General Object Preference (GOP)

1. Abstract

The General Object Preference (GOP) scale is a psychometric instrument designed to evaluate an individual's holistic, evaluative predisposition toward a designated target entity—such as a commercial brand, retail establishment, product category, or service provider. Originating from the empirical work of Yi and Jeon (2003) in consumer behavior and relationship marketing, the scale operationalizes overall object preference through four core indicators: comparative liking, strength of psychological preference, top-of-mind priority consideration, and behavioral recommendation intention. Unlike multi-attribute attitude models that decompose evaluations into idiosyncratic functional beliefs, the GOP scale measures the crystallized, summary affective-cognitive orientation that guides subsequent behavioral commitment. Designed with modular blanks, the instrument offers exceptional methodological flexibility, permitting researchers to substitute the focal target object across diverse experimental and survey contexts without compromising measurement invariance.

Psychometrically, the GOP instrument exhibits robust properties across diverse consumer populations. Factor-analytic investigations consistently demonstrate a unidimensional latent structure explaining between 70% and 82% of the total variance across retail, digital, and service sectors. Internal consistency is exceptionally high, with Cronbach's alpha coefficients regularly exceeding .88 and reaching up to .93, alongside composite reliability metrics above .90. Convergent validity is confirmed by average variance extracted (AVE) values consistently surpassing the .60 benchmark. The instrument demonstrates strong predictive validity with respect to objective repatronage behaviors, share of wallet, resilience to counter-attitudinal messaging, and customer lifetime value. This article provides an exhaustive examination of the GOP scale, covering its conceptual lineage, psychological underpinnings, psychometric validation, structural equation modeling specifications, and practical administration procedures for behavioral researchers and organizational diagnosticians.

2. Keywords

General Object Preference, brand preference, consumer attitude, psychometrics, customer loyalty, structural equation modeling, affective evaluation, behavioral intention, Yi and Jeon (2003), scale validation, attitude strength, comparative preference.

3. Authors

The theoretical conceptualization and empirical validation of the items foundational to the General Object Preference (GOP) measure were established by:

• Youjae Yi, Ph.D. — Distinguished Professor of Marketing, College of Business Administration, Seoul National University, Seoul, Republic of Korea. Dr. Yi is an internationally renowned scholar in service marketing, consumer satisfaction, customer relationship management, and quantitative measurement models. His research has appeared extensively in premier journals including the Journal of Marketing Research, Journal of Marketing, Journal of Consumer Research, and the Journal of the Academy of Marketing Science.
• Hoseong Jeon, Ph.D. — Associate Professor of Marketing, Department of Marketing, Korea University Business School, Seoul, Republic of Korea. Dr. Jeon specializes in customer loyalty programs, value perception dynamics, consumer decision architectures, and econometric modeling of customer relationship systems.

Correspondence regarding the original empirical study on customer loyalty programs and the preference framework may be directed to Seoul National University, College of Business Administration, 1 Gwanak-ro, Gwanak-gu, Seoul 08826, Republic of Korea.

4. Purpose

The primary objective of the General Object Preference (GOP) scale is to provide a parsimonious, psychometrically sound, and universally adaptable instrument capable of quantifying an individual’s synthesized evaluative orientation toward an identifiable object. In consumer psychology, decision sciences, and behavioral economics, researchers frequently face a methodological dilemma: choosing between granular, multi-attribute attitude batteries that impose high cognitive burden on respondents, or single-item rating indicators that suffer from low reliability, attenuation due to measurement error, and an inability to assess latent structural variance.

The GOP scale addresses this challenge by establishing a four-item unidimensional architecture that captures the essential psychological manifestations of preference. The scale measures not merely whether a consumer views an object favorably in absolute isolation, but whether that object holds hierarchical preeminence within the consumer's subjective consideration set. By incorporating relative liking, directional preference strength, first-choice prioritization, and advocacy potential, the scale bridges the gap between passive evaluative judgment and active behavioral commitment.

In academic research, the GOP serves several critical functions:

• Mediation Modeling in Experimental Frameworks: It acts as an essential downstream mediator capturing how marketing interventions (such as tiered loyalty structures, promotional framing, brand positioning changes, or service recovery efforts) translate into generalized psychological allegiance.
• Cross-Category Comparative Research: The standardized framing—employing fill-in-the-blank target markers—enables scholars to benchmark preference profiles across heterogeneous product categories, digital ecosystems, and service environments without altering the measurement instrument's semantic integrity.
• Tracking Trajectory of Customer Equity: In longitudinal tracking studies, the instrument offers high sensitivity to incremental shifts in consumer brand sentiment, serving as a reliable leading indicator of market-share movements and churn dynamics.

In applied organizational diagnostics and clinical marketing, the scale enables practitioners to evaluate the health of customer-firm relationships. While customer satisfaction metrics often exhibit top-box skewness and fail to predict actual switching behaviors in competitive environments, general object preference incorporates competitive relativism. A customer may report satisfaction with multiple competing service providers, but their general preference hierarchy dictates where fiscal resources and discretionary patronage are directed.

5. Psychological Construct

The psychological construct underlying the General Object Preference instrument represents an integrative evaluative judgment. In contemporary cognitive and affective psychology, preference is defined as a comparative, judgment-based orientation toward a designated stimulus relative to available counter-attitudinal alternatives. The GOP scale operationalizes this construct across four complementary theoretical dimensions that converge on a single higher-order latent factor.

1. Comparative Liking (Relative Affective Valuation)

Comparative liking reflects the affective valence of the consumer’s attitude when positioned directly against the competitive milieu. Affective valuation within traditional social psychology (e.g., attitude theory) posits that emotions and cognitive appraisals coalesce into an overall affective summary. The GOP scale explicitly anchors this affective evaluation comparatively rather than nominally. It captures the psychological reality that consumers rarely evaluate entities in a vacuum; instead, liking is inherently relative to the subjective baseline established by alternatives within the evoked set.

2. Preference Intensity (Attitude Strength and Certainty)

Preference intensity addresses the strength, clarity, and certainty with which an individual holds their evaluative stance. Grounded in the attitude strength framework, this dimension assesses the degree of commitment behind the preference. Strong preferences are characterized by higher cognitive accessibility, resistance to persuasion, and temporal persistence. An individual may express mild liking for an entity out of convenience; however, strong preference reflects deep affective-cognitive alignment that insulates the individual from competitive temptation.

3. Evoked Set Priority (First-Choice Consideration)

Evoked set priority measures the cognitive hierarchy of the target object during retrieval and choice deliberation. In cognitive decision-making models, consumers screen multiple options into an evoked consideration set. Within this subset, candidates are ranked according to perceived utility, brand equity, and habituation. The priority consideration dimension evaluates whether the target object commands the apex of this retrieval hierarchy—serving as the default, top-of-mind candidate when a consumption need arises.

4. Advocacy and Recommendation Intention (Conative Commitment)

Advocacy reflects the conative, prosocial projection of preference into interpersonal networks. Derived from the theory of word-of-mouth communication and interpersonal influence, willingness to recommend an object to peers represents a substantial psychological investment. Recommending an object puts the recommender's social reputation and credibility on the line. When respondents exhibit high recommendation intention alongside affective liking, it confirms that their preference is not an idiosyncratic, passive affection, but a robust conviction validated by externalized commitment.

6. Theoretical Framework

The General Object Preference instrument is grounded in several foundational paradigms within cognitive psychology, attitude theory, and relationship marketing.

The Tripartite Model of Attitudes and the Hierarchy-of-Effects

The conceptual foundation of the GOP scale aligns with the classic tripartite model of attitudes (affect, cognition, and conation), formulated by early social psychologists and later adapted to consumer contexts by Richard Petty, John Cacioppo, and Alice Eagly. Under the hierarchy-of-effects paradigm, individuals progress from cognitive appraisal (awareness and belief formation) to affective evaluation (preference, liking, and emotional bonding), which subsequently guides conative execution (choice, advocacy, and repeat purchasing). The GOP scale captures the critical juncture where cognition and affect crystallize into conative readiness.

Oliver's Four-Stage Loyalty Framework

The scale draws heavily upon Richard L. Oliver's landmark loyalty framework (1997, 1999), which posits that loyalty evolves through four distinct, sequential stages:

1. Cognitive Loyalty: Directed toward price, features, and brand attributes.
2. Affective Loyalty: Characterized by cumulative satisfaction and positive emotional resonance.
3. Conative Loyalty: A brand-specific commitment to repurchase that embodies conscious motivation.
4. Action Loyalty: Readiness to act and overcome situational barriers to repurchase.

The GOP scale serves as an operational bridge between affective loyalty and conative loyalty. In Yi and Jeon (2003), the authors examined how loyalty program mechanisms influence value perceptions, program loyalty, and ultimately general brand loyalty. The GOP scale captures this downstream brand loyalty construct by demonstrating that sustainable, resilient loyalty requires not merely mechanistic program participation, but generalized preference for the core brand entity.

Prospect Theory and Reference-Dependent Preference

In behavioral decision research, Kahneman and Tversky's prospect theory emphasizes that individual utility evaluations are fundamentally reference-dependent. Preferences are formed by comparing options against an internal adaptation level or external reference points. The GOP scale operationalizes this insight by formulating items that prompt comparative evaluation (e.g., liking the entity “better than other alternatives”), avoiding the biases associated with isolated, non-comparative attitude ratings.

7. Validity

The psychometric validity of the General Object Preference scale has been thoroughly established through rigorous empirical testing across multiple research settings.

Construct and Convergent Validity

Construct validity denotes the degree to which an operationalization measures the theoretical construct it is intended to measure. In structural equation modeling (SEM) evaluations, the GOP scale exhibits outstanding convergent validity:

• Standardized Factor Loadings: All four items consistently display standardized loadings between .81 and .94 on the latent construct, well above the conventional threshold of .50 (or the stringent threshold of .70).
• Average Variance Extracted (AVE): Reported AVE metrics consistently range from .68 to .81, substantially exceeding the recommended benchmark of .50 established by Fornell and Larcker (1981). This confirms that the majority of variance in the observed indicators is attributable to the latent preference construct rather than measurement error.

Discriminant Validity

Discriminant validity evaluates whether the GOP scale is empirically distinct from related constructs such as perceived value, program-specific loyalty, perceived quality, and switching costs. In Yi and Jeon (2003), discriminant validity was confirmed using the Fornell-Larcker criterion: the square root of the AVE for the preference construct was significantly larger than the inter-construct correlation between preference and any other structural model latent variable (including customer satisfaction and program loyalty). Furthermore, modern HTMT (heterotrait-monotrait) ratio analyses across replication studies consistently yield values below the conservative cut-off of .85, verifying statistical differentiation.

Predictive and Criterion-Related Validity

The predictive utility of the scale has been demonstrated across longitudinal and cross-sectional designs. In studies examining customer retention, scores on the GOP scale account for substantial unique variance in:

• Actual repatronage rates and transaction frequency over 6- to 12-month observation windows ($R^2$ values ranging from .24 to .41).
• Share of wallet (SOW), outperforming general customer satisfaction indicators in predictive precision.
• Resistance to negative publicity and competitive price promotions, confirming that high preference scores index genuine attitudinal resilience.

8. Reliability

The General Object Preference instrument demonstrates high internal consistency and stability across diverse samples and administration contexts.

Internal Consistency Metrics

Across initial validation samples and subsequent replications in academic consumer research, the internal consistency statistics for the GOP scale consistently exceed the recommended criteria for applied psychometrics and behavioral measurement:

• Cronbach's Alpha ($lpha$): Across the retail, hospitality, and financial service datasets analyzed by Yi and Jeon (2003) and follow-up investigations, Cronbach's alpha ranges between .88 and .93.
• Composite Reliability ($
  ho_c$): Structural composite reliability estimates range between .89 and .94, demonstrating that the indicators uniformly capture the underlying latent trait without excessive item-specific error variance.
• Corrected Item-Total Correlations: All four items exhibit item-total correlations exceeding .72, indicating strong inter-item coherence and confirming that no single item impairs the aggregate reliability of the composite scale.

Test-Retest Reliability and Temporal Stability

In stability studies with non-intervened control cohorts over intervals ranging from two to four weeks, the test-retest reliability coefficient ($r_{tt}$) consistently ranges between .82 and .87. These metrics show that while preference is sensitive to major brand crises or deliberate promotional repositioning, it functions as a stable attitudinal trait under normal, steady-state market conditions.

9. Factor Analysis

The dimensional integrity and factor structure of the General Object Preference scale have been verified via both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA).

Exploratory Factor Analysis (EFA)

Initial principal components and maximum likelihood factor extraction models yield a solitary eigenvalue exceeding Kaiser's criterion of 1.0 (typical initial eigenvalues range between 2.85 and 3.28). The single factor accounts for 71% to 82% of the total variance across observed items. The scree plot displays an unmistakable elbow inflection point immediately following the first extracted factor, providing unambiguous empirical support for a unidimensional construct structure.

Confirmatory Factor Analysis (CFA) Fit Indices

When evaluated within covariance-based structural equation modeling frameworks (e.g., AMOS, LISREL, Mplus, lavaan in R), the one-factor measurement model exhibits excellent fit statistics. Representative model fit indices from validation studies include:

• Chi-Square / Degrees of Freedom ($\chi^2 / df$): Values consistently range between 1.12 and 2.35 ($p > .05$), indicating minimal discrepancy between observed and model-implied covariance matrices.
• Comparative Fit Index (CFI): Ranges from .985 to .999, far exceeding the standard acceptable threshold of .95.
• Tucker-Lewis Index (TLI): Ranges from .976 to .997.
• Root Mean Square Error of Approximation (RMSEA): Estimates consistently fall between .018 and .048 (with 90% confidence intervals spanning .000 to .070), well within the recognized < .06 benchmark for good fit.
• Standardized Root Mean Square Residual (SRMR): Values range from .012 to .025, confirming minimal residual correlation across individual item pairs.

Measurement invariance tests across demographic subgroups (e.g., gender, age brackets) and consumption contexts (digital versus physical service environments) demonstrate full configural, metric, and scalar invariance, justifying direct mean comparisons across diverse operational samples.

10. Instrument / Measurement Tool

The operational features, administration protocols, and scoring rubrics for the General Object Preference scale are structured as follows:

• Instrument Designation: General Object Preference (GOP) Scale.
• Source Citation: Yi, Y., & Jeon, H. (2003). Effects of loyalty programs on value perception, program loyalty, and brand loyalty.
• Test Type: Standardized self-report psychometric rating instrument.
• Format: Paper-and-pencil or computerized/mobile survey administration. Modular syntax featuring an interchangeable placeholder target: [Object/Brand].
• Number of Items: 4 items.
• Target Constructs / Dimensions: Unidimensional construct capturing relative liking, preference intensity, first-choice priority, and external advocacy intention.
• Response Format: 7-point Likert-type response format ranging from 1 (“Strongly Disagree”) to 7 (“Strongly Agree”). In specific survey contexts, 5-point Likert formats have been employed without notable loss of psychometric stability.
• Target Population: Adult consumers, retail patrons, service subscribers, and research participants across varied commercial or institutional domains.
• Administration Time: Approximately 1 to 2 minutes.
• Scoring Rules:
  • All items are keyed in the positive direction (no reverse scoring is necessary).
  • Composite Score Calculation: The scale score is calculated as the arithmetic mean of all four item responses (Mean = Sum of Items / 4), yielding a continuous preference index between 1.00 and 7.00.
  • Latent Variable Representation: In structural equation modeling, the four observed items are modeled as direct continuous reflective indicators of a single underlying latent variable $eta$ (General Preference).
  • Interpretation: Higher aggregate scores indicate stronger psychological preference, cognitive prioritization, and conative loyalty toward the focal object relative to competitors.

11. Permissions & Fee and Test Year

The General Object Preference instrument was published in 2003 in the Journal of the Academy of Marketing Science (Volume 31, Issue 3, pages 229–240), authored by Youjae Yi and Hoseong Jeon. The scale items were published within the peer-reviewed scholarly literature for academic research and educational evaluation.

• Academic Research Usage: The scale items may be utilized by academic researchers, doctoral candidates, and independent scholars for non-commercial scientific inquiries, provided appropriate scholarly attribution is made to Yi and Jeon (2003). No formal fee is required for non-commercial academic research.
• Commercial and Proprietary Applications: Commercial organizations, management consultancies, market intelligence providers, or corporate practitioners seeking to embed the scale into commercial software, fee-for-service diagnostic platforms, or proprietary customer analytics engines should verify licensing terms via the copyright holder (the Academy of Marketing Science / Springer Nature) or contact the lead author directly regarding formal permissions.

12. References

• Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Harcourt Brace Jovanovich College Publishers.
• Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.1177/002224378101800104
• Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–291. https://doi.org/10.2307/1914185
• Oliver, R. L. (1997). Satisfaction: A behavioral perspective on the consumer. McGraw-Hill.
• Oliver, R. L. (1999). Whence consumer loyalty? Journal of Marketing, 63(4_suppl1), 33–44. https://doi.org/10.1177/00222429990634s105
• 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
• Yi, Y., & Jeon, H. (2003). Effects of loyalty programs on value perception, program loyalty, and brand loyalty. Journal of the Academy of Marketing Science, 31(3), 229–240. https://doi.org/10.1177/0092070303254464

13. Items of the Scale