NONATTITUDC

NONATTITUDE

Primary Disciplinary Field(s): Political Science, Public Opinion Research, Social Psychology

1. Core Definition

The term Nonattitude refers to unstable, often meaningless, survey responses given by individuals who lack a genuine, underlying predisposition or belief structure regarding the topic being queried. Originally conceptualized by political scientist Philip I. Converse in his seminal 1964 work, “The Nature of Belief Systems in Mass Publics,” the concept presents a critical challenge to the assumption that all respondents possess coherent and persistent political or social opinions ready to be measured. A nonattitude is characterized by its utter lack of stability over time; when asked the same question at different points, individuals expressing a nonattitude are likely to provide contradictory or random answers. These responses are often generated on the spot, driven by superficial cues, question wording, or immediate contextual factors, rather than being rooted in a durable belief system or ideological framework. Thus, a nonattitude represents an expression of temporary acceptance or rejection toward a survey solicitation that holds minimal or no predictive power regarding future behavior or complex political processing.

The core implication of the nonattitude thesis is the demonstration that public opinion, particularly regarding complex or abstract policy issues, is far more tenuous and poorly structured than democratic theory often presupposes. Converse argued that significant portions of the mass public—often labeled the inattentive public—do not engage in meaningful analysis of the political landscape necessary to form crystallized opinions. When forced to provide an answer in a survey setting, these individuals essentially “roll a die,” providing a response that masks the underlying absence of opinion. While superficially appearing as an attitude, the nonattitude fails the empirical test of persistence, coherence, and impact on subsequent data processing or actions. In severe cases, the response is merely an artifact of the survey instrument itself, reflecting compliance rather than conviction.

2. Etymology and Historical Development

The concept of the nonattitude emerged directly from intensive analysis of panel survey data conducted in the 1950s and 1960s, particularly the detailed interviews carried out by the American National Election Studies (ANES). Converse observed startling levels of response inconsistency when citizens were asked about specific policy issues, such as government involvement in housing or foreign aid, across multiple waves of the panel study. If attitudes were genuinely held, stability should have been high, yet the observed correlation across time points was often near zero for many demographic groups. This statistical instability led Converse to hypothesize that the measurable variance was not due to genuine attitude change, but rather to measurement error stemming from the respondents’ lack of underlying opinion structures.

Converse’s framework differentiated sharply between two types of publics: the politically sophisticated elites, who possessed complex and constrained belief systems where opinions logically connected to one another (high attitude stability), and the mass public, many of whom lacked such constraints (low attitude stability). The nonattitude hypothesis explained the instability observed in the latter group. This development was groundbreaking because it forced political science to confront the empirical reality that ideological thinking was not widespread, challenging previous assumptions rooted in democratic optimism about the informed electorate. The initial publication of this work generated immediate controversy, prompting decades of research attempting to either validate Converse’s findings or propose alternative models that could account for response instability without assuming a complete lack of opinion.

3. Key Characteristics

Nonattitudes possess several defining features that distinguish them from genuinely held opinions or even weakly held attitudes. These characteristics underscore why they are treated as statistical noise rather than substantive data points in political analysis.

• Low Persistence Over Time: The defining characteristic is temporal instability. When an individual with a nonattitude is re-interviewed, the probability of them giving the exact same response is only marginally higher than chance. This lack of test-retest reliability signals the absence of a fixed, internalized cognitive schema guiding the response.
• Minimal Impact on Data Processing and Actions: Since nonattitudes are generated randomly or contextually, they do not function as true attitudes should—namely, by serving as cognitive shortcuts to filter new information, organize political perceptions, or predict voting behavior. They have negligible predictive utility or behavioral consequence, contributing little to the individual’s overall political engagement.
• Low Constraint and Intercorrelation: Unlike genuine belief systems, nonattitudes do not correlate logically with other stated opinions. An individual holding a nonattitude on environmental policy is unlikely to have that position logically constrained by their stated position on taxation or military spending, revealing an unstructured outlook that lacks ideological coherence.
• Susceptibility to Solicitation and Framing: Individuals operating with nonattitudes are highly susceptible to changes in question wording, the order of questions, or minor contextual priming (solicitation). They tend to accept cues easily and demonstrate little intrinsic rejection capacity, meaning their “outlook” shifts dramatically based on how the query is framed or presented by the surveyor.

4. The Nonattitude as a Survey Artifact and Measurement Error

In practice, the nonattitude often manifests as a methodological artifact of the standard survey instrument. Surveys are designed under the implicit assumption that every citizen has an opinion on every public matter, regardless of how obscure or complex the topic may be. By forcing respondents to choose a stance (e.g., agree/disagree, liberal/conservative), the survey structure compels answers even from those who have not engaged in any meaningful analysis of the issue item, generating responses that meet the definition of nonattitudes.

This phenomenon means that traditional measures of public opinion frequently overestimate the degree of public engagement and ideological constraint. Converse suggested that responses derived from nonattitudes primarily contribute to classical measurement error, obscuring the true distribution of meaningful public opinion. The presence of nonattitudes necessitates statistical modeling techniques that explicitly account for this error component, such as latent class analysis or techniques designed to separate genuine attitude variance from random response variance. Failing to account for nonattitudes can lead analysts to misinterpret random fluctuations as evidence of widespread public volatility or rapid opinion change, thereby distorting conclusions about democratic stability.

5. Criticism and Alternative Models

While Converse’s work provided a powerful, pessimistic account of mass public opinion, it spurred significant scholarly debate that focused primarily on the interpretation of response instability. Critics generally accept the finding of low response stability but challenge the conclusion that this instability reflects an absolute lack of opinion (the “nonattitude”) rather than a probabilistic expression of weak opinions. Two major alternative frameworks emerged to address this conceptual challenge.

One prominent criticism, formalized by researchers like John G. Achen, suggested that much of the observed instability was attributable purely to classical, random measurement error in the survey instrument itself, rather than a lack of attitude in the respondent. Achen argued that even genuine attitudes might appear unstable if the measurement tools were imprecise or unreliable, implying that the underlying attitude existed but was simply measured poorly. Another, more substantive critique was advanced by John Zaller with his influential Receive-Accept-Sample (RAS) model. Zaller argued that instability is not due to nonattitudes, but rather due to citizens holding multiple, often conflicting, considerations (or “ambivalence”) simultaneously. When asked a survey question, the citizen samples randomly from these available considerations, leading to a probabilistic response that changes depending on which considerations are temporarily salient or primed by the survey context.

The RAS model therefore reframes the issue: instability reflects the probabilistic nature of attitude expression rather than the complete absence of a mental state. While Zaller’s model acknowledged that many people operate with poorly structured belief systems, it provided a more nuanced psychological mechanism than Converse’s stark dichotomy between attitude and nonattitude. Nonetheless, Zaller’s framework still supports the fundamental finding that citizens are often reacting to immediate, accessible information rather than drawing on deep, persistent ideological convictions, thereby affirming the limited depth of mass political engagement originally highlighted by the nonattitude concept.

6. Significance and Impact

The concept of the nonattitude fundamentally reshaped the field of public opinion research. Its primary impact was to introduce a profound skepticism regarding the reliability and coherence of aggregate survey data, particularly concerning low-salience political topics. By distinguishing between true attitudes and ephemeral responses, Converse provided a framework for understanding why public opinion often appears volatile.

First, it forced researchers to develop and apply more rigorous standards for assessing attitude quality and stability, moving beyond simple cross-sectional reporting to demanding evidence of temporal persistence and coherence through panel studies. Second, the nonattitude thesis provided the empirical foundation for subsequent theories of political sophistication, demonstrating that political knowledge and engagement are stratified, with major consequences for democratic stability and responsiveness. The finding implied that political outcomes are often driven by a relatively small segment of the ideologically constrained elite, while the large segment exhibiting nonattitudes is highly volatile and susceptible to short-term electoral campaigning and media framing. Therefore, the concept serves as a crucial reminder that survey responses must be treated cautiously, often necessitating methodological adjustments to filter out responses that represent mere noise in the political data landscape.

7. Further Reading

• Philip E. Converse – Wikipedia
• The Nature of Belief Systems in Mass Publics (Converse, 1964)
• Nonattitude – Wikipedia
• John Zaller – Wikipedia