Iyengar, S. (1987). Television news and citizen’s explanations of national affairs.
The American Political Science Review, 81(3), 815-832.

Citizens are only fleetingly acquainted with current events and very few utilize ideological precepts to organize their political beliefs (for a review of research, see Kinder and Sears 1985). The low level of political knowledge and the absence of ideological reasoning has lent credence to charges that popular control of government is illusory (with respect to U.S. public opinion, for example, see Schumpeter 1950; Toqueville 1954).

~snip~

Explanation is an essential ingredient of human knowledge. To explain events or outcomes is to understand them: to transform the "blooming, buzzing confusion" of today's world into orderly and meaningful patterns. Psychological research has demonstrated that causal relationships feature prominently in individuals' perceptions of social phenomena (Nisbett and Ross 1980; Weiner 1985). In fact, causal thinking is so ingrained in the human psyche that we even invent causation where none exists, as in purely random or chance events. (see Langer 1975; Wortman 1976).

Explanatory knowledge is important to political thinking for two reasons. First, answers to causal questions abound in popular culture, making the task of explanation relatively inexpensive. One need not devour the pages of the Wall Street Journal or study macroeconomics to "know" why there is chronic unemployment. Second and more important, explanatory knowledge is connotative knowledge. To "know" that unemployment occurs because of motivational deficiencies on the part of the unemployed is relevant to our attitudes toward the unemployed and our policy preferences regarding unemployment. In other words, explanatory knowledge is usable knowledge. Simple factual knowledge, on the other hand (e.g., the current rate of unemployment), does not so readily imply political attitudes and preferences. It is not surprising, then, that opinions, attitudes, feelings, and behaviors in a multitude of domains are organized around beliefs about causation (for illustrative research, see Schneider, Hastorf, and Ellsworth 1979, chap. 2). In fact, causal attributions exert such a powerful hold on behavior that "misattribution" techniques have proven effective in treating behavior disorders (see Fiske and Taylor 1984, 36-39), in inducing "prosocial" behaviors (see Schneider, Hastorf, and Ellsworth 1979, 93-95), and even in strengthening the general sense of psychological well-being (see Langer and Rodin 1976).