Sufficiency

Is there enough cited evidence to support the conclusion?

Generally, only “strongly” and not “weakly” supported conclusions should be accepted. The more controversial a claim is, the more evidence authors should provide before expecting an audience to accept it. If the evidence is not sufficient, the author may need to modify or qualify the claim, by stating that something is true ‘sometimes’ rather than ‘always’.

Typicality

Is the cited evidence typical or representative?

If an author makes a claim about a whole group but the evidence is based on a small or biased sample of that group, the evidence is not “typical.” Similar problems stem from relying just on personal experiences (anecdotal evidence) and from “cherry picking” data by citing only the parts that support a conclusion while ignoring parts that might challenge it.

Accuracy

Is the cited evidence up to date and accurate?

Authors using polls, studies and statistics must ask whether the data were produced in a biased way and also ask whether the sample was large and representative of its target population so that results were outside the “margin of error.” (Margin of error: If a sample is too small or not well chosen, results may be meaningless because they may represent random variation.)

Relevance

Is the cited evidence directly relevant to the claim(s) it is being used to support?

An author may supply lots of evidence, but the evidence may support something different from what the person is actually claiming. If the evidence is not relevant to the claim, the author may need to modify or qualify the claim—or even to acknowledge that the claim is indefensible.