Which sources project-diada.org uses, which it does not, and how a study is judged before it becomes an article.
Where stories come from
We monitor a fixed list of research sources rather than the general news cycle:
- Peer-reviewed journals publishing psychology, behavioural science and psychiatry
- Research announcements from universities, government health agencies and research funders
- Specialist research-reporting outlets that cover psychology studies rather than wellness trends
- Official statistical sources — national health agencies, statistical offices and international health organisations
The list is reviewed periodically. Sources that go stale, become unreliable, or drift into commercial content are removed.
The evidence hierarchy
Not all studies carry the same weight, and we do not pretend otherwise. Strongest to weakest:
- Meta-analyses and systematic reviews of randomised trials
- Large, preregistered randomised controlled trials
- Longitudinal cohort studies
- Cross-sectional surveys and correlational studies
- Lab experiments, typically on small student samples
- Case studies and case series
Where a new study contradicts a stronger body of prior evidence, the article says so. One paper does not overturn a literature.
What lowers our confidence
Each of these is noted in the article, and enough of them together mean we do not run the story:
- A sample too small for the claim being made
- No control or comparison group where the design needs one
- Self-reported outcomes presented as objective measurement
- A press release claim that does not appear in the paper
- Many outcomes measured and one reported
- Funding from a party with a direct commercial stake in the result
- A journal with no visible peer-review process
- Authors whose public statements are more confident than their own data
Preprints
We do sometimes cover preprints, because interesting work appears there first. When we do, the article states in the text that the paper has not been peer reviewed, and treats the finding as an early signal rather than an established result.
Sources we do not use
- Other outlets’ articles as a factual source
- Press releases as a substitute for the paper
- Predatory or unreviewed journals
- Commercial white papers and industry-funded reports presented as research
- Wellness sites, supplement companies and clinic marketing
- Social media claims, however widely shared
- Secondary citations of statistics — we go to the originating body
Citation
Every article ends with a linked source list, primary paper first, with authors, year, title, journal and DOI. Links go to the source itself, not to our own coverage of it. Where an open-access version exists we link that too, so readers can check our work without a paywall.
Corrections to sourcing
If we have cited something wrongly, misattributed a finding, or missed relevant contradicting evidence, tell us at [email protected]. Sourcing errors are corrected the same way factual errors are.