Free t-test calculator, ANOVA and chi-square test online
This hypothesis test calculator answers the everyday statistical questions of engineers and students: is the mean different from the target, do two machines or suppliers produce the same mean, did a change really improve the process, do three lines differ, is the spread the same, and are two categories related. It covers the one-sample, two-sample (Welch) and paired t-test, the F-test for variances, one-way ANOVA and the chi-square test of independence.
Instead of a bare p-value you get a plain-language verdict, the confidence interval of the difference, a dot plot of the data and all statistics in one table, without Excel, Minitab or R.
How the Hypothesis Tests works
Pick the test, paste one column per group and set the significance level. The t-tests use the Student t distribution, the Welch test the Welch–Satterthwaite degrees of freedom, ANOVA the F distribution of between- and within-group variance, and the chi-square test the expected counts of an independence model.
Every result is validated against SciPy in the automated test suite, so the p-values match what statistics packages report.
When to use it
Use a t-test to compare a supplier against a reference, two machines, two materials or a process before and after a change; ANOVA to compare several cavities, lines or shifts; the F-test to check whether a new process is less variable; and the chi-square test to see whether defect types depend on the shift, machine or supplier. It is also a quick way to back an engineering decision with data instead of opinion.
Remember that a significant result shows a difference exists, not that it matters: always look at the size of the difference and its confidence interval.
PDF report in one click
Export the test as a PDF report with your header, the question tested, the verdict, the statistics table, the confidence interval and the dot plot, ready for an engineering change, a supplier comparison or a thesis appendix.
Private: everything runs in your browser
All tests are calculated in your browser. Your data is never uploaded, and you can use the calculator without an account.
How to use the hypothesis test calculator
- Choose the test that matches your question.
- Paste the data, one column per group, and name the columns.
- Set α, usually 0.05.
- Read the verdict, the p-value and the confidence interval; check the dot plot for outliers.
Typical uses: does the new supplier deliver the same diameter (two-sample t), did the process change reduce the value (paired t), do three machines produce the same mean (ANOVA), is the defect type related to the shift (chi-square). Before comparing means, check normality with the histogram tool.
Frequently asked questions
›Which test should I use?
Comparing one mean with a target: one-sample t-test. Two independent groups: two-sample t-test. The same parts measured twice: paired t-test. Three or more groups: one-way ANOVA. Comparing spreads: F-test. Counts in categories: chi-square test.
›What does the p-value mean?
The probability of seeing a difference at least this large if there were really no difference. Below the significance level α (usually 0.05) the difference is called statistically significant.
›Why Welch and not Student t-test?
The Welch test does not assume equal variances in both groups. It is as good as the classic test when variances are equal and much more reliable when they are not, so it is the recommended default.
›Are the results validated?
Yes. All tests are checked against SciPy (ttest_1samp, ttest_ind, ttest_rel, f_oneway, chi2_contingency) in the automated test suite.