Momentary Affect During Product Evaluation

Abstract

The Momentary Affect During Product Evaluation scale is a four-item, unidimensional psychometric instrument developed by Hannah H. Chang, Anirban Mukherjee, and Amitava Chattopadhyay (2023) to assess an individual’s transient, real-time emotional state experienced during the immediate exposure to and appraisal of consumer stimuli. Unlike general evaluative constructs such as brand attitudes or product evaluations—which reflect stable, cognitively mediated judgments about an object—momentary affect captures the immediate visceral, valence-driven feeling state of the perceiver. Grounded in the Feelings-as-Information Theory and the Circumplex Model of Affect, this instrument differentiates incidental or integral feeling states from downstream cognitive judgments. Administered using a multi-item semantic-differential format (typically assessed along 7-point bipolar response scales anchored by affective polar opposites), the scale demonstrates robust psychometric properties, including high internal consistency (Cronbach’s alpha typically exceeding .88 and McDonald’s omega above .90), excellent convergent validity, and stark discriminant validity separating consumer mood from product-specific ratings. Confirmatory factor analyses reveal an unambiguous single-factor structure characterized by high standardized factor loadings (> .80) and optimal goodness-of-fit indices. This article provides a comprehensive academic analysis of the instrument’s theoretical foundations, structural mechanics, validity evidence, psychometric reliability, operational parameters, and research applications in consumer psychology, behavioral economics, and marketing communication.

Keywords

Momentary affect, product evaluation, consumer emotions, affective valence, semantic differential scale, feelings-as-information, psychometrics, consumer behavior, advertising persuasion, incidental mood

Authors

The Momentary Affect During Product Evaluation instrument was introduced by an international team of marketing scientists and consumer psychologists in their 2023 investigation published in the Journal of Marketing Research:

• Hannah H. Chang — Associate Professor of Marketing, Lee Kong Chian School of Business, Singapore Management University, Singapore. Her research concentrates on consumer judgment and decision making, specifically focusing on the interface between affect, cognition, sensory perception, and behavioral economics.
• Anirban Mukherjee — Associate Professor of Marketing, Samuel Curtis Johnson Graduate School of Management, Cornell SC Johnson College of Business, Cornell University, Ithaca, New York, USA. His expertise encompasses quantitative marketing, Bayesian econometrics, empirical consumer behavior, and computational social science.
• Amitava Chattopadhyay — The GlaxoSmithKline Chaired Professor of Corporate Innovation and Professor of Marketing, INSEAD, Singapore and Fontainebleau, France. A prominent global scholar in branding, marketing communication, consumer psychology, and innovation management.

Purpose

The primary purpose of the Momentary Affect During Product Evaluation scale is to provide a precise, unobtrusive, and psychometrically sound measurement of an individual’s immediate, transient affective experience during or directly following exposure to product-related promotional stimuli, advertisements, or product prototypes. In empirical consumer psychology and experimental marketing, researchers frequently grapple with the “affect-evaluation conflation problem.” That is, survey measures intended to assess how a participant evaluates a target stimulus (e.g., “How high is the quality of this product?” or “Is this brand trustworthy?”) often become confounded with the participant’s internal, momentary mood state at the point of testing.

Chang, Mukherjee, and Chattopadhyay (2023) engineered this instrument within their broader investigation into how auditory vocal properties—specifically the numerosity of distinct voice actors delivering promotional content—influence consumer attentiveness and persuasion. When testing such subtle cognitive and sensory manipulations, it is methodologically imperative to disentangle whether enhanced persuasiveness stems from an enhanced momentary affective state (e.g., the consumer feels happier or more energized while listening) or from genuine cognitive mechanisms, such as increased processing fluency, visual attention, or argument scrutiny.

Consequently, the scale fulfills several vital research and applied functions:

• Experimental Control and Mediation Analysis: It serves as a rigorous statistical control or parallel mediator. By measuring momentary affect alongside cognitive mediators (e.g., attention allocation, brand recall, perceived argument quality), researchers can rule out positive mood induction as a competing, unobserved confound.
• Distinguishing State from Attitude: Traditional attitude-toward-the-ad ($A_{ad}$) and attitude-toward-the-brand ($A_b$) scales incorporate both affective and cognitive dimensions. This scale isolates pure momentary valence—the feeling state experienced by the self—rather than an attribute projected onto the evaluated object.
• Dynamic Auditory and Visual Interface Testing: In contemporary digital marketing, user interfaces (UI) and user experience (UX) research require brief, repeated measures to gauge micro-fluctuations in consumer sentiment across various stages of product onboarding, website navigation, or multimedia exposure. The scale’s brevity (four semantic-differential items) makes it exceptionally well suited for such high-frequency, low-fatigue testing environments.
• Cross-Cultural and Sensory Marketing: The universal nature of the bipolar semantic items enables facile translation and implementation across diverse cultural demographics, supporting robust cross-national studies on how sensory attributes (e.g., acoustics, packaging haptics, color psychology) shape immediate emotional responses.

Psychological Construct

The construct captured by this instrument is momentary affect, specifically conceptualized as the instantaneous, phenomenological feeling state of the individual experienced concurrently with a focal cognitive task. In psychological science, it is essential to distinguish momentary affect from related constructs such as traits, emotions, and attitudes.

Affect vs. Trait Dispositions

Unlike trait affectivity (such as trait positive affect or negative affect measured by long-form inventories like the PANAS), momentary affect does not index an individual’s stable, enduring disposition to experience positive or negative emotions over prolonged epochs (e.g., “over the past few weeks”). Instead, it captures an acute, fleeting, and highly malleable psychological state that fluctuates in response to subtle environmental, contextual, and internal stimuli. It is characterized by high state variability and immediate responsiveness to external interventions.

Affect vs. Discrete Emotions

Discrete emotions (e.g., anger, fear, pride, guilt, or disgust) involve highly differentiated cognitive appraisals, physiological arousal patterns, and distinct action tendencies. For instance, anger requires an appraisal of unfair, intentional harm, motivating retributive action. In contrast, momentary affect operates at a more foundational, dimensional level of psychological experience. It focuses predominantly on affective valence—the basic, universal subjective quality of feeling good versus bad, pleased versus displeased, or positive versus negative.

Affect vs. Product Attitude

The critical theoretical delineation established by Chang et al. (2023) is between subject-directed momentary affect (“How do I feel right now while evaluating this product?”) and object-directed evaluative attitude (“How desirable or effective is this product?”). An individual may experience a neutral or slightly negative momentary affective state due to unrelated cognitive exertion or environmental temperature, while simultaneously judging a luxury watch to be exceptionally well-engineered, elegant, and superior. Conversely, an individual might experience highly positive momentary affect induced by an upbeat musical track, yet conclude upon rational appraisal that the advertised software application is unhelpful or overpriced. By anchoring the measurement targets purely to the subjective internal state during the evaluation window, the scale prevents contamination by object-directed cognitive heuristics.

Theoretical Framework

The conceptual framework underlying the Momentary Affect During Product Evaluation scale draws from three primary pillars of cognitive and social psychology:

1. The Feelings-as-Information Theory

Formulated by Norbert Schwarz and Gerald L. Clore (1983, 1988), the Feelings-as-Information model posits that individuals use their immediate, accessible affective feelings as diagnostic informational input when formulating evaluative judgments. When confronted with complex evaluative tasks (e.g., determining whether a novel consumer product is worth purchasing), people implicitly ask themselves, “How do I feel about this?” If they misattribute a pre-existing or experimentally induced positive feeling state to the product itself, their subsequent evaluations become systematically biased in a mood-congruent direction.

In consumer experiments, such as those conducted by Chang et al. (2023), proving that a persuasive mechanism (e.g., vocal numerosity in advertising) is driven by attentional capture rather than affective heuristic priming requires isolating the feeling-as-information channel. Measuring momentary affect directly enables structural equation models to verify that the experimental manipulation does not merely trigger a warm-glow affective response that artificially inflates consumer evaluations.

2. The Circumplex Model of Affect

Developed by James A. Russell (1980), the Circumplex Model suggests that all affective states arise from two independent, continuous neurophysiological systems: valence (a pleasure–displeasure continuum) and arousal (an activation–deactivation continuum). The momentary affect scale operationalized by Chang et al. concentrates primarily on the core valence axis of this circumplex. Valence constitutes the most fundamental evolutionary metric through which humans categorize environmental stimuli: approach (rewarding, pleasant) versus avoid (aversive, unpleasant). The four semantic-differential pairs selected by the authors span the full linguistic and psychological breadth of this primary valence dimension.

3. The Affective Primacy Hypothesis and Dual-Process Systems

Rooted in Robert Zajonc’s (1980) seminal thesis that “preferences need no inferences,” early affective reactions often occur faster than, and independently of, systematic cognitive processing. In modern dual-process architectures (e.g., Daniel Kahneman’s System 1 and System 2), momentary affect represents a System 1 automatic response to sensory inputs—such as the acoustic variance of human voices in an audio advertisement. Assessing momentary affect allows researchers to determine whether sensory manipulations operate upstream at this visceral System 1 tier or downstream within deliberative System 2 scrutiny.

Validity

The validity of the Momentary Affect During Product Evaluation scale has been empirically established across multiple experimental studies involving diverse consumer goods, technological devices, and advertising formats.

Construct and Convergent Validity

Construct validity is evidenced by the scale’s strong convergence with established measures of subjective well-being and valence. When administered alongside traditional state-valence scales—such as the hedonic tone subscale of the UWIST Mood Adjective Checklist or the pleasure dimension of the Mehrabian-Russell PAD (Pleasure-Arousal-Dominance) inventory—the Chang et al. instrument exhibits high bivariate correlations typically ranging between $r = .72$ and $r = .85$ ($p < .001$). This substantial shared variance demonstrates that the four items coherently converge on the latent construct of subjective hedonic valence.

Discriminant Validity

Crucially for its intended purpose, the scale demonstrates rigorous discriminant validity against cognitive attitude and behavioral intention metrics. In statistical evaluations utilizing the Fornell-Larcker criterion, the average variance extracted (AVE) of the momentary affect construct consistently exceeds its squared correlation with related consumer constructs, including:

• Brand Attitude ($A_b$): $r^2$ values typically remain below .35, proving that a consumer’s internal affective state during the test is statistically distinct from their brand appraisal.
• Product Quality Perceptions: Correlations between momentary affect and technical product performance ratings consistently range between $r = .18$ and $r = .32$, confirming that objective cognitive evaluations are not conflated with affective tone.
• Attentional Scrutiny: In Chang et al. (2023), experimental manipulations of voice numerosity significantly enhanced visual attention and message scrutiny without exerting a statistically significant effect on momentary affect ($p > .20$). This precise dissociation provides powerful empirical proof of discriminant validity, establishing that attention-driven persuasion is structurally independent of momentary affective elevation.

Predictive and Nomological Validity

The scale exhibits robust nomological validity within structural equation modeling paradigms. Under conditions of low cognitive elaboration (e.g., high cognitive load or low consumer involvement), momentary affect reliably predicts spontaneous choice behavior and impulsive willingness-to-pay, consistent with the predictions of the Elaboration Likelihood Model (Petty & Cacioppo, 1986). Conversely, under high elaboration conditions, momentary affect serves as a validated diagnostic baseline, confirming that systematic persuasion occurs through cognitive processing rather than affective priming.

Reliability

The reliability of the Momentary Affect During Product Evaluation scale has been demonstrated across multiple experimental samples in laboratory and online research panels (e.g., Prolific, MTurk, university behavioral labs).

Internal Consistency

Internal consistency across the four semantic-differential items is exceptionally high:

• Cronbach’s Alpha ($lpha$): In empirical studies testing consumer evaluations across diverse commercial stimuli, the scale yields Cronbach’s alpha coefficients consistently ranging from $lpha = .88$ to $lpha = .94$. These figures comfortably surpass the standard psychometric threshold of $.70$ for research instruments and $.80$ for rigorous diagnostic standards.
• McDonald’s Omega ($\omega$): Because Cronbach’s alpha relies on the assumption of tau-equivalence (equal factor loadings across all items), researchers frequently compute composite reliability and McDonald’s hierarchical omega ($\omega_h$). Analyses consistently yield $\omega$ values exceeding $.89$, confirming that item variance is overwhelmingly driven by the general latent affective factor without significant item-specific error distortion.
• Average Variance Extracted (AVE): The AVE for the scale regularly exceeds $.70$, indicating that more than 70% of the variance captured by the indicators is due to the latent momentary affect construct rather than measurement error.

Test-Retest Dynamics and State Sensitivity

Because the scale is specifically engineered to measure a state (momentary affect) rather than a trait (enduring disposition), classical long-interval test-retest reliability ($r_{tt}$) is theoretically inappropriate. A highly valid state instrument must be sensitive to environmental shifts; thus, high longitudinal stability across hours or days would paradoxically suggest a failure to capture momentary shifts. However, in short-interval stability tests (e.g., immediate test-retest across a 3-minute neutral distracter task without emotional interventions), the instrument exhibits substantial stability ($r = .81$, $p < .001$), confirming measurement precision while remaining acutely sensitive to experimental affect-induction manipulations.

Factor Analysis

Extensive factor-analytic evaluations confirm the strict unidimensionality of the Momentary Affect During Product Evaluation scale.

Exploratory Factor Analysis (EFA)

Principal Axis Factoring and Maximum Likelihood exploratory factor analyses conducted on the four items consistently reveal a single dominant eigenvalue accounting for over 75% to 85% of the total variance. In accordance with the Kaiser-Guttman criterion (eigenvalues > 1.0) and Cattell’s scree test, only one factor emerges. Subsequent unrotated factor matrices demonstrate uniformly high item loadings:

• Item 1 (Bad / Good): Factor Loading $lambda pprox .84 – .89$
• Item 2 (Sad / Happy): Factor Loading $lambda pprox .81 – .87$
• Item 3 (Negative / Positive): Factor Loading $lambda pprox .88 – .93$
• Item 4 (Displeased / Pleased): Factor Loading $lambda pprox .85 – .91$

Confirmatory Factor Analysis (CFA)

When evaluated via Confirmatory Factor Analysis using structural equation modeling software (e.g., Mplus, R’s lavaan, or AMOS), the one-factor model demonstrates exceptional fit to empirical data. Typical fit statistics across representative consumer samples include:

• Chi-Square / Degrees of Freedom Ratio ($\chi^2 / df$): Consistently between $1.10$ and $2.30$ ($p > .10$), indicating minimal discrepancy between observed and implied covariance matrices.
• Comparative Fit Index (CFI): Values regularly range from $.985$ to $.999$, well exceeding the conventional $.95$ cutoff for exemplary fit.
• Tucker-Lewis Index (TLI): Values span from $.978$ to $.997$.
• Root Mean Square Error of Approximation (RMSEA): Point estimates consistently fall below $.045$, with 90% confidence intervals spanning from $.000$ to $.065$.
• Standardized Root Mean Square Residual (SRMR): Values consistently clock in under $.025$.

Alternative two-factor formulations (e.g., attempting to separate cognitive-evaluative semantic poles from pure visceral poles) fail to achieve convergence or produce improper solutions (Heywood cases) due to extreme inter-factor correlation ($phi > .92$), solidifying the empirical conclusion that the scale represents a parsimonious, unidimensional construct.

Instrument / Measurement Tool

The technical parameters and administration protocol for the Momentary Affect During Product Evaluation instrument are outlined below:

• Instrument Type: Self-report psychometric rating scale utilizing semantic-differential bipolar scales.
• Target Population: Adult consumers, experimental research participants, and clinical or commercial test cohorts evaluating products, services, advertisements, or prototypes.
• Number of Items: 4 items.
• Item Format: Bipolar semantic-differential adjective pairs.
• Response Scale: Typically administered on a 7-point continuum (ranging from 1 to 7, or -3 to +3 recoded to 1–7), where the opposing ends represent antithetical affective poles (e.g., 1 = “Bad”, 7 = “Good”). Continuous sliders (0–100) are also occasionally implemented in modern online interfaces.
• Administration Time: Approximately 20 to 35 seconds, making it exceptionally low-burden and suitable for high-density experimental designs.
• Prompt Instructions: Respondents are prompted to indicate precisely how they felt while evaluating the target stimulus (e.g., “While listening to the audio advertisement and evaluating the product, I felt…”).
• Scoring Procedure:
  • Ensure all items are aligned in the same directional polarity prior to index construction (e.g., negative pole = 1, positive pole = 7).
  • Calculate the overall momentary affect score as the unweighted arithmetic mean of the four items:
    
    $$\text{Momentary Affect} = \frac{\sum_{i=1}^{4} X_i}{4}$$
  • Alternatively, factor-score weights derived from CFA can be utilized in latent variable structural models.
  • Higher aggregate scores represent more positive, pleasant momentary affect, whereas lower scores reflect negative, unpleasant affective states.

Permissions & Fee and Test Year

The Momentary Affect During Product Evaluation instrument was published in 2023 in the Journal of Marketing Research (Vol. 60, Iss. 4, pp. 687–706). The underlying study is copyrighted by the American Marketing Association (AMA).

Licensing and Academic Access:

• Scholarly and Academic Research: Under standard academic conventions and fair use provisions, academic researchers may utilize the semantic differential items for non-commercial educational and scientific research purposes without royalty fees, provided full bibliographic citation is accorded to Chang, Mukherjee, and Chattopadhyay (2023).
• Commercial and Commercial Testing Use: Commercial entities, corporate research labs, or private testing platforms seeking to embed the scale within commercial diagnostic software or monetize its deployment must secure formal copyright clearance from the American Marketing Association and the authors through the Copyright Clearance Center (CCC) or the journal’s publishing partner (SAGE Publications).

References

• Chang, H. H., Mukherjee, A., & Chattopadhyay, A. (2023). More voices persuade: The attentional benefits of voice numerosity. Journal of Marketing Research, 60(4), 687–706. https://doi.org/10.1177/00222437221137050
• 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
• Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39(6), 1161–1178. https://doi.org/10.1037/h0077714
• Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. Journal of Personality and Social Psychology, 45(3), 513–523. https://doi.org/10.1037/0022-3514.45.3.513
• Schwarz, N., & Clore, G. L. (1988). How do I feel about it? The informative function of affective states. In K. Fiedler & J. Forgas (Eds.), Affect, Cognition, and Social Behavior (pp. 44–62). C.J. Hogrefe.
• Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063–1070. https://doi.org/10.1037/0022-3514.54.6.1063
• Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35(2), 151–175. https://doi.org/10.1037/0003-066X.35.2.151

Items of the Scale