Report

Help us improve this tool

P Value Calculator

Find one-tailed or two-tailed p-values from z, t, chi-square, and F test statistics with decision guidance.

O M T

What This P Value Calculator Does

This calculator converts a test statistic into a p-value. Choose the distribution, set the tail type, enter your score, and compare the result with a significance level. It works for z, t, chi-square, and F statistics.

P Value Formula Basics

A p-value measures how extreme your observed statistic is under the null hypothesis. In cumulative distribution function form:

$$p_{\text{left}} = F(x), \qquad p_{\text{right}} = 1 - F(x), \qquad p_{\text{two}} = 2\min(F(x), 1-F(x))$$

For symmetric distributions like the normal and t distribution, the two-tailed p-value is often written as $2P(X \ge |x|)$. For skewed distributions like chi-square and F, the two-tailed version is based on the smaller tail area.

When To Use Each Distribution

  • Use z-score when the test statistic follows a standard normal model.
  • Use t-score when population variance is unknown and sample-based inference is used.
  • Use chi-square for variance checks and contingency-table style tests.
  • Use F-score for variance ratios and overall regression significance tests.

How To Interpret The Result

Compare the p-value with your chosen significance level $\alpha$. If $p \le \alpha$, you reject the null hypothesis. If $p > \alpha$, you do not have enough evidence to reject it. For related workflows, try our Hypothesis Testing Calculator, F Statistic Calculator, and Z Score Calculator.

Frequently Asked Questions

What does a small p-value mean?

It means your observed statistic would be unlikely if the null hypothesis were true, so the data provide stronger evidence against that null model.

Can a p-value be greater than 1?

No. P-values are probabilities, so they always stay between 0 and 1.

When should I use a one-tailed test?

Use a one-tailed test only when your alternative hypothesis is directional before you inspect the data.

Why do t, chi-square, and F inputs need degrees of freedom?

Those distributions change shape based on degrees of freedom, so the same statistic can map to different p-values under different sample structures.