Free · no signup

p-value calculator (from z, t, chi-square, or F)

Q: How do I read the result into an APA sentence?

The calculator writes it for you, e.g. "t(24) = 2.31, p = .030, two-tailed." APA style: italic statistic letter, exact p to 2–3 decimals with no leading zero, p < .001 for anything smaller, and an effect size alongside.

Q: Why do χ² and F have no one-/two-tailed choice?

Both statistics are squared quantities, so deviations in either direction push them the same way — into the upper tail. The upper-tail p is the standard test. (Lower-tail χ² tests exist for variance hypotheses, but they are rare.)

Q: Can I get the critical value instead of the p-value?

Yes — see the t-table, chi-square table, z-table and F-table pages in the free tools. Each includes an exact critical-value lookup alongside the printed table.

Turn a test statistic into an exact p-value — the lookup you’d otherwise do with a printed table. Handles t, z, χ² and F, one- or two-tailed, and formats the APA sentence for you.

Let the AI do this — sign up freeUpload your data, get the write-up back.

Or skip the manual work

In the app, you chat with your data — the AI runs the analysis and writes it up.

Kahubi AILive in the app

Typing numbers in by hand works — but inside Kahubi you just upload your dataset and ask. I check the assumptions, run the right test, and hand you an APA-ready results section. Try me.

Free accounts include AI chat, data analysis, and every tool on this site.

When to use this

Use this whenever you have a test statistic but not its p-value: older papers that report "t(24) = 2.31, p < .05" when you need the exact p for a meta-analysis, software output that truncates p-values, hand-calculated statistics from homework or a methods exam, or checking a manuscript you are reviewing. It replaces the printed z/t/χ²/F tables at the back of statistics textbooks — with exact values instead of bracketing between table rows.

Key assumptions

The statistic really follows the claimed distribution under H0 (the original test’s assumptions did hold).
Degrees of freedom are entered as reported — Welch-corrected dfs can be fractional, which is fine.
One-tailed p-values are only valid if the direction was specified before seeing the data.

Common mistakes

Halving a two-tailed p after seeing the result went "the right way" — that is p-hacking.
Reading p = .06 as "no effect" and p = .04 as "an effect" — the difference between those two is not itself significant.
Reporting p = .000 from software output — write p < .001.

Frequently asked questions

How do I read the result into an APA sentence?

The calculator writes it for you, e.g. "t(24) = 2.31, p = .030, two-tailed." APA style: italic statistic letter, exact p to 2–3 decimals with no leading zero, p < .001 for anything smaller, and an effect size alongside.

Why do χ² and F have no one-/two-tailed choice?

Both statistics are squared quantities, so deviations in either direction push them the same way — into the upper tail. The upper-tail p is the standard test. (Lower-tail χ² tests exist for variance hypotheses, but they are rare.)

Can I get the critical value instead of the p-value?

Yes — see the t-table, chi-square table, z-table and F-table pages in the free tools. Each includes an exact critical-value lookup alongside the printed table.

Related free tools

Sample size calculator Sample size for two groups Effect size (Cohen's d)Descriptive statistics One-sample t-test Independent t-test Paired t-test z-table (standard normal)t-table (critical values)Chi-square table

Stop copying numbers between tools

Inside Kahubi, the AI agent runs this analysis directly on your uploaded dataset — then writes the results section in your own writing style, with the statistics reported correctly.

Try Kahubi free More free tools