1. a type of program evaluation, either a formative evaluation or a summative evaluation, carried out at several sites. Each site has the same evaluation objectives, which are assessed in a coordinated effort by different evaluators in a continuous process. Information so obtained is then shared to enable common program outcomes to be assessed and to identify elements that contributed to the failures or successes of the program.

2. a strategy for accumulating information in evaluation research that involves combining and reconciling studies with somewhat different conclusions. This approach suggests criteria for determining when data from dissimilar studies can be pooled.

3. an approach in which individual evaluators in separate projects collaborate with an overarching evaluator. See also metaevaluation.