Run t tests from raw values or summaries. Pick alternatives, set alpha, and add weights. See results above, then export for your next report.
| # | Group | Score | Weight | Paired A | Paired B |
|---|---|---|---|---|---|
| 1 | A | 51 | 1.2 | 70 | 68 |
| 2 | A | 49 | 0.8 | 72 | 71 |
| 3 | B | 55 | 1.5 | 69 | 66 |
| 4 | B | 52 | 1.0 | 75 | 73 |
| 5 | A | 50 | 1.1 | 71 | 70 |
Use group columns for two-sample tests, and A/B columns for paired tests.
p-values come from the Student t distribution. Confidence intervals use t critical values consistent with the selected alternative.
Survey data often includes sampling error and unequal representation. A t test helps quantify mean differences with uncertainty. This calculator supports practical survey analysis workflows for quick decisions.
Use one sample testing when you compare a survey mean to a fixed target. Enter summary statistics for speed during reporting. Paste raw values to detect entry mistakes and outliers.
Compare two independent groups such as districts, cohorts, or customer segments. Select Welch when variances differ or sizes are unbalanced. Enable pooled variance when spreads are similar and justified.
Paired testing suits repeated measures from the same respondents over time. It uses differences between matched answers for each person. This approach reduces noise from stable individual traits and bias.
Survey weights adjust influence when respondents represent different population shares. Large weights can inflate precision if ignored. The tool uses Kish effective n for approximate degrees of freedom. This reduces overconfidence when weights vary strongly across records.
Confidence intervals show plausible mean differences at your chosen alpha. The p value summarizes evidence against the null under assumptions. Review the interval direction for one sided tests. Export CSV and PDF outputs for audit friendly reporting and sharing.
Choose alpha values that match your risk tolerance and standards. Track t, df, p, and intervals in your report. Use charts to explain skew and outliers clearly. If results surprise you, verify coding, missing values, and scales first. Save exports with dates for review trails.
A survey t test compares a mean or mean difference with sampling uncertainty. With weights, this tool approximates degrees of freedom using effective sample size for practical reporting.
Use a paired test when the same respondents answer twice, like pre and post surveys. The test works on within person differences, which often increases power and reduces noise.
Welch handles unequal variances and unbalanced group sizes more safely. It adjusts the degrees of freedom to keep error rates more stable than pooled variance under mismatched spreads.
Weights change the mean and variance by giving some responses more influence. The calculator uses Kish effective n to reduce inflated precision when weights vary strongly across respondents.
Charts require the individual values to show distributions and pair patterns. Summary inputs cannot show shape. You still get t, df, p, and intervals with summary statistics.
This is a helpful approximation for many weighted surveys. Clustered or stratified sampling needs design based variance estimation. For critical work, confirm results with dedicated survey methods software.
Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.