Statistics guide

Probability and Statistics Methods

Statistics calculators need method labels. A z-score, p-value, confidence interval and sample-size estimate answer different questions even when they use some of the same inputs.

Named methods and formulas

Use z = (x - mean) / standard deviation to standardize one value. Use combinations n! / (r!(n-r)!) when order does not matter, and permutations n! / (n-r)! when order matters. For a simple proportion margin of error, a common approximation is z x sqrt(p(1-p)/n).

Concrete example

If a score is 85, the mean is 75 and the standard deviation is 10, the z-score is 1.0. If 5 people are selected from 20 and order does not matter, the combination count is 15,504. If order matters, the permutation count is 1,860,480.

What changes the result most

  • Sample size n changes margin of error through the square root, so quadrupling n roughly halves margin of error.
  • A 95% confidence interval commonly uses a z critical value near 1.96 under normal-approximation assumptions.
  • P-values depend on the null hypothesis and test method, not only the observed difference.

Common mistakes

  • Using combinations when order matters.
  • Treating a confidence interval as a guarantee.
  • Using p = 0.5 for sample-size planning without explaining that it is a conservative default for proportions.
  • Ignoring sampling bias, which no formula can fix after the fact.

Use the calculators

FAQ

When should I use combinations instead of permutations?

Use combinations when order does not matter. Use permutations when different orders count as different outcomes.

Does a 95% confidence interval mean 95% probability for one finished interval?

Not exactly. The 95% refers to the long-run behavior of the method under its assumptions.

References

Last reviewed: 2026-05-15.