Desire for Certainty Test

Abstract

The Desire for Certainty Test, developed by sociologist and social psychologist Orville G. Brim, Jr. (1955) and further evaluated in collaboration with David B. Hoff (1957), is a pioneering psychometric instrument designed to assess individual differences in epistemic drive toward cognitive determinacy, subjective probability preferences, and certainty orientation. Emerging from early post-war research into decision theory, cognitive consistency, and the psychological structure of attitudes, the instrument operationalizes the desire for certainty not merely as a self-reported personality trait, but through empirical estimation behavior across uncertain social, behavioral, economic, and physical events. The test consists of 32 items spanning eight distinct life domains: education, recreation, politics, economics, religion, health, family, and transportation/communication. Respondents provide both a subjective numerical probability estimate (from 0 to 100) for each event and a meta-cognitive confidence judgment evaluated on a 5-point Likert-type scale ranging from 1 (very sure) to 5 (not sure at all). Psychometrically, the instrument yields indices of probability extremity (the magnitude of deviation from equiprobability, or 50/50 chance) and subjective confidence. Across early validation studies, the measure exhibited split-half reliability coefficients ranging between .68 and .84 across sub-domains, demonstrating moderate-to-high internal consistency for an assessment capturing broad social forecasting. Construct and convergent validity analyses revealed significant correlations with intolerance of ambiguity, cognitive rigidity, authoritarianism, and defensive confidence in decision-making under risk. The instrument remains a landmark psychometric foundation in judgment and decision-making (JDM) and behavioral economics, illustrating how cognitive closure and risk aversion manifest in subjective probability distributions.

Keywords

Desire for Certainty Test, subjective probability, cognitive certainty, Orville G. Brim, intolerance of ambiguity, decision-making under uncertainty, epistemic motivation, extremity bias, probability estimation, cognitive style.

Authors and Historical Context

The Desire for Certainty Test was formulated by Orville Gilbert Brim, Jr. (1923–2016), a prominent American sociologist whose scholarship bridged sociology, child development, and social psychology. Brim earned his doctorate in sociology from Yale University and served in key academic and leadership roles, including executive positions at the Russell Sage Foundation and the Social Science Research Council. In his 1955 seminal publication in the American Sociological Review, titled “Attitude content – intensity and probability expectations,” Brim sought to deconstruct attitudes into cognitive expectations regarding the likelihood of events rather than simple evaluative valence (positive versus negative affect). Brim observed that standard attitude measurement tools failed to capture how certain individuals felt about their knowledge of the world and the extent to which they gravitated toward definitive, deterministic outcomes.

To further isolate the personality and situational determinants of certainty-seeking behavior, Brim collaborated with David B. Hoff. Their joint 1957 paper in The Journal of Abnormal and Social Psychology, “Individual and situational differences in desire for certainty,” empirically tested how motivation, situational stress, and generalized personality predispositions modulate the avoidance of uncertainty. The scale was later archived and reviewed by social psychologists John P. Robinson and Phillip R. Shaver (1969) in their reference compendium Measures of Social Psychological Attitudes (published by the Institute for Social Research at the University of Michigan), solidifying its position in the history of attitude and personality measurement.

Purpose and Operational Rationale

The primary purpose of the Desire for Certainty Test is to measure an individual’s chronic tendency to eliminate cognitive ambivalence by leaning toward definitive, extreme probability judgments and expressing elevated confidence in uncertain social forecasts. In everyday life, human beings operate under conditions of incomplete information where absolute certainty is statistically unattainable. While normative models of rational choice, such as Bayesian probability theory, suggest that individuals should assign probabilities near the base rate or acknowledge substantial uncertainty when evidence is sparse, individuals vary widely in their comfort with ambiguity.

Brim recognized that traditional self-report inventories measuring traits such as intolerance of ambiguity were susceptible to social desirability bias, acquiescence, and semantic drift. When participants are asked directly whether they “like clear-cut answers” or “dislike ambiguous situations,” their responses frequently reflect cultural norms regarding intellectual flexibility rather than their actual cognitive processing style. To circumvent these methodological limitations, Brim designed a behavioral estimation task disguised as a general knowledge or forecasting inventory. By prompting participants to provide numerical odds for ambiguous empirical events—such as whether an adult male will earn a specific income, whether a marriage will end in divorce, or whether a criminal case will be solved—the instrument captures cognitive certainty in action.

In both research and applied clinical or organizational settings, the scale addresses crucial questions:

• Cognitive Style and Heuristics: How prone is the individual to black-and-white thinking, premature closure, and the neglect of probabilistic base rates?
• Risk Assessment: In financial, managerial, and military decision contexts, does an exaggerated desire for certainty cause decision-makers to treat moderate risks as either impossible (0%) or inevitable (100%)?
• Interpersonal and Clinical Dynamics: In clinical psychology, an inability to tolerate intermediate probabilities is closely linked to generalized anxiety disorder, obsessive-compulsive traits, and catastrophic misinterpretation. The scale allows researchers to isolate whether high subjective confidence serves as a psychological defense mechanism against underlying existential or situational anxiety.

Psychological Construct and Dimensionality

The Desire for Certainty Test operationalizes a complex construct that bridges cognitive psychology, psychometrics, and personality theory. Rather than treating certainty as a single unidimensional attitude, Brim identified two interrelated psychological phenomena: probabilistic extremity and metacognitive subjective confidence.

1. Probability Extremity (Deviation from Equiprobability)

The core cognitive dimension evaluated by the test is the degree to which an individual avoids intermediate, equivocal probability assignments (specifically the 50/50 midpoint). When confronted with uncertain outcomes—such as whether a child whose parents are divorced will develop a neurosis, or whether a small business will fail within two years—a perfectly neutral or uncommitted stance is reflected by a probability of .50 (equiprobability). A strong desire for certainty manifests psychologically as an epistemic push toward the polar anchors of the continuum: certainty of non-occurrence (approaching 0%) or certainty of occurrence (approaching 100%). Individuals high in the desire for certainty exhibit an “extremity bias” across varied content domains, demonstrating a cognitive need to structure an unpredictable social universe into deterministic categories.

2. Metacognitive Confidence (Subjective Sureness)

The second dimension reflects the respondent’s self-appraisal of their own judgment, measured on a 5-point rating scale from 1 (very sure) to 5 (not sure at all). In normatively calibrated cognitive systems, subjective confidence should correlate directly with objective knowledge or empirical evidence. However, Brim and Hoff demonstrated that individuals with a high generalized desire for certainty often report high subjective sureness (ratings of 1 or 2) even when their probability estimates are demonstrably inaccurate, biased, or founded on pure conjecture. This metacognitive overconfidence serves an emotional regulatory function, reducing the cognitive dissonance and discomfort generated by ambiguity.

3. Content Domain Universality

The test evaluates whether the desire for certainty functions as a domain-general cognitive trait or a domain-specific attitude. Brim deliberately embedded items across eight structural domains of everyday human experience:

• Education: e.g., college attendance, law school attrition, public school teacher demographics.
• Recreation: e.g., baseball pennant victories, sports subsidization, card games.
• Politics: e.g., presidential education, gubernatorial re-elections, civic organizations.
• Economics: e.g., income levels, small business survival, elderly financial dependency.
• Religion: e.g., belief in God, church attendance, religious endogamy.
• Health: e.g., longevity, fatal physical trauma, childhood nutrition.
• Family: e.g., divorce rates, childhood discipline, intergenerational career transmission.
• Transportation and Communication: e.g., telephone ownership, busy signals, automobile accident rates.

By assessing whether an individual exhibits extremity and high sureness across widely divergent content areas, the scale determines the presence of a generalized cognitive disposition toward determinism.

Theoretical Framework and Foundations

The Desire for Certainty Test is grounded in the confluence of mid-century psychoanalytic ego psychology, gestalt perceptual theory, and early behavioral decision theory. Three foundational theoretical paradigms directly informed Brim’s formulation of the instrument:

Frenkel-Brunswik’s Intolerance of Ambiguity

Following the influential work of Else Frenkel-Brunswik (1949), intolerance of ambiguity was recognized as a central emotional and cognitive personality variable. Frenkel-Brunswik posited that individuals with rigid psychological defenses demonstrate an aversion to ambiguous stimuli across perceptual, interpersonal, and cognitive realms. Brim operationalized Frenkel-Brunswik’s qualitative personality construct into precise, quantitative mathematical judgments. In Brim’s paradigm, an individual who cannot tolerate ambiguity manifests this intolerance by converting gray-scale probabilities into subjective black-and-white certainties.

Festinger’s Cognitive Dissonance and Consistency Theories

The instrument was developed concurrently with Leon Festinger‘s formulation of cognitive dissonance theory. Dissonance theory posits that psychological discomfort arises when an individual holds contradictory cognitions or experiences uncertainty regarding important behavioral choices. Brim viewed the desire for certainty as an anticipatory dissonance-reduction mechanism: by deciding in advance that an outcome is either guaranteed or impossible, the individual shields the ego from the tension inherent in suspense, doubt, and unpredictable futures.

Subjective Expected Utility and Decision Theory

Brim drew extensively upon the mathematical decision theories articulated by Savage (1954) and Edwards (1954). Classical economics presumed that humans assign subjective probabilities based on objective frequencies and rational Bayesian updates. Brim demonstrated that subjective probability distributions are systematically distorted by personality needs. The desire for certainty represents a systematic cognitive bias where internal utility functions favor determinacy over statistical accuracy.

Validity Evidence

Validity evaluations for the Desire for Certainty Test across the original Brim (1955) investigations and subsequent replications (Brim & Hoff, 1957; Robinson & Shaver, 1969) provide robust empirical support for its psychometric utility.

Construct and Convergent Validity

Construct validity was established by examining the correspondence between probability extremity scores and established personality inventories measuring rigidity and dogmatism. Brim and Hoff (1957) administered the test alongside measures of California Authoritarianism (F-Scale) and generalized rigidity. The findings indicated statistically significant positive correlations (ranging from r = .31 to r = .47, p < .01) between probability extremity and authoritarianism, demonstrating that individuals with authoritarian personality traits consistently avoid intermediate probability judgments.

Furthermore, convergent validity was confirmed against behavioral risk-taking tasks. Participants exhibiting elevated extremity scores on the 32 items showed a marked aversion to entering randomized lotteries where odds were explicitly presented as 50/50, preferring instead tasks framed with unambiguous (even if statistically disadvantageous) odds.

Discriminant Validity

Discriminant validity was established by demonstrating that probability extremity was statistically distinguishable from general intellectual ability (IQ) and academic achievement. While knowledge of actual statistical base rates (e.g., actual national divorce statistics or mortality rates) modestly correlated with educational attainment, the variance accounted for by the subjective certainty ratings (1–5 sureness) and the directional polarization away from 50% remained independent of general intelligence (r values typically between -.08 and .12, non-significant). This verified that the Desire for Certainty Test captures an epistemic motivation and cognitive style rather than objective knowledge deficits.

Predictive and Situational Validity

Brim and Hoff (1957) demonstrated experimental predictive validity by manipulating situational threat and ego-involvement. When participants were exposed to stress-inducing instructions (e.g., being informed that the task was a high-stakes diagnostic test of social insight), their desire for certainty increased significantly: probability estimates shifted further toward the extremes (0% or 100%), and confidence ratings polarized. This established that while certainty-seeking operates as a stable baseline trait, it is also sensitive to environmental stressors.

Reliability and Measurement Precision

The internal consistency and temporal stability of the Desire for Certainty Test have been evaluated across several psychometric trials:

• Split-Half Reliability: In Brim’s (1955) initial cohort of undergraduate students, split-half reliability coefficients for the extremity scores across the 32 items, adjusted via the Spearman-Brown prophecy formula, yielded coefficients between .74 and .84. This demonstrated that despite the diverse subject matter across eight different life domains, individuals maintained a consistent cognitive style of extremity.
• Subscale Consistency: Metacognitive certainty ratings (the 1-to-5 sureness ratings) exhibited an average split-half reliability of .79, reflecting high internal concordance in how participants evaluate their own subjective certainty.
• Test-Retest Stability: Over a four-week test-retest interval, stability coefficients for the overall extremity index clustered around r = .71 to .76, indicating adequate trait-level stability over time under standardized, neutral testing conditions.

Factor Structure and Psychometric Evaluation

Early psychometric analyses of the 32-item instrument focused on domain universality versus domain specificity through exploratory inter-item and inter-domain correlation matrices. In Brim’s foundational 1955 study, correlation matrices computed across the eight sub-areas (economics, politics, health, etc.) yielded universally positive inter-correlations ranging from .24 to .62.

Subsequent psychometric factor evaluations revealed that the instrument is best understood through a hierarchical factor model:

• Higher-Order General Factor (g-Certainty): A predominant general factor accounts for approximately 38% to 44% of the total variance across estimation extremity. This general factor reflects a domain-transcendent epistemic drive to avoid ambiguity, validating Brim’s assertion that certainty-seeking operates as a generalized personality dimension.
• Domain-Specific Primary Factors: Subordinate factors emerge corresponding to specific life domains, most notably:
1. Institutional / Societal Forecasting: Items assessing macro-level political and legal outcomes (e.g., presidential college education, gubernatorial re-election, law school dropout rates).
2. Interpersonal / Behavioral Determinism: Items assessing human relationship dynamics (e.g., divorce rates, parental divorce causing childhood neurosis, spanking yielding truthfulness).
3. Biological / Physical Vulnerability: Items centering on health risks, mortality, and physical survival (e.g., survival after broken neck, infant malnutrition, longevity past 55).

Modern psychometric critiques note that because the items were formulated in the mid-1950s, several item baselines (such as earning $4,000 annually or families lacking a home telephone) reflect historical demographic benchmarks of mid-20th-century America. However, psychometricians emphasize that the validational strength of the scale resides not in whether the respondent’s estimate matches modern demographic facts, but in the structural extremity of their probability estimates and their subjective metacognitive confidence.

Instrument / Measurement Tool

• Instrument Name: Desire for Certainty Test
• Alternative Titles: Brim Desire for Certainty Scale; Probability Extremity and Certainty Inventory
• Author: Orville G. Brim, Jr. (1955); revised with David B. Hoff (1957)
• Format: 32-item paper-and-pencil self-report forecasting assessment
• Administration Time: Approximately 20 to 30 minutes
• Content Domains (8 areas, 4 items each):
  • Education
  • Recreation
  • Politics
  • Economics
  • Religion
  • Health
  • Family
  • Transportation and Communication
• Response Modality: Twofold response per item:
  1. Probability Estimation: Numeric probability estimation (expressed as chances in 100, from 0 to 100).
  2. Subjective Confidence / Sureness: 5-point rating scale evaluating how confident the respondent feels about their estimate:
     • 1 = Very sure
     • 2 = Fairly sure
     • 3 = Moderately sure
     • 4 = Slightly sure
     • 5 = Not sure at all
• Scoring Procedures:
  • Extremity Score (Certainty Bias): Calculated as the absolute deviation of the probability estimate from equiprobability (|Estimate – 50|). Deviations range from 0 (complete uncertainty / 50-50 chance) to 50 (absolute certainty / 0% or 100%). Scores are summed across all 32 items; higher composite extremity scores reflect a greater desire for certainty.
  • Metacognitive Confidence Score: Calculated by summing or averaging the 1-to-5 sureness ratings. To align directionality, items are frequently reverse-scored (where 5 = Very sure and 1 = Not sure at all).
  • Certainty-Extremity Index: Researchers often compute the product or bivariate correlation between a participant’s extremity score and their sureness rating to evaluate calibration accuracy.

Permissions, Fee, and Test Year

The Desire for Certainty Test was originally developed and published in 1955 by Orville G. Brim, Jr., within an academic article published in the American Sociological Review. Additional experimental validation procedures were published in 1957 in The Journal of Abnormal and Social Psychology. The test was subsequently compiled as a non-commercial reference instrument in the public-domain academic monograph Measures of Political Attitudes (1969) by John P. Robinson and Phillip R. Shaver at the Institute for Social Research (ISR), University of Michigan, Ann Arbor.

Consistent with standard academic fair-use guidelines and historic social science inventories published in public scholarly journals prior to contemporary commercial copyright practices, the instrument is available free of charge for non-commercial academic, educational, and psychological research. Researchers utilizing the instrument should provide formal bibliographic attribution to Brim (1955), Brim and Hoff (1957), and the Institute for Social Research archive.

References

• Brim, O. G., Jr. (1955). Attitude content – intensity and probability expectations. American Sociological Review, 20(1), 68–76. https://doi.org/10.2307/2088323
• Brim, O. G., Jr., & Hoff, D. B. (1957). Individual and situational differences in desire for certainty. The Journal of Abnormal and Social Psychology, 54(2), 225–229. https://doi.org/10.1037/h0043818
• Edwards, W. (1954). The theory of decision making. Psychological Bulletin, 51(4), 380–417. https://doi.org/10.1037/h0053870
• Frenkel-Brunswik, E. (1949). Intolerance of ambiguity as an emotional and perceptual personality variable. Journal of Personality, 18(1), 108–143. https://doi.org/10.1111/j.1467-6494.1949.tb01236.x
• Robinson, J. P., & Shaver, P. R. (1969). Measures of Political Attitudes. Survey Research Center, Institute for Social Research, University of Michigan. ISR Resource Document
• Savage, L. J. (1954). The Foundations of Statistics. John Wiley & Sons.

Items of the Scale