Questions about StatEVN
Short answers to what people ask most. Something else on your mind? Ask through the contact form, or read the step-by-step Help.
How is this different from asking ChatGPT?
A chatbot writes code you can't see and can be confidently wrong about the numbers. StatEVN runs real statistical models on your data, checks them the way a statistician would, and tells you when a result can't be trusted. The same file gives the same answer every time.
Is an AI making up my results?
No. Every number comes from the fitted model, and every sentence of the explanation is built from those numbers by fixed rules. Nothing is guessed or invented.
Do I need to know statistics?
No. Type what you want to find out, like What drives revenue? In Amateur mode StatEVN picks the columns and the model and explains the result in plain language. If you do know statistics, the full output (coefficients, p-values, diagnostics) is one click away, and Advanced mode lets you choose everything yourself.
How do I know I can trust a result?
Before it answers, StatEVN runs the standard checks, such as uneven errors, overlapping inputs and errors that carry over in time, and it tests the model on rows it never saw. Each result says how much to trust it, and you get a clear warning when something is off.
What happens to my file?
It is used only to run your analysis and is not kept, unless you choose to go deeper with more files. Then it is kept for a limited time, and you can delete it whenever you like. It is never sold or used to train anything. See the Privacy Policy.
What data can I use?
Any table in CSV, Excel or JSON, or a table pasted from Excel or Google Sheets, with one row per observation and the column names in the first row. Plans allow from 5,000 to 500,000 rows per file. Analyst and Pro can also read from a PostgreSQL, MySQL or SQL Server database. For an A/B test (from Starter) you only need your totals.
Can I choose the model myself?
Yes. In Advanced mode (Analyst and Pro) you choose how much StatEVN does: Check and prepare (StatEVN ticks the fixes it would make; you keep, change or untick each one, then pick the columns and the model), or the Pro workbench, where you edit the table and write Python yourself. Every result lists the steps behind it.
Can I use it for my thesis or research?
Yes. You get the full statistics and the checks your supervisor will ask about, with an explanation of what each one means. Students get a free plan with 30 analyses a month.
What does it cost right now?
Nothing. StatEVN is in beta: the example needs no sign-up, and every beta tester gets Pro at no cost. Paid plans are not open yet.
How do I start?
Run the example first: one click runs a real analysis on a sample file. To connect your own data, ask for beta access and you will get a personal login by email.