What is Source Evaluation?
Source evaluation is the process of judging whether a source is credible, accurate and relevant enough to use in your work. It decides which sources go into a literature review or an essay, and how much weight each one carries.
The CRAAP Test
The most widely taught checklist, from the Meriam Library at California State University, Chico, asks five questions:
- Currency: when was it published or updated, and is that recent enough for your topic?
- Relevance: does it address your question, at the right level for academic work?
- Authority: who wrote it, what are their qualifications, and who published it?
- Accuracy: is it supported by evidence and references, and has it been through peer review?
- Purpose: was it written to inform, to persuade or to sell, and does that create bias?
Lateral Reading
For websites and news, checking a source against itself is not enough. Lateral reading means leaving the page and searching for what others say about the author or organisation, the approach summarised in the SIFT method: stop, investigate the source, find better coverage, and trace claims back to their original context.
Source Evaluation Example
A blog post claims a new study shows that one cup of coffee a day improves memory. Evaluating it: the post names no author and links no study (authority, accuracy); it sells a coffee subscription (purpose). Tracing the claim finds the original paper, a small study of 40 adults published in a peer-reviewed journal last year (currency, accuracy). The paper is the source to cite, with its sample size noted as a limitation; the blog post is not.
Primary or Secondary?
Part of evaluation is knowing what kind of source you have. A primary source is original evidence, such as a study's data or a historical document; a secondary source interprets primary sources. Cite the primary source when you rely on its findings.
Best Practices
- Start from scholarly sources found in academic databases rather than general web searches.
- Cross-check important claims against more than one source.
- Be cautious with AI-generated summaries: check every claim and reference against the original.
- Record why you included or excluded a source; it makes the methods section of a review easier to write.