This hypothesis test calculator runs the common statistical tests on data you paste. Choose a one-sample, two-sample or paired t-test, one-way ANOVA, chi-square test or correlation, enter the values, and get the test statistic, exact two-sided p-value, a significance decision at your α and an effect size. Paste a table with one column per variable to test them all and see a correlation matrix. It runs in your browser.
| Test | Degrees of freedom | Critical value |
|---|---|---|
| t | 10 | 2.228 |
| t | ∞ | 1.960 |
| F | 2, 12 | 3.885 |
| χ² | 1 | 3.841 |
| χ² | 10 | 18.307 |
Two-sample for independent groups (Welch by default), paired for before/after on the same subjects, one-sample to compare with a known value.
The probability of a result at least this extreme if the null hypothesis were true; below α (often 0.05) the result is called significant.
For ranked data, outliers or relationships that are monotonic but not linear.
The approximation is unreliable; use Fisher's exact test or combine categories.
No. Everything runs in your browser.
t-tests compute t from the means and standard errors (Welch's approximation for unequal variances, or the pooled variance) and the two-sided p-value from the t distribution via the regularised incomplete beta function. One-way ANOVA compares between- and within-group sums of squares as an F ratio. Chi-square compares observed and expected counts for a contingency table or goodness of fit, using the regularised gamma function. Pearson and Spearman correlations are tested with t = r√((n − 2) ÷ (1 − r²)). Effect sizes (Cohen's d, η², Cramér's V, r²) are reported, and a table of variables gets descriptives, one-sample tests and a correlation matrix.