To tell if our training program was a success, we must evaluate it.  Whether transfer occurred is part of this evaluation.  A more classic perspective is Kirkpatricks.  He came up with a 4-level way to evaluate whether training worked.  His most basic level was reactions - were the reactions of the trainees generally positive, or generally negative?  Did they LIKE the training?  Then there was learning.  Did they learn anything from the training?  Next came behavior, whether there was a change in the behavior of the trainees as a result of the trainees (this is where we come closest to the idea of transfer in his system).  Finally, he had results as a criterion for success - and here, we were asking the question, Was there a change in the bottom line as a result of training?  It could be asked of the organization - we implemented a new training program, are we making more money? - or it could be asked of the individual - I went through advanced training on how to wait tables, am I getting more tips?  You can break Kirkpatricks standards down into criteria that are internal to the trainee - how did they react to the training, and did they learn? - and ones that are external - does the behavior change, and are the results different.  We can argue pretty forcefully that for the company, the external standards are more important - but theyre also the hardest to leverage (that is, affect) and the hardest to develop.

Recently, researchers and practitioners have started moving past Kirkpatricks criteria and talking more about a holistic approach to training effectiveness.  At a broad level, did it work?  This includes multiple perspectives.  The trainer, trainees, organization, customers, and all sorts of folks are part of the assessment of whether training made a difference, and whether the goals of training were met.