Peer review rarely involves a single reviewer. While a single paper submitted to a journal may be reviewed by just one person, the collection of all submissions to the journal usually has a great many reviewers. Individual grant proposals are almost always reviewed by several people, and those reviewers often form a subset of a larger group of people reviewing many of the submissions. Because of the large number of people involved in the process, fairness dictates that the reviewers apply similar standards. When the proposal or paper is sent to a reviewer, it should come with instructions (often in the form of questions) that make clear the basis for judging the merit of the submission. While the reviewer may disagree with the standards, it is not fair to unilaterally change the standards. That would result in some of the submissions being treated differently from others. Changes to standards should only be made in conjunction with all the other reviewers and, usually need to be made prior to the solicitation of submissions.

When applying standards, the issue of bias naturally arises. Some biases are appropriate. It is fair to give a grant application a lower score if it is clear the applicant does not understand the subject of the proposal. Other biases, however, are inappropriate, and limiting their influence can be quite a challenge to the reviewer. Biases that are the result of a direct conflict of interest are discussed in a later section. Biases may also have a more academic origin. If there are two competing theories and you are a strong proponent of one, you may find it difficult to be objective in reviewing a paper that supports the other theory. If one of the submitters is an acquaintance whom you do not particularly care for, it again may be difficult to maintain your objectivity. In cases like these, it may be useful (where possible) to preface your review with a statement of your particular perspective. Disclosure allows the person overseeing the review process to decide if an additional reviewer should be brought in.