Critical Value Calculator
Find the critical value to compare against your test statistic in any hypothesis test.
📖 What is a Critical Value?
A critical value is the boundary point that separates the rejection region from the non-rejection region in a hypothesis test. If the computed test statistic falls beyond the critical value (into the rejection region), the null hypothesis is rejected at the chosen significance level α.
Critical values depend on three factors: the probability distribution of the test statistic (Z, t, F, or χ²), the significance level α (the acceptable probability of a Type I error - falsely rejecting H₀), and the direction of the test (one-tailed or two-tailed).
The most widely used critical value is z = 1.96, the two-tailed critical value for the standard normal distribution at α = 0.05. This appears in confidence interval formulas (95% CI: estimate ± 1.96 × SE) and in Z-tests for proportions and large-sample means. For the t-distribution, the critical value is higher and decreases as degrees of freedom increase, reflecting the heavier tails and greater uncertainty in small samples.
Critical values are equivalent to the quantile (inverse CDF) of the distribution. For example, the Z critical value at α = 0.05 (two-tailed) is the 97.5th percentile of the standard normal distribution: Φ⁻¹(0.975) = 1.96.
📐 Formulas
One-tailed right: z_α = Φ⁻¹(1 − α)
One-tailed left: z_α = Φ⁻¹(α) = −Φ⁻¹(1 − α)
t critical value: t_(α/2, df) = quantile of t-distribution with df degrees of freedom
Chi-square critical value: χ²_(α, df) = quantile of chi-square distribution (right-tailed)
F critical value: F_(α, df₁, df₂) = quantile of F-distribution (right-tailed)
Common Z critical values: α = 0.10 → 1.645 | α = 0.05 → 1.960 | α = 0.01 → 2.576 | α = 0.001 → 3.291 (all two-tailed)
📖 How to Use This Calculator
📝 Example Calculations
Example 1 - Z Critical Value (α = 0.05, two-tailed)
Example 2 - t Critical Value (α = 0.05, df = 19, two-tailed)
Example 3 - Chi-square Critical Value (α = 0.05, df = 4)
Example 4 - F Critical Value (α = 0.05, df₁ = 3, df₂ = 36)
Example 5 - Z Critical Value (α = 0.01, one-tailed right)
❓ Frequently Asked Questions
🔗 Related Calculators
What is a critical value in statistics?
A critical value is the threshold that a test statistic must exceed to reject the null hypothesis. It is determined by the distribution of the test statistic, the significance level (α), and the tail direction. For a two-tailed Z-test at α = 0.05, the critical values are ±1.96 - any Z-score outside this range leads to rejection of H₀.
What is the critical value for Z at α = 0.05?
For a two-tailed Z-test at α = 0.05: z_crit = ±1.96. For a one-tailed right test at α = 0.05: z_crit = +1.645. For a one-tailed left test: z_crit = −1.645. These values come from the inverse of the standard normal CDF (the quantile function Φ⁻¹).
How does the critical value change with sample size?
For t-tests, the critical value depends on degrees of freedom (df = n − 1 for one-sample). As n increases, df increases, and the t-distribution approaches the normal distribution - so the critical value decreases toward the Z critical value. For n > 30, t and Z critical values are very close.
What is the relationship between critical value and p-value?
They convey the same information from different directions. If |test statistic| > critical value, then p < α, and you reject H₀. If |test statistic| < critical value, then p > α, and you fail to reject H₀. Both approaches always give the same conclusion.
What critical values are used for confidence intervals?
Confidence intervals use the same critical values. A 95% CI uses z* = 1.96 (for Z) or t*(df) for t-intervals. A 99% CI uses z* = 2.576. The CI is: estimate ± critical_value × standard_error.
What is the F critical value used for?
The F critical value is used in ANOVA and regression F-tests. If the computed F-statistic exceeds the critical F value (at df₁, df₂, α), the null hypothesis that all group means are equal (ANOVA) or that all regression coefficients are zero (regression F-test) is rejected.
What critical value should I use for 95% confidence?
For a two-tailed test at 95% confidence: use z = 1.96 (normal distribution, large samples), or the t critical value for small samples. For one-tailed at 95%: z = 1.645. At 99% confidence two-tailed: z = 2.576. These values define the rejection regions - test statistics beyond these thresholds lead to rejecting the null hypothesis.
How does degrees of freedom affect the critical value?
For the t-distribution, lower degrees of freedom produce larger critical values (wider tails). With df = 5, the 95% two-tailed critical value is t = 2.571. With df = 30, it is 2.042. With df infinity (large sample), it converges to z = 1.96. This is why small-sample t-tests are more conservative - they require stronger evidence to reject the null.