What this calculator does
Give it a test statistic from any common test and it returns the exact p-value, so you do not need a printed table. It works for z, t, chi-square, F and correlation r, with two-tailed, right-tailed or left-tailed options where they apply. Chi-square and F tests are right-tailed by construction, so the option is hidden for them.
Worked example
A t-test gives t = 2.10 with 15 degrees of freedom. The two-tailed p-value is .053, just above the usual .05 cut-off. The same value of 2.10 treated as a z-score gives p = .036, which shows why small samples need the t-distribution: its heavier tails make the same statistic less surprising. The full method is in how to calculate a p-value.
Which tail should I choose?
Use two-tailed unless you chose the direction before seeing the data. A one-tailed p-value is half the two-tailed value when the result goes in the predicted direction, which is why switching after the fact is not legitimate. See one-tailed vs two-tailed tests.
Accuracy
All p-values were checked against SciPy across a grid of statistics and degrees of freedom, with differences below 0.0000001.
Frequently asked questions
How do I find a p-value from a t statistic?
Enter t and the degrees of freedom, and choose the tail. For t = 2.10 with df = 15, the two-tailed p-value is .053.
Why is there no tail option for chi-square and F?
Both statistics measure squared deviations, so larger values always mean stronger evidence. Their p-value is the right-tail area.
What does the p-value mean?
It is the probability of a result at least this extreme if the null hypothesis is true. It is not the probability that the null hypothesis is true.
How do I calculate this in Excel?
Use T.DIST.2T(ABS(t), df) for t, 2*(1-NORM.S.DIST(ABS(z),TRUE)) for z, CHISQ.DIST.RT(x, df) for chi-square and F.DIST.RT(x, df1, df2) for F.