How to take your results out as a report file
There is a second wall after the statistics. Six tables are on your screen, your supervisor wants a draft, the conference template takes Word, and the institutional report takes HWPX. Statory’s combined report screen does that work.
Seven steps
- Run a few analyses first. With no results there is nothing to put in the report.
- Go to /report.
- Reorder — move each item up and down with ▲▼. A paper’s order is not the order you ran things in.
- Edit the content — write the title, summary and notes yourself. "Save" keeps that edit with the analysis record, so the title you fixed follows the result the next time you put it in a report.
- Save new version — freezes the current state as one version. Versions accumulate, and you can pick one from the list and load it again. Earlier versions are not deleted.
- Compare with prior studies (optional) — pressing it matches each analysis against the paper cards and attaches two sections. It costs no credits — the comparison does not call an AI; it counts our own table.
- Export — choose PDF, Word or HWPX, with an academic or a business template.
What goes into those two sections
Theoretical discussion gives how many prior studies cover the same construct relationship, how many of them point the same way as your result and how many the opposite way, the median effect size, and verbatim excerpts from those papers. Policy implications carries the policy implications those papers state themselves, plus one line on how far your own result agrees in direction.
It costs no credits — the comparison does not call an AI; it counts our own table. What the engines were checked against is in the comparison table.
Three things worth knowing before they surprise you
With no evidence there is no section. We write neither an empty section nor a sentence saying "nothing found" — if a section is there, evidence is there.
The comparison needs to know what each variable measures. Names like Q1 or v23 match no paper. Names are read in this order: ① a construct written on the variable → ② the name you gave that scale group on the scale card → ③ the variable label → ④ the variable name. ① and ② count as "Confirmed"; ③ and ④ as "Unconfirmed" — only a name a person wrote counts as confirmed.
"n prior studies" means n in the Statory library. It is not n in the literature; it is the number of paper cards we hold. The screen and the exported file both say so.
A draft is saved in two places
This device (your browser) first, then your account. Clearing your browser data removes the first; deleting your account removes the second. The reason for two places is so you can carry on from another device — a draft made on your home computer should be visible on your lab computer.
If one of them fails, the screen says so. The same holds when the list cannot be read — it shows what this device has and says that is what you are seeing. Nothing is passed over quietly: a list that silently gets shorter can only be read as "my drafts are gone".
If you work without logging in there is no account, so it is saved on this device only. You are told that each time you save, too.
Three common snags
"Nothing shows up in the report." You need analysis results before there is anything to put in. Run one analysis first.
"My draft isn’t on the other computer." Check that you are logged in to the same account, and think back to whether the line "not saved to your account" appeared when you saved. If it did, that draft is only on the machine where you made it.
"Can I roll a version back?" Yes. Pick one from the version list and press "Load" to return to that order, those edits and that template. Earlier versions are not deleted.