How to Evaluate Sources When Research Gets Messy

Ask how close the source is to the fact you are trying to establish. A primary record may be stronger for what happened on a particular date, while a later synthesis may be better for broader context. “Primary” does not automatically mean “complete.”

Older material can be excellent historical evidence and terrible evidence for a current condition. If the subject can change, recency matters. Preserve older sources when they explain chronology, but do not silently treat them as current.

Multiple articles may all trace back to one original report. Count independent evidence, not the number of webpages repeating it. Follow citations backward when the distinction matters.

Two sources can appear to conflict because they measure different populations, time periods, categories or outcomes. Before deciding one is wrong, compare the definitions and methodology behind the numbers or claims.

If the available sources cannot resolve the disagreement, say so. Good research does not manufacture confidence. It shows what is supported, what is disputed, what is missing and what additional evidence would resolve the question.

Explore Source Evaluation & Research Quality →