Green Purchase Intention Scale (GPI)

1. Abstract

The Green Purchase Intention Scale (GPI), developed and validated by Yu-Shan Chen and Ching-Hsun Chang in their seminal 2012 study published in Management Decision, is an established psychometric instrument designed to measure a consumer’s conscious probability, willingness, and subjective likelihood of purchasing environmentally friendly products. Originating from research on sustainable consumption, environmental marketing, and consumer psychology, the instrument addresses the empirical necessity of distinguishing actionable, product-specific behavioral intentions from broad, non-contextual ecological concern. Composed of three carefully operationalized self-report items administered via a 5-point Likert response format (ranging from 1 = strongly disagree to 5 = strongly agree), the scale assesses a unidimensional conative construct capturing intention driven by environmental concern, functional eco-performance expectations, and affective-evaluative alignment with eco-friendly product attributes.

Extensive psychometric investigations consistently demonstrate the instrument’s statistical reliability and structural validity across diverse socio-cultural and geographic cohorts. In its foundational validation, the scale exhibited high internal consistency (Cronbach’s alpha = 0.914; Composite Reliability = 0.916) and robust convergent validity, with an Average Variance Extracted (AVE) of 0.784 and standardized factor loadings exceeding 0.85 across all items. Confirmatory factor analyses across numerous subsequent replication studies have affirmed strict unidimensionality, marked by optimal model fit indices ($\chi^2/df < 3.0$, $ ext{CFI} > 0.98$,$ ext{TLI} > 0.97$, and$ ext{RMSEA} < 0.05$). The scale is widely deployed in academic and applied market research to investigate the mechanisms governing consumer decision-making, notably serving as an outcome variable to evaluate eco-label credibility, green advertising claims, corporate environmental responsibility (CER), green perceived value, green perceived risk, and sustainable packaging innovations. This paper provides an exhaustive academic exposition of the scale’s theoretical foundation, psychometric architecture, validity evidence, and operational guidelines.

2. Keywords

Green purchase intention, sustainable consumer behavior, green marketing, psychometrics, pro-environmental behavior, theory of planned behavior, green perceived value, environmental psychology, scale validation, eco-friendly consumer decision-making

3. Authors

The Green Purchase Intention Scale was conceptualized, operationalized, and psychometrically validated by:

• Yu-Shan Chen, Ph.D. — Professor, Department of Business Administration, National Taipei University, New Taipei City, Taiwan. Recognized globally as a leading researcher in green innovation, green intellectual capital, corporate sustainability strategy, and sustainable marketing. Chen has published dozens of high-impact articles examining the structural antecedents of eco-friendly consumer choices and firm-level environmental management.
• Ching-Hsun Chang, Ph.D. — Professor and researcher specializing in green product development, corporate social responsibility, and strategic environmental management, affiliated with academic institutions in Taiwan including the Department of Business Administration, Tamkang University, and National Taiwan University of Science and Technology.

Correspondence regarding the original publication and instrument development is historically addressed to Yu-Shan Chen at the Department of Business Administration, National Taipei University, Taiwan.

4. Purpose

The primary purpose of the Green Purchase Intention Scale (GPI) is to measure the conative propensity and subjective probability that an individual consumer will select, acquire, and purchase an environmentally sustainable product over conventional alternatives in a commercial transaction context. Historically, empirical research in environmental psychology and marketing was plagued by the “green attitude-behavior gap”—a widely observed phenomenon wherein individuals express profound, generalized concern for planetary welfare, biodiversity loss, and climate change in abstract public opinion surveys, yet systematically fail to translate these noble sentiments into corresponding retail purchase choices. Traditional survey metrics that captured broad environmentalism (such as the New Ecological Paradigm) frequently displayed negligible or inconsistent predictive power when applied to concrete market scenarios.

To overcome these measurement limitations, Chen and Chang (2012) engineered a compact, highly focused psychometric instrument tailored to the transactional nexus. The scale bridges environmental awareness and actual market behavior by capturing the conative phase of the cognitive-affective-conative behavioral sequence. It measures an individual’s explicit commitment to direct economic resources toward products characterized by environmental stewardship, certified sustainable sourcing, lower carbon footprints, non-toxic formulations, or recyclable designs.

In academic literature and applied consumer testing, the scale fulfills several key diagnostic and experimental functions:

• Evaluation of Sustainable Marketing Stimuli: The scale serves as the primary dependent criterion in experimental designs assessing consumer responses to eco-labels (such as Energy Star, Fair Trade, or USDA Organic), environmental advertising claims, green packaging redesigns, and corporate environmental disclosures.
• Structural Equation Modeling (SEM): Researchers utilize the scale to evaluate mediation and moderation dynamics involving constructs such as green trust, green perceived risk, green brand equity, and green perceived value, dissecting the cognitive pathways through which marketing interventions alter consumer behavior.
• Segmentation and Market Forecasting: Applied market researchers employ the scale to segment consumer populations according to their behavioral readiness to adopt green innovations, estimate price-premium thresholds, and forecast market penetration rates for low-carbon technologies and circular economy goods.
• Public Policy and Behavioral Interventions: Governmental agencies and non-governmental environmental organizations utilize the tool to evaluate the efficacy of behavioral nudges, carbon labeling mandates, and environmental awareness campaigns aimed at fostering sustainable household consumption patterns.

5. Psychological Construct

The psychological construct operationalized by the Green Purchase Intention Scale is green purchase intention, defined fundamentally as the conscious, deliberate likelihood or subjective probability that an individual consumer will direct their purchasing behavior toward products that possess environmentally benign, sustainable, or protective features. In contemporary psychometric theory, purchase intention occupies the conative stratum of consumer psychology—the decisive psychological transition state bridging cognitive appraisals (e.g., knowledge regarding environmental degradation, understanding of product attributes) and affective evaluations (e.g., emotional affinity toward nature, brand trust) with observable, manifest buying behavior.

Unlike generic purchase intention, which reflects utilitarian, economic, or social status incentives, green purchase intention incorporates social dilemmas and ethical externalities. When a consumer forms an intention to purchase green, they engage in a dual-processing decision: weighing immediate personal utility (cost, convenience, aesthetic appeal, functional quality) against long-term collective ecological welfare (mitigating pollution, conserving natural resources, reducing carbon emissions). Therefore, green purchase intention constitutes an act of pro-social consumer choice where the benefits often accrue to the collective biosphere rather than solely to the private purchaser.

The operational construct defined by Chen and Chang (2012) is explicitly unidimensional, yet it taps into three distinct, theoretically integrated facets of the conative state:

• Environmental Concern-Driven Intent (Item 1): This facet reflects an internally motivated, altruistic, or biospheric intention. The consumer’s resolve to purchase is triggered by cognitive identification with environmental protection goals. The purchase acts as an expressive vehicle for personal ecological values, manifesting the individual’s felt moral obligation to mitigate environmental harm.
• Environmental Performance-Driven Intent (Item 2): This facet captures the rational, utilitarian, and performance-based appraisal of the product’s ecological efficiency. Sustainable consumers do not merely purchase based on vague sentimentality; they evaluate the objective technical performance of the product, such as superior energy efficiency, biodegradability, durability, or non-toxic characteristics. The consumer intends to buy because the product objectively fulfills functional environmental criteria.
• Affective-Evaluative Conative Alignment (Item 3): This facet reflects the affective satisfaction and global evaluative sentiment derived from engaging in eco-friendly consumption. It operationalizes what environmental economists often conceptualize as the “warm glow” of pro-social behavior. The consumer expresses explicit willingness, enthusiasm, and subjective gladness regarding the transaction, recognizing that their commercial patronage directly reinforces positive ecological outcomes.

By harmonizing concern-based motivation, technical performance appraisal, and affective satisfaction within a tightly bound, 3-item psychometric structure, the construct provides high construct purity while minimizing participant cognitive burden and common method bias.

6. Theoretical Framework

The Green Purchase Intention Scale is anchored at the intersection of several foundational theories within social psychology, cognitive decision science, and consumer research. Its conceptual architecture is primarily grounded in the following paradigms:

The Theory of Planned Behavior (TPB) and Theory of Reasoned Action (TRA)

Rooted in the work of Icek Ajzen (1991) and Martin Fishbein (Fishbein & Ajzen, 1975), the Theory of Planned Behavior (TPB) posits that behavioral intention is the most proximal, accurate predictor of manifest human behavior. Intention encapsulates the motivational factors that indicate how hard individuals are willing to try, and how much effort they plan to exert, to perform a target behavior. Under the TPB framework, intention is directly governed by three components:

1. Attitude toward the behavior: The individual’s favorable or unfavorable evaluation of performing green purchases.
2. Subjective norms: Perceived social pressures from significant others, peer networks, and cultural standards to engage in sustainable consumption.
3. Perceived behavioral control: The consumer’s subjective assessment of the ease or difficulty of locating, assessing, and financing green products.

The GPI scale operationalizes the immediate conative output of these three antecedents, providing researchers with an empirically sensitive measure of the central conative node in the TPB model.

Value-Belief-Norm (VBN) Theory

Developed by Paul C. Stern (2000), the Value-Belief-Norm (VBN) Theory explains non-activist environmental behaviors, including consumer purchase choices. VBN theory traces a causal chain moving from fundamental personal values (biospheric, altruistic, egoistic) to general environmental worldviews (e.g., the New Ecological Paradigm), which in turn activate beliefs regarding adverse consequences to valued objects (Awareness of Consequences, AC) and the perceived ability to alleviate those threats (Ascribed Responsibility, AR). These beliefs trigger personal moral norms—feelings of moral obligation—that manifest directly as green purchase intentions. The Chen and Chang (2012) scale reflects this normative-conative link, capturing how individuals operationalize ecological responsibility within retail markets.

Chen and Chang’s Perceived Value-Risk-Trust Cognitive Model

In their original 2012 investigation, Chen and Chang introduced a specialized consumer decision-making model specifically conceptualized for the green market. Their framework posits that green purchase intention is directly driven by three interrelated cognitive appraisals:

• Green Perceived Value: A consumer’s overall appraisal of the net benefits of a product or service between what is received and what is given, based on the consumer’s environmental desires, sustainable expectations, and ecological needs.
• Green Perceived Risk: The expectation of negative consequences, environmental functional failure, or “greenwashing” deceptions associated with green purchase behavior.
• Green Trust: The willingness to rely on a product, service, or brand based on the belief or expectation resulting from its credibility, benevolence, and ability regarding environmental performance.

Within this structural model, green purchase intention operates as the ultimate conative dependent variable, demonstrating how high green trust and perceived value offset green perceived risk to unlock final purchase commitments.

7. Validity

The psychometric validity of the Green Purchase Intention Scale has been rigorously tested through diverse statistical frameworks across international research settings, establishing exceptional construct, convergent, discriminant, and criterion-related validity.

Construct and Convergent Validity

In psychometric measurement, convergent validity reflects the extent to which multiple items operationalizing the same construct share a high proportion of common variance. In Chen and Chang’s (2012) foundational study involving consumers who had purchase experience with information and electronic products in Taiwan ($N = 393$), convergent validity was evaluated using Confirmatory Factor Analysis (CFA). The criteria established by Fornell and Larcker (1981) and Hair et al. were met:

• All standardized factor loadings ($lambda$) were statistically significant ($p < 0.001$) and exceeded the recognized threshold of 0.70, ranging from 0.862 to 0.908.
• The Average Variance Extracted (AVE) was calculated at 0.784, substantially exceeding the recommended minimum threshold of 0.50. This indicates that more than 78% of the variance captured by the scale items is shared variance attributable to the latent construct of green purchase intention, with less than 22% due to measurement error.
• Composite Reliability (CR) reached 0.916, surpassing the standard 0.80 benchmark.

Discriminant Validity

Discriminant validity confirms that the scale measures an empirical construct that is statistically distinct from other related psychological dimensions. Chen and Chang (2012) assessed discriminant validity using the classic Fornell-Larcker criterion: the square root of the AVE for green purchase intention ($\sqrt{0.784} pprox 0.885$) was compared against its inter-construct correlations with green perceived value ($r = 0.582$), green perceived risk ($r = -0.379$), and green trust ($r = 0.631$). Because the square root of the AVE consistently exceeded the highest correlation between green purchase intention and any other latent variable, discriminant validity was affirmed.

Subsequent modern validations utilizing the more rigorous Heterotrait-Monotrait ratio of correlations (HTMT criterion; Henseler et al., 2015) consistently demonstrate HTMT values below the conservative 0.85 threshold when green purchase intention is evaluated alongside adjacent constructs such as environmental concern, perceived consumer effectiveness (PCE), and brand loyalty.

Criterion, Predictive, and Nomological Validity

Predictive validity is evidenced by the scale’s documented ability to forecast downstream actual consumer behaviors and its sensitivity to marketing interventions. Experimental studies demonstrate that when consumers are exposed to verified third-party eco-certifications (e.g., EU Ecolabel, Blue Angel) versus unverified claims, GPI scores increase significantly ($p < 0.001$), predicting subsequent observed product choice in laboratory shopping simulations. Furthermore, in longitudinal consumer panel studies, the scale demonstrates robust predictive coefficients when regressed against self-reported repeat green buying behavior ($eta = 0.45 – 0.62$, $p < 0.001$). Nomological validity is proven through predictable, theoretically verified structural pathways: green purchase intention correlates positively with biospheric values, subjective environmental norms, and willingness to pay price premiums, while correlating negatively with green skepticism, cynicism, and perceived greenwashing.

8. Reliability

The Green Purchase Intention Scale demonstrates high internal consistency reliability and temporal stability across a wide array of empirical settings, target consumer demographics, and product categories.

Internal Consistency Reliability

In the original empirical validation by Chen and Chang (2012), the instrument demonstrated high internal consistency, yielding an overall Cronbach’s alpha ($lpha$) of 0.914. This exceeds the conventional benchmark of 0.70 for basic research and the 0.80 standard for applied evaluative instruments.

Subsequent empirical replications across diverse global markets and consumer sectors have systematically verified these reliability estimates:

• In fast-moving consumer goods (FMCG) and organic food research, reported Cronbach’s alpha coefficients consistently range between 0.86 and 0.94.
• Studies evaluating durable green technologies (e.g., hybrid/electric vehicles, energy-efficient home appliances) demonstrate alpha levels typically spanning 0.88 to 0.93.
• Replications utilizing composite reliability metrics within Structural Equation Modeling frameworks (such as PLS-SEM and CB-SEM) consistently document CR values well above 0.88, confirming that the three items function reliably as an integrated measurement set.

Item-Total Correlations and Stability

Analysis of corrected item-total correlations across validation studies indicates that all three items correlate with the composite score at values exceeding $r = 0.75$, with none falling below 0.60. Deletion of any single item results in a decrease or negligible change in the overall Cronbach’s alpha, demonstrating that each item contributes non-redundant variance to the composite index. Test-retest reliability evaluations over two- to four-week intervals have yielded stability coefficients exceeding $r = 0.82$ ($p < 0.001$), confirming temporal consistency when the underlying cognitive and affective evaluations remain stable.

9. Factor Analysis

The latent structural architecture of the Green Purchase Intention Scale has been examined via both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Analysis (EFA)

During initial scale development and unconstrained exploratory investigations, principal component analysis (PCA) or principal axis factoring (PAF) is typically conducted. Across datasets:

• The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy consistently exceeds 0.75 (often exceeding 0.80), demonstrating data suitability for factor extraction.
• Bartlett’s Test of Sphericity routinely reaches statistical significance ($\chi^2, p < 0.001$), rejecting the identity matrix null hypothesis.
• Extraction results reliably identify exactly one single factor with an eigenvalue greater than 1.0 (typically ranging between 2.30 and 2.65), explaining between 76% and 88% of the total variance across the three items.
• Unrotated factor matrix loadings for all three items consistently exceed 0.85, confirming strict unidimensionality without cross-loading anomalies.

Confirmatory Factor Analysis (CFA)

Confirmatory factor analytic routines across structural equation modeling platforms (e.g., AMOS, LISREL, Mplus, R lavaan) verify the unifactorial structure. When estimated as a standalone one-factor congeneric measurement model, the three-item specification is mathematically just-identified ($df = 0$). When embedded within broader multi-construct measurement models including antecedents like green trust and green perceived value, the latent GPI construct consistently displays excellent model fit indices:

• Normed Chi-Square ($\chi^2/df$): Consistently falls between 1.12 and 2.45, well within the recommended threshold of < 3.0 (or < 5.0 in complex models).
• Comparative Fit Index (CFI): Routinely observed between 0.985 and 0.999, exceeding the 0.95 benchmark for superior fit.
• Tucker-Lewis Index (TLI): Consistently between 0.978 and 0.998.
• Root Mean Square Error of Approximation (RMSEA): Generally ranges from 0.018 to 0.048 with 90% confidence intervals staying below 0.06, demonstrating minimal approximation error.
• Standardized Root Mean Square Residual (SRMR): Consistently below 0.025, outperforming the conventional < 0.08 cutoff.

Standardized CFA factor loadings ($lambda$) across empirical studies are systematically high:

• Item 1 (“concern”): $lambda = 0.86 – 0.91$
• Item 2 (“performance”): $lambda = 0.88 – 0.93$
• Item 3 (“glad / friendly”): $lambda = 0.85 – 0.90$

Multi-group CFA has further established measurement invariance (configural, metric, and scalar invariance) across diverse respondent groups (e.g., gender, generational cohorts such as Millennials vs. Generation Z, and cross-national samples), confirming that the instrument operationalizes the same psychological construct across varied consumer populations.

10. Instrument / Measurement Tool

The Green Purchase Intention Scale is structured as an efficient, self-administered questionnaire. Its operational parameters are detailed below:

• Instrument Name: Green Purchase Intention Scale (GPI)
• Authors / Originators: Yu-Shan Chen and Ching-Hsun Chang (2012)
• Assessment Type: Self-report psychometric rating scale
• Construct Measured: Consumer intention to purchase eco-friendly/environmentally sustainable products
• Number of Items: 3 items (unidimensional)
• Administration Format: Paper-and-pencil, computer-assisted web interviewing (CAWI), mobile survey interface, or integrated laboratory experiment module
• Estimated Completion Time: Approximately 30 to 60 seconds
• Response Format: 5-point Likert scale (1 = strongly disagree, 5 = strongly agree)
• Item Phrasing and Targets: Items can be deployed in relation to a specific focal product (using the prompt “this product”) or adapted for generic product categories (e.g., “green electronics,” “organic foods,” or “electric vehicles”) without compromising construct integrity.
• Scoring Guidelines:
  • Reverse Scoring: None. All three items are positively worded; there are no reverse-coded items.
  • Composite Score Calculation: The overall green purchase intention score is derived either by computing the arithmetic mean across the three items (ranging from 1.00 to 5.00) or by summing raw item scores (ranging from 3 to 15).
  • Score Interpretation: Higher composite scores indicate stronger psychological intention and higher subjective probability of buying green products. Mean scores around 3.0 represent a neutral conative stance, scores below 2.5 signify low behavioral readiness or reluctance, and scores exceeding 4.0 demonstrate high green purchase intention. In structural equation modeling, the construct is typically operationalized as a continuous latent variable.

11. Permissions & Fee and Test Year

The Green Purchase Intention Scale was originally published in 2012 in the peer-reviewed scholarly journal Management Decision (Emerald Group Publishing). The scale was developed within an academic research context and is documented in the public academic literature.

Licensing and Usage Permissions:

• Academic and Non-Commercial Research: The scale may be utilized by independent scholars, university faculty, graduate researchers, and non-profit research organizations for non-commercial educational and scientific research without payment of licensing fees, provided that formal academic attribution and citation are accorded to the foundational work of Chen and Chang (2012).
• Commercial Applications: Commercial entities, corporate research divisions, market intelligence agencies, and proprietary polling organizations seeking to integrate the instrument into commercial diagnostic toolkits or syndicated client deliverables should review Emerald Group Publishing’s copyright policies and contact the original authors regarding proprietary or for-profit deployment.

12. References

• Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
• Chen, Y. S., & Chang, C. H. (2012). Enhance green purchase intentions: The roles of green perceived value, green perceived risk, and green trust. Management Decision, 50(3), 502–520. https://doi.org/10.1108/00251741211216250
• Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Addison-Wesley.
• 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
• Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2019). Multivariate data analysis (8th ed.). Cengage Learning.
• Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science, 43(1), 115–135. https://doi.org/10.1007/s11747-014-0403-8
• Stern, P. C. (2000). Toward a coherent theory of environmentally significant behavior. Journal of Social Issues, 56(3), 407–424. https://doi.org/10.1111/0022-4537.00175

13. Items of the Scale