Compute the test statistic with a Test Statistic Calculator
A test statistic summarizes how far your sample result is from the hypothesized (null) value, measured in standard error units. This calculator computes a z-test statistic when the population standard deviation is known, or a t-test statistic when it is unknown.
Enter the sample mean, null mean, and the right variability inputs. The calculator outputs the test statistic and (optionally) an interpretation-ready value you can compare against critical values or p-values.
What “test statistic” means
A test statistic converts raw data into a standardized number. That number tells you whether the observed difference is large compared with expected random variation under the null hypothesis.
- Null hypothesis (H0): the hypothesized mean value (often μ0).
- Sample mean (x̄): the average from your sample.
- Standard error: the typical size of sampling variation.
- Test statistic: the standardized distance between x̄ and μ0.
Core formulas (z and t)
The calculator uses the standard one-sample mean test forms.
Z test statistic (population standard deviation known)
Use this when you know the population standard deviation σ (or you treat it as known). The formula is:
z = (x̄ − μ0) / (σ / √n)
- x̄ = sample mean
- μ0 = null (hypothesized) mean
- σ = population standard deviation
- n = sample size
T test statistic (population standard deviation unknown)
Use this when σ is unknown and you rely on the sample standard deviation s. The formula is:
t = (x̄ − μ0) / (s / √n)
- s = sample standard deviation
- n = sample size
How to choose z vs. t
Pick the version that matches your assumptions.
- Choose z when σ is known from a credible source (for example, a controlled process with a stable, well-established variance).
- Choose t when σ is unknown and you estimate variability using s from your sample.
If you’re unsure, the t approach is the standard default for one-sample mean problems where σ is not given.
Variables, units, and what stays consistent
Test statistics are unit-free because the numerator and denominator share the same units. Still, you must enter consistent measurement units for means and standard deviations.
- If your data are in meters, enter x̄, μ0, and σ or s in meters.
- If your data are in seconds, enter x̄, μ0, and σ or s in seconds.
- If you switch units, the calculator converts variability inputs to match the selected unit.
Practical example 1: Quality control (z test)
A factory measures the average thickness of a coating. The target thickness is μ0 = 50.0 μm. A sample of n = 36 parts has x̄ = 50.8 μm. Historical data show the population standard deviation is σ = 2.0 μm.
Compute the standard error: σ/√n = 2.0/√36 = 2.0/6 = 0.333. Then z = (50.8 − 50.0)/0.333 ≈ 2.40. A z value this large suggests the mean may be higher than the target.
Practical example 2: Lab measurement (t test)
A lab tests whether the mean concentration matches a reference value. The null hypothesis is μ0 = 10.0 mg/L. You collect n = 25 samples and find x̄ = 9.2 mg/L with sample standard deviation s = 1.6 mg/L.
Standard error is s/√n = 1.6/5 = 0.32. Then t = (9.2 − 10.0)/0.32 = −2.50. The negative sign indicates the sample mean is below the null value; you compare the magnitude against t critical values for the appropriate degrees of freedom.
Interpreting the test statistic (quick guide)
The test statistic’s meaning depends on the alternative hypothesis:
- Two-sided: large positive or large negative values are evidence against H0.
- Right-sided: large positive values are evidence against H0.
- Left-sided: large negative values are evidence against H0.
In practice, you compare the statistic to a critical value (from a z or t table) or convert it to a p-value using a distribution.
Common mistakes to avoid
- Mixing σ and s: using σ in a t test formula (or s in a z test) breaks the assumptions.
- Forgetting √n: the standard error shrinks as sample size grows; omitting √n makes the statistic too large or too small.
- Inconsistent units: means and standard deviations must represent the same measurement scale.
- Using n = 0 or n = 1: sample size must be valid; t tests require enough data to estimate variability.
Frequently Asked Questions
What does a higher test statistic mean?
A higher absolute test statistic means your sample mean is farther from the null mean relative to the standard error. That indicates stronger evidence against the null hypothesis. Whether it is “higher” in a positive or negative direction depends on your alternative hypothesis.
Is the test statistic the same as the p-value?
No. The test statistic is the standardized number computed from your data. The p-value is the probability, under the null hypothesis, of getting a test statistic at least as extreme as the one observed. You use the distribution to convert between them.
When should I use a z test instead of a t test?
Use a z test when the population standard deviation σ is known and you can reasonably treat it as fixed. Use a t test when σ is unknown and you estimate variability using the sample standard deviation s. Most real-world cases default to t.
Do test statistics have units?
No. Because the numerator (x̄ − μ0) and the denominator (standard error) share the same units, they cancel out. For example, if both are in mg/L, the ratio becomes unitless. You still must enter consistent units for inputs.
What happens if the sample mean equals the null mean?
If x̄ equals μ0 exactly, the numerator is 0, so the test statistic is 0. That means your data match the null mean and provide no standardized evidence against H0. In many tests, a statistic near 0 leads to a large p-value.
Next steps
Use the Test Statistic Calculator to get the correct z or t value quickly. Then compare it to the right critical value or compute a p-value for your chosen alternative hypothesis.
If you share your inputs (x̄, μ0, n, and either σ or s), you can verify the result and check whether the z or t assumption fits your situation.