Confidence (General) (CONF)

1. Abstract

The Confidence (General) (CONF) Scale is a streamlined, psychometrically robust four-item self-report instrument developed to evaluate the subjective sense of metacognitive confidence, certainty, and validity individuals attach to their cognitive responses, judgments, or emergent attitudes. Introduced by Pablo Briñol, Richard E. Petty, and Zakary L. Tormala in their seminal 2004 publication on the self-validation hypothesis in consumer psychology and social cognition, the instrument operationalizes the subjective metacognitive appraisal of generated thoughts. Rather than assessing the valence (favorable versus unfavorable) or the primary quantity (number of thoughts generated) of cognitive processing, the CONF scale captures the critical metacognitive tier: the extent to which an individual believes their thoughts are valid, dependable, and actionable.

The scale employs four unipolar semantic descriptors—specifically evaluating the degree to which judgments or thoughts feel confident, certain, valid, and sure. The standard response format leverages an expanded unipolar Likert-type rating continuum, typically ranging from 1 (“Not at all”) to 9 (“Extremely”), although 1-to-7 variations are occasionally implemented across diverse experimental paradigms. The structural integrity of the scale is characterized by a unidimensional factor structure, demonstrating high factor loadings (routinely exceeding λ = .80) and exceptional internal consistency reliability across multiple empirical demonstrations, with Cronbach’s alpha coefficients typically ranging between α = .86 and α = .94.

Construct validity is thoroughly documented in social-psychological and consumer cognition research: the instrument reliably predicts whether cognitive elaboration translates into stable, durable attitude change and behavior, functioning as a primary empirical tool to test the self-validation hypothesis. High scores on the CONF scale predict strong primary-thought-to-attitude correspondence, whereas attenuated scores reflect metacognitive doubt, dissolving the persuasive impact of generated thoughts. Its brief administration profile, generalizability across domains, and exceptional psychometric stability render it an indispensable diagnostic tool for experimental psychology, behavioral economics, metacognition research, and marketing communication assessment.

2. Keywords

Confidence (General) Scale, metacognition, self-validation hypothesis, attitude certainty, thought confidence, cognitive responses, persuasion, attitude change, social cognition, psychometrics

3. Authors

The Confidence (General) (CONF) Scale was originated and empirically validated by a team of leading social psychologists and persuasion theorists:

• Pablo Briñol, Ph.D. — Professor of Psychology at the Universidad Autónoma de Madrid (Spain). Dr. Briñol is an internationally recognized scholar in the areas of attitudes, persuasion, and metacognition. His pioneering research centers on how metacognitive factors modulate judgment, communication dynamics, and behavioral change. (Email: [email protected])
• Richard E. Petty, Ph.D. — Distinguished University Professor Emeritus of Psychology at The Ohio State University (Columbus, Ohio, USA). Co-originator of the canonical Elaboration Likelihood Model (ELM) of persuasion, Dr. Petty is among the most heavily cited social psychologists globally, having fundamentally shaped modern theories of attitude formation, cognitive response theory, and self-validation. (Email: [email protected])
• Zakary L. Tormala, Ph.D. — Professor of Marketing at the Stanford Graduate School of Business, Stanford University (Stanford, California, USA). Dr. Tormala investigates consumer judgment, attitude certainty, source credibility, and persuasive framing, with a primary focus on how psychological certainty governs consumer decision-making and resistance to influence. (Email: [email protected])

4. Purpose

The fundamental purpose of the Confidence (General) (CONF) Scale is to quantify the subjective, metacognitive assessment of thought validity that occurs downstream of cognitive elaboration. For decades, classical dual-process frameworks—such as the Elaboration Likelihood Model (ELM) developed by Petty and Cacioppo (1986) and the Heuristic-Systematic Model (HSM) introduced by Chaiken (1980)—posited that persuasion was fundamentally driven by the valence and quantity of cognitive responses. When exposed to a persuasive message, individuals engaged in cognitive elaboration, generating favorable thoughts (proattitudinal/support arguments) or unfavorable thoughts (counterarguments). Under high-elaboration conditions, the net valence of these primary cognitions was assumed to directly determine post-message attitudes.

However, theoretical advancements in metacognition indicated that primary thoughts alone are insufficient to account for attitude formation and behavioral execution. The self-validation hypothesis proposed that individuals do not merely generate thoughts; they also reflect upon them, formulating secondary, metacognitive appraisals regarding whether those primary cognitions are correct, trustworthy, and valid. If individuals doubt the validity of their favorable thoughts, those thoughts will not be relied upon, preventing persuasion. Conversely, if individuals doubt their counterarguments, the persuasive impact of the message is preserved or amplified. The CONF scale was specifically designed to capture this critical secondary cognitive appraisal.

Beyond its initial derivation in cognitive response paradigms, the CONF scale serves several vital research and clinical applications:

• Experimental Persuasion and Communication Research: Researchers employ the scale to isolate the exact mechanism through which experimental manipulations (e.g., source credibility, emotional framing, bodily feedback, power induction, head nodding) moderate persuasive outcomes. It provides an empirical check on whether an experimental induction operates via primary thought generation or metacognitive self-validation.
• Consumer Judgment and Marketing Science: In evaluating commercial advertisements, product claims, or brand repositioning campaigns, the CONF scale enables marketers to establish whether consumers truly believe their positive brand impressions, allowing organizations to differentiate between superficial processing and resilient, confident brand attitudes that predict buying behavior.
• Metacognitive and Educational Psychology: The CONF framework is widely utilized to assess subjective certainty in academic problem-solving, intuitive reasoning versus deliberate computation, and metacognitive calibration (the alignment between subjective confidence and objective epistemic accuracy).
• Clinical and Cognitive Therapy Implications: In Cognitive Behavioral Therapy (CBT), psychological distress often stems not merely from the presence of maladaptive automatic thoughts, but from the unexamined confidence with which clients embrace those dysfunctional cognitions. The conceptual mechanics operationalized by the CONF scale parallel the cognitive therapy process of introducing doubt into cognitive distortions, reducing the subjective validity of catastrophic or depressive rumination.

5. Psychological Construct

The psychological construct assessed by the CONF instrument is metacognitive thought confidence (also referred to across social cognition literature as attitude certainty, subjective validity, or metacognitive certainty). Metacognition, fundamentally defined as “cognition about cognition” or “knowing about knowing,” encompasses an extensive constellation of self-reflective processes. Within attitude theory, a distinct line is drawn between first-order (primary) cognitions and second-order (metacognitive) cognitions.

Primary cognitions represent the immediate psychological reactions that surface when an individual confronts an object, idea, or argument (e.g., “This new environmental policy will damage industrial competitiveness” or “This electric vehicle has remarkable safety ratings”). In contrast, secondary cognitions involve an evaluative monitoring of those initial thoughts (e.g., “Am I really sure about that conclusion?” or “That thought seems completely valid and dependable”). The CONF scale explicitly targets this secondary tier of cognitive appraisal.

The Single Factor: Metacognitive Certainty

Unlike multidimensional attitude instruments that partition constructs into affective, behavioral, and cognitive domains, the CONF scale is purposefully unipolar and unidimensional. It operationalizes a homogeneous psychological continuum anchored by absolute doubt at the lower bound and unyielding subjective conviction at the upper bound. This construct is delineated by four complementary semantic facets:

• Confidence (Subjective Boldness and Faith in Judgment): Reflects an internal state of trust in one’s cognitive output. A consumer who reads a product description and feels confident in their derived opinion possesses a high readiness to commit cognitive and material resources based upon that belief.
• Certainty (Freedom from Doubt and Indecision): Represents the cognitive stability of the response. Certainty indicates that the individual experiences minimal ambiguity, ambivalence, or internal hesitation regarding the judgment.
• Validity (Perceived Truth Value and Epistemic Soundness): Evaluates whether the cognitive response is perceived to align with objective reality, logic, and factual accuracy. Validity is the epistemic pillar of the construct, transforming an internal impression into an authoritative truth claim.
• Sureness (Immediate Perceptual Conviction): Captures the intuitive, resolute firmness of the thought. Sureness bridges reflective epistemic appraisal with spontaneous intuitive conviction, ensuring that both deliberate and visceral dimensions of confidence are encompassed.

A crucial conceptual nuance of metacognitive confidence is its orthogonal independence from attitude valence. An individual can hold a highly unfavorable attitude toward a proposal with either total certainty or profound hesitation. Similarly, an individual can harbor highly favorable thoughts accompanied by either strong confidence or deep skepticism. By decoupling cognitive valence from metacognitive confidence, the CONF construct provides the requisite empirical architecture to explain why two people holding identical primary attitudes frequently demonstrate radically divergent behaviors: only the individual whose attitudes are characterized by high metacognitive validity will exhibit attitude-behavior consistency, resilience against counterpersuasion, and temporal persistence.

6. Theoretical Framework

The Confidence (General) Scale is grounded theoretically in the convergence of two foundational paradigms in psychological science: the Cognitive Response Theory of persuasion and the Self-Validation Hypothesis.

1. The Cognitive Response Approach and the ELM

Traditional persuasion research initiated by the Yale Communication Research Program (Hovland, Janis, & Kelley, 1953) emphasized message learning: attitude change occurred when an individual attended to, comprehended, and retained the persuasive argument. However, Greenwald (1968) and subsequently Petty, Ostrom, and Brock (1981) demonstrated that message retention correlates weakly with persuasion. Instead, the primary cognitive responses generated by the recipient during message exposure dictate attitude change.

Petty and Cacioppo (1986) integrated these observations into the Elaboration Likelihood Model, distinguishing between the central route to persuasion (requiring high motivation and ability to scrutinize argument quality) and the peripheral route (relying on low-effort heuristics). While the ELM thoroughly mapped how the quantity and valence of cognitive responses predict judgment under central-route processing, it left unanswered a fundamental anomaly: why do individuals who generate equivalent numbers of positive or negative arguments frequently display marked discrepancies in resulting attitude strength?

2. The Self-Validation Hypothesis (SVH)

To resolve this theoretical limitation, Petty, Briñol, and Tormala (2002) introduced the Self-Validation Hypothesis (SVH). The core tenet of the SVH is that generating thoughts is a necessary, but insufficient, condition for persuasion; those thoughts must also be psychologically validated to impact subsequent attitudes. Metacognitive thought confidence operates as a psychological amplifier or dampener:

• When Thought Confidence is High: Generated cognitions exert full downstream impact on attitudes. Favorable thoughts lead to exceptionally positive attitudes, whereas counterarguments lead to resolute resistance or negative evaluations.
• When Thought Confidence is Low (Doubt): Generated cognitions are mentally discounted. Positive thoughts fail to produce favorable attitudes, and negative thoughts fail to drive resistance, causing attitudes to revert toward neutrality or default heuristic baselines.

Briñol, Petty, and Tormala (2004) extended this framework to consumer psychology and advertising, systematically introducing the four-item CONF scale to capture this process. Their empirical investigations demonstrated that variables previously assumed to alter primary thought generation (e.g., source credibility, emotional state, message framing) could instead exert their persuasive influence downstream by altering thought confidence after thoughts had already been generated.

3. Foundational Metacognitive Assumptions

The CONF instrument assumes three primary psychological mechanisms:

1. Metacognitive Accessibility: Individuals possess sufficient introspective access to evaluate their immediate feelings of certainty regarding their cognitive responses.
2. Continuum of Credence: Confidence is not a binary, all-or-nothing cognitive state; it exists on a continuous, unipolar spectrum from absolute doubt to maximum conviction.
3. Behavioral Consequentiality: Subjective confidence directly governs the translation of cognitive contents into overt behavioral intentions, willingness to advocate, and resistance to counter-attitudinal attacks.

7. Validity

The psychometric validity of the Confidence (General) Scale has been extensively corroborated across dozens of empirical studies in social cognition, consumer research, and communication science. The evidence establishes strong construct, convergent, discriminant, and predictive validity.

Construct and Predictive Validity

Construct validity for the CONF scale is primarily demonstrated through rigorous experimental interaction paradigms. Under the self-validation framework, the interaction between thought valence (manipulated via strong vs. weak arguments) and thought confidence (measured via the CONF scale) must predict post-communication attitudes.

In the foundational validation experiments by Briñol, Petty, and Tormala (2004), participants evaluated an advertisement for a consumer product (e.g., a fictitious pain reliever or an automobile) and subsequently completed thought-listing procedures followed by the four-item CONF scale. Multiple regression analyses yielded a statistically significant interaction between the primary thought index and the CONF scale scores (β ranging from .32 to .48, p < .001). Specifically, among participants scoring high on the CONF scale, the correlation between thought favorability and post-message attitude approached r = .80 (p < .001). Conversely, among participants exhibiting low scores on the CONF scale, this correlation dropped dramatically (r < .20, p = ns), confirming that thoughts without metacognitive confidence do not govern attitude formation.

Convergent Validity

The CONF scale exhibits robust convergent validity through strong positive correlations with established indices of metacognitive certainty, attitude strength, and subjective clarity:

• It correlates strongly (r = .68 to .79, p < .001) with the Attitude Certainty Scale developed by Petrocelli, Tormala, and Rucker (2007), specifically its attitude correctness and attitude clarity subdimensions.
• It demonstrates high convergence with single-item direct measures of judgment confidence (r > .80) while yielding superior psychometric stability and reduced measurement error.
• It aligns closely with objective indicators of processing certainty, such as shortened response latencies when reporting attitudes via computer-based reaction time protocols (Fazio, 1990).

Discriminant Validity

Critically, the CONF scale exhibits sharp discriminant validity against potentially confounding psychological constructs:

• Cognitive Valence: Factor-analytic and correlational investigations confirm that CONF scores are empirically decoupled from the favorability or unfavorability of primary thoughts (|r| typically < .12, non-significant). Individuals can be highly confident in negative thoughts or highly uncertain about positive thoughts.
• Thought Quantity (Elaboration Depth): The total number of thoughts generated in cognitive-response listing protocols exhibits near-zero correlation with CONF scores (r ranging from .04 to .09), indicating that metacognitive certainty does not merely mirror cognitive productivity or verbal fluency.
• General Self-Esteem and Trait Confidence: While situational CONF scores can be influenced by trait factors under specific boundary conditions, the CONF instrument functions distinctly as a state-based, thought-focused measure. Correlational analyses with the Rosenberg Self-Esteem Scale routinely show low coefficients (r < .25), demonstrating that the scale measures epistemic certainty rather than global personal self-worth.

8. Reliability

The internal consistency and measurement precision of the four-item Confidence (General) Scale have been systematically documented across diverse samples, experimental contexts, and cross-cultural adaptations.

Internal Consistency (Cronbach’s Alpha)

Across the foundational studies reported by Briñol, Petty, and Tormala (2004) and dozens of subsequent empirical investigations, the CONF scale has consistently yielded exceptional internal consistency coefficients:

• In Briñol et al. (2004, Study 1, Consumer Advertising Context): Cronbach’s α = .89.
• In Briñol et al. (2004, Study 2, Source Credibility Paradigm): Cronbach’s α = .91.
• In Briñol et al. (2004, Study 4, Commercial Product Evaluation): Cronbach’s α = .93.
• In subsequent cross-cultural research across Spanish, English, and German samples (e.g., Briñol & Petty, 2009): Cronbach’s α consistently fell within the range of .86 to .94.

These values demonstrate that despite containing only four items, the scale achieves an optimal balance between brief administration and high measurement fidelity, safely exceeding the canonical psychometric threshold of α ≥ .80 recommended by Nunnally and Bernstein (1994) for empirical research.

Item-Total Statistics and Composite Reliability

Detailed psychometric evaluations indicate that corrected item-total correlations for all four items routinely surpass .70, typically ranging from .72 to .85. Furthermore, calculating McDonald’s omega (ω) as a more robust metric of composite reliability yields values consistently aligning with or exceeding Cronbach’s alpha (ω = .88 – .94), reflecting the absence of severe tau-equivalence violations.

Test-Retest Stability

Because the CONF scale is predominantly utilized as a state-based measure of thought confidence tied to an immediate cognitive stimulus, test-retest reliability across long intervals is theoretically inappropriate: thought confidence fluctuates as individuals encounter new counter-arguments or forgotten information. However, within immediate post-experimental testing intervals (e.g., 20-minute delays without intervening cognitive interference), test-retest correlations remain exceptionally high (r = .82 – .87), indicating high temporal stability of the constructed metacognitive judgment in the absence of external perturbations.

9. Factor Analysis

Extensive exploratory and confirmatory factor analyses validate the singular, robust dimensionality of the CONF instrument.

Exploratory Factor Analysis (EFA)

When the four descriptors (confident, certain, valid, sure) are subjected to exploratory factor analysis (Principal Axis Factoring or Maximum Likelihood extraction with unrotated solutions):

• A single dominant eigenvalue is uniformly extracted, typically exceeding 3.10, accounting for 75% to 85% of the total variance across the four items.
• The scree test clearly displays an unmistakable drop after the first factor, with the second eigenvalue routinely registering below 0.35, firmly rejecting multidimensionality.
• Standardized factor loadings on this single latent dimension are uniformly high, typically loading as follows:

Confirmatory Factor Analysis (CFA)

Confirmatory factor analytic investigations examining the one-factor measurement model have confirmed superior goodness-of-fit indices across varying sample sizes (N > 250). Standard structural equation modeling parameter evaluations reveal:

• Model Chi-Square (χ²): Non-significant or demonstrating a χ²/df ratio under 2.0 (e.g., χ²(2) = 2.84, p = .24), indicating excellent data-to-model consistency.
• Comparative Fit Index (CFI): Routinely > .99 (frequently reaching 1.000).
• Tucker-Lewis Index (TLI): Routinely > .98.
• Root Mean Square Error of Approximation (RMSEA): Consistently < .04 (90% CI [.00, .08]).
• Standardized Root Mean Square Residual (SRMR): Consistently < .02.

Attempts to fit alternative multi-factor models fail because the covariance among items is almost entirely accounted for by the singular latent construct of metacognitive confidence. Measurement invariance testing across gender, age cohorts, and experimental conditions (e.g., high vs. low message involvement) confirms strict metric and scalar invariance, supporting direct cross-group mean comparisons.

10. Instrument / Measurement Tool

The Confidence (General) (CONF) Scale is structured as an efficient, easily implementable self-report tool. Below is the structural profile of the instrument:

• Instrument Designation: Confidence (General) Scale (Acronym: CONF).
• Primary Reference Authors: Pablo Briñol, Richard E. Petty, and Zakary L. Tormala (2004).
• Construct Measured: Metacognitive certainty and subjective validity of generated cognitive responses, evaluations, or emerging attitudes.
• Test Type: Unidimensional psychological rating scale / state metacognitive assessment.
• Item Count: Exactly 4 items.
• Response Continuum: Unipolar Likert-type scale, typically administered on a 1-to-9 numerical continuum (where 1 = “Not at all” and 9 = “Extremely”), though 1-to-7 variations are also common in computerized social cognition surveys.
• Target Population: Adolescents and adults participating in psychological, sociological, consumer, or political judgment studies.
• Administration Modality: Self-administered; compatible with pencil-and-paper questionnaires, online web surveys (Qualtrics, REDCap, Gorilla), and laboratory computer terminals.
• Estimated Completion Time: Less than 60 seconds (typically 30–45 seconds).
• Scoring Algorithm:
  • All 4 items are positively keyed (unipolar format; zero reverse-scored items).
  • Individual item scores are summed or averaged across the four descriptors:
  • Composite Score Formula: Mean CONF = (Confident + Certain + Valid + Sure) / 4
  • Higher numerical averages indicate greater metacognitive confidence, signaling that the respondent relies heavily upon their primary cognitive reactions.

11. Permissions & Fee and Test Year

• Year of Formal Publication: 2004 (introduced in Journal of Consumer Research, Volume 30, Issue 4).
• Copyright & Intellectual Rights: The original empirical manuscript is copyrighted by the Journal of Consumer Research, Inc. and Oxford University Press. However, the four-item scale itself was placed in the scientific public domain by the authors for academic, non-commercial, and clinical research purposes.
• Usage Fee: Free ($0.00). There is no royalty or license purchase required for non-commercial scientific, educational, or experimental inquiry.
• Commercial Applications: Commercial organizations, market research firms, or consulting agencies seeking to embed the scale into proprietary commercial consumer intelligence platforms should seek permission or credit the original authors according to standard fair-use and intellectual property norms.
• Citation Expectation: Any scholarly article, thesis, or technical report deploying the scale is expected to cite Briñol, Petty, and Tormala (2004) alongside Petty, Briñol, and Tormala (2002).

12. References

The theoretical, empirical, and psychometric foundations of the Confidence (General) Scale are established in the following peer-reviewed literature:

• 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(09)00402-4
• Briñol, P., Petty, R. E., & Tormala, Z. L. (2004). Self-validation of cognitive responses to advertisements. Journal of Consumer Research, 30(4), 559–573. https://doi.org/10.1086/380283
• Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology, 39(5), 752–766. https://doi.org/10.1037/0022-3514.39.5.752
• Fazio, R. H. (1990). Multiple processes by which attitudes guide behavior: The MODE model as an integrative framework. Advances in Experimental Social Psychology, 23, 75–109. https://doi.org/10.1016/S0065-2601(08)60318-4
• Greenwald, A. G. (1968). Cognitive learning, cognitive response to persuasion, and attitude change. In A. G. Greenwald, T. C. Brock, & T. M. Ostrom (Eds.), Psychological Foundations of Attitudes (pp. 147–170). Academic Press. https://doi.org/10.1016/B978-1-4832-3071-9.50012-X
• Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and Persuasion: Psychological Studies of Opinion Change. Yale University Press.
• Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric Theory (3rd ed.). McGraw-Hill.
• Petrocelli, J. V., Tormala, Z. L., & Rucker, D. D. (2007). Unpacking attitude certainty: Attitude clarity and attitude correctness. Journal of Personality and Social Psychology, 93(1), 30–41. https://doi.org/10.1037/0022-3514.93.1.30
• 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., 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., Ostrom, T. M., & Brock, T. C. (Eds.). (1981). Cognitive Responses in Persuasion. Lawrence Erlbaum Associates.
• Tormala, Z. L., & Rucker, D. D. (2018). Attitude certainty: Antecedents, consequences, and new directions. Consumer Psychology Review, 1(1), 72–89. https://doi.org/10.1002/arcp.1004

13. Items of the Scale