Replication

From EvaluationWiki

Successful programs are often emulated. The question of whether the emulations are in fact a replication (a.k.a., implementation) of the original is a difficult one to answer; even the designer or manager of the original version may not be a reliable judge of replication. Good program evaluation should try to cover the issue of generalizability of the original (a.k.a., external validity) but it is extremely expensive as well as difficult to determine this with any certainty. The issue of course centers on the two questions whether (i) the original staff, or (ii) the original environment, were major contributors to its success, and if so, in exactly what respects, and, to make the issue harder, one also needs to know (iii) to what extent were those two factors interdependent.