Hypothesis Testing Calculator - Z-Test, T-Test & P-Value

Hypothesis testing is the backbone of inferential statistics. It provides a principled, probabilistic framework for deciding whether the data you have collected is consistent with a theoretical claim — the null hypothesis — or whether the evidence is strong enough to reject that claim in favour of an alternative. Every experiment in medicine, psychology, economics, engineering quality control, and A/B website testing ultimately comes down to some form of hypothesis test. The null hypothesis (H₀) is the default assumption: nothing happened, the treatment has no effect, the process is on target, or the proportions are unchanged. The alternative hypothesis (H₁) is what you are trying to detect: the mean has shifted, the proportion has changed, or one treatment is better than another. The significance level α — usually 0.05 or 0.01 — is the probability of incorrectly rejecting H₀ when it is actually true (a Type I error). If the p-value returned by the test is less than α, you reject H₀. The Z-test for means is appropriate when the population standard deviation σ is known and either the sample is large (n ≥ 30) or the population is normally distributed. The test statistic is Z = (x̄ − μ₀) / (σ / √n). Because σ is known, the statistic follows the standard normal distribution exactly, and the p-value is read from the normal table. The T-test for means applies when σ is unknown, which is the realistic situation in most real-world research. The sample standard deviation s is used instead, and the test statistic T = (x̄ − μ₀) / (s / √n) follows a t-distribution with df = n − 1 degrees of freedom. With small samples the t-distribution has heavier tails than the normal, making it harder to reach significance — a sensible penalty for the extra uncertainty about σ. The Z-test for proportions tests whether an observed sample proportion p̂ is consistent with a hypothesised population proportion p₀. The standard error is √(p₀(1 − p₀) / n) and the test statistic is Z = (p̂ − p₀) / SE. This test is widely used in A/B testing, clinical trial primary endpoints, and quality-control fraction-defective charts. For a two-tailed test you reject H₀ when |statistic| > critical value, capturing deviations in either direction. For a one-tailed test (left or right) you specify the direction in advance; this gives more power to detect a shift in that direction but cannot flag an unexpected shift the other way. The critical value displayed is for the right-tail boundary; for a left-tailed test the relevant boundary is its negative. The p-value is the probability of observing a test statistic at least as extreme as the one computed, assuming H₀ is true. A p-value of 0.03 does not mean there is a 3% chance the null is true; it means that if H₀ were true, there would be only a 3% chance of seeing data this extreme or more extreme by random sampling alone. Statistical significance is not the same as practical significance: a tiny effect can be highly significant with a large n, while a large effect may fail to reach significance with a small n. Always pair the p-value with an effect size and a confidence interval.