In Excel: use the appropriate test function, such as =T.TEST(array1,array2,2,2) for a two-tailed two-sample t-test, and interpret the result against your chosen alpha level.
On this page7
Choose the statistical test before writing the formula.
Put each sample in a clean numeric range.
Use the matching Excel test function and tail/type arguments.
Report p-value with sample sizes and the test used.
Need it as an auditable file?
Ships inside the linked template — formula-driven, unlocked, audit-ready.
What this does
A p-value is not a generic spreadsheet statistic; it comes from a specific hypothesis test. Excel can calculate p-values with T.TEST, Z.TEST, CHISQ.TEST and Data Analysis ToolPak outputs. The right function depends on the design, distribution assumptions and whether the test is one-tailed or two-tailed. Most people learn this as a sequence of clicks and forget it by next week; learning it as a pattern instead is what lets you apply it to the next, slightly different version of the problem without starting from scratch. That is the difference this page is trying to make. Treat “do p value in excel” as a small repeatable workflow rather than a one-off click you hope to remember next time. Use a small test block before the live file, so any surprise in the affected cells shows up while it is still harmless. When a formula is involved, keep the inputs labelled beside it, reference cells instead of typing values, and apply number formatting only after the result checks out. That turns a calculation you can defend to a CFO or an auditor into a method you can reuse, explain, and defend when the workbook leaves your screen.
A worked example
Scores before training are in A2:A21 and after training in B2:B21 for separate groups. =T.TEST(A2:A21,B2:B21,2,2) returns the two-tailed p-value for a two-sample equal-variance t-test. If p is 0.03 and alpha is 0.05, the difference is statistically significant under that test. P-value content needs caution because users can easily produce a number that looks authoritative but comes from the wrong test. One habit worth forming early: name the cells that hold your inputs, so the formula reads in plain language instead of a string of cell addresses. A reviewer — or you in three months — can then follow the logic without decoding what B7 and D2 were supposed to mean, which is most of what makes a sheet maintainable.
In Google Sheets
If you are in Google Sheets rather than Excel, the good news is that the formula shown here is identical and the workflow barely changes — menus sit across the top instead of in a ribbon, and a few function names differ slightly, but anything you build here moves across with little or no rework. The aim was to get you unstuck fast and leave you a little more capable than a copy-paste would. The answer is at the top and the detail above shows why it holds — so the next time a colleague asks, you can answer without reaching for search. If you take one thing from this page on “do p value in excel”, make it the habit rather than the keystrokes: set the problem up with labelled inputs, reference those cells, and let Excel do the recomputing. Bookmark the page for the syntax, but do the example once in a blank sheet and check it against the tool above — five minutes of hands-on practice fixes the method in memory far better than re-reading, and it surfaces the small snags while they are still harmless. After that the technique is genuinely yours: faster than searching for it again, and reliable enough to drop into work that other people depend on.
Common mistakes
- Using
T.TESTfor every p-value question. - Ignoring paired versus unpaired samples.
- Treating p-value as effect size.
- Changing one-tailed versus two-tailed after seeing results.
Frequently asked questions
Which Excel function gives a p-value?
It depends on the test: T.TEST, Z.TEST, CHISQ.TEST and ToolPak tests can return p-values.
Is p < 0.05 always significant?
Only relative to a prechosen alpha and suitable test assumptions.
Does p-value show importance?
No. It is evidence against a null hypothesis, not effect size.