Communication of science information.

There are obvious limitations to normative studies of the kind described. Many of the studies dealt with special content areas-for example, mental health-and thus may not provide a sound basis for generalization. Selection of the various samples presented still greater problems. In some cases, the universe was so ill-defined (for example, that of science writers) that conventional sampling procedures were of doubtful significance. There were also limitations of subject availability and funds. Thus, we tended to proceed on an ad hoc basis, using more-or-less intact groups where we could find them-a group of scientists or a group of science writers attending an NSF-sponsored seminar; news editors from papers in the State of Wisconsin; and so on.