A p-value is not the probability that your result is true — it is the probability of seeing data this extreme if nothing were actually going on. That single misunderstanding has sunk a thousand research conclusions. The p-value starts from the null hypothesis (the boring claim that nothing changed) and asks: if the boring claim were true, how surprising would my data be? A p-value of 0.03 says "this surprising or more happens 3% of the time under the boring claim" — surprising enough that most fields reject the boring claim at the standard 5% significance level.

Website A/B tests live on this logic. A startup tests a new checkout button on 10,000 visitors, computes a z-score of 2.1 from the conversion difference, and gets a p-value near 0.036 — below 0.05, so the new button ships. The same z-score with a stricter 1% level would not ship; the significance level is a policy choice, not a law of nature.

This calculator converts your test statistic into p-values. Enter the Z-Score and your Significance Level α, and it returns the one- and two-tailed p-values plus whether the result clears your significance bar.

With the pre-filled defaults — Z-Score 1.96, Significance Level α 0.05 — the P Value Calculator returns P-Value (one-tailed) 0.024998, P-Value (two-tailed) 0.049996, Significant at 95%? 1, Significant at 99%? 0, Critical Z (95%) 1.9600, Critical Z (99%) 2.5758. Change any input and those results update instantly, so you can see exactly how each number you enter moves the answer.