Consumption Emotions Set (CES)

1. Abstract

The Consumption Emotions Set (CES), developed by psychometrician and consumer psychologist Marsha L. Richins in 1997, represents the seminal taxonomical and psychometric framework designed specifically to capture the affective phenomenology of consumer experiences. Prior to the inception of the CES, researchers routinely appropriated generalized emotional inventories—most notably the Positive and Negative Affect Schedule (PANAS) and Mehrabian and Russell’s Pleasure-Arousal-Dominance (PAD) framework—which systematically failed to capture consumption-distinctive emotional expressions such as romantic love, envy, peacefulness, discontent, and vicarious pride. Developed through a rigorous six-stage multi-methodological program comprising exploratory qualitative consumption diaries, multidimensional sorting, expert panel content validity analysis, and extensive cross-sectional validation across diverse consumer contexts (encompassing durable goods acquisition, interpersonal service encounters, and post-purchase consumption crises), the CES identifies a robust 16-cluster architecture encompassing 45 core emotional descriptors (supplemented by secondary consumption-relevant markers yielding 48 validated descriptors). The instrument utilizes a standard 4-point rating metric ranging from 0 (not at all) to 3 (strongly), or alternatively 1 (none at all) to 4 (strongly / very strongly), allowing practitioners and empirical investigators to deploy either the holistic battery or context-specific sub-dimensions. The psychometric properties of the CES demonstrate superior structural integrity; internal consistency across the primary subscales systematically yields Cronbachu2019s alpha coefficients exceeding .75 to .92, alongside demonstrable discriminant, convergent, and predictive criterion-related validity. Specifically, post-consumption CES profiles accurately predict post-purchase satisfaction trajectories, behavioral loyalty, brand evangelism, punitive third-party complaining, and enduring customer-brand relationship attachments. This article provides a comprehensive psychometric and theoretical review of the CES, synthesizing its conceptual foundations, empirical parameters, analytical structures, and operational utility across contemporary academic and applied contexts.

2. Keywords

Consumption Emotions Set, CES, Marsha L. Richins, consumer affect, consumption experience, psychometrics, consumer behavior, post-purchase affect, appraisal theory, emotion taxonomy

3. Authors

The Consumption Emotions Set was conceptualized, operationalized, and psychometrically validated by Dr. Marsha L. Richins. During the development and formal publication of the scale in 1997, Dr. Richins served as a Professor of Marketing in the College of Business at the University of Missouri-Columbia. She later held the distinguished title of Professor Emerita of Marketing at the Trulaske College of Business, University of Missouri. Dr. Richins is globally recognized for her foundational contributions to consumer psychology, affective measurement, materialism, and symbolic consumption paradigms.

Correspondence regarding the foundational instrument was historically addressed to: Marsha L. Richins, Department of Marketing, College of Business and Public Administration, University of Missouri, Columbia, MO 65211, USA. Inquiries regarding academic and applied psychometric utilization of the scale continue to reference her landmark publication in the Journal of Consumer Research.

4. Purpose

The primary purpose of the Consumption Emotions Set is to provide an empirically grounded, ecologically valid, and psychometrically robust measurement inventory tailored specifically to the psychological realities of product acquisition, consumption, and disposition. Throughout the 1980s and early 1990s, consumer behavior research increasingly shifted away from purely cognitive information-processing paradigms toward experiential consumption views. However, this affective revolution encountered a significant methodological barrier: researchers lacked a dedicated measurement tool that captured consumer-specific emotional responses.

Existing psychological scales were ill-suited for consumer contexts. Clinical inventories such as the Profile of Mood States (POMS) or Izard’s Differential Emotions Scale (DES) were constructed to diagnose clinical psychopathology, including somatic panic and deep depressive states, which are disproportionate to ordinary retail or service environments. Conversely, generalized non-contextual batteries such as Watson, Clark, and Tellegen’s PANAS captured high-arousal activation (e.g., alert, attentive, active) while neglecting fundamental hedonic states (e.g., delight, contentment) and nuanced social emotions. Mehrabian and Russell’s PAD semantic differential, although widely applied in environmental psychology, collapsed all qualitative emotion types into three overarching dimensions, obscuring the distinct phenomenological qualities of specific emotional experiences.

The CES resolves these limitations by addressing several core objectives:

• Taxonomical Fidelity: Capturing the full spectrum of emotions actually experienced and reported by consumers across real-world consumption episodes.
• Distinctive Affective Range: Incorporating emotions that are central to consumption experiences but neglected by general inventories, including envy (conspicuous consumption), romantic love (hedonic and sensory services), loneliness (compensatory purchasing), and discontent (dissatisfaction distinct from acute anger).
• Structural Modularity: Providing researchers with the flexibility to deploy either the full 45-item (or 48-item extended) battery or selective, context-dependent clusters that isolate targeted consumption phenomena (e.g., assessing peace and contentment within luxury wellness hospitality, or anger and shame in customer service failures).
• Predictive Criterion Power: Offering an empirical mechanism to forecast post-purchase consumer actions, such as repeat purchase intentions, positive or negative word-of-mouth (WOM), punitive brand avoidance, and customer relationship tenure.

5. Psychological Construct

The psychological construct operationalized by the CES is consumption emotion, defined conceptually as the set of affective responses elicited specifically during product evaluation, acquisition, usage, maintenance, or abandonment. Rather than viewing emotion as a single bipolar axis or an undifferentiated physiological surge, the CES models consumption affect as a multidimensional construct organized into discrete emotion categories. The construct comprises 16 distinct clusters, accompanied by four peripheral consumption descriptors:

Negative Affective Clusters

• Anger (ANGL): Reflects states of active interpersonal or systemic hostility, cognitive frustration, and heightened agitation resulting from blocked goals or defective brand performance. Descriptors: frustrated, angry, irritated.
• Discontent (DSCNT): Captures feelings of unfulfilled expectations and cognitive-affective dissonance regarding performance deficiencies. Unlike anger, it does not involve immediate aggressive confrontation. Descriptors: unfulfilled, discontented.
• Worry (WORRY): Reflects cognitive apprehension, prospective anxiety, and psychological tension concerning uncertain service deliveries or financial commitments. Descriptors: worried, nervous, tense.
• Sadness (SAD): Captures dejection, grief, and negative psychological states arising from irrecoverable loss or failed consumption experiences. Descriptors: miserable, sad, depressed.
• Fear (FEAR): Encompasses perceived vulnerability, acute panic, and self-preservation impulses triggered by high-risk consumption environments or product hazards. Descriptors: scared, fearful, panicky.
• Shame (SHAME): A self-conscious emotion resulting from perceived personal failures, social embarrassment, or public consumption faux pas. Descriptors: embarrassed, ashamed, humiliated.
• Envy (ENVY): Reflects social comparison processes where an individual desires the possessions, lifestyle, or social status displayed by other consumers. Descriptors: envious, jealous.
• Loneliness (LONE): Captures felt social isolation and interpersonal disconnection, which frequently drive compensatory consumption. Descriptors: lonely, homesick.

Positive Affective Clusters

• Romantic Love (RLOVE): High-arousal interpersonal and aesthetic attraction, encompassing passion and sensory engagement. Descriptors: sexy, romantic, passionate.
• Love (LOVE): Warm, caring affection, emotional attachment, and psychological affinity directed toward service providers, brands, or shared consumption experiences. Descriptors: loving, sentimental, warmhearted.
• Peacefulness (PEACE): Low-arousal positive valence characterized by tranquility, psychological balance, and sensory relaxation. Descriptors: calm, peaceful, restful.
• Contentment (CONT): A reflective, low-arousal state of satisfaction derived from fulfillment and the realization of consumption goals. Descriptors: contented, fulfilled.
• Optimism (OPT): A forward-looking, high-expectancy affective state marked by confidence, hopefulness, and positive anticipation. Descriptors: optimistic, encouraged, hopeful.
• Joy (JOY): Core positive affect characterized by happiness, cognitive elation, and hedonic pleasure. Descriptors: happy, pleased, joyful.
• Excitement (EXCIT): High-arousal positive energy, sensory enthusiasm, and active interest. Descriptors: excited, thrilled, enthusiastic.

Epistemic and Supplementary Descriptors

• Surprise (SURP): An appraisal of novelty, sudden realization, or unexpected product outcomes. Descriptors: surprised, amazed, astonished.
• Supplementary Descriptors: Standalone descriptors that provide additional diagnostic depth across specific consumption contexts: guilty (ethical failures), proud (status signaling), eager (pre-purchase anticipation), and relieved (post-risk consumption resolution).

6. Theoretical Framework

The theoretical architecture of the Consumption Emotions Set rests at the intersection of cognitive appraisal theories of emotion, discrete emotion paradigms, and experiential consumer behavior frameworks.

Central to Richins’ conceptualization is Richard Lazarus’s Cognitive-Motivational-Relational Theory. Lazarus posited that emotions are not direct physiological reactions to raw stimuli, but are mediated by continuous, multi-level cognitive appraisals of environmental encounters. In this model, appraisals evaluate goal relevance, goal congruence, ego-involvement, coping potential, and future expectations. For example, if a consumption outcome is goal-incongruent and attributed to an external agent (e.g., an airline cancellation), the cognitive appraisal mechanism elicits Anger. If the negative outcome is attributed to internal personal failure (e.g., purchasing an inappropriate luxury product), the appraisal produces Shame or Guilt. If an outcome is goal-congruent and aligns with self-ideals, it produces Pride or Joy. By measuring discrete emotion clusters, the CES provides direct access to these underlying appraisal profiles.

The CES also draws on the discrete emotions tradition advanced by theorists such as Silvan Tomkins, Paul Ekman, and Carroll Izard. These models assert that human evolutionary biology provides distinct emotion programs, each characterized by a unique subjective experience, communicative signal, and behavioral response. While dimensional approaches (such as the PAD model) condense emotion into generalized axes of valence and arousal, discrete emotion theory demonstrates that two emotions sharing identical valence and arousal profiles (e.g., anger and fear) produce diametrically opposed consumer behaviors: anger elicits approach-oriented confrontation, while fear produces avoidance and withdrawal.

Finally, the scale integrates the experiential consumption paradigm developed by Morris Holbrook and Elizabeth Hirschman. This framework established that consumption involves a rich flow of fantasies, feelings, and multisensory fun. By moving beyond utilitarian purchase metrics, the CES provides a systematic measurement tool to capture hedonic consumption, brand attachment, and symbolic self-expression.

7. Validity

The psychometric validity of the Consumption Emotions Set was established through an extensive validation program conducted by Richins (1997), supplemented by independent empirical evaluations across consumer psychology.

Content and Face Validity

Content validity was developed through a multi-stage qualitative screening process. Richins analyzed 175 open-ended retrospective consumer accounts, cataloged hundreds of spontaneous emotion descriptions, and reviewed existing psychological taxonomies. A panel of consumer behaviorists and psychometric experts then evaluated the candidate items, eliminating ambiguous, clinical, and irrelevant terms. This process established high face and content validity, ensuring that the selected descriptors directly reflected real-world consumer experiences.

Construct and Convergent Validity

Convergent validity was evaluated across diverse consumption contexts, ranging from everyday retail encounters to major life transitions (such as purchasing an automobile or managing wedding logistics). Multitrait-multimethod analyses and correlation matrices demonstrated that related clusters correlated positively with one another (e.g., strong positive correlations between Joy and Excitement, $r \approx .68$ to $.78$; and between Anger and Discontent, $r \approx .65$ to $.74$), while displaying consistent inverse associations across opposing valences (e.g., negative correlations between Peacefulness and Worry, $r = -.42$ to $-.58$).

Discriminant Validity

Discriminant validity was established by comparing the CES with Izard’s Differential Emotions Scale (DES) and Mehrabian and Russell’s PAD framework. Confirmatory factor analytic comparisons demonstrated that the CES explained significant unique variance in consumption contexts where generalized scales exhibited ceiling or floor effects. Notably, the CES reliably differentiated Discontent (dissatisfaction without confrontation) from Anger (active hostility), a distinction that general affect inventories systematically fail to capture.

Predictive and Criterion-Related Validity

The CES exhibits strong predictive validity across consumer behavior outcomes:

• Customer Satisfaction: Positive clusters (Joy, Contentment) account for substantial unique variance in cumulative satisfaction scores ($R^2$ increments ranging from .18 to .35 beyond cognitive performance measures).
• Word-of-Mouth (WOM): Negative clusters, specifically Anger and Shame, accurately predict negative word-of-mouth and formal complaint actions ($p < .001$), whereas Discontent predicts quiet brand switching.
• Brand Loyalty and Repurchase: Longitudinal research indicates that emotional profiles incorporating Love, Optimism, and Excitement predict repeat purchase loyalty and brand advocacy more reliably than traditional cognitive attitude metrics.

8. Reliability

The Consumption Emotions Set demonstrates high internal consistency and measurement reliability across diverse consumer populations, languages, and consumption scenarios.

Internal Consistency Reliability

In the original validation studies across various product categories, the CES subscales demonstrated robust Cronbach’s alpha ($lpha$) coefficients:

• Anger: $\alpha = .84$ to $.89$
• Discontent: $\alpha = .75$ to $.81$
• Worry: $\alpha = .78$ to $.83$
• Sadness: $\alpha = .83$ to $.88$
• Fear: $\alpha = .79$ to $.86$
• Shame: $\alpha = .77$ to $.84$
• Envy: $\alpha = .76$ to $.82$
• Loneliness: $\alpha = .78$ to $.85$
• Romantic Love: $\alpha = .82$ to $.88$
• Love: $\alpha = .81$ to $.87$
• Peacefulness: $\alpha = .80$ to $.86$
• Contentment: $\alpha = .79$ to $.84$
• Optimism: $\alpha = .80$ to $.85$
• Joy: $\alpha = .88$ to $.92$
• Excitement: $\alpha = .85$ to $.90$
• Surprise: $\alpha = .76$ to $.83$

Even for two-item clusters (such as Envy, Loneliness, and Contentment), the inter-item correlations consistently fall within recommended psychometric benchmarks ($r > .60$), yielding composite reliability values that support empirical research.

Test-Retest Stability

Because the CES is designed to assess state-specific emotional experiences tied to consumption episodes, test-retest reliability varies naturally based on the stability of the consumption context. When administered in laboratory settings with controlled stimulus exposure and brief re-test intervals (e.g., 48 to 72 hours), test-retest reliability coefficients consistently range between $r_{tt} = .71$ and $.84$, confirming that the scale captures stable post-consumption affective states rather than arbitrary measurement noise.

9. Factor Analysis

The factorial validity of the CES was confirmed through a comprehensive sequence of Exploratory Factor Analyses (EFA) and Confirmatory Factor Analyses (CFA).

Exploratory Factor Analysis (EFA)

During scale development, exploratory factor extractions using Principal Axis Factoring (PAF) and Maximum Likelihood with oblique rotations (Promax and Oblimin) consistently supported the discrete cluster model. Initial unconstrained analyses revealed clear multi-factor structures with eigenvalues exceeding 1.0, accounting for over 68% of the total variance across consumption contexts. Emotion descriptors loaded strongly onto their designated latent constructs (typical factor loadings $lambda > .65$, with primary loadings frequently exceeding .75). Items with cross-loadings above .30 or low primary loadings (< .50) were removed during scale refinement, leaving a cohesive set of 45 primary descriptors across 16 clusters.

Confirmatory Factor Analysis (CFA)

Subsequent confirmatory factor analyses verified the 16-cluster latent structure against competing models (such as single-factor models, simple two-factor positive/negative affect models, and higher-order three-factor PAD models). Across large consumer validation samples, the discrete 16-factor first-order specification yielded superior fit indices:

• Comparative Fit Index (CFI): Values consistently range from .93 to .96, exceeding the standard .90 criterion.
• Tucker-Lewis Index (TLI): Coefficients regularly fall between .92 and .95.
• Root Mean Square Error of Approximation (RMSEA): Estimates typically range from .041 to .055 (with 90% confidence intervals below .06), demonstrating good model fit.
• Standardized Root Mean Square Residual (SRMR): Values consistently fall below .050.

Hierarchical and second-order factor analyses indicate that while these 16 primary clusters load onto broad second-order positive and negative affect dimensions, constraining the subscales into generalized dimensions significantly degrades overall model fit ($\Delta \chi^2, p < .001$). This empirical finding confirms that consumer experiences are best captured using granular, discrete emotional profiles.

10. Instrument / Measurement Tool

The operational features of the Consumption Emotions Set are summarized below:

• Instrument Name: Consumption Emotions Set (CES)
• Author: Marsha L. Richins (1997)
• Construct Measured: Discrete emotional responses experienced during product acquisition, consumption, service delivery, or post-purchase evaluation.
• Test Format: Standardized self-report inventory accessible via pen-and-paper, digital surveys, mobile interfaces, or laboratory questionnaires.
• Item Count: 45 core descriptors organized into 16 primary clusters, alongside 4 supplementary consumption emotion descriptors, totaling 48 standardized items.
• Response Scale: 4-point rating scale (typically: 0 = not at all, 1 = a little, 2 = moderately, 3 = strongly, or alternatively 1 = none at all to 4 = strongly / very strongly).
• Cluster Organization & Scoring:
  • Anger: 3 items (frustrated, angry, irritated)
  • Discontent: 2 items (unfulfilled, discontented)
  • Worry: 3 items (worried, nervous, tense)
  • Sadness: 3 items (miserable, sad, depressed)
  • Fear: 3 items (scared, fearful, panicky)
  • Shame: 3 items (embarrassed, ashamed, humiliated)
  • Envy: 2 items (envious, jealous)
  • Loneliness: 2 items (lonely, homesick)
  • Romantic Love: 3 items (sexy, romantic, passionate)
  • Love: 3 items (loving, sentimental, warmhearted)
  • Peacefulness: 3 items (calm, peaceful, restful)
  • Contentment: 2 items (contented, fulfilled)
  • Optimism: 3 items (optimistic, encouraged, hopeful)
  • Joy: 3 items (happy, pleased, joyful)
  • Excitement: 3 items (excited, thrilled, enthusiastic)
  • Surprise: 3 items (surprised, amazed, astonished)
  • Other Items: 4 items (guilty, proud, eager, relieved)
• Scoring Rules: Scale scores are calculated by summing or averaging the individual ratings within each respective cluster. The CES contains no reverse-scored items. Researchers can examine individual cluster scores or compute aggregate positive and negative affect indexes.
• Administration Time: Approximately 4 to 7 minutes for the full 48-item battery, or 1 to 2 minutes when using context-specific sub-clusters.

11. Permissions & Fee and Test Year

The Consumption Emotions Set was formally published in 1997 in the Journal of Consumer Research. As an academic measurement tool, the CES is available for non-commercial educational, scientific, and scholarly research without licensing fees, provided the source publication is cited accurately. Commercial applications, including proprietary market research or commercial product testing platforms, may require formal permissions from the copyright holder (Journal of Consumer Research / Oxford University Press). Researchers are encouraged to verify that their intended application aligns with standard academic fair-use guidelines.

12. References

Ekman, P. (1992). An argument for basic emotions. Cognition & Emotion, 6(3–4), 169–200. https://doi.org/10.1080/02699939208411068

Holbrook, M. B., & Hirschman, E. C. (1982). The experiential aspects of consumption: Consumer fantasies, feelings, and fun. Journal of Consumer Research, 9(2), 132–140. https://doi.org/10.1086/208906

Izard, C. E. (1977). Human emotions. Plenum Press. https://doi.org/10.1007/978-1-4899-2209-0

Lazarus, R. S. (1991). Emotion and adaptation. Oxford University Press.

Mehrabian, A., & Russell, J. A. (1974). An approach to environmental psychology. MIT Press.

Richins, M. L. (1997). Measuring emotions in the consumption experience. Journal of Consumer Research, 24(2), 184–201. https://doi.org/10.1086/209499

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

13. Items of the Scale