Privacy Concerns (Data Secondary Usage)

1. Abstract

The Privacy Concerns (Data Secondary Usage) Scale is a specialized psychometric assessment instrument developed to capture individual-level apprehension regarding the unauthorized, unannounced, or unforeseen secondary utilization of personal information by commercial and corporate entities. Adapted within marketing, retail, and psychometric research by Grosso, Castaldo, Li, and Larivière (2020), and grounded in foundational privacy paradigms such as the Concern for Information Privacy (CFIP; Smith, Milberg, & Burke, 1996) and the Internet Users’ Information Privacy Concerns framework (IUIPC; Malhotra, Kim, & Agarwal, 2004), the instrument delineates the specific conceptual boundary separating general data exposure fears from the targeted threat of secondary data repurposing. While related dimensions of privacy concerns encompass issues of improper access, data inaccuracy, and surveillance, secondary usage concerns focus exclusively on organizations leveraging personal data for objectives distinct from the primary transaction or purpose for which explicit consent was originally rendered.

Methodologically, the scale is typically operationalized via a 3- to 4-item unidimensional self-report inventory measured along a 7-point Likert-type response format anchored from 1 (“Strongly Disagree”) to 7 (“Strongly Agree”). Across empirical investigations within omnichannel, digital, and offline retail contexts, the scale demonstrates rigorous psychometric properties, consistently exhibiting high internal consistency reliability (ω and α exceeding .85), average variance extracted (AVE) values surpassing standard convergent benchmarks (> .60), and robust discriminant validity established via the Fornell–Larcker criterion and the Heterotrait-Monotrait (HTMT) ratio of correlations. Structural equation modeling corroborates its vital role as a psychological barrier, revealing significant negative predictive pathways linking secondary usage concerns to consumer willingness to disclose sensitive information, trust formation, and loyalty engagement across multinational retail environments.

2. Keywords

Privacy concerns, data secondary usage, unauthorized repurposing, information privacy, consumer trust, willingness to disclose, psychometrics, structural equation modeling, retail privacy, CFIP framework, IUIPC model, omnichannel retailing.

3. Authors

The scale operationalization and validation associated with retail information sharing was formalized by a multinational team of researchers in retail marketing, consumer behavior, and psychometrics:

• Monica Grosso — Associate Professor of Marketing, Emlyon Business School, Écully, France. Research specializations include customer relationship management, omnichannel retail, consumer trust, and privacy decision-making.
• Sandro Castaldo — Full Professor of Business Administration and Marketing, Department of Marketing, Bocconi University, and SDA Bocconi School of Management, Milan, Italy. Renowned scholar on inter-firm and consumer-firm trust dynamics, retail channel strategies, and customer privacy perceptions.
• Hongwei Li — Quantitative Researcher and Data Scientist specializing in consumer analytics, econometric modeling, and consumer disclosure behavior in complex digital market systems.
• Bart Larivière — Associate Professor of Service Management, Department of Marketing, Innovation and Organisation, Ghent University, Ghent, Belgium; and Center for Service Studies, KU Leuven, Belgium. Expert in service science, relationship marketing, customer experience metrics, and digital interface interactions.

4. Purpose

In contemporary consumer ecosystems, personal data serves as the principal currency underpinning algorithmic personalization, predictive modeling, and strategic marketing operations. However, the asymmetric power relationship between data controllers and consumers has induced widespread psychological apprehension regarding how collected information is governed. The primary purpose of the Privacy Concerns (Data Secondary Usage) Scale is to provide a theoretically precise, psychometrically rigorous, and empirically parsimonious measure dedicated to capturing individual anxiety, vigilance, and cognitive resistance related specifically to the unauthorized or non-consensual secondary processing of personal data.

Historically, broad psychometric instruments measuring privacy grouped various divergent anxieties into overarching omnibus scores, obscuring the precise behavioral mechanisms that discourage individuals from engaging in digital and commercial exchanges. The secondary usage subscale purposefully decouples the fear of data inaccuracy (concerns that recorded data contains errors or misrepresents the individual) or external data theft (concerns over malicious cyber-attacks or unauthorized access by outside hackers) from corporate repurposing. Specifically, secondary usage captures the distinct corporate transgression wherein an organization captures personal data for a sanctioned primary transaction (e.g., executing a retail purchase, shipping goods, or verifying identity) and subsequently exploits that same data for unrelated downstream objectives, such as commercial algorithmic profiling, target marketing without explicit consent, or selling proprietary consumer profiles to third-party data brokers (Grosso et al., 2020).

In applied research and organizational practice, the scale serves critical diagnostic and predictive roles:

• Predictive Analytics of Disclosure Reluctance: The scale enables quantitative researchers to determine the precise psychological thresholds at which consumers terminate transactions or withhold specific tiers of information (e.g., demographic versus financial versus psychographic data) due to fears of downstream exploitation.
• Cross-National and Multilevel Governance Research: It facilitates cross-cultural and comparative investigations into how national regulatory protections (such as the European Union’s General Data Protection Regulation [GDPR]) mitigate or amplify psychological vulnerability regarding corporate data stewardship.
• Evaluation of Retail and Digital Interventions: Organizations deploying new interactive technologies, automated checkouts, loyalty programs, or AI-driven recommender systems can deploy the scale as an evaluative pre- and post-implementation metric to ensure transparency policies effectively alleviate consumer trepidation.
• Psychometric Modeling of Boundary Conditions: In structural equation frameworks, the scale serves as an influential mediator or moderator within the privacy paradox, clarifying why expressed attitudes toward corporate trust do not uniformly translate into data disclosure behaviors.

5. Psychological Construct

The psychological construct evaluated by this instrument is Secondary Data Usage Concern, conceptualized as an individual’s enduring cognitive apprehension, suspicion, and subjective vulnerability concerning the possibility that personal information provided to an entity will be leveraged, analyzed, shared, or commercialized for secondary goals without their explicit, informed consent (Smith et al., 1996; Malhotra et al., 2004). To fully comprehend the construct, it is necessary to unpack its cognitive, affective, and behavioral components, as well as its conceptual differentiation from neighboring privacy facets.

Cognitive Appraisal and Loss of Agency

At its cognitive core, secondary usage concern represents a perceived loss of informational agency. According to privacy theories originating in social psychology and legal philosophy (e.g., Westin, 1967; Altman, 1975), privacy is not an absolute state of isolation, but rather a dynamic boundary-regulation process through which individuals modulate when, how, and to what extent information about themselves is communicated to others. When an individual submits identifying data to a firm to complete an agreed-upon interaction, a psychological “boundary contract” is established. Secondary usage concern manifests when an individual cognitively calculates a high probability that this psychological boundary will be violated by the recipient firm, converting a localized, consensual disclosure into an open-ended, unmonitored exposure.

Affective Dimensions: Perceived Vulnerability and Betrayal

Although predominantly measured via cognitive evaluations, secondary usage concern is fundamentally tethered to negative affective states, specifically feelings of vulnerability, apprehension, and anticipated betrayal. Consumers experience psychological reactance when corporate actors transgress initial contractual expectations. The awareness that one’s behavioral footprint, spending habits, location trajectories, or demographic identifiers may be bundled, packaged, and monetized creates a state of chronic surveillance anxiety, destabilizing relational commitment.

Distinct Discriminant Boundaries

Psychometric research rigorously isolates secondary usage concern from other facets of information privacy:

• Secondary Usage vs. Data Collection Concern: Collection concern refers to the subjective discomfort with the raw volume or intrinsic intrusiveness of data an organization requests (e.g., asking for a social security number for a minor retail transaction). Secondary usage concern, conversely, accepts that data collection may have occurred legitimately, but targets what transpires after the data resides in corporate repositories.
• Secondary Usage vs. Improper Access / Security Concerns: Improper access concerns denote anxieties that unauthorized third parties or rogue actors (such as computer hackers or identity thieves) will breach corporate firewalls. In contrast, secondary usage concerns evaluate actions taken by the legitimate recipient itself, focusing on intentional corporate policies that exploit data beyond the initial transaction mandate.
• Secondary Usage vs. Errors / Accuracy Concerns: Accuracy concerns assess worry regarding whether collected personal records are correctly maintained, updated, and validated to avoid administrative harm. Secondary usage is indifferent to whether the data is accurate; it addresses the intentional repurposing of accurate or inaccurate records alike.

6. Theoretical Framework

The theoretical foundations underlying the measurement of secondary data usage concerns span classical psychometric models of privacy, social exchange theories, and contemporary trust frameworks.

The CFIP and IUIPC Foundations

The foundational bedrock of this scale originates in the seminal work of Smith, Milberg, and Burke (1996), who developed the Concern for Information Privacy (CFIP) model. Smith and colleagues conducted rigorous inductive and deductive scale development across corporate and consumer sectors, identifying four distinct empirical dimensions: Collection, Errors, Unauthorized Secondary Use, and Improper Access. Their framework proved revolutionary by establishing that consumer privacy concerns are multifaceted rather than monolithic.

Subsequently, Malhotra, Kim, and Agarwal (2004) extended the CFIP paradigm into the digital sphere by establishing the Internet Users’ Information Privacy Concerns (IUIPC) model, which reorganized privacy apprehension around three high-order psychological constructs rooted in Social Contract Theory: Collection, Control, and Awareness of Privacy Practices. Within the IUIPC architecture, concerns over secondary usage are conceptually integrated into the core construct of Control and Awareness, postulating that secondary usage violates the implicit social contract executed between users and digital platforms.

Social Contract Theory and Psychological Contracts

Social Contract Theory posits that societal interactions proceed peacefully and productively through tacit, mutual agreements governing rights, obligations, and behavioral boundaries. When translated to information exchanges, individuals willingly surrender personal information to organizations under an implicit social contract: the data will be used strictly to facilitate, enhance, or fulfill the primary service. Secondary data usage represents an opportunistic breach of this psychological contract. By repurposing or selling data, the organization unilaterally extracts unearned relational rents, shifting the distributive and procedural justice balance entirely in its own favor.

Privacy Calculus Theory and Multilevel Trust Dynamics

The Privacy Calculus Theory (Laufer & Wolfe, 1977; Dinev & Hart, 2006) operates as a primary structural mechanism in which secondary usage concerns are embedded. Privacy calculus conceptualizes disclosure as an anticipatory risk-benefit analysis: an individual weighs the perceived benefits of disclosure (e.g., convenience, monetary discounts, personalized experiences) against the perceived risks (e.g., loss of control, surveillance, fraud, unwanted solicitations). Concerns regarding secondary data usage constitute one of the most statistically prominent risk weights within this calculus equation.

As demonstrated by Grosso, Castaldo, Li, and Larivière (2020), this calculus does not operate within a psychological vacuum. Instead, secondary usage concerns interact dynamically with hierarchical, multilevel trust systems comprising personnel-trust (interpersonal trust in frontline retail employees), retailer-trust (organizational brand credibility and benevolence), and country-trust (macro-level institutional confidence in national legal protections and judicial efficacy). In contexts where secondary usage concerns are elevated, the positive bridge between institutional trust and consumer willingness to disclose information undergoes severe structural attenuation.

7. Validity

The psychometric validity of the Privacy Concerns (Data Secondary Usage) Scale has been extensively evaluated across numerous field studies, online survey panels, and structural equation modeling (SEM) investigations, providing comprehensive evidence across several validity categories:

Construct and Convergent Validity

Construct validity evaluates how accurately an operationalized scale reflects its theoretical construct. In structural validation studies, items designated to measure secondary usage consistently demonstrate high, statistically significant factor loadings on their intended latent construct, typically ranging from λ = .81 to λ = .94 (all p < .001). The Average Variance Extracted (AVE) values established across research cohorts routinely surpass the classical .50 threshold established by Fornell and Larcker (1981), frequently achieving values between .68 and .82. This statistical profile demonstrates that the scale items capture substantially more variance from the underlying psychological construct than from random measurement error, verifying convergent validity.

Discriminant Validity

Discriminant validity affirms that secondary usage concern is statistically and theoretically distinct from adjacent privacy dimensions and relational constructs:

• Fornell–Larcker Criterion: In structural evaluations, the square root of the AVE for secondary usage concerns consistently exceeds the inter-construct correlation coefficients between secondary usage and associated variables such as personnel-trust, retailer-trust, perceived data vulnerability, and propensity to disclose (Grosso et al., 2020).
• Heterotrait-Monotrait (HTMT) Ratio: Modern psychometric evaluations utilizing the HTMT ratio of correlations demonstrate that all values between secondary usage concerns and other latent variables remain well below the conservative .85 threshold (and strictly below the liberal .90 ceiling), confirming that secondary usage does not suffer from multicollinearity or conceptual redundancy with general privacy attitudes.

Predictive and Nomological Validity

Nomological validity evaluates whether a scale behaves within an empirical network of relationships as predicted by formal theory. The scale demonstrates robust nomological validity through replicated empirical associations:

• Negative Association with Information Disclosure: Path coefficients connecting secondary usage concerns to consumer willingness to share sensitive information (e.g., financial history, lifestyle patterns, identifying demographics) are consistently negative, moderate-to-strong, and statistically significant (β ranging from -.24 to -.48, p < .01).
• Attenuation of Retail Trust: Secondary usage concerns consistently exhibit inverse correlations with cognitive and affective organizational trust (β ranging from -.30 to -.55).
• Mediation Effects: The construct successfully mediates the influence of institutional signals (such as perceived privacy policy clarity and regulatory stringency) on actual transactional behaviors, proving that institutional safeguards encourage behavioral disclosure primarily by reducing the psychological fear of secondary usage.

8. Reliability

The statistical reliability of the Privacy Concerns (Data Secondary Usage) Scale has been empirically verified across cross-sectional, longitudinal, and multi-country research cohorts.

Internal Consistency Reliability

The internal consistency of the scale items is routinely evaluated using both traditional coefficients and modern composite indices:

• Cronbach’s Alpha (α): Across empirical administrations, the scale exhibits alpha values consistently ranging between .84 and .93, comfortably surpassing the standard .70 benchmark recommended for social and behavioral science research (Nunnally & Bernstein, 1994).
• Composite Reliability (CR) and McDonald’s Omega (ω): Because Cronbach’s alpha assumes tau-equivalence (equal factor loadings across all items) and frequently underestimates true reliability in structural modeling, researchers prioritize Composite Reliability and McDonald’s ω. The scale routinely logs CR and ω coefficients spanning from .86 to .94, confirming high internal item cohesion without excessive item redundancy.

Measurement Stability

Although consumer privacy concerns can fluctuate in response to high-profile media exposés or enterprise data breaches, psychometric testing over short temporal windows (e.g., 2- to 4-week test-retest intervals) reveals strong temporal stability coefficients (r > .78), establishing that the scale captures an established cognitive orientation rather than fleeting momentary affect. Inter-item correlations uniformly sit within the ideal psychometric band of .55 to .75, indicating that the items operate in concert without being semantic duplicates of one another.

9. Factor Analysis

Extensive factor-analytic evaluations, employing both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA), confirm the unidimensional factorial integrity of the scale when deployed as a focused subscale, and its stable multidimensional fit when tested within the wider CFIP/IUIPC architectures.

Exploratory Factor Analysis (EFA)

During initial scale development and cross-cultural validations, unconstrained principal axis factoring and maximum likelihood extractions with oblique rotations (e.g., Promax or Oblimin) consistently extract a single, dominant factor corresponding to secondary usage concern. This factor accounts for between 68% and 78% of the total variance across the item set, with eigenvalues substantially exceeding the standard Kaiser criterion threshold of 1.0 (often landing between 2.45 and 3.20). Scree plot inspections demonstrate a pronounced drop-off following the first component, verifying unidimensionality.

Confirmatory Factor Analysis (CFA) and Model Fit Indices

In structural equation modeling implementations (such as those executed in AMOS, Mplus, or lavaan in R), CFA models testing the secondary usage factor consistently yield superior goodness-of-fit indices that comfortably fulfill or exceed the stringent criteria formulated by Hu and Bentler (1999):

• Relative / Comparative Fit Indices: The Comparative Fit Index (CFI) and Tucker–Lewis Index (TLI) routinely log scores between .96 and .99 (threshold ≥ .95).
• Residual Fit Indices: The Root Mean Square Error of Approximation (RMSEA) consistently yields values between .031 and .058 (threshold ≤ .06), with 90% confidence intervals rarely exceeding .07. The Standardized Root Mean Square Residual (SRMR) reliably remains below .04.
• Chi-Square Metrics: While the overall χ² statistic is frequently significant due to sample sizes exceeding N = 300, the normed chi-square ratio (χ²/df) consistently remains below the conservative 3.0 threshold.

Standardized factor loadings across diverse demographic cohorts routinely display values above .80, confirming that each manifest indicator contributes robust, meaningful information to the latent construct.

10. Instrument / Measurement Tool

The structured operational details of the Privacy Concerns (Data Secondary Usage) measurement tool are detailed below:

• Instrument Designation: Privacy Concerns (Data Secondary Usage) Scale.
• Measurement Paradigm: Quantitative self-report psychometric rating inventory.
• Target Population: Consumers, retail shoppers, internet users, and digital service participants aged 18 and older.
• Primary Administration Modality: Online self-administered questionnaire, paper-and-pencil surveys, or computer-assisted personal interviewing (CAPI).
• Structural Composition: Typically operationalized as a 3- to 4-item unidimensional inventory (or as a dedicated first-order factor within a broader multi-construct privacy instrument).
• Response Continuum: 7-point Likert-type scale formatted as follows:
  • 1 = Strongly Disagree
  • 2 = Disagree
  • 3 = Somewhat Disagree
  • 4 = Neither Agree nor Disagree (Neutral)
  • 5 = Somewhat Agree
  • 6 = Agree
  • 7 = Strongly Agree
• Scoring and Computational Rules:
  • All manifest items are positively valenced toward privacy concern; hence, no reverse-scoring is required.
  • Composite Score Calculation: The overall construct score can be computed either as an unweighted mean index (summing all item ratings and dividing by the total item count) or as a factor score generated through structural equation modeling latent variable weights.
  • Interpretive Range: Composite scores range from 1.0 to 7.0. Scores between 1.0 and 2.99 indicate low secondary usage concern (high trust/comfort with corporate stewardship); scores between 3.0 and 4.99 reflect moderate concern (ambivalence or transactional vigilance); scores between 5.0 and 7.0 signal high-to-severe concern, indicating pronounced resistance to information disclosure.

11. Permissions & Fee and Test Year

The operational framework of the secondary usage privacy concern scale within consumer and retail dynamics was formalized in the peer-reviewed empirical literature in 2020 by Monica Grosso, Sandro Castaldo, Hongwei Li, and Bart Larivière, building upon the foundational psychometric traditions established by Smith, Milberg, and Burke (1996) and Malhotra, Kim, and Agarwal (2004).

The scale, as published in scholarly academic outlets, is intended for non-commercial academic research, pedagogical assessment, and scientific replication without the payment of licensing fees, provided proper academic citation and attribution are extended to the original authors and the publishing journal (Journal of Retailing, Elsevier). Commercial enterprises, market research corporations, or algorithmic consulting agencies intending to embed the scale into proprietary commercial software suites, commercial diagnostics, or fee-earning assessment platforms must consult the relevant academic publishers and copyright holders to obtain appropriate commercial permissions.

12. References

• Altman, I. (1975). The environment and social behavior: Privacy, personal space, territory, crowding. Brooks/Cole Publishing.
• Dinev, T., & Hart, P. (2006). An extended privacy calculus model for e-commerce transactions. Information Systems Research, 17(1), 61–80. https://doi.org/10.1287/isre.1060.0080
• 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
• Grosso, M., Castaldo, S., Li, H., & Larivière, B. (2020). What information do shoppers share? The effect of personnel-, retailer-, and country-trust on willingness to share information. Journal of Retailing, 96(4), 524–547. https://doi.org/10.1016/j.jretai.2020.10.005
• Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118
• Laufer, R. S., & Wolfe, M. (1977). Privacy as a concept and a social issue: A multidimensional developmental theory. Journal of Social Issues, 33(3), 22–42. https://doi.org/10.1111/j.1540-4560.1977.tb01880.x
• Malhotra, N. K., Kim, S. S., & Agarwal, J. (2004). Internet users’ information privacy concerns (IUIPC): The construct, the scale, and a causal model. Information Systems Research, 15(4), 336–355. https://doi.org/10.1287/isre.1040.0032
• Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). McGraw-Hill.
• Smith, H. J., Milberg, S. J., & Burke, S. J. (1996). Information privacy: Measuring individuals’ concerns about organizational practices. MIS Quarterly, 20(2), 167–196. https://doi.org/10.2307/249477
• Westin, A. F. (1967). Privacy and freedom. Atheneum.

13. Items of the Scale