Preprint servers and what screening they do posts 91–107
This is a continuation of a long topic, addressed by post number rather than by page. Start at post 1.
Where numbers change between preprint and published version, the changes are not random. They tend toward a smaller effect and a more carefully hedged claim.
Collapsed as off-topic by two members at trust level 3 or above
Post #91 and I disagree about the size of the effect, not about the direction.
Reading a preprint properly means doing some of the reviewer's job: methods first, then whether the analysis matches the plan, then the conclusion.
Same conclusion as the reply above, reached differently, which is mildly reassuring.
Everything in post #95 holds. The case it does not cover is the one I have.
Say it is a preprint in the first line rather than the last. Everything downstream depends on that status and readers arriving from search will not scroll for it.
A note on scope: what I am saying about preprint servers applies to the case in the first post and I would not extend it further without checking.
Post #96 is right about the mechanism and I think understates the practical bit.
Always label it as a preprint in any discussion. A preprint is not a published trial and it is not settled knowledge. Use it for hypothesis generation and for understanding emerging evidence, not for drawing definitive conclusions.
Genuinely open to being wrong about this one.
How to cite a preprint: author last name, first initial, et al. Title. Server name [preprint]. Posted date. Doi. Include the server name and note that it is a preprint.
One of those cases where knowing the mechanism does not help the decision.
Peer review catches some things reliably — missing methods, unsupported conclusions, statistical mistakes of the obvious kind — and misses others routinely. Knowing which is which makes preprint reading easier.
Not the whole picture, but the part of it I can speak to.
Picking up post #102: that is the part I would want checked first.
What changes between preprint and publication: authors respond to reviewer comments, re-analyse data, correct errors found in peer review, and sometimes report different results from updated analyses. The published version is not the same as the preprint.
Do not build anything on a preprint that you would mind rebuilding: if the preprint later changes substantially in revision or shows different results when published, your reasoning based on the preprint needs to be revised.
That matches what I was told, which is not the same as knowing it.
Where I have landed on preprint servers, having got it wrong once in public: the direction is clear, the magnitude is not, and anyone quoting a precise magnitude has borrowed it from somewhere that did not measure it.
This is the sort of exchange that makes the archive worth searching.
This topic was referenced in
- When a preprint is the best available evidence and we say soEvidence › Preprints · 160 replies
Suggested topics
| Topic | Participants | Replies | Views | Activity |
|---|---|---|---|---|
|
A preprint whose numbers changed substantially at publication — what changed since
On the subject in the title: A preprint whose numbers changed substantially at publication — what changed since Working notes rather than a conclusion. A preprint question rather than a question about the…
|
+3 | 7 | 125 | 8mo |
|
Coming back to: Why a preprint's supplementary material is often the best part
Why a preprint's supplementary material is often the best part — that is the question, and I have not found it answered plainly anywhere I have looked. A question about how this site should cite something…
|
+109 | 116 | 36k | 2d |
|
When a preprint is the best available evidence and we say so — a second dataset
When a preprint is the best available evidence and we say so — a second dataset I have a specific reason for asking rather than idle curiosity, and the context is below. A preprint question rather than a…
|
+20 | 24 | 36k | 14mo |
|
Tracking a preprint to publication: a worked example — the long version
Posting this under the heading it deserves: Tracking a preprint to publication: a worked example — the long version Everything below is what sits behind that. A preprint question rather than a question about…
|
+38 | 47 | 560 | 22mo |
|
[2026 update] Conference abstracts as evidence: the weakest common citation
Conference abstracts as evidence: the weakest common citation — setting out what I have, and where I think it stops being reliable. Reading an unrefereed manuscript properly, and asking whether I am doing it…
|
2 | 48k | 7mo |
Related topics — sharing the tags preprint, effect size, needs a source
| Topic | Participants | Replies | Views | Activity |
|---|---|---|---|---|
|
Levothyroxine timing: a real interaction with a simple fix
Levothyroxine timing: a real interaction with a simple fix Writing it up because I had to work it out twice and would rather nobody else did. A narrow question about levothyroxine timing, deliberately narrow,…
|
+20 | 24 | 60k | 18mo |
|
Regression to the mean in progress reports
On the subject in the title: Regression to the mean in progress reports Working notes rather than a conclusion. Something about regression does not reconcile and I would like a second pair of eyes before I…
|
+30 | 34 | 49k | 12mo |
|
How this community labels and discusses unrefereed work — the long version
Asking directly, because I could not find a straight answer: How this community labels and discusses unrefereed work — the long version Reading an unrefereed manuscript properly, and asking whether I am doing…
|
+18 | 23 | 21k | 2mo |
|
AOD-9604 and the fragment hypothesis, revisited
Posting this under the heading it deserves: AOD-9604 and the fragment hypothesis, revisited Everything below is what sits behind that. Reading back through what has been written here about AOD-9604, three…
|
+40 | 44 | 2.8k | 6mo |
|
BPC-157: what the rodent literature actually shows, and what it does not — a second dataset
On the subject in the title: BPC-157: what the rodent literature actually shows, and what it does not — a second dataset Working notes rather than a conclusion. The question about BPC-157 that I actually want…
|
+77 | 84 | 54k | 11mo |