Loyalty Program Preference

Abstract

The Loyalty Program Preference (LPP) scale is a psychometric instrument developed by Youjae Yi and Hoseong Jeon (2003) to assess consumer comparative evaluation, relative attitudinal preference, and conative advocacy toward customer reward schemes. In modern relationship marketing, firms deploy customer loyalty programs (such as point collection systems, tiered membership programs, and cash-back structures) to increase retention, raise switching costs, and foster long-term customer equity. However, consumer engagement is heavily contingent upon how favorable a specific program architecture is perceived in contrast to competing marketplace offerings. The LPP scale captures this psychological phenomenon through a parsimonious, unidimensional, three-item self-report battery administered via a 7-point Likert or semantic differential response format. The instrument evaluates three tightly linked facets of consumer decision-making: comparative cognitive evaluation, direct affective preference, and behavioral recommendation intentions (advocacy/word-of-mouth). Methodologically, the scale demonstrates exceptional psychometric properties across diverse retail and service industry settings, yielding high internal consistency reliability (typically with Cronbach’s alpha exceeding .85 and composite reliability values well above benchmark thresholds), robust unidimensionality supported by confirmatory factor analysis (CFA), and distinct discriminant validity separating program loyalty from brand loyalty and perceived value. By isolating consumer preference for the loyalty mechanism itself rather than the core brand, the LPP provides researchers and consumer psychologists with a precise diagnostic tool to evaluate program design efficacy, reward timing (immediate versus delayed), and value congruence across high-involvement and low-involvement product categories.

Keywords

Loyalty Program Preference, Program Loyalty, Brand Loyalty, Customer Relationship Management, Value Perception, Reward Schemes, Consumer Decision Making, Psychometrics, Structural Equation Modeling, Behavioral Intention

Authors

The Loyalty Program Preference (LPP) operationalization was established by academic researchers in marketing science and consumer psychology:

• Youjae Yi, Ph.D. — Distinguished Professor of Marketing, College of Business Administration, Seoul National University, Seoul, Republic of Korea. Recognized internationally for seminal contributions to customer satisfaction, service quality measurement, customer engagement, and structural equation modeling in consumer behavior.
• Hoseong Jeon, Ph.D. — Researcher and marketing scholar affiliated with Seoul National University at the time of the scale’s development, specializing in customer relationship management (CRM) systems, sales promotion architecture, and quantitative consumer choice modeling.

Purpose

The primary purpose of the Loyalty Program Preference (LPP) scale is to empirically quantify the degree to which an individual consumer favors, prefers, and intends to advocate for a specific customer loyalty program over existing competitive alternatives. Loyalty programs represent substantial financial investments for corporations across airline, hospitality, retail, telecommunications, and financial service industries. Despite their ubiquity, marketing researchers long struggled with a critical theoretical and empirical confound: whether repeat purchasing behavior stemmed from genuine devotion to the underlying brand (brand loyalty) or merely from opportunistic engagement with the incentives offered by the rewards program (program loyalty).

To resolve this theoretical problem, Yi and Jeon (2003) constructed the LPP measure within a comprehensive structural framework examining the differential effects of loyalty programs on customer value perception, program loyalty, and ultimate brand loyalty. The scale fulfills multiple strategic and academic research objectives:

• Isolating the Incentive Mechanism: The LPP measures attitudes toward the program as an independent structural entity. This allows researchers to isolate whether consumers find a loyalty program compelling on its own merits, independent of baseline brand sentiment.
• Comparing Program Designs: In experimental and applied market research, the scale serves as a sensitive dependent variable when contrasting diverse reward architectures, such as direct versus indirect rewards (e.g., airline miles redeemable for flights versus third-party catalog gifts) and immediate versus delayed rewards (e.g., immediate price markdowns versus long-term accumulated point thresholds).
• Analyzing Product Involvement Moderation: The instrument facilitates empirical investigation into how consumer product involvement (high-involvement goods like electronics versus low-involvement consumer packaged goods) moderates the pathway from program preference to brand loyalty.
• Diagnostic Optimization for Customer Relationship Management (CRM): Practitioners use the LPP to audit existing programs, benchmark prospective loyalty structures prior to full commercial rollout, and prevent program churn by identifying deficits in comparative attractiveness.

Psychological Construct

The psychological construct underlying the Loyalty Program Preference scale resides at the intersection of attitude theory, comparative consumer judgment, and behavioral intentions. Rather than conceptualizing program loyalty solely as repeat card-swiping or reward redemption frequency (behavioral loyalty), the LPP captures attitudinal program loyalty, which consists of three interrelated psychological components:

1. Comparative Cognitive Evaluation

Human decision-making operates through comparative reference points rather than absolute evaluation. When a consumer evaluates a customer rewards scheme, they activate internal standards and mental representations of competing programs available in the marketplace (e.g., competing airline frequent-flyer tiers or grocery store point systems). The cognitive component of the LPP reflects the consumer’s rational assessment of whether Program A offers superior utility, better redemption ratios, lower qualification thresholds, and greater transaction efficiency than Program B. This cognitive assessment aligns with mental accounting theory, where customers weigh the perceived transaction utility of the reward against the effort costs of participation.

2. Relative Affective Preference

Beyond rational calculation, consumers develop an emotional affinity or psychological leaning toward particular loyalty schemes. The affective preference dimension reflects positive valence, enthusiasm, and an overall psychological commitment to using one program over another. This dimension accounts for the feeling of being a “valued member,” the gamification appeal of accumulating status badges or tier levels, and the anticipated joy associated with reward goal attainment. When affective preference is high, the program is not merely tolerated as a utility; it is prioritized during consumption decisions.

3. Conative Advocacy and Recommendation Intentions

The conative facet translates cognitive appraisal and affective preference into future behavioral commitments. Specifically, the willingness to recommend a loyalty program to peers, colleagues, or family members represents a high-level psychological commitment. Because recommending a service puts the consumer’s personal social capital at risk, affirmative recommendation intentions signify strong psychological attachment and robust perceived superiority of the program. Within the LPP framework, this advocacy dimension ensures that preference is not merely passive acceptance, but an active driver of customer-led promotion.

Theoretical Framework

The development of the Loyalty Program Preference scale is grounded in foundational social science and consumer psychology theories, primarily combining Expectancy-Value Theory, Social Exchange Theory, and Richard Oliver’s Four-Stage Loyalty Framework.

Expectancy-Value Theory and Utilitarian Mechanics

Expectancy-value models (e.g., Fishbein & Ajzen) posit that an individual’s attitude toward an object is a function of the beliefs held about the object and the evaluative aspects of those beliefs. In the context of loyalty programs, consumers develop expectations regarding the probability of reward attainment multiplied by the perceived value of those rewards. Yi and Jeon (2003) integrated this theoretical foundation to argue that loyalty programs alter consumer value perceptions by augmenting total perceived customer value (combining functional value, monetary value, and psychological benefits). The LPP measures the resulting evaluative attitude that emerges when a customer concludes that the expected return from a specific program exceeds that of alternative systems.

Social Exchange and Equity Theory

Social Exchange Theory suggests that human interactions and relational partnerships are formed by the use of a subjective cost-benefit analysis and the comparison of alternatives. Consumers invest cognitive effort, repeat visits, personal data, and financial expenditure into a firm; in return, they expect proportional, fair, and equitable compensation through the loyalty program. When the perceived equity of the exchange is superior to competing relationship opportunities, preference crystallizes. The LPP captures the psychological equilibrium wherein the consumer perceives the firm’s reward exchange mechanism as exceptionally fair and advantageous.

Oliver’s Loyalty Hierarchy: From Program Loyalty to Brand Loyalty

A crucial theoretical contribution of Yi and Jeon’s framework is clarifying where program preference sits within Oliver’s hierarchy of loyalty: cognitive, affective, conative, and action loyalty. Yi and Jeon demonstrated that program loyalty is structurally antecedent to brand loyalty under specific conditions (notably in low-involvement settings). When a consumer forms an attitudinal preference for a loyalty program, this program-specific affinity can spill over to enhance loyalty toward the parent brand itself. Conversely, under high-involvement conditions, brand loyalty can develop independently, and the loyalty program acts as a secondary reinforcement mechanism rather than the primary driver of brand devotion.

Validity

The psychometric validity of the Loyalty Program Preference instrument has been empirically substantiated across multiple laboratory experiments and field surveys in consumer marketing research.

Construct and Convergent Validity

In the original validation study by Yi and Jeon (2003), construct validity was evaluated within a controlled experimental design involving diverse reward structures (direct vs. indirect, immediate vs. delayed) across both high-involvement (personal computers) and low-involvement (cleansing cream/toiletries) product categories. Using structural equation modeling (SEM) and confirmatory factor analysis (CFA), the items demonstrated exceptionally high standardized factor loadings onto the program preference construct (all standardized $lambda > .80$, $p < .001$). The average variance extracted (AVE) substantially exceeded the recommended .50 benchmark (typically surpassing .75), indicating that the scale items capture a large majority of common construct variance rather than measurement error.

Discriminant Validity

A critical psychometric hurdle in CRM research is establishing that program preference is empirically distinct from brand loyalty and overall brand value perception. Yi and Jeon (2003) conducted rigorous discriminant validity testing utilizing the Fornell and Larcker (1981) criterion. The squared correlation between the Loyalty Program Preference construct and Brand Loyalty was significantly lower than the AVE of either individual construct. Furthermore, chi-square difference tests comparing an unconstrained two-factor model against a constrained one-factor model (where the correlation between program loyalty and brand loyalty was fixed to 1.0) yielded highly significant chi-square increases ($\Delta \chi^2 > 35.0, p < .001$). These findings conclusively established that consumers cognitively separate their feelings toward a reward system from their feelings toward the brand providing it.

Predictive and Nomological Validity

Nomological validity has been demonstrated by embedding the LPP within broader theoretical models. The scale successfully predicts critical downstream outcome variables, including share of wallet, cross-buying intentions, program redemption rates, and resilience to competitive price promotions. In particular, the scale validated theoretical hypotheses regarding moderation: the path coefficient from program preference (program loyalty) to brand loyalty was strong and statistically significant in low-involvement markets ($\gamma = .42, p < .01$), while brand loyalty was more directly influenced by intrinsic brand perceptions in high-involvement settings, confirming theoretical expectations.

Reliability

The Loyalty Program Preference scale demonstrates consistently high internal consistency across empirical investigations:

• Internal Consistency: In the foundational Yi and Jeon (2003) investigation, the instrument achieved a Cronbach’s alpha ($\alpha$) of .89 to .93 across experimental conditions and product categories, well above the conventional academic threshold of .70 recommended for psychometric research.
• Composite Reliability (CR): Structural equation modeling evaluations report composite reliability values exceeding .90, demonstrating exceptional internal cohesion among the three operational items without psychometric redundancy.
• Item-Total Correlations: Corrected item-total correlations for each of the three indicator items routinely range between .72 and .86, confirming that each item contributes strongly and uniquely to the measurement of the underlying latent dimension.
• Cross-Sample Stability: Subsequent replication studies examining airline frequent-flyer programs, hotel rewards, and omnichannel retail loyalty programs have mirrored these findings, reporting internal consistency estimates consistently between .86 and .94, confirming the tool’s structural stability across differing cultural and commercial contexts.

Factor Analysis

The latent dimensionality of the Loyalty Program Preference instrument has been subjected to rigorous factor analytic procedures to verify its structural integrity.

Exploratory Factor Analysis (EFA)

During initial scale development, principal components analysis with varimax rotation reliably extracts a single eigenvalue greater than 1.0 (typically accounting for 75% to 84% of the total variance across items). The scree plot clearly displays a severe elbow after the first factor, confirming the absence of multi-dimensional noise or secondary nuisance factors.

Confirmatory Factor Analysis (CFA)

In structural confirmatory factor models, the unidimensional three-item structure demonstrates exemplary goodness-of-fit indices. When embedded in full structural measurement models alongside value perception, brand loyalty, and involvement, typical fit metrics reported in the literature include:

• Comparative Fit Index (CFI): $ge .97$ (often exceeding .99)
• Tucker-Lewis Index (TLI): $ge .96$
• Root Mean Square Error of Approximation (RMSEA): $le .05$ (with 90% confidence intervals bounded within .00 to .08)
• Standardized Root Mean Square Residual (SRMR): $le .03$
• Normed Chi-Square ($\chi^2/df$): Frequently between 1.10 and 2.10, indicating satisfactory model-to-data congruence.

Because a saturated three-item single latent variable model has zero degrees of freedom ($df = 0$), CFA fit is typically evaluated either within a multi-construct measurement model or with equality constraints imposed across factor loadings. Across these evaluations, standardized factor loadings consistently fall in the range of $lambda = .82$ to $.94$, with all loadings statistically significant at $p < .001$.

Instrument / Measurement Tool

The technical specifications and administration parameters of the Loyalty Program Preference instrument are structured as follows:

• Instrument Name: Loyalty Program Preference (LPP) / Program Loyalty Scale
• Construct Measured: Attitudinal preference, comparative attractiveness, and recommendation advocacy toward a customer loyalty scheme.
• Item Count: 3 items.
• Response Format: 7-point Likert scale (ranging from 1 = Strongly Disagree to 7 = Strongly Agree) or 7-point bipolar semantic differential anchors (e.g., Not at all attractive to Very attractive; Definitely would not recommend to Definitely would recommend).
• Administration Time: Less than 2 minutes (ideal for inclusion in long CRM surveys, experimental lab procedures, or post-transaction feedback screens).
• Target Population: Consumers, retail patrons, service subscribers, and loyalty program members across business-to-consumer (B2C) and business-to-business (B2B) domains.
• Scoring Procedure:
  • All items are positively phrased; no reverse scoring is required.
  • A composite score is computed by calculating the arithmetic mean of the three completed items: $\text{LPP Score} = \frac{\text{Item}_1 + \text{Item}_2 + \text{Item}_3}{3}$.
  • Higher scores (ranging from 1.00 to 7.00) reflect stronger comparative preference, greater perceived program attractiveness, and higher conative loyalty toward the rewards program.

Permissions & Fee and Test Year

The Loyalty Program Preference scale was published in 2003 in the Journal of the Academy of Marketing Science:

• Original Publication: 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.
• Copyright Holder: Academy of Marketing Science / Springer Nature.
• Academic Usage & Licensing: The instrument items are freely accessible for scholarly research, educational purposes, and scientific replication within academic institutions, provided that standard academic attribution and citation are given to the original authors (Yi & Jeon, 2003).
• Commercial Applications: Commercial enterprises, management consultancies, and market research agencies seeking to integrate the scale into proprietary commercial software, fee-based benchmarking tools, or monetized consumer audits should verify copyright policies with the journal publisher or the authors.

References

The following academic sources provide theoretical, empirical, and methodological foundations for the scale:

• 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
• Oliver, R. L. (1999). Whence consumer loyalty? Journal of Marketing, 63(4_suppl1), 33–44. https://doi.org/10.1177/002224299906300405
• Roehm, M. L., Bolman Pullins, E., & Roehm, H. A. (2002). Designing loyalty program campaigns: Examination of reward types and customer characteristics. Journal of the Academy of Marketing Science, 30(2), 145–160. https://doi.org/10.1177/03079459994500
• Thaler, R. (1985). Mental accounting and consumer choice. Marketing Science, 4(3), 199–214. https://doi.org/10.1287/mksc.4.3.199
• 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/10.1177/0092070303031003002

Items of the Scale