Abstract
Q analysis shows that among newsmen desk-bound editors are least able to predict subscriber preferences; but on the whole newsmen and readers show high agreement in their preferences.