About NumberCalcs

We are statisticians, data scientists, and educators who kept running into the same problem: every calculator on the web gives you the number and nothing else.

That is a problem in every setting where the number is not the point. A student submitting a t-test for a dissertation needs to show the work, not just cite an output. A researcher cross-checking a chi-square result needs to verify the degrees of freedom, not trust a black box. A data analyst explaining a power analysis to a non-technical stakeholder needs to walk through the logic step by step.

So we built tools that show every step, in the order a textbook would show them.

Formulas and standards

The formulas follow published standards. Our Welch two-sample t-test follows Welch's 1947 derivation — the version that does not assume equal variances, which is the correct default for most real-world data. Chi-square tests follow Pearson's original formulation. Power analysis uses the Cohen conventions that appear in most graduate-level methodology courses.

What we cover

The tool set covers what actually gets assigned and used: descriptive statistics, probability distributions, hypothesis tests (t-test, z-test, chi-square, ANOVA, Mann-Whitney), confidence intervals, regression, and power analysis. We expand based on what people request, not on what is easy to build.

Privacy

Everything runs client-side. No account required, no data sent to a server. The numbers you enter stay in your browser and nowhere else. We did not build a login system because there was no reason to — the tools work without one, and building one would have required us to store data we have no interest in storing.

Who uses this

The audience is exactly who you would expect: undergraduate and graduate students, researchers who want a second opinion on their own calculations, and working analysts who need to explain a method to someone who did not run it.

Contact

Questions or formula errors to report: Contact us