Source evaluation involves assessing credibility (author expertise, bias, currency of information), relevance (connection to your specific claim), and reliability (evidence of rigorous research or peer review). Writers develop judgment about which sources strengthen their argument and which, despite being well-written, lack sufficient authority for academic composition. Evaluation happens throughout research as you discover contradictions or unexpected weaknesses in initially trusted sources.
Evaluate the same topic using a peer-reviewed journal article, a recent news article, and a blog post. Assess each along multiple dimensions: author credentials, publication venue, presence of citations, and date. Compare your conclusions about source reliability.
Source evaluation is not a checklist you run once — it is a judgment you refine continuously as you write. You already know from finding-and-evaluating-sources that not all sources are equal, and from source-credibility-assessment that credibility depends on expertise, transparency, and accountability. The next layer is learning to evaluate sources *in relation to your specific argument*, not in the abstract. A source that is perfectly credible for one claim may be useless or misleading for another, even on the same broad topic.
The three dimensions worth separating are credibility, relevance, and reliability. Credibility asks who is speaking and why you should trust them: Does the author have domain expertise? Is the publication venue peer-reviewed, institutionally accountable, or editorially selective? Does the source disclose its methods and funding? Relevance asks whether this source actually serves your argument: A 2005 study may be highly credible but irrelevant if your claim requires current data. A primary source is irreplaceable if you're making a textual claim, but a secondary synthesis may serve better if you need an overview. Reliability asks whether the evidence within the source is sound: Are claims supported? Are counterarguments addressed? Is the sample size or evidence base adequate?
These three often come apart in practice, and that separation is where judgment enters. A newspaper article may be highly relevant to a claim about public opinion but unreliable as evidence of scientific consensus. A blog post by a recognized expert may be more credible than an anonymous institutional report, even though the blog lacks formal peer review. A peer-reviewed study from a prestigious journal may have been retracted or substantially challenged — currency matters. Part of your job as an academic writer is to account for these tensions explicitly: when you use a source with a known limitation, you acknowledge that limitation and explain why the source still earns its place.
Experienced writers also know that source evaluation is *recursive*. A source you initially trust may lose credibility when you discover it contradicts three more rigorous studies. A source you dismissed as too old may turn out to be the original finding that all later work builds on. Strong research involves tracking the conversation among sources: Do they cite each other? Where do they agree or diverge? Who is responding to whom? This conversation-mapping reveals the intellectual landscape of your topic and helps you position your argument within it, rather than simply piling up authorities. When you cite a source, you are implicitly vouching for it — your credibility as a writer is partly delegated to the sources you choose.