf(X, 1E99, 3, 36) = 0.05 for X with the equation solver | Solve Fcdf(0, X, 3, 36) = 0.05 for X |
Common use cases
- Setting the rejection region for a one-way or two-way ANOVA.
- Checking the overall significance of a multiple regression model.
- Comparing two population variances with the F-test, for example the consistency of two machines or two lab methods.
- Comparing nested regression models, where df₁ is the number of added parameters.
Assumptions and limitations
- The F critical values give correct error rates only if the data are independent and normally distributed; the variance-ratio test in particular is very sensitive to non-normality. Levene’s test is a more robust check of equal variances.
- df₁ and df₂ are not interchangeable. Swapping them gives a different critical value (except through the reciprocal rule).
- For an observed F, the P-Value from F Calculator tells you how far past the cut-off you are, which a critical value alone does not.
Frequently asked questions
How do I find the F critical value?
What is the F critical value for ANOVA at α = 0.05?
What is the formula for the F value?
What if my F value is greater than the F critical value?
Why is the F-test usually one-tailed?
How do I get a lower-tail F critical value from a table?
How is the F critical value related to the t critical value?
Cite this page
Statculator. (2026, November 14). F Critical Value Calculator. Retrieved November 14, 2026, from https://statculator.com/p-values-critical-values/f-critical-value-calculator/