This was an issue considered by the WHO Working Group on Evaluation and is discussed in depth in a chapter by McQueen and Anderson.

They first mention several definitions that are commonly used.

- The first is a common understanding of the term as in the case where the fact that the lake is frozen solid is an indication or sign that temperatures have been "very cold''. The second often is used in a legal context in which.evidence is introduced in a way to prove that something has occurred with certainty - i.e. "beyond a reasonable doubt". They note that in this context, evidence is often a mixture of stories, witness accounts, policy testimony, expert opinions and forensic science. That is, it frequently comes from multiple sources and from persons with widely ranging expertise and requires interpretation. The third definition refers to the product of observation, sometimes over a long period of time, sometimes described as "empirical", although in some cases, the observations have a theoretical perspective such as the Darwinian observations leading to a theory of evolution. Finally, evidence is sometimes described as something coming from "experiment".

- In spite of the apparent relevance of evidence to science and the increasing use of the term "evidence-based" among researchers, planners and policymakers, McQueen and Anderson point out that the literature in the history and philosophy of science does not contain extensive discussion about the term itself and in the field of "evidence-based medicine" the word evidence still remains a primitive term, without further specification and definition. Nevertheless, The use of the term in clinical medicine is generally within a broad hierarchy of proof resting on experiment in general,and the RCT in particular,even though there are critics of this approach within the medical community.