Generalization

Generalization refers, broadly, both to the process of drawing a general conclusion from specific observations (e.g., generalizing about a large population from a much smaller sample) and to the conclusion that results (e.g., a generalization drawn about whether males or females more strongly prefer a specific political candidate). In the context of evaluation, the relevant question typically is: How well, based on a specific set of observations made in a particular evaluation conducted in a given setting at a particular time, can one draw more general conclusions about the policy, program, or practice in question?

Researchers often think about generalization in terms of generalizing from a sample, preferably a random sample, to a population. However, in evaluation (and elsewhere), the generalizations in which we are interested ...