Paste one section of your manuscript and get the first objection a Q1 reviewer would raise, with the exact sentence from your text. Free, no sign-up, on screen in seconds.
Paste the Method, the Results or whichever section you are writing, and I will tell you the first thing a reviewer would object to, with the exact sentence from your text.
Nothing to object to here. And that is not politeness: if there were something, I would say so.
Not a textbook list: these are the failures that keep coming back in each part of a paper. The reviewer picks the one that weighs most in your text and tells you, with the exact sentence.
Whether it promises something the design cannot support; whether it gives results without the numbers behind them (statistic, p, effect size, n); whether its conclusion is stronger than the paper’s own; whether the design or the sample are missing; whether it claims causation from a design that does not allow it.
Whether there is a real gap in the literature or just the claim that there is one; whether the hypotheses follow from the theory you have just laid out or appear out of nowhere; whether they are falsifiable and directional when they should be; whether cited papers actually support what you cite them for; whether the last paragraph announces aims the design never addresses.
Whether the sample size is justified; whether the instruments are validated in this population and language, with reliability reported in this sample rather than the manual’s; whether the design allows the inference you will want to make; whether inclusion and exclusion criteria are missing; whether the procedure could be replicated from what you wrote; whether ethics approval or consent are missing; whether obvious confounders go uncontrolled and unmentioned.
Whether every test comes with its statistic, degrees of freedom, p and effect size with an interval; whether assumptions are checked; whether there are multiple comparisons without adjustment; whether the numbers in the text match the tables; whether you interpret instead of report; whether a non-significant result is sold as a trend; whether the n of each analysis matches the one in the Method.
Whether the interpretation goes beyond what the data support; whether causal language appears with a correlational design; whether you generalize to populations you did not sample; whether the limitations acknowledge the real problem or only the harmless ones; whether hypotheses that were never tested are discussed.
It only judges the section you paste. If deciding something would require seeing another part of the manuscript, it says so separately, under “what I cannot see from here”, instead of turning it into an objection.
When you have the whole manuscript, Previum goes through it from start to finish with two simulated reviewers and a simulated editor. The Basic report, free, gives you the finding that decides with its quote and its fix, plus your references and p values checked; the Professional report, every objection and your own Word file commented in the margin.
Written by an AI from your text. Your section is not stored.