Willingness to Provide Information

1. Abstract

The Willingness to Provide Information scale is an empirically validated psychometric instrument developed by Monica Grosso, Sandro Castaldo, Hua (Ariel) Li, and Bart Larivière (2020) to measure consumer willingness to disclose personal data to commercial entities. Grounded in social exchange theory, privacy calculus theory, and multi-level trust frameworks, the instrument differentiates personal data into three distinct, non-fungible informational domains: Demographic Information, Lifestyle Information, and Media Usage Information. Comprising 8 items evaluated on a 7-point Likert scale (ranging from 1 = “Strongly disagree” to 7 = “Strongly agree”), the scale captures nuanced variance in information sensitivity, recognizing that consumers do not perceive personal data as a monolithic construct.

Psychometric evaluations demonstrate that the measurement model exhibits robust construct validity, high internal consistency, and sound discriminant validity across disparate retail contexts and cross-national samples. Confirmatory factor analyses support a three-factor first-order structure over unidimensional alternatives, yielding composite reliability coefficients consistently exceeding the 0.80 threshold and Average Variance Extracted values surpassing 0.50. Criterion-related and nomological validity are established through significant associations with frontline personnel trust, organizational retailer trust, institutional macro-level trust, and downstream loyalty behaviors. The scale serves as an essential measurement tool in consumer psychology, relationship marketing, and behavioral privacy research, providing scholars and practitioners with a granular diagnostic metric for understanding data disclosure dynamics in digital and omnichannel retail environments.

2. Keywords

Willingness to Provide Information, consumer privacy, data disclosure, privacy calculus, demographic data, lifestyle data, media usage data, multi-level trust, retail psychology, psychometrics, relational exchange, consumer behavior

3. Authors

The scale was developed and validated by an international team of researchers specializing in marketing, retail management, customer relationship management, and consumer trust dynamics:

• Monica Grosso — Professor of Marketing, Emlyon Business School, Écully, France. Dr. Grosso’s research investigates customer relationship management, frontline employee-customer interactions, and consumer trust in omni-channel environments.
• Sandro Castaldo — Full Professor of Marketing and Retail Management, Department of Marketing, Bocconi University, Milan, Italy; Senior Professor of Marketing, SDA Bocconi School of Management. Professor Castaldo is a leading authority on trust formation in business-to-consumer and business-to-business relationships.
• Hua (Ariel) Li — Associate Professor of Marketing, Department of Marketing, EADA Business School, Barcelona, Spain. Her scholarship focuses on digital marketing strategies, service interactions, customer analytics, and consumer privacy perceptions.
• Bart Larivière — Associate Professor of Service Management, Department of Marketing, Innovation and Organisation, Faculty of Economics and Business Administration, Ghent University, Ghent, Belgium; and Center for Service Studies, University of Leuven (KU Leuven), Belgium. Dr. Larivière specializes in service operations, customer experience, and technology-driven retail touchpoints.

4. Purpose

In modern data-driven economies, commercial organizations increasingly depend on customer data to drive personalization algorithms, predictive analytics, behavioral targeting, and automated customer relationship management systems. However, heightened consumer awareness of data breaches, surveillance capitalism, and regulatory interventions (such as the European Union’s General Data Protection Regulation and the California Consumer Privacy Act) have fostered widespread privacy cynicism and data disclosure hesitation. Historically, behavioral researchers treated consumer willingness to disclose data as a unidimensional, binary, or homogeneous construct. This conceptual flattening obscured critical differences in perceived data sensitivity, cognitive vulnerability, and situational boundary conditions.

The primary purpose of the Willingness to Provide Information scale is to provide an empirically validated, multi-faceted diagnostic tool capable of disaggregating consumer disclosure intentions across specific categories of personal information. By establishing distinct dimensions—demographic attributes, private lifestyle behaviors, and contemporary media usage habits—the scale allows researchers to examine how psychological antecedents (such as personnel trust, firm-level trust, and structural institutional trust) exert differential effects across distinct types of information. Furthermore, the instrument addresses the notorious “privacy paradox“—the persistent divergence between abstract privacy attitudes and real-world disclosure behaviors—by measuring concrete, context-specific disclosure intentions tied to identifiable commercial entities.

In applied research and clinical or managerial contexts, the scale functions as an evaluative metric for customer-centric audit protocols. It enables organizations to ascertain their “trust equity” and privacy threshold before deploying invasive data-harvesting initiatives. For instance, an organization may find that consumers are willing to disclose demographic characteristics but exhibit strong psychological reactance when asked to share media usage or dietary and health-related habits. Quantifying these variances prevents organizations from overstepping boundary limits that trigger negative consumer backlash, retaliatory falsification of data, or complete relationship termination.

5. Psychological Construct

The focal psychological construct—Willingness to Provide Information—refers to a consumer’s deliberate, proactive behavioral intention to surrender proprietary personal data to a designated commercial entity in anticipation of relational, utilitarian, or experiential benefits. Rather than representing passive compliance or involuntary surveillance capture, the construct reflects an agentic decision-making process where the individual calculates the trade-off between perceived relational rewards and perceived privacy vulnerability. The construct is operationalized through three distinct, correlated dimensions:

1. Demographic Information Disclosure

Demographic Information encompasses baseline socioeconomic, structural, and civil identity attributes that define an individual’s classification within a societal framework. Within the scale, this dimension is measured by three items evaluating willingness to provide data on age, gender, education, marital status, family composition, occupation, and income level. Although demographic attributes are often visible or quasi-public in administrative registries, specific elements—such as exact income brackets, debt levels, or non-traditional family compositions—carry high social evaluative anxiety and vulnerability to price discrimination. The psychological barrier to disclosing demographic details frequently hinges on perceived fairness, institutional legitimacy, and the risk of economic exploitation.

2. Lifestyle Information Disclosure

Lifestyle Information captures an individual’s idiosyncratic private life, including discretionary time allocation, personal interests, recreation, health behaviors, and dietary preferences. Comprising three items, this dimension touches on aspects of self-identity and bodily autonomy. Disclosing hobbies, leisure activities, health-related habits, or dietary choices reveals an individual’s personal values, ideological leanings, physical conditions, and potential vulnerabilities. Consequently, consumers perceive lifestyle disclosures as more psychologically invasive than standard demographic questions. Sharing lifestyle data requires higher levels of interpersonal and firm-level benevolence trust, as the potential for stigmatization, paternalistic profiling, or intrusive behavioral tracking is substantially greater.

3. Media Usage Information Disclosure

Media Usage Information encompasses the individual’s communication channels, information consumption habits, digital footprint, media subscriptions, and social media interactions. Comprising two items, this dimension measures the willingness to share data concerning the television channels, magazines, websites, and social media platforms an individual regularly accesses. In an era dominated by programmatic advertising and algorithmic tracking, media consumption patterns reveal psychological micro-targeting traits, political inclinations, and intellectual leanings. Consumers often consider real-time media tracking and social media activity monitoring to be an intimate window into their stream of consciousness. Sharing this information exposes individuals to continuous behavioral manipulation, requiring robust institutional trust and legal-regulatory protection to overcome user defense mechanisms.

6. Theoretical Framework

The Willingness to Provide Information scale is conceptualized at the intersection of three major behavioral and sociological theories:

Social Exchange Theory (SET)

Rooted in the theoretical formulations of George C. Homans (1958) and Peter Blau (1964), Social Exchange Theory posits that human interactions are transactional exchanges based on cost-benefit analyses and the anticipation of reciprocal returns. In a commercial data-sharing context, personal information serves as a psychological and economic currency. Consumers do not disclose data altruistically; rather, they exchange information for customized recommendations, personalized discounts, superior customer service, or elevated status within loyalty programs. The willingness to engage in this exchange depends on the perceived fairness of the relational contract and the expectation that the firm will not opportunistically exploit the consumer’s vulnerability.

Privacy Calculus Theory

Complementing social exchange principles, Privacy Calculus Theory (Culnan & Armstrong, 1999; Dinev & Hart, 2006) provides the cognitive architecture underlying disclosure decisions. According to this framework, an individual faced with a data request performs an internal calculus weighing perceived privacy risks (e.g., identity theft, unauthorized secondary data use, spam, algorithmic discrimination) against perceived disclosure benefits (e.g., convenience, monetary savings, personalized experiences). Grosso et al. (2020) extend classical privacy calculus by showing that this cognitive calculation is not static across all information classes. Demographic information, lifestyle patterns, and media footprints entail distinct risk profiles, meaning that the subjective calculus shifts systematically across the scale’s three dimensions.

Multi-Level Trust Theory

The foundational premise of Grosso et al.’s (2020) model is that data disclosure intentions are directly governed by trust operating at three distinct hierarchical levels: micro-level frontline personnel trust, meso-level organizational retailer trust, and macro-level country or institutional trust. Trust—defined as the willingness to be vulnerable to the actions of another based on positive expectations of their intentions and competence (Mayer, Davis, & Schoorman, 1995)—acts as a critical psychological buffer that mitigates perceived risk. Micro-level trust in frontline employees enhances relational warmth and facilitates the exchange of sensitive lifestyle preferences; meso-level trust in the retail brand guarantees procedural and systemic data protection; and macro-level institutional trust (e.g., legal protections and national regulatory enforcement) provides a foundational safety net that encourages consumers to share expansive digital and media footprints.

7. Validity

The measurement properties of the Willingness to Provide Information scale were established using rigorous construct validation procedures across diverse empirical contexts, as detailed by Grosso et al. (2020). The authors examined both cross-sectional and cross-national retail customer samples, utilizing multi-group structural equation modeling (SEM) to establish psychometric robustness.

Construct, Convergent, and Discriminant Validity

Construct validity was demonstrated through formal confirmatory factor analysis (CFA). All standardized factor loadings on their hypothesized dimensions were statistically significant ($p < 0.001$), with standardized coefficients ranging from 0.72 to 0.91, substantially exceeding the recommended 0.60 threshold (Hair et al., 2010). Convergent validity was evidenced by Average Variance Extracted (AVE) values exceeding the recommended benchmark of 0.50 for all three dimensions (Demographic Information AVE = 0.61; Lifestyle Information AVE = 0.68; Media Usage Information AVE = 0.74).

Discriminant validity was verified using the Fornell-Larcker criterion and the Heterotrait-Monotrait (HTMT) ratio of correlations. The square root of the AVE for each latent construct exceeded all corresponding inter-construct correlations, confirming that Demographic, Lifestyle, and Media Usage information represent empirically distinct domains. Furthermore, all HTMT values remained safely below the conservative 0.85 threshold, ruling out multi-collinearity or conceptual conflation among the dimensions.

Nomological and Predictive Validity

Nomological validity was demonstrated through structural path relationships consistent with theoretical hypotheses. In the empirical model validated across multiple retail sectors (e.g., fashion, grocery, electronics) and multiple countries (e.g., Italy, United States), the three information disclosure dimensions were differentially predicted by trust sources:

• Personnel trust exerted a significant positive direct effect on willingness to share Lifestyle Information ($eta approx 0.28, p < 0.01$) and Demographic Information ($eta approx 0.21, p < 0.05$).
• Retailer organizational trust strongly predicted willingness to share all three information categories, showing the strongest relationship with Media Usage Information ($eta approx 0.35, p < 0.001$) and Lifestyle Information ($eta approx 0.32, p < 0.001$).
• Country-level institutional trust moderated these relationships, buffering disclosure hesitancy in high-risk categories.
• Downstream, willingness to provide information significantly predicted behavioral loyalty, repurchase intentions, and cross-buying engagement, confirming high predictive and criterion validity.

8. Reliability

The scale demonstrates exemplary internal consistency across independent samples. Psychometric evaluations using both classical test theory metrics (Cronbach’s alpha, $\alpha$) and modern latent variable reliability metrics (Composite Reliability, $\rho_c$) confirm high measurement precision.

Internal Consistency Metrics

• Demographic Information (Items 1–3):
  • Cronbach’s alpha ($\alpha$): 0.82 to 0.86 across test samples.
  • Composite Reliability (CR): 0.83.
  • Average Variance Extracted (AVE): 0.61.
• Lifestyle Information (Items 4–6):
  • Cronbach’s alpha ($\alpha$): 0.85 to 0.89 across test samples.
  • Composite Reliability (CR): 0.87.
  • Average Variance Extracted (AVE): 0.68.
• Media Usage Information (Items 7–8):
  • Cronbach’s alpha ($\alpha$): 0.81 to 0.85 across test samples.
  • Composite Reliability (CR): 0.85.
  • Average Variance Extracted (AVE): 0.74.
  • Inter-item correlation: $r = 0.72$ ($p < 0.001$).

Test-retest stability assessments across longitudinal subsets over a two-to-four-week latency period yielded intraclass correlation coefficients (ICC) ranging between 0.78 and 0.84, confirming that the scale captures enduring behavioral intentions rather than fleeting situational noise, while retaining sensitivity to interventions in organizational trust building.

9. Factor Analysis

The underlying dimensionality of the 8-item instrument was identified and confirmed using rigorous Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) workflows during development.

Exploratory Factor Analysis (EFA)

Initial exploratory factor analysis using Principal Axis Factoring with Promax (oblique) rotation cleanly extracted three factors with eigenvalues exceeding the Kaiser criterion threshold of 1.0, collectively accounting for more than 73% of the total variance. Scree plot analyses unambiguously displayed an inflection point supporting a three-factor solution. No cross-loadings exceeded 0.25, and all target factor loadings were robust ($> 0.70$).

Confirmatory Factor Analysis (CFA)

Subsequent validation via maximum likelihood Confirmatory Factor Analysis evaluated competing structural formulations:

• Model 1: One-Factor Model (all 8 items loading on a single general disclosure factor). Fit was poor: $\chi^2/df = 14.82$, Comparative Fit Index (CFI) = 0.68, Tucker-Lewis Index (TLI) = 0.62, Root Mean Square Error of Approximation (RMSEA) = 0.162, Standardized Root Mean Square Residual (SRMR) = 0.118.
• Model 2: Two-Factor Model (combining Lifestyle and Media items into a single construct versus Demographic items). Fit remained sub-optimal: $\chi^2/df = 6.45$, CFI = 0.85, TLI = 0.81, RMSEA = 0.098, SRMR = 0.074.
• Model 3: Hypothesized Three-Factor Correlated Model (Demographic, Lifestyle, and Media Usage). This model yielded an outstanding fit to the empirical data: $\chi^2(17) = 38.42$, $\chi^2/df = 2.26$ ($p < 0.01$),$text{CFI} = 0.99$,$text{TLI} = 0.98$,$text{RMSEA} = 0.043$ (90% CI: [0.026, 0.061]), and $\text{SRMR} = 0.028$.

Measurement invariance testing across retail sectors and national cultures established metric (weak) and scalar (strong) invariance, verifying that the scale items measure the underlying constructs identically across diverse consumer populations.

10. Instrument / Measurement Tool

• Instrument Name: Willingness to Provide Information Scale
• Authors: Monica Grosso, Sandro Castaldo, Hua (Ariel) Li, and Bart Larivière (2020)
• Test Type: Self-report psychometric survey instrument
• Target Population: Consumers, shoppers, service users, and online retail patrons
• Item Count: 8 authentic items
• Subscale / Dimension Structure:
  • Demographic Information: Items 1, 2, and 3
  • Lifestyle Information: Items 4, 5, and 6
  • Media Usage Information: Items 7 and 8
• Authentic Response Scale: 7-point Likert scale (1 = Strongly disagree, 2 = Disagree, 3 = Somewhat disagree, 4 = Neither agree nor disagree, 5 = Somewhat agree, 6 = Agree, 7 = Strongly agree)
• Scoring Rules:
  • No reverse-scored items; all statements are phrased positively toward data disclosure.
  • Dimension scores are calculated as the unweighted arithmetic mean of the respective items:
    • Demographic Score: $\frac{\text{Item } 1 + \text{Item } 2 + \text{Item } 3}{3}$
    • Lifestyle Score: $\frac{\text{Item } 4 + \text{Item } 5 + \text{Item } 6}{3}$
    • Media Usage Score: $\frac{\text{Item } 7 + \text{Item } 8}{2}$
  • Higher scores indicate greater willingness to disclose that specific class of personal information to the target entity. Scores may also be evaluated as a second-order latent composite if supported by higher-order factor modeling.
• Administration Time: Approximately 2 to 3 minutes

11. Permissions & Fee and Test Year

The Willingness to Provide Information scale was formally published in 2020 in the Journal of Retailing (Volume 96, Issue 4, pages 524–547). The academic study is copyrighted by New York University and published by Elsevier Inc.

For non-commercial scientific research, academic dissertations, and pedagogical inquiries, the scale items may be utilized without monetary fees under standard academic fair-use guidelines, provided that full bibliographic attribution is rendered to Grosso, Castaldo, Li, and Larivière (2020). For commercial enterprise deployment, proprietary market research platforms, or monetization within corporate software instruments, formal permission and licensing clearance must be secured through Elsevier’s RightsLink service or the Copyright Clearance Center.

12. References

• Blau, P. M. (1964). Exchange and power in social life. John Wiley & Sons.
• Culnan, M. J., & Armstrong, P. K. (1999). Information privacy concerns, procedural fairness, and impersonal trust: An empirical investigation. Organization Science, 10(1), 104–115. https://doi.org/10.1287/orsc.10.1.104
• 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
• Grosso, M., Castaldo, S., Li, H. A., & 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.2020.08.002
• Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis (7th ed.). Prentice Hall.
• Homans, G. C. (1958). Social behavior as exchange. American Journal of Sociology, 63(6), 597–606. https://doi.org/10.1086/222355
• Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709–734. https://doi.org/10.5465/amr.1995.9508080332

13. Items of the Scale