About this calculator
Hypothesis Test Calculator helps students and analysts estimating one-sample z-tests or t-tests from summary statistics. Use it when you need a clear numerical result, the formula behind the result, and enough context to explain the answer in homework, a report, a spreadsheet, or a practical data check. It is designed for educational and analytical use, so it should support your reasoning rather than replace judgement about the data source, sampling method, or assumptions.
Hypothesis Test Calculator methodology
The calculator subtracts the hypothesised mean from the sample mean, divides by the standard error, then estimates a p-value using either the normal distribution or Student t distribution.
- test statistic = (sample mean - hypothesised mean) / (standard deviation / sqrt(n))
- t-test degrees of freedom = n - 1
- two-tailed p-value = 2 x smaller tail probability
How to use the Hypothesis Test Calculator
- Enter the data, counts, or parameters requested by the calculator.
- Remove labels, currency symbols, blank cells, and non-numeric notes before calculating.
- Check whether the problem asks for a sample result, population result, one-tail result, or two-tail result.
- Review the formula and make sure it matches the convention used by your course, worksheet, or report.
- Compare the result with the worked examples if you are learning the method.
- Round only at the final step unless your instructions require a specific precision.
- Keep a copy of the input data if the result needs to be checked later.
Worked examples
Z-test example
Input: Sample mean 52, hypothesised mean 50, standard deviation 10, n = 64
Calculation: z = (52 - 50) / (10 / sqrt(64)) = 1.6.
Result: Two-tailed p-value is about 0.11.
T-test example
Input: Sample mean 18, hypothesised mean 20, sample SD 4, n = 16
Calculation: t = (18 - 20) / (4 / sqrt(16)) = -2, df = 15.
Result: Use the t distribution because the population SD is not known.
What the result helps you decide
A hypothesis test helps you compare a sample result with a hypothesised value and decide whether the difference is large relative to expected sampling variation.
For school and university work, the result is often only one part of the answer. You may still need to state assumptions, show working, define variables, and interpret the result in words.
Z-test vs t-test
| Test | Typical use |
|---|---|
| Z-test | Known population SD or large-sample approximation |
| T-test | Sample SD with smaller or moderate sample size |
P-value interpretation
A p-value is the probability of seeing a result at least this extreme if the null hypothesis and model assumptions are true. It is not the probability that the null hypothesis is true.
Common mistakes to avoid
- Mistake 1
- Do not treat statistical significance as practical importance.
- Mistake 2
- Choose one-tailed tests only when the direction was planned before seeing the data.
- Mistake 3
- Biased samples can make a hypothesis test meaningless.
Edge cases
- Very small datasets can produce unstable summaries and wide uncertainty.
- Rounded inputs can slightly change final answers, especially in multi-step calculations.
- Different textbooks and software packages may use different percentile or quartile conventions.
- A statistically valid calculation can still be misleading if the data is biased or measured poorly.
Limitations
This calculator is for general educational information only. It does not prove that a statistical model is appropriate, that a sample is representative, or that a result is practically important.
- Check the formula convention required by your teacher, exam board, software package, or research method.
- For professional research, engineering, clinical, legal, or financial decisions, verify results with a qualified person.
- Use the calculator as a transparent estimate and keep the original data available for audit.
Frequently asked questions
What is the null hypothesis?
It is the baseline claim tested against the sample data.
What is a p-value?
It measures how unusual the sample result would be under the null hypothesis and model assumptions.
When should I use a t-test?
Use a t-test when the population standard deviation is unknown and you estimate it from the sample.
Does p < 0.05 prove a claim?
No. It is evidence under a chosen model, not proof.
Can I choose the tail after seeing the result?
No. Tail choice should be made before analysis to avoid bias.
Related calculators
- Confidence Interval Calculator
- Z-Score Calculator
- T-Distribution Calculator