StatEze

Statistics for clinical research, without a statistician

Statistics you can put your name on

Upload your dataset. StatEze picks the correct test through a transparent decision tree and computes every number itself. No statistic is ever guessed by an AI, and every p-value recomputes to the same decimal. You get a clean, journal-formatted analysis: Table 1, AMA-style Results, publication figures, and Word, PDF and LaTeX export. You stay the author. Start free. No credit card.

Start free See pricing

An AI never touches a number. Instead of reformatting SPSS output by hand or waiting weeks on a statistician, you get reproducible results a reviewer can recompute. See how we check our work

What StatEze won't do

It won't think for you, and it won't pretend to. That is the point.

StatEze does the mechanical part that is easy to get wrong, so you can spend your attention on the science.

Don't trust us. Recompute us.

Same data, same tests, same numbers. Every test family StatEze runs is checked against an independent reference value, produced outside StatEze's own code, before it ships. And you do not have to take our word for it: we publish worked examples on real, widely cited clinical datasets, each with the chart, the result StatEze computed, and the data. Download any of them, run it in SPSS, and reproduce every value yourself.

See the worked examples Read the validation summary

What you actually get

Upload your dataset, choose from the suggested analyses, and paste the written result straight into your manuscript. A Results sentence generated by StatEze:

Thickness was significantly higher in the ulcerated group (median (IQR), 3.54 (2.92); n = 90) than in the non-ulcerated group (median (IQR), 1.29 (1.29); n = 115); U = 8519.50; P < .001; r = -0.65, large effect; 95% CI, -0.75 to -0.52; Hodges-Lehmann median difference = 2.09; 95% CI, 1.61-2.58.

Generated by StatEze from a published clinical dataset (the Andersen et al. melanoma cohort). Recompute it yourself.
Start free

Free during open access. No credit card.

Common questions

Does AI calculate my statistics?
No. Every test statistic, p-value, effect size, and confidence interval is computed by SciPy and statsmodels, the same established libraries behind published research, through a fixed decision tree. An AI is never allowed to touch a number. An optional wording-refinement step, on by default, may smooth the prose of your Results; a language model can never change a value, and you can turn it off. That is why your results reproduce exactly and hold up in review.
What do I actually get?
A complete analysis of your dataset: the statistical tests, publication-quality figures, and Methods and Results paragraphs written in journal style. The output is built to be pasted into a manuscript, not exported and reworked.
Is it really free?
Yes. StatEze is in open access and every feature is currently free, with no credit card required. When billing starts, the Author plan is $29 a month or $200 a year, and founding members keep that rate for life.
How do I know the statistics are correct?
Every statistical method is checked against reference results computed independently of StatEze's own code before release. We also publish worked examples on real, widely cited clinical datasets from the published research literature, each with the chart, the result, and the data, so you can download any of them and reproduce every number yourself. Run your own analysis twice and you will get identical numbers each time. Read the validation summary.
Which analyses does StatEze run?
The everyday tests of clinical research: two-group and multi-group comparisons, paired analyses, correlations, regression, and categorical comparisons, in parametric and nonparametric forms. StatEze checks the assumptions and selects an appropriate test for your data.
Do I need to install anything?
No. StatEze runs in your browser. Create a free account, upload a dataset, and your first written results arrive in minutes.
Start free

Free during open access. No credit card.