Research › Evaluating sources: is it good enough to cite?
Evaluating sources: is it good enough to cite?
Not everything you can find is worth using. These checklists help you decide quickly whether a source is relevant, reliable and suitable for your purpose.
Why evaluation matters
Markers expect you to use evidence critically. A source can be well written and still be out of date, biased or unsuited to your question. Evaluating sources means asking questions about them, then deciding how much weight to give them, rather than rejecting everything that isn't perfect.
The CRAAP questions
The UCL library guide sets out the CRAAP questions for judging a source:
| Letter | Question |
|---|---|
| Currency | When was it published? Is recent material essential, or is older material acceptable? |
| Relevance | Does it address your topic, and is it at the right academic level? |
| Authority | Who are the author and publisher, and are they trustworthy? |
| Accuracy | Where does the information come from? Is there evidence, and can it be verified? |
| Purpose | Why was it produced? Is the purpose clear, and are there biases? |
The PROMPT questions
UCL also lists PROMPT, which asks about Presentation, Relevance, Objectivity, Method, Provenance and Timeliness:
- Presentation: Is the information clearly communicated in its language, layout and structure?
- Relevance: Does it match your needs?
- Objectivity: Does the author state a position? Is the language emotive? Are there hidden or vested interests?
- Method (for research reports): Is it clear how data was collected? Were the methods appropriate and trustworthy?
- Provenance: Is the source clear? Can you identify the authors or organisations, and how was it published?
- Timeliness: Is it clear when it was produced? Does its date meet your needs?
Types of source
| Type | Usually good for | Watch out for |
|---|---|---|
| Peer-reviewed journal article | Original research, evidence | Narrow focus, may be hard to read, can still contain errors |
| Textbook | Background, theory | May be out of date |
| Government or official report | UK statistics and policy | May reflect policy aims |
| News article | Current events, quotes | Speed over depth, headline spin |
| Company or campaign website | The organisation's own view | Selective use of evidence |
| Social media and blogs | Public opinion, examples | Anyone can post anything |
Peer review is not a guarantee
Peer review means other experts have checked the work before publication, and it is a useful signal. But it isn't perfect. Studies can be small, flawed or later contradicted. Treat peer review as one factor and still ask questions about the method and evidence.
Be cautious with metrics
The UCL guide advises caution about choosing sources on the basis of times cited, journal impact factor or author h-index. A paper with many citations might be famous for being wrong, and a little-cited paper might be perfect for your question. Judge each source on its own merits and check whether it is authoritative for the way you plan to use it.
Look for balance
The same guide encourages you to seek a balance of perspectives, including voices beyond predominantly Western, English-language sources, and to search widely for grey literature, unpublished documents and marginal publications. This is especially important for topics where one point of view dominates.
Check the claim
- Does the source give evidence for each key claim?
- Can you trace the numbers to an original source?
- Do other reliable sources agree or disagree?
- Who benefits from you believing it?
Writing about sources critically
Don't just summarise; compare. Say where sources agree, disagree and what the limits are. "Smith (2021) found X, but the sample was small and limited to one region, so..." shows critical thinking. See referencing for how to cite them, and searching the literature for finding them in the first place.
Frequently asked questions
What does CRAAP stand for?
Currency, Relevance, Authority, Accuracy and Purpose. They are questions to ask about any source.
Is a peer-reviewed article always reliable?
No. It is a useful signal, but studies can still be small, flawed or later contradicted. Always check the method and evidence.
Should I choose sources by citation count?
Be careful. UCL's guide advises caution with times cited, impact factor and h-index, and recommends judging each source on its merits.