Volatility

1. Abstract

The Volatility Scale is a psychometric measurement tool developed by Keith A. Botner, Arul Mishra, and Himanshu Mishra (2020) to quantify consumer and decision-maker perceptions of how consistent, stable, and predictable versus erratic, turbulent, and risky an object, event, entity, or stimulus is perceived to be. Rooted in cognitive psychology, linguistic phonosemantics, and behavioral decision theory, the scale addresses a critical gap in consumer behavior and risk literature: while objective volatility denotes statistical variance or dispersion across temporal horizons (such as standard deviation in asset returns or fluctuation in atmospheric pressure gradients), subjective or perceived volatility represents an evaluative cognitive appraisal driven by salient perceptual, semantic, and phonetic cues. Developed through a rigorous, multi-stage psychometric purification protocol involving initial semantic generation, expert panel curation, exploratory factor analysis (EFA), and confirmatory factor analysis (CFA), the resulting unidimensional measurement model captures the core semantic polarity between erratic turbulence and steady consistency. Typically operationalized through semantic differential or multi-item Likert-type response formats (ranging from 1 to 7), the instrument demonstrates high internal consistency (Cronbach’s α regularly exceeding .85 to .92 across experimental paradigms), robust test-retest reliability, and pronounced discriminant and convergent validity with respect to generalized risk, subjective uncertainty, and affective valence. By linking phonetic features (such as voiceless stop consonants and back vowels) directly to downstream risk estimations across disparate domains—including financial asset pricing, corporate branding, natural hazard appraisals (e.g., hurricane destructiveness), and pharmaceutical evaluation—the Volatility Scale serves as a foundational instrument for researchers investigating heuristics, sound symbolism, decision architecture, and subjective risk appraisal.

2. Keywords

Volatility Scale, Risk Perception, Sound Symbolism, Phonosemantics, Consumer Decision Making, Behavioral Economics, Perceived Stability, Psychometrics, Cognitive Appraisal, Brand Equity, Heuristics

3. Authors

The Volatility Scale was conceptualized, developed, and empirically validated by a team of researchers specializing in marketing, behavioral decision theory, and consumer psychology:

• Keith A. Botner, Ph.D. — Associate Professor of Marketing, Department of Marketing, College of Business, Lehigh University, Bethlehem, Pennsylvania, USA. Research specializations include sound symbolism, consumer cognition, linguistics in marketing communications, and behavioral judgment.
• Arul Mishra, Ph.D. — David Eccles Professor of Marketing, David Eccles School of Business, University of Utah, Salt Lake City, Utah, USA. Research interests focus on behavioral decision making, judgment and decision processes, heuristic processing, and consumer welfare.
• Himanshu Mishra, Ph.D. — David Eccles Professor of Marketing, David Eccles School of Business, University of Utah, Salt Lake City, Utah, USA. Research expertise encompasses behavioral decision theory, predictive judgment, consumer heuristics, and quantitative experimental design.

Correspondence regarding the foundational empirical research can be directed to the primary authors through the David Eccles School of Business at the University of Utah or Lehigh University College of Business.

4. Purpose

The primary purpose of the Volatility Scale is to provide a theoretically grounded, psychometrically standardized measurement instrument capable of assessing perceived volatility across human judgment contexts. In standard economic and financial paradigms, volatility is classically operationalized as an objective, historical, or implied mathematical dispersion parameter, most notably captured by the standard deviation or variance of an underlying distribution (e.g., the Black-Scholes implied volatility surface or realized asset variance). However, human decision makers rarely process complex risk profiles strictly through objective mathematical formulas. Instead, consumers, managers, and lay decision makers depend on subjective appraisals formed rapidly through heuristic processing, intuitive semantics, and associative memory networks.

Traditional subjective risk scales often measure perceived risk as an undifferentiated, unitary construct, frequently conflating expected severity of loss, probability of negative occurrence, emotional dread, and cognitive unfamiliarity (as seen in early psychometric risk paradigms pioneered by Paul Slovic and Baruch Fischhoff). While valuable, these omnibus risk indices fail to isolate the specific attribute of volatility—the perceived degree of rapid, erratic, unpredictable, and turbulent fluctuation over time versus calm, steady, reliable, and predictable progression. The Volatility Scale was explicitly created to fill this methodological void.

In consumer research and experimental psychology, isolating perceived volatility is critical because volatility serves as an upstream cognitive mediator that dictates subsequent risk evaluation and behavioral choice. For instance, when individuals evaluate a newly listed stock ticker, a novel cryptocurrency asset, an impending tropical storm, or a new pharmaceutical drug, the phonetic architecture of the target’s linguistic label activates latent phonemic associations. If the phonetic features convey harshness, abruptness, or turbulence, individuals automatically infer that the underlying entity operates in an erratic, unpredictable fashion. Measuring this specific perception allows researchers to determine whether observed changes in risk tolerance, willingness to pay, or protective behavior are mediated specifically by shifts in perceived erraticism rather than generalized affective dislike or baseline uncertainty.

Clinical and applied behavioral contexts benefit equally from this precise operationalization. In financial counseling, understanding how retail investors subjectively estimate asset volatility based on non-fundamental cues (such as corporate name changes or ticker redesigns) helps prevent irrational panic selling or speculative overexposure. In public policy and disaster risk communication, measuring how communities perceive the volatility of severe weather alerts enables meteorologists and crisis managers to craft warnings that do not inadvertently minimize perceived threat due to passive, soothing phonetics or evoke paralysis through exaggerated erraticism. Consequently, the Volatility Scale functions as both an experimental manipulation check and a core dependent or mediating construct across consumer research, applied psychology, behavioral economics, and communication studies.

5. Psychological Construct

The psychological construct captured by the Volatility Scale is defined as the subjective appraisal of the magnitude, frequency, and unpredictability of fluctuation exhibited by an entity, stimulus, or phenomenon. Rather than reflecting an objective mathematical property, this construct operates as a cognitive representation structured along a continuum ranging from extreme stability, continuity, and predictability on one anchor, to extreme turbulence, inconsistency, erraticism, and risk on the opposing anchor.

Although empirically validated as a parsimonious unidimensional construct, perceived volatility encompasses several distinct yet intimately correlated semantic facets that together constitute the subjective mental schema of volatility:

Erraticism vs. Consistency

Erraticism represents the degree to which an entity is perceived as prone to abrupt, irregular, and sudden shifts in state, behavior, or magnitude. When an object is perceived as erratic, an individual expects that its trajectory will defy smooth extrapolation. Conversely, consistency implies uniform, continuous, and steady progression where past behavior reliably mirrors future behavior. For example, in consumer branding, a product categorized as consistent generates an expectation of unwavering quality control, whereas an erratic rating signals fluctuating performance and unreliable outcomes.

Turbulence vs. Steadiness

Turbulence captures the perceived kinetic energy and chaotic instability inherent in the target stimulus. Rooted deeply in physical metaphors (such as atmospheric turbulence or violent wave dynamics), psychological turbulence reflects an expectation of agitation, disturbance, and violent swings. Steadiness, by contrast, conveys equilibrium, balance, and quiet endurance. In risk research, stimuli characterized by high turbulence evoke an intuitive sense that the environment or object is being buffeted by unpredictable forces beyond systemic control.

Unpredictability vs. Predictability

Subjective predictability reflects the degree of confidence a decision maker maintains regarding their ability to forecast the entity’s future trajectory. While mathematically related to variance, psychological unpredictability manifests as cognitive entropy—a high subjective information processing burden wherein individuals experience an inability to establish deterministic or probabilistic forecasts. An entity high on perceived volatility makes pattern recognition challenging, increasing feelings of subjective vulnerability and perceived risk.

Dynamic Instability vs. Inherent Safety

At the upper bound of the volatility spectrum, dynamic instability converges with operational risk. While stability creates psychological safety and permits heuristic complacency, perceived volatility signals an active, dynamic hazard requiring heightened vigilance, cognitive monitoring, and protective adjustments. Thus, perceived volatility acts as a cognitive gateway to risk appraisal: things that do not move, vary, or fluctuate are intuitively coded as benign, whereas entities perceived as shifting wildly across temporal states trigger danger detection schemas.

The integration of these dimensions into a cohesive construct ensures that the Volatility Scale measures not merely how dangerous an object is (which could stem from static, known toxicity), but precisely how dynamically unstable and erratic its ongoing state is perceived to be.

6. Theoretical Framework

The Volatility Scale is situated at the intersection of three major theoretical domains: Sound Symbolism (Phonosemantics), Heuristic and Associative Processing, and The Psychometric Paradigm of Risk Perception.

Sound Symbolism and Phonosemantics

For decades, conventional structural linguistics—grounded in the foundational assertions of Ferdinand de Saussure—posited that the relationship between the sound structure of a word (the signifier) and its underlying conceptual meaning (the signified) is entirely arbitrary. However, an extensive body of modern psycholinguistic and cognitive research has established that linguistic sounds possess intrinsic, non-arbitrary semantic associations, a phenomenon known as sound symbolism or phonosemantics. Classic experiments by Wolfgang Köhler (the famous “Bouba/Kiki” effect) and Edward Sapir demonstrated that human beings cross-modally map specific acoustic and articulatory properties of phonemes onto physical and psychological attributes, including size, shape, angularity, weight, and speed.

Botner, Mishra, and Mishra (2020) advanced this theoretical paradigm by integrating articulatory phonetics with cognitive risk appraisal. They posited that the phonetic architecture of language directly influences higher-order, abstract assessments of danger and stability. Specifically, phonemes are categorized along acoustic and articulatory continua:

• Voiceless vs. Voiced Consonants: Voiceless consonants (such as /p/, /t/, /k/, /s/, /f/) are produced without vocal cord vibration, characterized by sudden acoustic bursts, transient noise, and sharp energy transitions. In contrast, voiced consonants (such as /b/, /d/, /g/, /z/, /v/) involve sustained glottal pulsing, yielding acoustic waveforms that are continuous, smooth, and low-frequency.
• Acoustic Abruptness: The stop-consonant burst of voiceless plosives mirrors environmental sounds associated with sudden collisions, fractures, and explosive shifts. Consequently, the human auditory and conceptual systems intuitively bind voiceless phonemes to schemas of abruptness, harshness, and turbulence.

The Volatility Scale provides the precise empirical instrument necessary to measure this cross-modal phonosemantic transfer, proving that names featuring voiceless, abrupt phonemes trigger mental representations of erratic movement, which then translate directly into heightened subjective volatility ratings.

Heuristic and Dual-Process Decision Theory

The scale relies extensively on dual-process models of cognition (e.g., Daniel Kahneman and Amos Tversky’s System 1 and System 2 frameworks). When decision makers face novel, ambiguous, or complex stimuli (e.g., an unfamiliar corporate ticker on an exchange or an impending tropical storm), deliberate analytical evaluation (System 2) is often constrained by lack of data, cognitive fatigue, or temporal pressure. Consequently, rapid, autonomous, associative mechanisms (System 1) dominate judgment.

Through the affect heuristic and attribute substitution, individuals unconsciously substitute a difficult computational evaluation (“What is the historical variance and predictive distribution of this asset?”) with a readily accessible perceptual evaluation (“How does the name or visual presence of this entity feel? Does it feel harsh, abrupt, and jagged, or smooth and stable?”). Perceived volatility acts as the immediate cognitive bridge: phonetic and surface-level attributes activate sensory schemas of turbulence, which System 1 effortlessly converts into a subjective rating of volatility.

The Psychometric Risk Paradigm

Historically, the psychometric paradigm established by Paul Slovic, Baruch Fischhoff, and Sarah Lichtenstein demonstrated that psychological risk is multidimensional, determined by subjective factors such as dread, familiarity, controllability, and catastrophic potential. The Volatility Scale enriches this framework by isolating temporal dynamic variance as a distinct, measurable cognitive dimension. Rather than viewing risk solely through the prism of negative outcome valence or static severity, the theoretical framework of the Volatility Scale emphasizes that human beings are acutely sensitive to the temporal patterning and behavioral variance of stimuli. Dynamic instability is inherently perceived as less controllable and more cognitively demanding, thereby directly elevating perceived subjective risk.

7. Validity

The psychometric validity of the Volatility Scale has been rigorously established through multi-stage experimental and correlational methodologies conducted by Botner, Mishra, and Mishra (2020), demonstrating substantial construct, convergent, discriminant, and predictive (criterion) validity.

Construct and Content Validity

To establish initial content validity, the authors utilized a structured, multi-stage semantic extraction procedure. An expansive initial pool of English adjectives describing temporal dynamics, movement, stability, and turbulence was subjected to independent expert judges trained in psychometrics, linguistics, and consumer behavior. Words exhibiting ambiguity, low semantic clarity, or excessive colloquialism were systematically eliminated. The retained items were subjected to comprehensive pilot testing to confirm that the semantic pairs accurately reflected polar extremities of dynamic stability versus erratic volatility.

Convergent Validity

Convergent validity was evaluated by examining the scale’s association with established constructs measuring related dimensions of risk, uncertainty, and expected variance. Across multiple empirical studies, the Volatility Scale demonstrated robust, statistically significant positive correlations with:

• Overall Subjective Risk Assessments: r values consistently ranging between .58 and .74 (p < .001), indicating that while perceived volatility strongly informs risk evaluation, it remains conceptually differentiated from general risk.
• Expected Price Variance and Fluctuation: In simulated financial investment tasks, perceived volatility scores correlated strongly with participants’ quantitative estimates of anticipated high-low trading spreads (r ≈ .62, p < .001).
• Anticipated Threat and Destructiveness: In severe hazard studies evaluating tropical storm names, perceived volatility correlated positively with anticipated wind speed variability, barometric instability, and damage severity ratings (r > .50, p < .001).

Discriminant Validity

To ensure that the Volatility Scale does not merely capture generalized negative affect or semantic valence, the authors conducted discriminant validity assessments utilizing the Fornell-Larcker criterion and average variance extracted (AVE) comparisons. When modeled alongside measures of general likeability, aesthetic attractiveness, and subjective familiarity:

• The average variance extracted (AVE) for the Volatility factor consistently exceeded .68, well above the standard .50 psychometric threshold.
• The squared correlation between perceived volatility and generalized affective valence (likeability) was low (r² < .15), demonstrating that a stimulus can be perceived as highly volatile without being evaluated as intrinsically “bad” or unappealing (e.g., thrilling extreme sports products or high-growth speculative assets).
• Discriminant validity was further substantiated against measures of simple cognitive processing fluency, confirming that perceived volatility is driven by specific phonosemantic cues rather than difficulty in pronunciation or typographic legibility.

Predictive and Criterion Validity

The predictive power of the Volatility Scale was confirmed through incentive-compatible experimental choice tasks and field-data modeling:

• Financial Allocation Tasks: Higher scores on the Volatility Scale predicted significantly reduced portfolio allocations in risk-averse participants, mediated by subjective volatility rather than objective historical performance metrics.
• Real-World Asset Tickers: When applied to historical stock market data, simulated asset tickers containing voiceless, abrupt consonants yielded systematically elevated volatility ratings, which in turn predicted real consumer expectations of trading risk.
• Natural Disasters: In analyses of historical hurricane data, participants presented with names high in perceived volatility predicted significantly higher damage potential and reported greater willingness to evacuate or take protective action compared to names low in perceived volatility.

8. Reliability

The Volatility Scale has exhibited exceptional reliability metrics across numerous experimental studies, laboratory replications, and varied stimulus environments.

Internal Consistency

Internal consistency evaluates the degree to which all individual items within the instrument measure the identical underlying latent construct. Across the empirical studies reported by Botner, Mishra, and Mishra (2020), the scale exhibited high internal reliability coefficients:

• Cronbach’s Alpha (α): The scale consistently yields Cronbach’s α coefficients ranging from .85 to .93 across diverse experimental treatments (e.g., Study 1 α = .88; Study 2 α = .91; Study 3 α = .87; Study 4 α = .92). These values significantly exceed the conventional psychometric adequacy threshold of .70 and the rigorous threshold of .80 recommended for basic and applied research.
• Composite Reliability (CR): Structural equation modeling and CFA evaluations demonstrate composite reliability values consistently exceeding .89, confirming that the scale items reliably reflect the latent volatility factor without excessive measurement error.
• Item-Total Correlations: Corrected item-to-total correlations for all scale items consistently surpassed .65, with none falling below the recommended .40 cut-off, confirming that each item contributes meaningfully to the overall score.

Test-Retest and Cross-Stimulus Stability

Test-retest stability was evaluated by administering the scale to non-clinical participant cohorts across separated time intervals (e.g., 2-week intervals) using standardized target stimuli. Intraclass correlation coefficients (ICC) demonstrated robust temporal stability (ICC > .82, p < .001). Furthermore, the scale exhibits high stability across varied stimulus modalities, maintaining equivalent psychometric reliability whether administered in response to auditory acoustic stimuli (spoken words), visual text (orthographic brand names and tickers), or abstract conceptual scenarios.

9. Factor Analysis

The structural dimensionality of the Volatility Scale was rigorously evaluated through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA), confirming a parsimonious, robust unidimensional latent factor structure.

Exploratory Factor Analysis (EFA)

During initial instrument development, EFA was conducted using principal axis factoring and maximum likelihood estimation with both orthogonal (Varimax) and oblique (Promax) rotations to examine whether the semantic items bifurcated into separate sub-dimensions (such as separating physical movement from cognitive uncertainty):

• Eigenvalue Criteria: Examination of the initial scree plot revealed a distinct, sharp break after the first factor. Only a single factor demonstrated an initial eigenvalue greater than 1.0 (Kaiser-Guttman rule), with the primary factor accounting for over 65% to 72% of the total variance across datasets.
• Factor Loadings: All retained items exhibited strong, positive primary loadings onto the single underlying volatility factor, with standardized factor loadings ranging between .76 and .91. No significant cross-loadings or secondary factor structures emerged, establishing that the items tap directly into a unified cognitive construct of perceived volatility.

Confirmatory Factor Analysis (CFA)

To validate the unidimensional factor model in independent samples, CFA was executed using structural equation modeling software (SEM) with maximum likelihood estimation. The single-factor model demonstrated exceptional goodness-of-fit indices across diverse empirical studies:

• Model Chi-Square: The model yielded non-significant or acceptable normed chi-square statistics (χ²/df ratios typically falling between 1.15 and 2.20).
• Comparative Fit Index (CFI): Values consistently ranged between .97 and .99, surpassing the conservative .95 benchmark for superior model fit.
• Tucker-Lewis Index (TLI): Coefficients regularly exceeded .96, indicating excellent fit relative to the null baseline model.
• Root Mean Square Error of Approximation (RMSEA): Point estimates ranged from .028 to .052, with 90% confidence intervals well within the strict ≤ .06 psychometric threshold for excellent approximation error.
• Standardized Root Mean Square Residual (SRMR): Observed SRMR values remained below .035, confirming minimal residual discrepancy between the observed sample covariance matrix and the implied theoretical model matrix.

Alternative multi-factor models (e.g., bifurcating erraticism from predictability) failed to produce significant improvements in chi-square statistics and resulted in factor correlations exceeding .85, providing overwhelming statistical support for the parsimonious, unidimensional specification of the Volatility Scale.

10. Instrument / Measurement Tool

The Volatility Scale is designed as a standardized, brief, self-report measurement instrument that can be implemented in laboratory experiments, online panels, field surveys, and organizational assessments. Below is the structured operational specification of the instrument:

• Instrument Type: Self-administered psychometric questionnaire / semantic differential or Likert-type cognitive appraisal scale.
• Administration Format: Computer-assisted web interviewing (CAWI), paper-and-pencil questionnaire, or integrated software experiment interface (e.g., Qualtrics, PsychoPy, Gorilla).
• Item Count: 3 to 5 core semantic differential or rating pairs (commonly utilized as a 3-item or 4-item parsimonious battery to minimize respondent fatigue while maintaining α > .85).
• Response Format: Typically administered using a 7-point semantic differential scale or a 7-point Likert scale (e.g., 1 = Extremely Stable / Consistent / Predictable to 7 = Extremely Volatile / Erratic / Unpredictable). Formats using 5-point or 9-point scales have also been adapted with equivalent reliability.
• Scoring Protocol:
  • Individual item responses are coded such that higher numerical values systematically represent higher perceived volatility, erraticism, and dynamic risk.
  • Any reverse-coded anchor pairs (where the stable/predictable dimension is assigned the higher value) must be recoded prior to index computation using the standard transformation formula: Recoded Score = (Scale Maximum + 1) – Original Score.
  • An overall Perceived Volatility Index is calculated by computing the arithmetic mean of the item ratings for each respondent. Higher composite scores indicate greater perceived volatility, turbulence, and unpredictability.
• Administration Time: Rapid administration requiring approximately 30 to 60 seconds per target stimulus, making it ideal for repeated-measures and multi-trial experimental designs.

11. Permissions & Fee and Test Year

The Volatility Scale was formally introduced and published in the peer-reviewed psychological and consumer literature in 2020:

• Publication Year: 2020.
• Original Source: Journal of Consumer Research, Volume 47, Issue 1, Pages 128–145.
• Intellectual Property & Copyright: The conceptual and empirical publication is copyrighted by the Journal of Consumer Research, Inc., published by Oxford University Press.
• Academic Research Usage: Consistent with standard academic conventions, the measurement methodology, theoretical construct, and item structures described in the published journal article may be utilized by academic researchers and educators for non-commercial scientific research, academic dissertations, and university-based investigations without licensing fees, provided proper formal academic citation is rendered to Botner, Mishra, and Mishra (2020).
• Commercial Applications: Commercial entities, market research firms, and proprietary software developers seeking to integrate the scale or its derivative proprietary algorithms into commercial brand-naming tools, enterprise risk systems, or monetized testing platforms should consult the original authors and the copyright holders regarding permissions and commercial licensing requirements.

12. References

Below are primary academic references documenting the theoretical development, psychometric validation, and foundational literature supporting the Volatility Scale:

• Botner, K. A., Mishra, A., & Mishra, H. (2020). The influence of the phonetic elements of a name on risk assessment. Journal of Consumer Research, 47(1), 128–145. https://doi.org/10.1093/jcr/ucz054
• Fischhoff, B., Slovic, P., Lichtenstein, S., Read, S., & Combs, B. (1978). How safe is safe enough? A psychometric study of attitudes towards technological risks and benefits. Policy Sciences, 9(2), 127–152. https://doi.org/10.1007/BF00143739
• Kahneman, D. (2011). Thinking, fast and slow. Farrar, Straus and Giroux.
• Köhler, W. (1929). Gestalt psychology. Liveright.
• Klink, R. R. (2000). Creating brand names with meaning: The use of sound symbolism. Marketing Letters, 11(1), 5–20. https://doi.org/10.1023/A:1008183819686
• Lowrey, T. M., & Shrum, L. J. (2007). Phonetic symbolism and brand name preference. Journal of Consumer Research, 34(3), 406–414. https://doi.org/10.1086/518530
• Sapir, E. (1929). A study in phonetic symbolism. Journal of Experimental Psychology, 12(3), 225–239. https://doi.org/10.1037/h0070931
• Slovic, P. (1987). Perception of risk. Science, 236(4799), 280–285. https://doi.org/10.1126/science.3563507
• Yorkston, E., & Menon, G. (2004). A sound idea: Phonetic effects of brand names on consumer evaluations. Journal of Consumer Research, 31(1), 43–51. https://doi.org/10.1086/383422

13. Items of the Scale