Certainty (CER)

Abstract

The Certainty (CER) scale, conceptualized and formalized within social psychometrics by attitude strength scholars including Sharon R. Gross, Rolf Holtz, and Norman Miller (1995), serves as a foundational measurement instrument evaluating subjective confidence in an individual’s evaluation of an attitude object. Within the broader architecture of attitude strength delineated by Richard E. Petty and Jon A. Krosnick, attitude certainty is defined as a metacognitive appraisal reflecting the degree to which an individual views their personal attitude toward an object as firm, definite, valid, and resistant to alternative perspectives, rather than tentative, fragile, or easily swayed. The standard CER battery operationalizes this construct through a 4-item self-report questionnaire utilizing 7-point or 9-point semantic differential rating scales anchored by cognitive and epistemic assessments of confidence, correctness, conviction, and subjective certitude.

Extensive psychometric investigations have affirmed the scale’s high internal consistency, with Cronbach’s alpha values systematically ranging between α = .86 and α = .94 across diverse socio-political, environmental, clinical, and consumer domains. Confirmatory factor analyses typically establish a robust unidimensional structure for the core 4-item measure, while contemporary metacognitive models demonstrate its functional split or alignment with higher-order factors comprising attitude clarity and attitude correctness. Evaluated across decades of persuasion and behavioral research, the Certainty scale possesses documented predictive, convergent, and discriminant validity. It robustly accounts for variance in attitude-behavior consistency, temporal stability across longitudinal intervals, and resistance to counterattitudinal persuasion beyond the variance explained by attitude valence and extremity alone.

Keywords

Attitude Certainty, Attitude Strength, Metacognition, Social Cognition, Persuasion Resistance, Attitude-Behavior Consistency, Epistemic Conviction, Psychometrics, Semantic Differential, Construct Validity

Authors

The standard formulations of the Attitude Certainty assessment paradigm were codified and analyzed by:

• Sharon R. Gross, Ph.D. — Department of Psychology, University of Southern California; specializing in social cognition, intergroup relations, and attitude structure.
• Rolf Holtz, Ph.D. — Department of Psychology, University of Southern California; recognized for contributions to social psychology, minority influence, and the structural properties of attitudes.
• Norman Miller, Ph.D. — Professor Emeritus of Psychology, University of Southern California; pioneer in social psychology, meta-analysis methodology, and social influence theories.
• Richard E. Petty, Ph.D. & Jon A. Krosnick, Ph.D. — Co-editors and principal architects of the overarching structural taxonomy of attitude strength in which the CER paradigm was formally integrated and benchmarked against other cognitive dimensions.

Purpose

The primary purpose of the Certainty (CER) scale is to measure how self-assured an individual feels regarding an attitude they hold toward a specific attitude object. Historically, social psychologists recognized that two individuals could register identical evaluations on a standard bipolar attitude scale (e.g., both endorsing a “+3” on a 7-point scale ranging from -3 = extremely unfavorable to +3 = extremely favorable) yet diverge radically in their subsequent behavioral trajectories. One individual might abandon the attitude upon encountering minor counter-arguments, vote inconsistently with their stated preference, or exhibit minimal willingness to defend their stance in public discourse. The other individual might maintain their evaluation for decades, proselytize to peers, actively resist counterattitudinal propaganda, and engage in high-cost behaviors aligned with their view.

The CER scale bridges this empirical divide by capturing the metacognitive tier of attitude representation. Metacognitions are second-order cognitions (thoughts about thoughts or feelings about evaluations). Rather than assessing whether an attitude is positive or negative, the Certainty scale quantifies the individual’s subjective sense that their attitude is durable, correct, and appropriate to hold as a guide for action. Research paradigms in social psychology, political behavior, consumer choice, health communication, and clinical intervention employ the CER instrument for several core theoretical and applied purposes:

• Predicting Attitude-Behavior Consistency: Attitudes held with high certainty serve as vastly superior behavioral guides compared to attitudes held with ambivalence or doubt. CER scores reliably moderate the correlation between initial attitudes and observed behaviors, such as voting in municipal or national elections, adhering to clinical or pharmacotherapeutic regimens, purchasing sustainable consumer products, or participating in collective political protests.
• Forecasting Resistance to Persuasion: In public health campaigns and persuasion experiments, assessing CER permits researchers to determine whether individuals will assimilate or reject persuasive counter-messages. Attitudes accompanied by elevated certainty scores generate proactive counter-arguing when exposed to contradictory evidence, thereby exhibiting profound resistance to social influence.
• Evaluating Temporal Durability: Longitudinal studies utilize CER to gauge the temporal decay rate of newly formed or altered attitudes. Interventions that shift attitudes without fostering certainty often see the newly induced attitude regress to baseline over days or weeks, whereas interventions that elevate certainty establish durable cognitive structures.
• Information Processing and Selective Exposure: CER guides research into selective exposure mechanisms. Highly certain individuals often selectively seek out attitude-consistent media to reinforce their worldview or, under specific conditions of perceived threat, selectively expose themselves to counterattitudinal claims solely to dismantle them.

Psychological Construct

The psychological construct captured by the CER scale is attitude certainty, situated within the broader taxonomic domain of attitude strength. Conceptually, attitude strength refers to the extent to which an attitude manifests two primary structural attributes—durability (persistence over time and resistance to attack)—and produces two primary behavioral consequences: impactfulness on information processing and impact on behavior. Within this matrix, attitude certainty is an explicit metacognitive evaluation that qualifies the primary attitude.

To fully delineate the construct, psychometricians distinguish certainty from adjacent structural dimensions:

• Attitude Extremity: The deviation of an attitude from the neutral midpoint of an evaluative continuum (e.g., loving or hating an object vs. feeling lukewarm). Although extremity and certainty frequently correlate positively, they are conceptually and empirically distinct; an individual can hold an extreme attitude with substantial hesitation, or maintain a moderate, centrist attitude with absolute certainty.
• Attitude Importance: The psychological significance and personal relevance an individual attaches to an attitude object. An individual may believe environmental protection is vital to their identity (high importance) while acknowledging deep uncertainty regarding the efficacy of a specific cap-and-trade policy (low certainty).
• Attitude Accessibility: The ease and speed with which an evaluation is retrieved from long-term memory into working memory, conventionally measured via response latency paradigms. While accessible attitudes are often held with certainty, accessibility reflects retrieval fluency, whereas certainty reflects an explicit cognitive judgment of confidence and truth value.
• Structural Ambivalence: The simultaneous presence of both positive and negative evaluations toward the same attitude object. Whereas ambivalence introduces evaluative conflict and tension, low certainty can exist in the complete absence of positive/negative conflict, such as when an individual simply lacks sufficient information to feel confident.

Modern psychometric theory further dissects the attitude certainty construct into two distinct yet highly intercorrelated sub-dimensions, as articulated by Petrocelli, Tormala, and Rucker (2007):

1. Attitude Clarity: The subjective sense that one knows what one’s attitude actually is (e.g., “I have a clear, definite sense of where I stand”). Clarity reflects low subjective ambiguity regarding one’s own evaluative orientation.
2. Attitude Correctness: The subjective belief that one’s attitude represents the objectively valid, true, and moral viewpoint regarding the object (e.g., “My view is the correct view that any rational observer should hold”).

The standard 4-item Gross et al. (1995) Certainty scale samples this theoretical domain symmetrically, utilizing semantic anchors that capture certainty, correctness, conviction, and confidence, thereby synthesizing clarity and epistemic correctness into a coherent, highly reliable omnibus index.

Theoretical Framework

The CER scale is rooted in social-cognitive theories of persuasion and cognitive architecture, most notably the Elaboration Likelihood Model (ELM) of persuasion developed by Petty and John Cacioppo, and the subsequent Self-Validation Hypothesis formulated by Petty, Pablo Briñol, and Zakary Tormala.

The Elaboration Likelihood Model and Metacognition

The ELM posits that attitudes can be formed and modified along a continuum of cognitive elaboration. When individuals possess both the motivation and the cognitive ability to scrutinize issue-relevant arguments, persuasion proceeds via the central route. Attitudes forged through central route processing involve extensive issue-relevant thinking, counter-argumentation, and integration into existing knowledge networks. The ELM predicts that central-route attitudes inherently engender higher levels of attitude certainty because the individual possesses a dense network of supportive arguments. Conversely, attitudes formed via the peripheral route—governed by superficial heuristics, source attractiveness, or conditioning—typically manifest low CER scores. The CER scale provides the quantitative metric required to verify whether elaboration successfully crystallized into subjective certainty.

The Self-Validation Hypothesis

The Self-Validation Hypothesis expands classical dual-process frameworks by introducing metacognitive validation. Prior to this model, cognitive response theories assumed that persuasion was exclusively a function of the number and valence of thoughts generated in response to a message. Petty and colleagues established that thought generation alone is insufficient; individuals also assess their thoughts metacognitively, assigning levels of confidence or doubt to them. When individuals generate favorable thoughts toward a message but experience low confidence in those thoughts (e.g., due to source invalidity or somatic markers of doubt), the thoughts are discounted, and persuasion fails. The CER scale captures the cumulative, post-message crystallization of this self-validation process: the ultimate confidence an individual places in their synthesized perspective.

Antecedents of Attitude Certainty

The theoretical framework delineates multiple direct antecedents that drive scores on the CER instrument:

• Cognitive Elaboration and Direct Experience: Engaging in systematic analysis of an issue or acquiring direct, firsthand physical experience with an attitude object reliably enhances attitude certainty.
• Perceived Social Consensus: Learning that one’s attitude is shared by peers or ingroup members validates the perception of correctness, driving dramatic spikes in CER scores. Conversely, encountering unexpected consensus from an outgroup can stimulate validation through perceived cross-cutting truth.
• Successful Resistance to Persuasion: When individuals actively resist an attempt to change their view, and perceive that the counter-persuasive message was strong or came from an expert source, they infer that their own attitude must be remarkably sound. This metacognitive inference leads to a paradoxical increase in CER scores beyond pre-message baseline levels.
• Perceived Attitude Completeness: The subjective assessment that one has acquired all essential information required to form a definitive stance on the topic.

Validity

The validity of the CER scale has been established through decades of rigorous empirical research spanning hundreds of independent laboratory experiments and field investigations.

Construct Validity

Construct validity is evidenced by the scale’s predictable sensitivity to experimental manipulations designed to alter certainty without altering underlying evaluative valence. In classical experiments (e.g., Tormala & Petty, 2002), researchers presented participants with identical persuasive messages while manipulating metacognitive appraisals through feedback (e.g., informing participants that they had successfully resisted an expert persuasive attack vs. a non-expert attack). Across such studies, CER scores demonstrated significant, theory-congruent shifts (Δ η² typically ranging between .08 and .18) while primary attitude scores remained constant, confirming that the scale assesses a distinct metacognitive construct rather than primary evaluative valence.

Convergent and Discriminant Validity

Multitrait-multimethod matrices and correlational batteries confirm the convergent and discriminant profile of the CER scale:

• Convergent Associations: CER correlates moderately to strongly with measures of attitude importance (r = .35 to .62), perceived knowledge (r = .40 to .68), and attitude extremity (r = .30 to .55). These relationships confirm that certainty resides organically within the broader constellation of attitude strength properties.
• Discriminant Independence: Structural equation modeling consistently reveals that CER does not collapse into adjacent constructs. For example, Krosnick and Fabrigar (1995) demonstrated via exploratory and confirmatory factor modeling that attitude certainty, attitude importance, and perceived knowledge form three separable latent dimensions. Corrected latent correlations between certainty and importance rarely exceed r = .65, demonstrating sufficient unshared variance (> 55%) to affirm construct uniqueness.

Predictive and Criterion-Related Validity

The predictive validity of the CER scale is among the most robust in social psychology:

• Moderation of Attitude-Behavior Consistency: In meta-analytic reviews (e.g., Glasman & Albarracín, 2006; Kraus, 1995), attitude certainty emerged as a robust statistical moderator of the relationship between attitudes and subsequent behavior. The attitude-behavior correlation under high certainty frequently exceeds r = .60, whereas under conditions of low certainty, the correlation plummets to r ≤ .25.
• Resistance to Counter-Persuasion: Individuals scoring high on the CER scale exhibit significantly lower attitude change scores when subjected to counterattitudinal communication compared to low-CER peers (Cohen’s d > 0.70 across multiple replication studies).
• Temporal Stability: In longitudinal tracking spanning months or electoral cycles, attitudes accompanied by high baseline CER scores demonstrate test-retest stabilities exceeding r = .75, compared to r < .40 for attitudes characterized by baseline uncertainty.

Reliability

The Certainty scale exhibits exceptional psychometric reliability across diverse demographic groups, cultures, and topic areas.

Internal Consistency

Across numerous published studies, the 4-item CER battery displays high internal consistency:

• Cronbach’s Alpha (α): Standard published investigations report alpha coefficients consistently exceeding the conventional .80 psychometric threshold, typically falling between α = .86 and α = .94. In the foundational validation studies by Gross, Holtz, and Miller (1995), the items yielded internal consistencies above .88 across diverse political and social attitude targets.
• McDonald’s Omega (ω_t): Modern structural analyses report McDonald’s hierarchical and total omega coefficients closely mirroring alpha, with total omega frequently registering between ω = .89 and ω = .95, confirming that the scale items share substantial true score variance attributable to the general certainty factor.
• Inter-Item Correlations: Corrected item-total correlations for all four items routinely surpass r = .68, with average inter-item correlations (AIC) residing comfortably in the .62 to .78 range. These metrics indicate homogeneous item content without excessive semantic redundancy.

Test-Retest Stability

The stability of CER scores across time depends fundamentally on the presence or absence of intervening persuasive contexts:

• In neutral, non-interventional control conditions over short intervals (e.g., 1 to 2 weeks), CER displays robust test-retest reliability (r_tt = .78 to .88), indicating that individual baseline assessments of certainty are stable cognitive constructs.
• Over extended multi-month spans, test-retest reliability reflects meaningful evolutionary dynamics: certainty gradually climbs as individuals repeatedly encounter and selectively process attitude-consonant environmental cues.

Factor Analysis

The structural composition of the CER scale has been subjected to extensive exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across numerous sample cohorts.

Exploratory Factor Analysis (EFA)

When the four CER items are submitted to EFA with principal axis factoring or maximum likelihood extraction, extraction criteria (Kaiser criterion of eigenvalues > 1.0 and Cattell’s scree test) universally indicate a single dominant factor. The initial unrotated eigenvalue for this factor typically accounts for 68% to 82% of the total variance across items, with no secondary factor yielding an eigenvalue above 0.50.

Confirmatory Factor Analysis (CFA)

In CFA testing, the hypothesized unidimensional model provides an excellent fit to empirical data when assessing omnibus attitude certainty. Standard structural equations reveal:

• Standardized Factor Loadings (λ): Loadings are uniformly high and statistically significant (p < .001). Typical parameter estimates are:
  • Item 1 (Certainty): λ = .88 – .93
  • Item 2 (Convinced/Correct): λ = .81 – .87
  • Item 3 (Confident): λ = .89 – .94
  • Item 4 (Sure/Right): λ = .83 – .89
• Goodness-of-Fit Indices: Evaluated in structural models with single-factor operationalizations, fit parameters routinely surpass standard academic benchmarks:
  • Comparative Fit Index (CFI) ≥ .985
  • Tucker-Lewis Index (TLI) ≥ .975
  • Root Mean Square Error of Approximation (RMSEA) ≤ .045 (90% CI [.000, .072])
  • Standardized Root Mean Square Residual (SRMR) ≤ .020

Multidimensional Nuance: Clarity vs. Correctness

When expanded psychometric protocols test the distinction between Attitude Clarity and Attitude Correctness (Petrocelli et al., 2007), a two-factor correlated model often demonstrates superior fit to a strictly unidimensional model across larger item pools. In such analyses, Item 1 (“certain”) and Item 3 (“confident”) load preferentially onto the Latent Clarity dimension (λ > .85), while Item 2 (“convinced/correct”) and Item 4 (“sure/right”) load preferentially onto Latent Correctness (λ > .82). The latent correlation between these two dimensions remains high (typically r = .65 to .80), justifying the conventional practical usage of the 4-item Gross et al. scale as an integrated, unidimensional summary index of overall attitude certainty in general research applications.

Instrument / Measurement Tool

The CER is a self-report instrument administered following an evaluative assessment of a targeted attitude object (such as a public policy, political candidate, organizational decision, or consumer product).

Administrative Specifications

• Instrument Designation: Certainty Scale (CER) / Attitude Certainty Scale
• Originating Framework: Gross, Holtz, & Miller (1995); Petty & Krosnick (1995)
• Measurement Format: Standardized self-report rating scale
• Item Inventory: 4 items
• Target Object: Flexible stem; adapted to any attitude object (e.g., [attitude object] = capital punishment, universal healthcare, brand choice, environmental regulations)
• Response Modality: 7-point or 9-point semantic differential / rating scale (e.g., 1 to 7 or 1 to 9), anchored from low certainty to high certainty
• Administration Time: Approximately 1 to 2 minutes
• Target Population: Adolescents and adults (ages 14+); valid across diverse student, clinical, organizational, and general community populations

Scoring Protocol and Data Processing

• Item Polarity: All 4 items are keyed in the direct, positive direction. There are no reverse-scored items in the authentic instrument.
• Score Computation: Scores across all 4 items are summed or mathematically averaged:

  $$\text{Certainty Index} = \frac{\sum_{i=1}^{4} \text{Item}_i}{4}$$

• Scale Range: For a 7-point scale, mean index scores range from 1.00 to 7.00 (sum scores: 4 to 28). For a 9-point scale, mean index scores range from 1.00 to 9.00 (sum scores: 4 to 36).
• Score Interpretation:
  • Low Certainty (1.00 – 2.99 on 7-pt; 1.00 – 3.99 on 9-pt): The individual holds an attitude that is fragile, tentative, unstable, and highly vulnerable to persuasive counter-appeals. Attitude-behavior consistency is negligible.
  • Moderate Certainty (3.00 – 4.99 on 7-pt; 4.00 – 6.99 on 9-pt): The individual possesses a working evaluation of the object but experiences moderate hesitation, epistemic ambiguity, or unverified assumptions.
  • High Certainty (5.00 – 7.00 on 7-pt; 7.00 – 9.00 on 9-pt): The individual is exceptionally self-assured, viewing their position as firm, correct, and unassailable. Predicts high behavioral fidelity, proactive defense against opposing arguments, and longitudinal durability.

Permissions & Fee and Test Year

• Year of Formal Codification: 1995 (formalized in Attitude Strength: Antecedents and Consequences, edited by R. E. Petty and J. A. Krosnick).
• Licensing and Accessibility: The 4-item CER scale resides in the public academic domain for non-commercial research, academic, and clinical assessment purposes. No licensing fees or prior author authorization are required to utilize the instrument in empirical investigations.
• Usage Requirement: Proper bibliographic citation of Gross, Holtz, and Miller (1995) or the relevant primary validation literature in subsequent academic publications, dissertations, and technical reports.

References

• Briñol, P., & Petty, R. E. (2009). Persuasion: Insights from the self-validation hypothesis. Advances in Experimental Social Psychology, 41, 69–118. https://doi.org/10.1016/S0065-2601(08)00402-4
• Glasman, L. R., & Albarracín, D. (2006). Forming attitudes that predict future behavior: A meta-analysis of the attitude-behavior relation. Psychological Bulletin, 132(5), 778–822. https://doi.org/10.1037/0033-2909.132.5.778
• Gross, S. R., Holtz, R., & Miller, N. (1995). Attitude certainty. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences (pp. 215–245). Lawrence Erlbaum Associates, Inc.
• Kraus, S. J. (1995). Attitudes and the prediction of behavior: A meta-analysis of the empirical literature. Personality and Social Psychology Bulletin, 21(1), 58–75. https://doi.org/10.1177/0146167295211007
• Krosnick, J. A., & Fabrigar, L. R. (1995). Designing rating scales for effective measurement in surveys. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences (pp. 429–460). Lawrence Erlbaum Associates, Inc.
• Petrocelli, J. V., Tormala, Z. L., & Rucker, D. D. (2007). Unpacking attitude certainty: Attitude clarity and attitude correctness. Journal of Personality and Social Psychology, 92(1), 30–41. https://doi.org/10.1037/0022-3514.92.1.30
• Petty, R. E., Briñol, P., & Tormala, Z. L. (2002). Thought confidence as a determinant of persuasion: The self-validation hypothesis. Journal of Personality and Social Psychology, 82(5), 722–741. https://doi.org/10.1037/0022-3514.82.5.722
• Petty, R. E., & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of persuasion. Advances in Experimental Social Psychology, 19, 123–205. https://doi.org/10.1016/S0065-2601(08)60214-2
• Petty, R. E., & Krosnick, J. A. (Eds.). (1995). Attitude strength: Antecedents and consequences. Lawrence Erlbaum Associates, Inc.
• Rucker, D. D., Tormala, Z. L., Petty, R. E., & Briñol, P. (2014). Consumer conviction and certainty: Consequences for persuasion and consumption. Journal of Consumer Psychology, 24(1), 119–136. https://doi.org/10.1016/j.jcps.2013.07.001
• Tormala, Z. L., & Petty, R. E. (2002). What doesn’t kill me makes me stronger: The effects of resisting persuasion on attitude certainty. Journal of Personality and Social Psychology, 83(6), 1298–1313. https://doi.org/10.1037/0022-3514.83.6.1298

Items of the Scale