LIBRARY USERS 49 stantial differences in political opinion. In addition, the differences between library users and non-users were slight in comparison with differences associated with such factors as education or eco- nomic status or other predispositions. On the whole—and given the benefit of the statistical doubt—the library users were slightly more “alive” to the presidential campaign than the non-users, but ~other differences among people were much more important. . . From related investigations, the most probable interpretation of the differences in interest and activity between the two groups involves a general characteristic which might be called “cultural alertness.” Studies in this field have repeatedly identified a certain group of people who engage in all sorts of cultural activities, in the broad sense, more than does the rest of the community. They read more, and listen more, and talk more; they have more opin- ions and they feel more strongly about them: they join more or- ganizations and are more active in them; and they know more about what is going on. . . It is in this context that the differ- ences in political interest and activity between the library users and non-users can best be understood. Such differences probably do not arise because the former read more books than the latter, but rather because they are generally more sensitive and respon- sive to the culture in which they live—and book reading and library use are only one manifestation of that general character- iStic.t” All in all, the question of the relationship between library use on the one hand, and opinion formation and opinion leader- ship on the other, represents an undeveloped and promising area for library research. SUMMARY From the preceding analyses these conclusions may be drawn. The young use the library more than the old, the better educated more than the lesser educated, and women a little more than, and differently from, men. The public li- brary serves the middle class, defined either by occupation or “Berelson, 1945, pp. 297-98.