Opinion Formation Research Paper

A second divide is more explicitly political. The issue here is where to locate public opinion and, by implication, where to look in order to explain it. One view assumes the public is a composite phenomenon, an aggregate function of its individual citizens. In this view, political meaning and value are attributes of individuals. Analysis therefore focuses on the qualities of cognition and, in this light, how individuals process the experiences and social influences to which they are exposed. The opposing view presumes that the public is a collective entity, a corpus of its own. Here political meaning and value are emergent properties of the collective. They are viewed as essentially cultural phenomena and are explained with reference to large-scale social forces.

1. The Mainstream Perspective: Publics And Meaning As Aggregates

Although these debates remain central in contemporary political theory, a consensus emerged in the empirical study of public opinion formation. Its basic contours were defined by Allport’s (1935) seminal discussion of attitudes, the Yale persuasion studies (Hovland et al. 1953), and Converse’s (1964) analysis of political belief systems. Although pursuing different topics from different disciplinary perspectives, the research is oriented by the same political and epistemological vision: the public is an aggregate of its individual citizens, and political meaning is a matter of calculated or learned associations among representations and preferences. In this light, the formation of public opinion is understood to be a process that revolves around individuals. It begins with their exposure to politically relevant experiences and information. Each individual processes this information, thereby coming to a judgment that yields an attitude. The attitudes of different individuals are then aggregated, either through informal interactions or more formal mechanisms, such as elections or opinion polls.

For the most part, this research effort has focused on the second step in the opinion formation process, the social psychology of how individuals process information (either from direct or indirect experience) when forming an attitude. The critical outcome or dependent variable is an attitude. Typically, this is defined as an individual’s overall evaluation of an object with some degree of favor or disfavor. In this view, an attitude consists of three types of elements: cognitive (beliefs); affective (emotions or feelings); and behavioral (intent to act). These elements can affect one another. While they are usually consistent, they may not be. Attitudes are also related to one another.

In a way that has important consequences for attitude strength and stability, this interattitudinal structure may vary with regard to its elaboration, its coherence, and the degree to which it has a hierarchical structure. A central aim of the research is to examine how different conditions (both situational and dispositional) affect the processing of information and thus the resulting attitude. The research is designed on the presumption that all individuals operating under similar conditions will process information in the same way. Consequently subjects’ responses are aggregated, and inferences are made regarding how the ‘average’ person will form, maintain, or change their opinions under the conditions examined.

Adopting this general framework, a wide variety of competing theories or models of public opinion formation were presented from the 1950s to the 1980s. Some posit a rational, almost logical, integration of information. A prominent example is the value-expectancy model (e.g., Fishbein and Ajzen 1975). According to this model, an individual forms or changes an attitude by reviewing their beliefs about the attitude object. A weight is assigned to each belief by multiplying its assessed value by the probability that the belief correctly pertains to the object in question. The expected values of all the beliefs are then averaged, leading to an overall attitude or judgment being formed.

Other models accept the basic features of such rational computational models but focus more on considerations of memory and its effect on what beliefs will be considered. The suggestion here is that an individual may have a diverse set of contradictory beliefs or affects associated with an attitude object. Situational and internal cues affect which of these beliefs is likely to be remembered on any given occasion. Consequently, although individuals may process beliefs rationally, the ways in which memory functions will bias which beliefs are processed and therefore what judgment is finally made. How mass media prime candidate evaluations and the apparent instability of political opinions are readily explained in these terms (e.g., Zaller 1992). On-line processing models of political candidate evaluation de-emphasize the importance of recall (e.g., Lodge and Steenbergen 1995). In this view, individuals do not judge an object by recalling all relevant beliefs. Rather, they simply remember their prior evaluation of the object and adjust this in light of the new information provided. While this information may then be forgotten, the overall assessment is carried forward when further judgment is required.

Other models of information processing suggest that judgment is more irrational. One example is the family of cognitive consistency models which suggest that individuals relate objects to one another, not only on categorical or logical grounds, but also on the basis of evaluation or feeling. A particularly influential extension of this is Tajfel’s (1981) analysis of the consequences of group identification. His research suggests that a person’s desire to maintain self-esteem (positive evaluation) and their negative identification with an out-group (categorical link) will result in negatively biased evaluations and behaviors toward that group. Following a learning theory tradition, other research demonstrates how regularly pairing an unconditioned stimulus (an object already valued positively or negatively) with a conditioned stimulus (an object not valued) will shape the subsequent evaluation of the condition stimulus.

In recent years, there has been a shift to more eclectic dual processing models of attitude formation. Self-consciously integrating earlier conceptions, these models suggest that different processes will operate under different conditions. Two types of processes are distinguished. Central processes, such as the value expectancy model, are relatively demanding and involve the evaluation of relevant arguments and information. Peripheral processes are less demanding than central processes and include classical conditioned responses, reliance on peripheral situational cues (e.g., physical attractiveness of the source), or the use of simple heuristics (e.g., experts know what they are talking about). The focus of this research is to discover (a) what conditions foster what type of processing and (b) how different kinds of processes affect attitude formation under different conditions. This has led to a reconsideration of the early research on the persuasive effect of source variables (e.g., credibility, expertise), message variables (e.g., argument quality, personal relevance), recipient variables (e.g., prior attitude, personality), and context variables (e.g., distractions, audience reaction). The net result is a more complex and contingent understanding of how attitudes are formed, sustained, and changed (Petty and Wegener 1998).

Political scientists have drawn upon these models in their attempt to understand opinion formation in mass publics. Thus they have focused on who is presenting what information (the agents of socialization—family, workplace, school, social groups, and the mass media), to whom, and with what effect on individuals’ judgment and knowledge of political leaders, issues, and events. The research on agent effects on opinion formation suggest that whereas the family is not as influential as initially believed, it does provide individuals with one of their most enduring characteristics: their political identity. Supporting early survey research, contemporary work indicates that group identification can influence powerfully individuals’ attitudes toward issues, especially those emphasized by the group. Research also indicates that the mass media, while not the powerful persuader it was initially feared to be, nonetheless is able to affect people’s sense of political priorities and to prime the specific considerations they may bring to bear when judging a particular issue or candidate.

In the case of knowledge and sophistication, the evidence suggests overwhelmingly that the vast majority of people have little accurate information about political leaders, current legislation, or foreign affairs. Nor do their judgments reflect much sophistication or consideration. What attitudes they express do not cohere in any apparent ideological fashion, nor are they particularly reliable over time. Even slight changes in how questions are posed can produce different attitudinal responses. Some studies do show correlations between apparently abstract and more concrete attitude items. This result is interpreted to suggest people do base particular judgments on more abstract core values. However evidence from open-ended interviews indicates that whereas a few people do think about politics in this deductive way, most do not. Their responses are generally ill-considered, unjustified, and readily changed by new considerations that pop to mind. Overall, the evidence suggests that citizens lack the information, understanding, or considered judgment typically called for in democratic political theory.

2. Alternative Direction: Constructivism, Discourse, And Collective Publics

Questions have been raised regarding the basic assumptions orienting this dominant approach to the study of public opinion. The results of open-ended interviewing have sparked an increasing interest in taking a more constructive, holistic view of political thinking. This has lead to a renewed interest in constructivist psychologies that consider basic differences in how individuals reason (e.g., Rosenberg 2002). In this view, how an individual thinks about social or political events will determine the kinds of interrelationships and the kinds of representations (and interests) they can construct. Moreover, different individuals may think in qualitatively different ways. For example, some people tend to focus on a simple causal or categorical relationship cued by the situation when formulating an understanding, and they draw primarily on social convention or group interests to make their judgments. Other people focus on only the concrete contours of the immediate situation and draw primarily on present feelings to make a judgment. Still others infer quite abstract issues that are only implicit in the situation and judge them with reference to some notion of the general interest or abstract principle. Contrary to the common assumption of the ‘average citizen,’ this work suggests that people may differ significantly, both in how they are affected by the same social influences and in their capacity to respond to different kinds of institutional regulation.

This interest in reasoning dovetails with the dissatisfaction expressed by some political theorists regarding representative democracy and its emphasis on electoral mechanisms of preference aggregation. Instead, they advocate more deliberative forms of democracy that require individuals to cooperate in the generation and resolution of political issues (e.g., Habermas 1996). In this light, considerations of the requirements of citizenship shift. Rather than personal judgment, the focus is on cooperation with others in order to come to common understanding and judgment. This involves taking the perspective of another, relating it to one’s own, and then critically considering both. The latter not only entails compromise, but the self-conscious construction of new positions that incorporate and go beyond the original ones. Research motivated by these concerns seeks to determine how differing institutional arrangements impact on deliberation and with what effect on the collective judgments that ensue. It does so by eschewing a focus on individuals in favor of a direct examination of the quality of deliberations themselves and the social representations and group opinions they produce.

Finally, there is renewed interest in a more holistic, constructivist view of the public. Rejecting what is referred to as methodological individualism, social theorists have argued that consciousness and meaning making are essentially collectively structured phenomena that are best thought of as a culture or mode of discourse. Some view culture as singly structured and coherent, and others argue that it is essentially an arena of conflict. However, all agree that the individual’s understanding is a generally poor reflection of a larger social process and an inappropriate object of inquiry. In this more sociological vision, changes in public opinion reflect changes in the underlying structure of collective meaning. Examples include the shift in public discourse from appeals to the sacred to reasoned argument regarding the profane, or a population coming to imagine itself as a nation (e.g., Anderson 1991). Such changes are explained with reference to collective forces such as the form of the relations of production, the mode of information, or a new geographical positioning of a population. With such changes, alternative forms of institutional regulation become possible (e.g., rule by personal power vs. rule by law) and new kinds of political values are constructed (social tradition vs. principled critique). The challenge to this line of inquiry is to develop an empirical methodology as sophisticated as the theorizing.

Bibliography:

1. Allport G W 1935 Attitudes. In: Murchison C A (ed.) Handbook of Social Psychology. Clark University Press, Worcester, MA
2. Anderson B R O’G 1991 Imagined Communities: Reflections on the Origin and Spread of Nationalism. Verso, London
3. Converse P E 1964 The nature of belief systems in mass publics. In: Apter D E (ed.) Ideology and Discontent. Free Press, New York
4. Fishbein M, Ajzen I 1975 Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research. Addison-Wesley, Reading, MA
5. Habermas J 1996 Between Norms and Facts. MIT Press, Cambridge, MA
6. Hovland C I, Janis I L, Kelley H H 1953 Communication and Persuasion: Psychological Studies of Opinion Change. Yale University Press, New Haven, CT
7. Lodge M, Steenbergen M 1995 The responsive voter: Campaign information and the dynamics of candidate evaluation. American Political Science Review 83: 309–26
8. Petty R E, Wegener D T 1998 Attitude change: Mutiple roles for persuasion variables. In: Gilbert D T, Fiske S T, Lindzey G (eds.) Handbook of Social Psychology, 4th edn. McGraw-Hill, New York, Vol. 1
9. Rosenberg S W 2002 The Not So Common Sense: Different Ways People Judge Social and Political Events. Yale University Press, New Haven, CT, Chap. 8, pp. 323–90
10. Tajfel H 1981 Human Groups and Social Categories: Studies in Social Psychology. Cambridge University Press, Cambridge, UK
11. Zaller J R 1992 The Nature and Origins of Mass Opinion. Cambridge University Press, Cambridge, UK