1. A practice which practitioners know is feasible to implement because they have implemented it.
2. A practice which practitioners think probably improves outcomes (but they are not making a strong high-level outcomes/impact evaluation attributional claim for it).
3. A practice which independent evaluators (or reviewers) of some sort think probably improves outcomes (but they are not making a strong high-level outcomes/impact evaluation attributional claim for it).4. A practice for which someone has made a strong high-level outcomes/impact attributional claim (i.e. they have claimed that they have proof that the practice improves high-level outcomes).
Dr. Duignan argues that best practice is a combination of all four definitions, however, according to him, it is mostly like the first definition.
He also goes on to state that people occasionally criticize the use of the term 'best practice' because it suggests that there is only one way of doing things. He claims that those who argue from that point of view sometimes look at it in terms of 'good' rather than 'best' practice and 'practices' rather than just using the singular term 'practice'. This critique, however, is a matter of wording rather than an actual critique of the notion.
The strongest critique that Dr. Duignan points out is that best practice is difficult to transfer to other settings without starting the process over again. Basically, one cannot assume that just because it is a best practice in one setting, that it will be the best practice in another. This is why it all depends upon the context in which it is being applied.
Finally, Dr. Duignan points out that one of the greatest challenges with best practice is actually getting practitioners to implement it once the practice has been identified. Best practice studies are constantly being conducted. However, Duignan states that the findings of these studies "are not systematically taken up by the bulk of practitioners undertaking programs and interventions in an area."