Evidence and referencing policy

What counts as evidence here

The Hub prefers sources in this order:

1. Systematic reviews and meta-analyses of high-quality randomised controlled trials — the strongest evidence for whether an intervention works in practice.
2. Individual randomised controlled trials — useful when they address a specific population or question not yet covered by a synthesis.
3. Position stands, consensus statements and evidence-based guidelines from recognised professional or scientific bodies — for example the American College of Sports Medicine, the International Society of Sports Nutrition, and national physical activity guidelines.
4. Observational studies, including prospective cohort, case-control and cross-sectional studies — used to identify associations, generate hypotheses and describe real-world patterns, not to prove causation.
5. Mechanistic and physiological research — explains how an effect might occur, but is not treated as proof that the effect happens in people without supporting human outcome data.
6. Textbooks, narrative reviews and expert opinion, used primarily for established background and context and never to support a numeric recommendation on their own.

Referencing style: Cite Them Right Harvard

The Hub uses the Cite Them Right Harvard system throughout, as set out in the Open University’s quick guide to Harvard referencing. This applies consistently to in-text citations, reference lists, author formatting and source-specific reference formatting, so every article on the Hub references the same way.

• In-text citations are author–date in brackets at the point the source is relied upon — for example (Morton et al., 2018) — using “and” for two authors and “et al.” for four or more.
• Journal articles are referenced as: Author(s) (Year) ‘Article title’, Journal Title, volume(issue), page range or article number, with a DOI or link where one exists.
• Organisational web pages and reports are referenced under the corporate author, with the publisher, the link and the date accessed.
• Where several works by one author share a year, they are distinguished as 2024a, 2024b, and so on.

How references appear on the page

Substantive pages end with a reference list in alphabetical order by author. Each entry gives the full Cite Them Right Harvard reference and a link to the paper, abstract or official document, so you can check the claim rather than trust it. Every source in the list is cited in the text, and every citation in the text has a full entry in the list — the Hub does not pad a list to look authoritative.

Numbers come from one shared reference

Recurring figures — weekly sets per muscle, useful rep ranges, protein per kilogram of bodyweight, sensible rates of weight gain or loss, rest between sets — are held once in a single internal reference used by every page and by the Bodywise application. This prevents the common failure where two articles on the same site quote different numbers for the same thing.

When the evidence disagrees

Where good studies conflict, the Hub gives the range rather than picking a side, states what the disagreement is about, and tells you which practical choice matters and which does not. Where a popular claim is not supported, the page says so plainly instead of hedging.

What is never used as a source

• Supplement or equipment manufacturer marketing material.
• Anecdote, coaching folklore or social-media claims presented as evidence.
• Predatory or unindexed journals, and papers the author has not read beyond the title.
• Other content sites summarising research second-hand.

Limits of the evidence base

Much of the training literature uses small samples, short durations, and young or already-trained participants. Nutrition studies frequently rely on self-reported intake. The Hub states effect sizes as ranges for that reason, and treats individual variation as expected rather than as failure.

Reviewing and re-checking

Pages are re-read when relevant new evidence appears, when a reader reports a problem, or during a periodic sweep of the library. The date shown on a page is the date it was last genuinely reviewed.

Report a sourcing problem

If a claim looks unsupported, a link is broken, or a citation does not say what the page says it says, tell us through the contact page and it will be handled under the corrections policy.