Guides to reading science honestly
A paper being published does not mean it is right. These are the checks a reader can actually run, who screens science for problems, and where each safeguard stops. All free, all sourced, none of it assuming you are a scientist.
Get the next issue free
Five papers free every issue, about 15 emails a month. No spam. Leave in one click.
All guides
11 pieces
Can you trust this study?
Start here if you have a specific paper or a headline in front of you.
- How to tell if a scientific study is trustworthy The checks you can run yourself in a few minutes, and the ones that need an expert. What a journal name does and does not tell you.
- How to check if a paper was retracted The free ways to find out, step by step, plus why a paper can be wrong for years before anyone withdraws it.
- What a retraction is, and what it means for you What happens when a journal pulls a paper, who decides, and why a retraction is not always an accusation of fraud.
How often science goes wrong
The scale of it, with the sources shown.
- How many papers get retracted? The sourced numbers The real figures and the trend behind them, and what those numbers honestly do and do not prove about science.
- Why retracted papers are still cited A withdrawn paper keeps being cited for years, and the AI research tools people now trust repeat it without a word of warning. What the studies actually found, and how to avoid citing one yourself.
Who checks science, and where they stop
The safeguards that exist, named, including their limits.
- Who checks research integrity? The tools, named Papermill Alarm, the STM Integrity Hub, Proofig, PubPeer and the Retraction Watch dataset: who screens what, and the gap none of them covers.
- How Lucent checks every paper before you read it Our own method written out in full, including, plainly, what it cannot tell you.
- We re-checked every paper we have published The whole archive re-screened against Crossref, with the full list and the date of the run. Including why a clean sweep proves less than it looks like it does.
Where to read new science
Honest comparisons, including where we are not the right answer.
- The best plain-language science newsletters What is out there and who each one is actually for, written without pretending we win every comparison.
- Lucent vs Retraction Watch: what each one does Where the two overlap, where they do not, and which one you actually want for your question.
- Lucent vs Nature Briefing: what is different Two different jobs, compared without spin.