Gen Z Characteristics and Sustainable Consumption: Bridging the Intention–Behavior Gap

Abstract

Generation Z, a cohort defined by digital connectivity, sensitivity to social influence, and environmental awareness, has attracted considerable scholarly attention in sustainable consumption research. Yet a persistent gap between their expressed pro-sustainability attitudes and actual purchasing decisions remains well-documented. This study examines whether Gen Z characteristics help bridge that gap by directly influencing sustainable purchase behavior and by moderating the role of purchase intention in that process. A quantitative design was employed using survey responses from 302 Gen Z consumers. The findings suggest that while Gen Z characteristics significantly predicted actual sustainable purchasing and purchase intention exerted a positive direct effect, the interaction between the two was negative and statistically significant. Conditional effects analysis further revealed that the influence of generational characteristics on purchasing behavior is stronger at lower levels of purchase intention and progressively weaker as intention increases. These results suggest that traits such as digital responsiveness, social embeddedness, and environmental orientation do not merely reinforce existing intentions but appear to compensate for their absence, activating sustainability-aligned behavior even when motivational commitment is limited. The study repositions the intention–behavior gap among Gen Z as something modulated by generational characteristics that drive purchasing behavior when intention alone falls short.

1. Introduction

Sustainable consumption has emerged as a central preoccupation across marketing, public policy, and consumer research, driven by mounting pressure to address environmental degradation at a societal scale [1]. Yet despite widespread awareness of climate change, resource depletion, and concerns over ethical production, the conversion of positive environmental attitudes into sustained purchasing behavior continues to elude researchers and practitioners alike [2]. A substantial body of research [3,4,5] documents this persistent intention–behavior gap, in which consumers articulate strong support for environmentally responsible products yet consistently fail to act on those stated intentions. Identifying the conditions under which this gap narrows or widens is therefore not merely an academic exercise but it carries direct implications for both theoretical development and practical intervention in sustainability-oriented markets.

Generation Z occupies a particularly significant position within this conversation [6]. As the first cohort to have grown up entirely within digital environments, Gen Z consumers are deeply enmeshed in social media ecosystems, routinely exposed to peer reviews and online influencers, and often active participants in brand communities [7]. Their environmental awareness is frequently cited as a defining generational trait [8]. Collectively, these characteristics suggest that Gen Z may represent a meaningful force in the broader shift toward sustainable consumption. That said, this cohort is also characterized by pragmatism and price sensitivity, which complicates any straightforward optimism about their purchasing behavior [9]. The interplay of digital fluency, social responsiveness, economic constraint, and environmental values renders Gen Z a particularly instructive group through which to examine how sustainable consumption actually unfolds [6].

The Greek context offers a particularly informative setting for this investigation. Greek Gen Z consumers came of age during a prolonged economic crisis and its aftermath, which has produced a generational profile in which strong digital immersion and elevated environmental awareness coexist with pronounced price sensitivity and constrained discretionary spending [6]. Recent studies on Greek Gen Z indicate above-average reliance on peer-generated content, online reviews, and short-form video for product evaluation, alongside notably value-driven attitudes toward sustainability [6,7]. At the same time, the relatively limited availability and higher relative price of certified sustainable products in the Greek market amplify the friction between intention and behavior. Examining Gen Z characteristics in this setting therefore allows the intention–behavior gap to be observed under conditions where motivational, social, and economic forces are unusually salient simultaneously, providing a context-rich test of the proposed framework.

Despite the breadth of existing research, several important gaps persist. Much of the prior literature has concentrated disproportionately on purchase intention rather than observable purchasing behavior [10,11,12,13]. Intention is, of course, a meaningful predictor of action, but an exclusive reliance on intention-based measures tends to overstate the effectiveness of sustainability-related messaging while underweighting the structural and contextual factors that drive real purchasing decisions. A second limitation concerns the fragmented treatment of generational characteristics. Existing studies tend to isolate individual determinants of sustainable consumption, such as environmental concern, perceived consumer effectiveness, or price sensitivity, without accounting for the broader constellation of traits that collectively define a generational cohort [14,15]. In the case of Gen Z, attributes such as digital immersion, social influence responsiveness, participation in online communities, and visual orientation are often discussed separately and descriptively, rarely modeled as an integrated set of generational characteristics with collective explanatory power [16,17,18]. This fragmentation obscures how such traits operate in combination to shape market behavior.

Perhaps the most consequential oversight, however, concerns the intention–behavior gap itself. Existing models have examined this gap predominantly through motivational and cognitive variables [5,19,20], with relatively little attention to the possibility that generational characteristics may condition how strongly intention translates into action. The prevailing assumption is that the intention–behavior link is relatively stable, moderated primarily by situational constraints or perceptions of behavioral control [21,22]. What remains largely unexplored is whether generational traits function as a contextual layer that either amplifies or attenuates the behavioral impact of intention. The degree to which sustainable intentions convert into actual purchases may, in other words, vary systematically depending on how strongly consumers embody the characteristics associated with their generation.

The present study addresses these gaps by examining how Gen Z characteristics shape actual sustainable purchasing behavior and by testing their moderating role in the relationship between purchase intention and behavior. Drawing on survey data from 302 Gen Z consumers and employing moderation analysis, the study investigates whether generational traits can help close the intention–behavior gap by activating purchasing, particularly among consumers whose intentions are not yet firmly established. This study makes three connected contributions, of which the first is the principal theoretical novelty. Specifically, at the theoretical level, the study introduces and tests a compensatory mechanism within the intention–behavior gap: Gen Z characteristics substitute for, rather than amplify, motivational commitment, activating sustainable purchase behavior most strongly when intention is weak. This reverses the standard assumption that contextual factors reinforce strong intentions, and reframes the intention–behavior link as a variable rather than a fixed quantity. The mechanism is identified by integrating TPB and S-O-R into a single architecture, with intention specified as a moderator rather than a mediator. At the conceptual level, the study reconceptualizes Gen Z characteristics as a higher-order, multidimensional construct integrating seven previously isolated dimensions, providing a unified operationalization where the prior literature offered fragmented predictors. At the empirical level, the study tests these relationships against actual self-reported purchase behavior rather than intention alone, in a Greek Gen Z sample, addressing repeated calls for behavioral evidence in sustainable consumption research. In this way, the study contributes to a more context-sensitive understanding of sustainable consumption within the contemporary digital landscape.

2. Literature Review

2.1. Sustainable Consumption and the Intention Behavior Gap

Sustainable consumption is broadly understood as a pattern of purchasing and use that minimizes environmental harm, upholds principles of social equity, and supports long-term economic viability without compromising the capacity to meet individual needs [23,24]. In consumer research, it is typically operationalized through the acquisition of products that are environmentally friendly, ethically produced, and socially responsible [25,26]. As ecological pressures mount and public scrutiny of corporate conduct intensifies, sustainable consumption has established itself as a central preoccupation in both marketing scholarship and industry practice [27]. Yet the relationship between awareness and action remains fraught. A well-documented body of research points to the persistence of an intention–behavior gap, capturing the recurring inconsistency between what consumers say they intend to do and what they actually do in the marketplace [20,21,22]. Even among those who express genuine environmental concern, contextual barriers, competing priorities, and situational pressures frequently erode the link between intention and follow-through [28]. This raises a fundamental question about the sufficiency of intention as an explanatory variable, particularly for younger cohorts whose decision-making is deeply embedded in digital and socially mediated environments [29].

The Theory of Planned Behavior (TPB) provides a well-established account of sustainable purchasing decisions [30]. It positions intention as the proximate determinant of behavior, formed through attitudes, subjective norms, and perceived behavioral control [31]. Applied to sustainable consumption, this implies that favorable attitudes toward sustainability, perceived social expectations, and a sense of behavioral capacity should produce intention, which in turn predicts purchasing. Empirically, however, the intention–behavior link is probabilistic rather than deterministic, attenuated by habit, contextual constraints, and economic considerations [4,32,33,34]. Gen Z fits this framework naturally: digital immersion shapes subjective norms, environmental values shape attitudes, and access to online information influences perceived control [35,36,37]. At the same time, TPB itself acknowledges that perceived control and contextual factors can affect behavior directly, bypassing intention [38].

The Stimulus–Organism–Response (S-O-R) framework complements TPB by emphasizing that external stimuli (digital exposure, peer cues, visual content) act on internal states such as intention, which in turn drive behavior [39]. Read together, the two frameworks suggest a clear division of labor. TPB captures the motivational pathway; S-O-R captures the stimulus-driven activation pathway. Where intention is firmly established, behavior is primarily motivational; where intention is weak, digital and social stimuli can activate behavior directly. Gen Z characteristics are therefore relevant not only to forming intention but also to executing behavior, and the intention–behavior gap can be reframed as a phenomenon that varies with generational context rather than as a fixed quantity. Read together, the two frameworks supply the scaffolding for the empirical model tested below, namely that Gen Z characteristics predict both purchase intention and actual behavior, and that the strength of the link to behavior depends on how firmly intention is already in place.

2.2. Gen Z Characteristics in the Digital Era

Generation Z is widely recognized as the first fully digital cohort, having come of age in an environment defined by social media platforms, mobile connectivity, algorithmic content curation, and near-constant online interaction [6]. What distinguishes this generation from its predecessors is not simply the use of digital tools but a thoroughgoing embeddedness within digitally mediated ecosystems that organize how information is accessed, products are evaluated, preferences are formed, and identity is performed [40,41,42]. These conditions have given rise to a distinctive behavioral configuration that bears directly on marketplace decision-making, especially in domains like sustainable consumption where social signaling and value expression carry particular weight [43].

Digital nativeness stands as perhaps the most fundamental of these characteristics [44]. Growing up with continuous access to digital interfaces has normalized the use of online reviews, influencer endorsements, visual content, and viral trends as the default sources of product information [45,46,47]. Rather than being shaped primarily by conventional advertising, Gen Z consumers tend to privilege peer-generated content and digitally mediated cues when forming brand evaluations [7,48]. Within the Theory of Planned Behavior, such pervasive digital exposure plays a central role in configuring subjective norms, given that individuals are continuously absorbing the opinions and behaviors of others in their networks [49]. The influence of online reviews exemplifies this dynamic since reviews operate as decentralized information signals that reduce uncertainty and provide a form of social validation grounded in collective experience [50]. Influencer impact, meanwhile, speaks to the role of parasocial relationships and perceived authenticity in shaping attitudes and behavioral intentions, a process consistent with Social Influence Theory’s emphasis on the informational and normative dimensions of consumer decision-making [51]. Online buzz impact captures something slightly different; the way in which the visibility and virality of sustainability-related content reinforce collective awareness and social expectations, making sustainable behavior feel not just desirable but socially normative [52]. Visual element impact rounds out this picture by reflecting Gen Z’s pronounced orientation toward visual communication formats, including short-form video, aesthetic branding, and symbolic sustainability cues [53]. Such visual signals tend to function as heuristic shortcuts that shape attitudes rapidly, a process well captured by the S-O-R framework’s account of how external stimuli condition internal evaluations and the behavioral responses that follow [54]. Taken together, these dimensions articulate the depth of Gen Z’s digital immersion and provide strong justification for their inclusion as core components of the Gen Z characteristics construct. After reviewing each dimension in detail, it is useful to indicate how the seven sub-dimensions group together conceptually. The normative dimensions (online reviews, influencers, online buzz) capture the informational and social-proof channels through which subjective norms are formed. The cognitive-perceptual dimension (visual elements) captures the heuristic processing of aesthetic and symbolic stimuli emphasized by the S-O-R framework. The environmental-value dimension (environmental responsibility) captures the value-based attitudinal component of TPB and the moral norm activation posited by Value–Belief–Norm theory. The contextual–resource dimensions (product price, brand community) capture, respectively, perceived behavioral control and identity-based group membership. This four-fold mapping clarifies that the construct integrates normative, cognitive, environmental, and contextual mechanisms rather than enumerating loosely related traits.

Beyond their digital orientation, Gen Z consumers are commonly described as value-driven, socially conscious, and community-embedded. They attach significant weight to authenticity, ethical positioning, and environmental responsibility in their consumption [55,56,57], and their decision-making is closely tied to peer networks, brand communities, and social-identity processes [58,59,60,61,62,63]. Within the TPB framework, value-based orientations shape attitudes toward sustainable behavior [64,65,66], while community embeddedness and peer signaling shape subjective norms and perceptions of group endorsement, in line with Social Identity Theory and the logic of social proof [67,68,69,70]. The environmental responsibility and brand community dimensions capture, respectively, the value-based and identity-based facets of this configuration. Economic considerations form a further, distinct layer. Many Gen Z consumers are still in education or in early career stages, and sustainable purchasing often involves negotiating a tension between ethical preference and economic practicality [6,71,72]. The product price dimension captures this friction and maps directly onto perceived behavioral control in TPB [73], since elevated prices can constrain the translation of favorable attitudes and norms into purchase.

Considered collectively, the dimensions of online review impact, influencer impact, online buzz impact, visual element impact, environmental responsibility impact, product price impact, and brand community impact constitute a multidimensional operationalization of Gen Z characteristics in the digital era. These components capture the principal features that distinguish Gen Z as a consumer cohort that illustrates the depth of their digital immersion, their responsiveness to social influence, their value-driven orientations, their embeddedness in community structures, and their economic calculus. By integrating these elements into a higher-order construct labeled Gen Z characteristics, this study treats generational traits not as background demographic attributes but as behaviorally consequential drivers of sustainable consumption. This conceptualization aligns coherently with both the Theory of Planned Behavior and the Stimulus-Organism-Response framework. Gen Z characteristics can be understood as external and social stimuli that configure internal evaluative states such as attitudes and purchase intention, while simultaneously exerting a direct influence on behavior through contextual and socially reinforced mechanisms. This dual role underscores the importance of situating sustainable consumption research within a generational context and provides the conceptual grounding for understanding how digitally embedded characteristics shape both intention formation and actual purchasing conduct.

The selection of the seven dimensions is therefore not the result of an ad hoc aggregation of prior measures but a theory-driven mapping in which each dimension corresponds to a specific mechanism within the integrated TPB–S-O-R framework. Online review impact and influencer impact operationalize the informational and normative components of subjective norms (TPB) and of Social Influence Theory. Online buzz impact and brand community impact capture the normative pressure arising from group membership and collective visibility, consistent with Social Identity Theory and the S-O-R notion of socially mediated stimuli. Visual element impact reflects the perceptual-affective stimuli that, in the S-O-R framework, shape internal evaluative states through aesthetic and symbolic cues. Environmental responsibility impact operationalizes the value-based attitudinal component of TPB and the moral norm activation posited by Value-Belief-Norm theory. Product price impact corresponds to perceived behavioral control in TPB, capturing the resource constraints that condition whether favorable attitudes translate into purchase. Five of the seven dimensions therefore capture digital and social stimuli (S-O-R) that act through subjective norms (TPB), while the remaining two anchor the construct in the value-based (attitudes) and resource-based (PBC) components of TPB. This mapping provides a single, consistent theoretical foundation for treating the seven dimensions as facets of one higher-order construct.

Although a substantial body of research has examined Gen Z sustainable consumption, the evidence remains fragmented across geographical contexts, methodological approaches, and conceptualizations of the constructs at stake. Studies have variously emphasized environmental concern, social media exposure, influencer effects, price sensitivity, or value alignment, but they rarely integrate these elements into a single analytical framework, and even more rarely do they situate Gen Z findings within the specific socioeconomic conditions of Southern European economies such as Greece. Table 1 synthesizes the principal empirical studies on Gen Z and sustainable consumption across contexts, summarizing their settings, methods, and main findings. The table makes three patterns visible. First, almost all prior studies operationalize sustainable engagement through purchase intention rather than actual purchase behavior. Second, the antecedents tested are typically isolated determinants (a single dimension at a time) rather than an integrated higher-order characterization of the cohort. Third, no prior study, to our knowledge, has examined the moderating role of purchase intention in the relationship between generational characteristics and actual purchasing behavior. The present study addresses these gaps by examining a Greek Gen Z sample, by integrating seven dimensions into a higher-order Gen Z characteristics construct, and by testing intention as a moderator on actual purchasing behavior.

Table 1. Summary of prior empirical studies on Gen Z and sustainable consumption across contexts.

Study	Country/Region	Focus	Method	Key Findings
Pinho & Gomes (2024) [8]	Portugal/EU	Generational comparison of ecological commitment	Survey	Gen Z more ecologically committed than X/Y but with weaker behavioral follow-through
Dragolea et al. (2023) [35]	Romania	Determinants of Gen Z sustainable behavior	SEM	Environmental concern and social influence drive intention; price weakens behavior
Kara & Min (2024) [43]	USA	Influencers and moderators of sustainable consumption	Regression	Influencer credibility and social norms shape intention; behavior depends on PBC
Confetto et al. (2023) [61]	Italy	Social-media content and sustainability advocacy	Survey	Digital content significantly drives sustainable behaviors
Anjum et al. (2023) [71]	Malaysia	Green consumption intentions	SEM	Sustainable marketing influences green intentions
Hudayah et al. (2023) [16]	Indonesia	Green perceived value & intention	Survey	Environmental concern moderates value–intention link
Choi et al. (2025) [17]	Hong Kong	Socially responsible consumption	Survey	Environmental knowledge and PCE drive responsible consumption
Theocharis & Tsekouropoulos (2025) [6]	Greece	Branding and Gen Z technology adoption	Survey	Brand dimensions shape Gen Z sustainable behavior
Theocharis (2025) [7]	Greece	Peer dynamics in digital marketing	Survey	Peer influence varies by product type
Present study	Greece	Gen Z characteristics, intention, and actual behavior; moderation of intention	Survey + moderation (PROCESS)	Gen Z characteristics directly affect actual purchase; intention moderates negatively (compensatory effect at low intention)

In summary, the seven sub-dimensions form a single higher-order construct that brings normative, cognitive, environmental, and contextual mechanisms under one analytical roof, with each facet anchored to a specific component of TPB or to the S-O-R logic of stimulus-driven activation. This conceptualization is what allows the empirical model that follows to test both the direct effects of Gen Z characteristics on intention and behavior and the question of whether intention conditions how strongly those characteristics translate into purchasing. The following section develops the hypotheses that guide the empirical investigation.

2.3. Hypotheses Development

The theoretical discussion leads to three concrete empirical questions. The first is whether Gen Z characteristics, taken as an integrated construct, do in fact predict sustainable purchase intention, and how much of the variance in intention they account for. The second is whether the same construct predicts actual sustainable purchase behavior, and not only intention, given that much of the prior literature has stopped at intention as a near-sufficient proxy. The third, and the one most directly aimed at the intention–behavior gap, is whether the strength of the link between Gen Z characteristics and actual behavior depends on the level of purchase intention itself, or whether the two operate independently of one another. These three questions, in turn, shape the three hypotheses developed in this section. Drawing on the theoretical discussion above, three hypotheses are developed to test the role of Gen Z characteristics in sustainable consumption. We treat Gen Z characteristics as a multidimensional construct that brings together digital immersion, sensitivity to social influence, environmental orientation, community engagement, and economic evaluation, and we expect each of these facets to feed into both purchase intention and actual sustainable purchasing. Since the analysis is regression-based, the hypotheses are stated in predictive terms, capturing both the direction of the effect and the share of variance the construct explains in the outcome. Two of the hypotheses concern direct effects on purchase intention and on actual purchase behavior, while the third tests whether purchase intention moderates the link between Gen Z characteristics and behavior.

Within the Theory of Planned Behavior, purchase intention occupies the position closest to behavior, formed through the antecedents of attitudes, subjective norms, and perceived behavioral control [74,75,76]. Gen Z characteristics are theoretically well-positioned to shape each of these components. Environmental responsibility and value-driven consumption orientations strengthen favorable attitudes toward sustainable products, since alignment between personal values and product attributes enhances the positive evaluation of environmentally responsible choices [77,78]. The social and digital embeddedness of Gen Z, manifested through exposure to online reviews, influencers, online buzz, and brand communities, amplifies subjective norms by reinforcing perceived social expectations and lending legitimacy to sustainable purchasing [7,79]. Access to online information and visual cues may additionally enhance perceived behavioral control by reducing informational uncertainty and supporting more confident decision-making [80]. The Stimulus–Organism–Response framework complements this picture [39,40] and Gen Z characteristics can be positioned as external stimuli that configure internal cognitive and affective states, including purchase intention. Digital exposure and social validation mechanisms act upon the organism state, which then steers behavioral responses [39,49]. Taken together, these converging theoretical arguments imply that Gen Z characteristics should not only exert a directional effect on sustainable purchase intention but also account for a meaningful share of its variance, a dual expectation that motivates the first hypothesis in predictive form.

H1: 

Gen Z characteristics positively predict sustainable purchase intention.

While purchase intention functions as a key predictor of behavior, sustainable purchasing is also subject to contextual and generational influences that may act on behavior directly, bypassing or supplementing the intentional pathway [14,15]. The Theory of Planned Behavior itself acknowledges that perceived behavioral control and situational factors can shape behavior independently of intention [30,31,32]. In digitally embedded environments, exposure to online reviews, influencer endorsements, brand communities, and sustainability-related visual content may activate purchasing behavior even when underlying intention remains moderate [6,35,36]. The Stimulus–Organism–Response framework reinforces this point suggesting that external stimuli do not operate exclusively through the organism state but may also produce more direct behavioral reactions [54]. Prominent online buzz or community endorsement, for instance, can trigger socially reinforced purchasing decisions, while price considerations may either facilitate or constrain behavior regardless of how firmly intention is established. Given that Gen Z characteristics encompass both motivational and contextual dimensions, they are expected to predict actual sustainable purchasing behavior directly, both in the sense of exerting a positive effect and in the sense of accounting for variance in the behavioral outcome.

H2: 

Gen Z characteristics positively predict actual sustainable purchase behavior.

Although the Theory of Planned Behavior positions intention as the primary proximate determinant of behavior, the extent to which generational traits translate into action may itself depend on the level of purchase intention. When intention is high, behavior is largely driven by motivational commitment, and the additional explanatory contribution of generational characteristics may be correspondingly limited. When intention is lower or not yet firmly established, however, traits such as digital immersion, social influence sensitivity, community orientation, and environmental responsibility may compensate for weaker motivation, activating behavior through normative reinforcement and contextual stimulation. The Stimulus–Organism–Response perspective supports this reasoning since intention represents the organism state, while Gen Z characteristics constitute the stimulus conditions. The interaction between these two elements shapes the behavioral response. Where intention is weak, stimuli associated with Gen Z characteristics may play a stronger activating role; where intention is strong, the incremental influence of those stimuli is likely to diminish. This argument speaks directly to the intention–behavior gap in sustainable consumption, proposing that generational characteristics condition how effectively intention converts into action.

H3: 

Purchase intention moderates the relationship between Gen Z characteristics and actual purchase behavior.

2.4. Conceptual Model

The model brings together three elements. Gen Z characteristics enter as the independent variable, actual sustainable purchase behavior as the dependent variable, and purchase intention as the moderator that shapes how strongly the two are linked. Intention also appears, separately, as an outcome of Gen Z characteristics, but this second relationship is examined on its own rather than as part of a mediation chain feeding into behavior. Keeping the two analyses apart matters because intention does double duty in the paper, once as something Gen Z characteristics predict, and once as the variable that conditions how those characteristics translate into purchasing. The two roles are estimated in separate regressions, and the moderation model does not estimate, nor does the paper claim, a mediation effect. Treating intention as a moderator rather than a mediator follows from how TPB and the S-O-R perspective are read together here. A strict TPB account would place intention as the link that carries antecedents through to behavior. Once S-O-R is brought in alongside TPB, however, Gen Z characteristics are no longer only antecedents to intention; they also function as contextual stimuli that can act on behavior directly, without first passing through the intentional pathway. The empirical question worth asking in that reading is not whether intention transmits the influence of generational traits, but whether the strength of that influence shifts with the level of intention itself. That is a moderation question, and specifying it as such is what allows the compensatory mechanism, which we argue is the paper’s principal theoretical contribution, to be put to the test. A moderated-mediation specification could clearly add further insight and can be a productive future research direction.

A moderating mechanism is incorporated into the model specifically to address the intention–behavior gap. Purchase intention is proposed to moderate the relationship between Gen Z characteristics and actual purchase behavior, with the underlying premise that the strength of the link between generational traits and behavioral outcomes is not fixed but varies with the level of motivational commitment. When intention is high, behavior is expected to be predominantly steered by that motivational commitment, with generational traits contributing less additional explanatory weight. When intention is lower, however, Gen Z characteristics may take on a more pronounced compensatory role, activating behavior through social influence, digital engagement, and contextual stimulation. Taken as a whole, the proposed model (Figure 1) integrates motivational, contextual, and generational perspectives into a unified explanatory framework for sustainable consumption among Gen Z. By examining the effects of Gen Z characteristics on both purchase intention and actual behavior, while also modeling the moderating role of intention, the framework moves beyond prior research that has concentrated predominantly on intention-based explanations. The model is empirically tested using survey data from Gen Z consumers, employing moderation analysis to evaluate both direct effects and the interaction between generational characteristics and motivational commitment.

Figure 1. Conceptual model.

3. Materials and Methods

3.1. Research Design

This study adopted a quantitative research design to investigate the relationships among Gen Z characteristics, purchase intention, and actual sustainable purchase behavior. Data were collected through a cross-sectional survey, a method well-suited to gathering standardized information from a defined population and supporting statistical tests of hypothesized relationships among constructs [81]. The research is explanatory in orientation, concerned with establishing both direct and moderating effects within the proposed conceptual framework.

3.2. Sample and Data Collection

The target population comprised Generation Z consumers, defined in this study as individuals born between 1997 and 2012 [6,7,82]. The sampling frame consisted of undergraduate students aged 18 years or older, a criterion that ensured all participants satisfied both the generational boundary and ethical eligibility requirements for participation. Initial selection relied on convenience sampling, a practical choice given the researchers’ direct access to the relevant population and a method commonly employed in behavioral research where access to specific groups is necessary [83,84]. To introduce additional structure into the selection process, the convenience sample was supplemented by a fixed-pattern selection procedure within the sampling frame. The frame consisted of an ordered enrollment list of undergraduate students in a Private University in Thessaloniki, Greece, randomized by removing alphabetical ordering. From this list, participants were selected following a repeating positional pattern (1st, 3rd, 6th, and 9th positions within each block of nine). However, since the intervals between selected positions are not constant, this procedure does not constitute systematic sampling in the strict probabilistic sense, and the sample should not be treated as a fully representative probability sample of the Greek Gen Z population. The procedure is best characterized as structured convenience sampling, with the corresponding generalizability constraints discussed as limitation. The final sample comprised 302 participants. Sample adequacy was evaluated through an a priori power analysis using G Power analysis [85,86]. With a medium effect size (f² = 0.15), a significance threshold of α = 0.05, statistical power set at 0.95, and seven predictors entered into the model, the minimum sample required for multiple regression was estimated at approximately 153 respondents. The obtained sample of 302 comfortably exceeds this benchmark, providing statistical power to detect both main effects and interaction effects. This is of particular relevance in moderation analysis, where detecting reliable interaction terms typically requires larger samples [87]. Finally, because the sampling frame is restricted to Greek undergraduate students aged 18 and over, generalizability of the findings is limited to this sub-population of Gen Z. Working, non-student, and younger Gen Z consumers may differ on purchasing autonomy, financial independence, and digital media use, and the findings should not be extrapolated to the full generational spectrum without replication.

3.3. Measures

All constructs were assessed using structured questionnaire items adapted from established instruments in the sustainable consumption and digital consumer behavior literatures. Items were measured on Likert-type scales, a format extensively used in behavioral research for capturing attitudes, perceptions, and behavioral tendencies [88]. Gen Z characteristics were operationalized as a multidimensional construct comprising seven dimensions: online review impact, influencer impact, online buzz impact, visual element impact, environmental responsibility impact, product price impact, and brand community impact. These dimensions collectively capture the digital, social, value-driven, and economic aspects of Gen Z consumer behavior, and a composite score was computed to represent the overall construct. Purchase intention [89,90] and actual purchase behavior [91] were each assessed using well-validated scales with established application in consumer behavior and sustainable consumption research. Purchase intention reflected the degree of willingness and commitment to engage in sustainable purchasing, while actual purchase behavior was assessed through self-reported engagement in environmentally responsible purchasing within a specified reference period. The use of validated scales supports the reliability and cross-study comparability of the measurements.

3.4. Data Analysis and Analytical Model

The analytical model is a moderation model in the strict statistical sense, estimated as a single regression equation with Gen Z characteristics as the independent variable, purchase intention as the moderator, actual purchase behavior as the dependent variable, and no mediator. The parallel regression of purchase intention on Gen Z characteristics is reported separately, as a descriptive relationship of theoretical interest, and is not part of a tested mediation pathway. Data analysis was carried out in SPSS (version 27) following a structured sequential procedure. Descriptive statistics were first computed to characterize the sample and examine distributional properties. Internal consistency was assessed using Cronbach’s alpha, and exploratory factor analysis was performed to evaluate construct validity and examine the dimensional structure of the Gen Z characteristics construct [88]. Construct validity was examined through Exploratory Factor Analysis given the partly novel multidimensional configuration of the Gen Z characteristics construct, which integrates dimensions previously studied in isolation. Confirmatory Factor Analysis and PLS-SEM were not adopted in this study because the focus of the hypotheses is on direct and moderating effects of a higher-order composite rather than on a full measurement-and-structural model. To complement Cronbach’s alpha, we computed McDonald’s omega and composite reliability for each construct, values were broadly consistent with the alpha estimates. Hypothesis testing was conducted using regression-based moderation analysis via the PROCESS macro developed by Hayes [92]. The effects of Gen Z characteristics on purchase intention and actual purchase behavior were examined to test both directional and predictive relationships. The moderating role of purchase intention was assessed by introducing an interaction term between Gen Z characteristics and purchase intention as a predictor of actual purchase behavior. Bootstrapping with 5000 resamples was used to generate bias-corrected confidence intervals for the interaction effects, strengthening the robustness of the inferences drawn [92]. Prior to computing the interaction term, all variables were mean-centered to mitigate potential multicollinearity [87]. A statistically significant interaction coefficient was taken as evidence of moderation, and conditional effects were subsequently examined at low, moderate, and high levels of purchase intention to characterize the nature of the interaction. Overall model performance was evaluated using the coefficient of determination (R²), F-statistics, and incremental variance explained by inclusion of the interaction term.

3.5. Ethical Considerations

The study was conducted in accordance with established ethical standards for social science research, including the principles set out in the Declaration of Helsinki [93]. Participation was entirely voluntary, and all respondents were briefed on the aims of the research before completing the survey. Informed consent was obtained from each participant, and strict confidentiality and anonymity were maintained throughout. No personally identifiable information was collected or retained, and participants were explicitly informed of their right to withdraw at any point without consequence. Data were used solely for academic purposes, consistent with recognized ethical guidelines for social science inquiry [94].

4. Results

4.1. Gen Z Characteristics Scale Reliability and Validity Assessment

The construct validity of the Gen Z characteristics measures was examined through Exploratory Factor Analysis using Principal Component Analysis [95]. Before proceeding with factor extraction, the dataset was assessed for suitability using two standard diagnostics: the Kaiser-Meyer-Olkin measure of sampling adequacy and Bartlett’s Test of Sphericity. As shown in Table 2, the KMO value was 0.728, comfortably above the recommended minimum of 0.60, indicating that the variables share sufficient common variance and that the sample is appropriate for factor analysis. Bartlett’s Test of Sphericity returned a statistically significant result (χ² = 2714.382, df = 253, p < 0.001), confirming that the correlation matrix departs meaningfully from an identity matrix. Taken together, these diagnostics support proceeding with factor extraction. Seven factors were identified, corresponding to the seven conceptual dimensions used to operationalize Gen Z characteristics: online review impact, influencer impact, online buzz impact, visual element impact, environmental responsibility impact, product price impact, and brand community impact. The solution is therefore fully consistent with the theoretical structure underpinning the construct. Collectively, the seven factors accounted for 67.195% of total variance, a figure that surpasses the 60% threshold widely regarded as acceptable in social science research. This level of explained variance indicates that the factor solution captures a substantial share of the variability in the observed items and provides empirical support for the multidimensional structure of the Gen Z characteristics construct.

Table 2. KMO and Bartlett’s Test for Gen Z characteristics.

KMO and Bartlett’s Test
Kaiser–Meyer–Olkin Measure of Sampling Adequacy.		0.728
Bartlett’s Test of Sphericity	Approx. Chi-Square	2714.382
	df	253
	Sig.	0.000
Factors extracted		7
Variance Explained Factor 1	%	13.104
Variance Explained Factor 2	%	11.892
Variance Explained Factor 3	%	10.447
Variance Explained Factor 4	%	9.216
Variance Explained Factor 5	%	8.763
Variance Explained Factor 6	%	7.681
Variance Explained Factor 7	%	6.092
Total Variance Explained	%	67.195

Communalities and the rotated component matrix for the construct items are reported in Table 3. Communality values across items range from 0.502 to 0.784, indicating that each item shares a satisfactory proportion of variance with the extracted factors. Values above 0.50 are generally regarded as acceptable in the literature, and the present results confirm that all items contribute meaningfully to the underlying factor structure. Inspection of the rotated component matrix reveals strong item-to-factor relationships throughout. The majority of factor loadings exceed 0.70, reflecting clear and robust associations between items and their designated constructs. Items 1 and 2, for instance, load prominently on Component 1, with values of 0.863 and 0.833 respectively, while Items 3 to 5 cluster coherently on Component 2, a pattern that is broadly replicated across the remaining components. The consistency of these high loadings constitutes evidence of convergent validity: items designed to capture the same conceptual dimension demonstrate strong co-variation with a common factor. Equally important is what the factor structure reveals about discriminant validity. The components exhibit clear separation, with items loading predominantly on a single factor and showing little evidence of substantial cross-loadings on others. This clean structure supports the interpretation that the seven dimensions, while related, represent conceptually distinct facets of the Gen Z characteristics construct rather than overlapping or redundant indicators of the same underlying phenomenon.

Table 3. Communalities and component matrix for Gen Z characteristics.

Communalities			Rotated Component Matrix
	Initial	Extraction	Comp. 1	Comp. 2	Comp. 3	Comp. 4	Comp. 5	Comp. 6	Comp. 7
Item 1	1.000	0.784	0.863
Item 2	1.000	0.732	0.833
Item 3	1.000	0.753		0.844
Item 4	1.000	0.662		0.786
Item 5	1.000	0.734		0.837
Item 6	1.000	0.686			0.805
Item 7	1.000	0.605			0.736
Item 8	1.000	0.615				0.724
Item 9	1.000	0.701				0.708
Item 10	1.000	0.703				0.563
Item 11	1.000	0.654					0.782
Item 12	1.000	0.712					0.765
Item 13	1.000	0.502					0.525
Item 14	1.000	0.723						0.808
Item 15	1.000	0.706						0.804
Item 16	1.000	0.524							0.440
Item 17	1.000	0.547							0.521
Item 18	1.000	0.575							0.703

Extraction Method: Principal Component Analysis.

Considered as a whole, the exploratory factor analysis offers compelling support for the validity of the Gen Z characteristics construct. The satisfactory KMO value, the significant Bartlett’s Test result, the substantial proportion of explained variance, adequate communality values, and consistently high factor loadings converge in confirming that the measurement items reliably capture the multidimensional structure of Gen Z characteristics within the sustainable consumption context. Regarding internal consistency, Cronbach’s alpha, McDonald’s omega, and composite reliability (CR) were computed for each construct and are reported in Table 4. Alpha values range from 0.595 to 0.934. Most constructs exceed the 0.70 threshold, with several showing notably high coherence among items. Two dimensions, environmental responsibility impact (α = 0.595) and influencer impact (α = 0.611), return alpha values below 0.70. We acknowledge this as a measurement limitation, and five considerations contextualize the result. First, both scales were operationalized with only three items each, a condition under which alpha is known to underestimate reliability since alpha depends jointly on average inter-item correlation and on scale length; under such conditions, values in the region of 0.60 are considered defensible in exploratory contexts [96]. Second, McDonald’s omega and CR values for these constructs (ω = 0.66/0.67; CR = 0.69/0.70) sit closer to the 0.70 benchmark, indicating that the alpha-based estimate is conservative. Third, applying a Spearman–Brown projection to a hypothetical doubling of items yields projected alpha values of approximately 0.75 for environmental responsibility and 0.76 for influencers, suggesting that the underlying coherence is acceptable and that the low observed alpha is partly an artefact of scale brevity rather than of construct heterogeneity. Fourth, the Average Variance Extracted (AVE) values for both constructs approach but do not exceed the 0.50 threshold (AVE ≈ 0.49 for environmental responsibility and ≈0.45 for influencers); read alongside the omega and CR values, this pattern points to a borderline but defensible measurement structure for exploratory work, rather than to a breakdown of convergent validity. Fifth, the study is exploratory in orientation, integrating dimensions that have previously been examined separately in the literature; in such contexts, alpha values in the region of 0.60 are widely regarded as defensible provided that other reliability and validity indicators (omega, CR, AVE) align directionally, as they do here. We nevertheless flag this as a measurement limitation and commit, in Section 6.2, to expanding the item pools for these two dimensions and to applying confirmatory factor analysis in future replications. On balance, the reliability analysis indicates that the measurement scales perform satisfactorily and provide a sound basis for the subsequent statistical analyses. The items measuring Gen Z characteristics, purchase intention, and actual purchase behavior each demonstrate sufficient internal consistency to be regarded as coherent reflections of their respective constructs.

Table 4. Reliability analysis.

Scale	Cronbach’s Alpha	Items
Online review impact	0.732	2
Influencer impact	0.611	3
Online buzz impact	0.707	2
Visual element impact	0.754	3
Environmental responsibility impact	0.595	3
Product price impact	0.934	2
Brand community impact	0.919	3
Purchase intention	0.801	5
Actual purchase	0.827	4

Common Method Bias

Because all variables were measured using self-reports from a single survey, the potential for common method bias was evaluated using Harman’s single-factor test. An unrotated principal-components extraction constraining all items to a single factor accounted for less than 50% of the total variance (≈28.4%), below the threshold typically used to flag common method bias as a substantive concern. Procedural remedies were also implemented at the design stage: respondent anonymity was guaranteed, items measuring predictors and outcomes were separated in the questionnaire, response scales varied across blocks, and items were phrased to reduce evaluation apprehension. Taken together, these checks suggest that common method variance is unlikely to have materially biased the relationships reported.

4.2. Descriptive Analysis

Table 5 summarizes the demographic characteristics of the study sample. Gender distribution is reasonably balanced, with female respondents accounting for 52.3% of participants and male respondents making up the remaining 47.7%. This near-equal representation is a useful feature of the sample, as it reduces the risk of gender-related bias in the findings and supports broader applicability within the Gen Z population. In terms of educational attainment, the sample is dominated by current university students (85.4%), a profile that aligns closely with the typical life stage of Generation Z. A smaller share of respondents hold a Bachelor’s degree (11.2%), while 2.7% have completed postgraduate education at the Master’s level, and 0.7% entered the study having completed secondary school only. The overall educational composition reflects the transitional academic and early career stage that characterizes much of this cohort. The distribution of family income is reasonably spread across the available categories. The largest single group reports household income in the range of 10,000 to 20,000 euros (41.7%), followed by those with income above 20,000 euros (39.1%), while 19.2% of respondents come from households earning below 10,000 euros annually. This variation across income bands means the sample is not skewed toward any particular socioeconomic stratum, which strengthens confidence in the diversity of the respondent pool. Participants ranged in age with a mean of 20.52 years and a standard deviation of 2.35, confirming that the sample falls squarely within the Generation Z age bracket. Taken together, the demographic profile of the sample is consistent with the defining characteristics of this generational cohort and supports the suitability of the dataset for examining sustainable consumption behavior among Gen Z consumers.

Table 5. Demographic characteristics.

Variable	Categories	Percentage
Gender	Male	47.7
	Female	52.3
Educational level	High school	0.7
	University student	85.4
	Bachelor’s degree	11.2
	Master’s degree	2.7
Family income	<10.000 euros	19.2
	10–20.000 euros	41.7
	>20.000 euros	39.1
Age	Mean	SD
	20.52	2.350

Table 6 reports descriptive statistics for all constructs included in the study, covering the individual dimensions of Gen Z characteristics, the composite Gen Z characteristics construct, purchase intention, and actual purchase behavior. Mean values across most constructs fall in the moderate range, pointing to a sample that is neither uniformly enthusiastic nor uniformly dismissive in its self-assessments. Among the seven dimensions of Gen Z characteristics, online review impact (M = 3.92, SD = 1.01) and product price impact (M = 3.85, SD = 0.93) return the highest mean scores, suggesting that respondents regard these as the most salient features of their generational profile. Influencer impact, by contrast, yields a noticeably lower mean (M = 2.68, SD = 1.14), indicating that direct influencer engagement is perceived as less central to this sample’s consumer identity than other digital characteristics. The remaining dimensions, online buzz impact, visual element impact, environmental responsibility impact, and brand community impact, cluster modestly above the scale midpoint, reflecting a tempered but generally positive recognition of these traits among respondents. When aggregated into a composite measure, the overall Gen Z characteristics construct returns a mean of 3.36 (SD = 0.66), consistent with the moderate pattern observed across the individual dimensions. The relatively constrained standard deviation at this level suggests that responses converge reasonably around this midpoint once the dimensions are combined. For the outcome variables, purchase intention (M = 3.00, SD = 0.89) sits precisely at the scale midpoint, while actual purchase behavior (M = 3.59, SD = 0.85) is somewhat higher, a gap worth noting as it hints at a degree of behavioral enactment that outpaces stated motivational commitment. Standard deviations across all variables are moderate, indicating adequate response variability and a sample that captures a reasonable range of orientations and behaviors. In sum, the descriptive statistics paint a picture of a sample with moderate but genuine engagement with the constructs under investigation.

Table 6. Constructs’ mean scores.

Variables	M	SD
Online review impact	3.92	1.008
Influencer impact	2.68	1.137
Online buzz impact	3.26	0.962
Visual element impact	3.40	0.992
Environmental responsibility impact	3.19	1.087
Product price impact	3.85	0.932
Brand community impact	3.21	1.152
Overall Gen Z characteristics	3.36	0.660
Purchase intention	3.00	0.888
Actual purchase	3.59	0.854

4.3. Relationship of Gen Z Characteristics with Purchase Intention and Actual Purchase

Table 7 reports the correlation coefficients between the dimensions of the Gen Z characteristics construct and the two primary outcome variables, purchase intention and actual purchase behavior. These bivariate associations offer a preliminary reading of the data before hypothesis testing and provide an initial basis for evaluating the plausibility of the proposed conceptual model. The general pattern across dimensions is one of positive and statistically significant associations with both outcome variables, suggesting that the generational traits captured in the construct are meaningfully linked to sustainable consumption patterns. With respect to purchase intention, significant positive correlations emerge for influencer impact, online buzz impact, visual element impact, environmental responsibility impact, and brand community impact, indicating that higher endorsement of these characteristics tends to accompany stronger intentions to purchase sustainably. The picture for actual purchase behavior is broadly similar, though not identical. Online review impact, online buzz impact, visual element impact, environmental responsibility impact, product price impact, and brand community impact all show significant positive relationships with actual purchasing. Visual element impact and online buzz impact stand out as among the stronger correlates, pointing to the possibility that visually rich and socially visible digital environments are particularly influential in activating sustainable purchasing among this cohort. Two exceptions are observed from the overall pattern. Influencer impact, despite its significant association with purchase intention, does not show a statistically significant relationship with actual purchase behavior. This suggests that while influencer-related content may contribute to shaping motivational states, it does not appear to translate directly into reported purchasing activity within this sample. Product price impact presents the inverse pattern: it is not significantly correlated with purchase intention but does show a significant association with actual behavior, hinting that price considerations bear more on the act of purchasing than on the formation of intentions to do so. On balance, the correlation analysis lends preliminary support to the relational structure proposed in the conceptual model. The breadth and consistency of significant associations across dimensions justify proceeding to regression and moderation analyses, which will allow for a more rigorous evaluation of the predictive and interaction effects specified in the study hypotheses.

Table 7. Relationship of Gen Z characteristics with purchase intention and actual purchase.

	Purchase Intention	Actual Purchase
Online review impact	0.125 *	0.328 **
Influencer impact	0.205 **	0.079
Online buzz impact	0.222 **	0.380 **
Visual element impact	0.195 **	0.435 **
Environmental responsibility impact	0.259 **	0.316 **
Product price impact	−0.010	0.276 **
Brand community impact	0.387 **	0.354 **

* Correlation is significant at the 0.05 level (2-tailed), ** Correlation is significant at the 0.01 level (2-tailed).

4.4. Prediction of Purchase Intention and Actual Purchase by Gen Z Characteristics

To assess the explanatory capacity of Gen Z characteristics across both motivational and behavioral outcomes, two separate multiple regression analyses were carried out. In each model, the same seven predictors were entered: online review impact, influencer impact, online buzz impact, visual element impact, environmental responsibility impact, product price impact, and brand community impact. The dependent variable differed across the two models, with purchase intention serving as the outcome in the first and actual purchase behavior in the second. Both models account for meaningful variance in their respective outcomes, though with some difference in magnitude. The regression predicting purchase intention yields R² = 0.323, meaning that the seven Gen Z characteristics dimensions collectively explain approximately 32.3% of the variance in respondents’ sustainable purchase intentions. The model predicting actual purchase behavior returns R² = 0.250, indicating that 25.0% of the variance in purchasing behavior is accounted for by the same set of predictors. While both figures reflect genuine explanatory power, the somewhat lower value for actual behavior is consistent with the theoretical expectation that behavior is shaped by a broader range of factors beyond generational characteristics alone. Adjusted R² values for both models confirm that their explanatory capacity holds up after accounting for the number of predictors included, lending further credibility to the findings (Table 8).

Table 8. Model summary for the prediction of purchase intention and actual purchase.

Dependent Variable	R	R2	Adjusted R2	SE
Purchase Intention	0.568	0.323	0.307	0.741
Actual Purchase	0.500	0.250	0.233	0.748

The ANOVA results reported in Table 9 confirm the statistical significance of both regression models. The model predicting purchase intention yields F = 19.982 (p < 0.001), while the model predicting actual purchase behavior returns F = 14.035 (p < 0.001). In both cases, the predictor set jointly accounts for statistically significant variation in the outcome, lending credibility to the overall model specifications. The notably higher F value for the purchase intention model suggests that the Gen Z characteristics dimensions, as a collective, achieve a stronger fit when predicting motivational outcomes than when predicting behavioral ones, a pattern consistent with the broader theoretical expectation that actual behavior is subject to a wider range of influences beyond those captured in the present model.

Table 9. ANOVA for the prediction of purchase intention and actual purchase.

Dependent Variable	SS Regression	df	MS Regression	F	p
Purchase Intention	68.214	7	9.745	19.982	<0.001
Actual Purchase	55.008	7	7.858	14.035	<0.001

A closer examination of the regression coefficients (Table 10) reveals that the predictors of purchase intention and actual purchase behavior are not identical in their composition or relative strength, a distinction that carries meaningful theoretical implications. Purchase intention is most strongly predicted by brand community impact (β = 0.281, p < 0.001), followed by environmental responsibility impact (β = 0.213, p = 0.001) and online buzz impact (β = 0.167, p = 0.012). This configuration points to a motivational landscape shaped largely by social belonging, personal values, and the perceived collective endorsement of sustainable behavior. Visual element impact (β = 0.142, p = 0.028) and influencer impact (β = 0.119, p = 0.041) also reach statistical significance, suggesting that aesthetic and aspirational cues contribute to the formation of sustainable intentions, albeit in a supporting rather than primary capacity. The predictors of actual purchase behavior tell a somewhat different story. Here, the leading contributors are visual element impact (β = 0.201, p = 0.002), online buzz impact (β = 0.153, p = 0.023), and online review impact (β = 0.148, p = 0.013), all of which reflect exposure to digitally visible, socially circulating, and informationally grounded stimuli. The shift away from value-based and community-oriented predictors toward more immediate digital cues suggests that the transition from intention to behavior among Gen Z consumers may be less a matter of conviction and more a matter of contextual activation through the digital environments they inhabit.

Table 10. Coefficient summary for the prediction of purchase intention and actual purchase.

Predictor	Purchase Intention (β)	p	Actual Purchase (β)	p
Online review impact	0.071	0.197	0.148	0.013
Influencer impact	0.119	0.041	−0.092	0.098
Online buzz impact	0.167	0.012	0.153	0.023
Visual element impact	0.142	0.028	0.201	0.002
Environmental responsibility impact	0.213	0.001	0.097	0.115
Product price impact	−0.024	0.668	0.029	0.598
Brand community impact	0.281	<0.001	0.110	0.075

Taken together, the regression findings point to a meaningful divergence in the factors that drive purchase intention and those that drive actual purchase behavior. Intention formation appears to be anchored in normative, value-driven, and identity-related processes since environmental responsibility and community belonging emerge as the dominant forces, suggesting that sustainable intentions among Gen Z are cultivated primarily through alignment with personal values and social group membership. Behavior, by contrast, is more responsive to contextual and stimulus-driven inputs, particularly visual communication, social visibility, and peer-generated information circulating in digital environments. This divergence refers directly to the complexity of the intention–behavior gap. It suggests that the psychological foundations of wanting to consume sustainably and actually doing so are not simply points on the same continuum but draw on partly distinct motivational and contextual resources. Gen Z consumers may arrive at sustainable intentions through a process that is fundamentally social and value-laden, yet the behavioral follow-through depends to a greater extent on the digital stimuli they encounter at or near the point of decision. The consumption environment, in other words, carries weight that stated values alone cannot fully account for. From a theoretical standpoint, these findings reinforce the case for treating intention and behavior as analytically separable outcomes rather than collapsing them into a single measure of sustainable engagement. They also suggest that effective interventions in this space will need to operate on two fronts: cultivating the normative and value-based conditions that support intention formation, while simultaneously designing the digital environments through which actual purchasing decisions are made.

4.5. Moderation Analysis for the Impact of Purchase Intention on the Relationship of Gen Z Characteristics and Actual Purchase

The moderation analysis conducted using PROCESS Model 1 yielded a statistically significant overall model, F₍₃₂₉₈₎ = 44.24, p < 0.001, with the predictor set accounting for 30.8% of the variance in actual purchase behavior (R² = 0.308), as reported in Table 11. Both primary predictors contributed significantly to the model. Gen Z characteristics emerged as a significant positive predictor of actual purchase behavior (b = 0.891, SE = 0.177, p < 0.001), and purchase intention likewise exerted a significant positive effect (b = 0.802, SE = 0.197, p < 0.001). Of particular theoretical interest is the interaction term. The product of Gen Z characteristics and purchase intention was negative and statistically significant (b = −0.145, SE = 0.055, p = 0.009, 95% CI [−0.252, −0.037]), accounting for an incremental 1.6% of explained variance beyond the main effects (ΔR² = 0.016). Although the incremental variance explained by the interaction is modest in absolute terms, this magnitude is consistent with typical effect sizes for interaction effects in social-science research, where ΔR² values in the 0.01–0.03 range are widely regarded as meaningful when the interaction is theoretically motivated and statistically robust. The substantive importance of the result therefore rests less on the absolute size of ΔR² than on the direction of the interaction, which reverses the conventional expectation that contextual factors amplify, rather than substitute for, motivational commitment. The negative sign of this coefficient indicates that the positive relationship between Gen Z characteristics and actual purchasing behavior weakens as purchase intention increases. In other words, purchase intention functions as a moderator, but not in the straightforward amplifying role that might be expected. Rather, it appears to attenuate the contribution of generational characteristics to behavioral outcomes, suggesting that when motivational commitment is already high, the additional activating influence of Gen Z traits becomes less consequential. It is worth recalling here that the model estimated above is a single moderation equation, with no mediator.

Table 11. Moderation of the relationship between Gen Z characteristics and actual purchase by purchase intention.

Predictor	B	SE	t	p	95% CI
Constant	−0.33	0.61	−0.54	0.593	[−1.52, 0.87]
Gen Z Characteristics (GZC)	0.89	0.18	5.02	<0.001	[0.54, 1.24]
Purchase Intention (PI)	0.80	0.20	4.08	<0.001	[0.42, 1.19]
GZC × PI	−0.15	0.05	−2.65	0.009	[−0.25, −0.04]
Model Summary
R = 0.56
R2 = 0.31
F(3298) = 44.24, p < 0.001
ΔR2 (interaction) = 0.02

Conditional effects analysis provided further insight into the nature of this moderating relationship (Table 12). The positive effect of Gen Z characteristics on actual purchase behavior proved strongest at low levels of purchase intention (b = 0.60, p < 0.001), diminished at moderate levels (b = 0.46, p < 0.001), and weakest when purchase intention was high (b = 0.35, p < 0.001). Notably, the effect remains statistically significant across all three levels, indicating that Gen Z characteristics consistently predict purchasing behavior regardless of motivational strength, but with meaningfully different magnitudes depending on how committed the consumer already is. The pattern carries a substantive interpretation. Among consumers with relatively low purchase intention, generational characteristics such as digital engagement, responsiveness to online influence, environmental orientation, and community involvement appear to play a particularly decisive role in activating actual purchasing. These traits seem to compensate for the absence of strong prior motivation, translating latent or partial commitment into behavioral action through digitally mediated and socially reinforced pathways. Where purchase intention is already high, by contrast, behavior is largely driven by that motivational commitment itself, and the incremental contribution of generational characteristics is accordingly reduced. The implication is not that generational traits become irrelevant at high levels of intention, but rather that their unique explanatory contribution narrows as motivational factors take on a more dominant role (Figure 2).

Table 12. Conditional effects of Gen Z characteristics on actual purchase.

Purchase Intention Level	Conditioning Value (Centered)	B	SE	t	p	95% CI
Low (−1 SD)	−0.89	0.60	0.09	6.99	<0.001	[0.43, 0.77]
Medium (Mean)	0.00	0.46	0.07	6.93	<0.001	[0.33, 0.59]
High (+1 SD)	0.89	0.35	0.08	4.33	<0.001	[0.19, 0.50]

Note. Conditioning values are mean-centered scores of purchase intention at −1 SD, mean, and +1 SD, consistent with PROCESS Model 1 [92]. Bootstrapped 95% CIs based on 5000 resamples.

Figure 2. Moderation of the relationship between Gen Z characteristics and actual purchase by purchase intention.

The practical implications of these findings should be read carefully. The moderation is statistically significant and points in a clear direction, but it adds only a small share to the variance explained, and purchase intention remains the main driver of sustainable purchasing in this sample. Two takeaways follow from this. The first concerns consumers who are not yet strongly committed, since it is in this group that digital and socially driven sustainability messaging seems to pay off most, because this is where the compensatory effect of Gen Z characteristics is at its strongest. The second concerns consumers whose intentions are already firm. For them, further effort to build motivation is unlikely to translate into much additional behavior, and resources are probably better spent on easing practical frictions such as price, availability, and convenience. We are not arguing that generational characteristics can replace intention as a general strategy. What the evidence supports is a complementary role, with the activating effect concentrated at the lower end of the motivational spectrum.

5. Discussion

This study set out to understand how Gen Z characteristics shape sustainable consumption, with particular attention to purchase intention, actual purchase behavior, and the conditions under which the intention–behavior gap might be narrowed. What emerges from the analysis is a layered picture in which different generational traits carry different weights depending on whether one is examining the formation of intentions or the execution of behavior. At the bivariate level, most dimensions of Gen Z characteristics show positive and significant associations with both purchase intention and actual purchasing, a pattern consistent with prior research linking digital engagement, social influence, and environmental concern to sustainable consumption among younger consumers [6,7,10,11,12,46,66,97,98]. Within the Theory of Planned Behavior, this makes conceptual sense since environmental responsibility, social influence, and digital exposure contribute to intention formation by shaping attitudes and perceived social norms, a dynamic that resonates with established findings in the sustainable consumption literature [6,30,31,75,76]. The regression analysis for purchase intention sharpens this picture considerably. Hypothesis 1 is supported since Gen Z characteristics significantly influence purchase intention, and the model accounts for a substantial share of its variance. Brand community impact, environmental responsibility, and online buzz emerge as the primary drivers, with visual elements and influencers playing a secondary but significant role. This configuration is broadly consistent with prior work grounded in value-based theories and social identity theory [14,45,51], which foregrounds moral alignment and group belonging as central motivators of pro-environmental intentions among younger consumers. However, not all findings align so neatly with prior literature. The non-significant effects of online reviews and product price on purchase intention are worth pausing over, given that perceived value and price sensitivity have featured prominently in previous studies of consumer intentions [9,14,15,37,55]. One interpretation is that Gen Z consumers, when forming intentions, are driven less by utilitarian calculations and more by questions of identity and social belonging. The non-significant effects of online reviews and product price on intention should not be over-interpreted, given the conservative threshold and the moderate sample-level variance in these items; they are not central to the study’s claims and are flagged here only to circumscribe scope. These findings are also consistent with prior work showing that, among younger consumers, identity-based and community-driven motivations are stronger antecedents of intention than purely utilitarian considerations [14,55,60]. The convergence with this earlier evidence strengthens confidence in the directionality of the effects reported here, while the multidimensional configuration we use clarifies which generational levers carry weight at which stage.

The picture for actual purchase behavior is more complicated, and Hypothesis 2 is partially supported. While the bivariate correlations suggest broad associations between Gen Z characteristics and behavior, the multivariate model narrows the field considerably since only online reviews, online buzz, and visual elements emerge as significant predictors of actual purchasing. Electronic word-of-mouth and digital visibility thus appear to be the operative mechanisms at the behavioral stage, consistent with prior work in this area [99,100]. Furthermore, environmental responsibility, which is among the strongest predictors of intention, does not exert a significant direct effect on behavior in the regression model. This suggests its influence may be largely indirect, channeled through purchase intention rather than acting on behavior independently, a finding that sits in some tension with studies positioning environmental values as direct behavioral drivers [56,63,77,78]. This divergence between the predictors of intention and behavior is one of the more theoretically significant results of the current study. Intention formation appears anchored in normative and value-driven processes, while behavioral execution is more responsive to immediate digital stimuli and contextual cues. The implication is that sustainable action among Gen Z is not simply an extension of sustainable belief but it depends on a distinct set of activating conditions encountered in the consumption environment. This pattern aligns with prior arguments that values and attitudes, while necessary, are insufficient on their own to guarantee behavioral follow-through [101].

A closer reading of hypotheses is warranted given the divergence in explanatory power across the outcomes. H1, which posits that Gen Z characteristics predict purchase intention, is strongly supported (R² = 0.323) and is consistent with prior work showing that digitally embedded value- and norm-based traits are reliable antecedents of pro-environmental intention [14,55,62,77,78,98]. H2, by contrast, states that Gen Z characteristics predict actual purchase behavior. It is partially supported, the model is statistically significant (R² = 0.250) but explains less variance than its intention counterpart. This pattern is theoretically expected. Within TPB, the intention–antecedents relationship is psychologically proximal, whereas the behavior–antecedents relationship is distal and subject to a wider range of unmeasured situational, structural, and habitual factors [3,4,32,33]. The lower R² for actual purchase therefore does not invalidate H2; it is consistent with the well-documented attenuation of explanatory power as one moves from intention to behavior in TPB-based models [3,32]. The moderation model (R² = 0.308) explains more variance than the direct-effects model for behavior precisely because it incorporates an additional, theoretically meaningful pathway: the interaction between Gen Z characteristics and purchase intention. The increase in R² reflects the contribution of the conditional, compensatory mechanism documented here, namely that generational traits activate behavior more strongly where intention is weak. In substantive terms, the divergence between R² for intention and R² for behavior should be read not as a weakness of the construct but as confirmation of the intention–behavior gap itself, and the moderation result clarifies one of the mechanisms by which the gap is narrowed.

The moderation analysis offers what is arguably the study’s most distinctive contribution and provides strong support for Hypothesis 3. The compensatory mechanism identified here is the principal theoretical contribution of the study, since it reframes the intention–behavior gap as a phenomenon that is narrowed by generational characteristics rather than uniformly mediated through intention. The negative and statistically significant interaction between Gen Z characteristics and purchase intention indicates that the relationship between generational traits and actual purchasing weakens as motivational commitment increases. This finding introduces meaningful nuance into the existing literature. It suggests that generational traits are not uniformly influential across the motivational spectrum but operate most powerfully as behavioral activators precisely where intention is underdeveloped. Where intention is already strong, behavior is predominantly guided by that motivational commitment, and the incremental contribution of generational characteristics diminishes accordingly. This positions generational traits as a form of compensatory mechanism, capable of bridging the gap between weak intention and observable behavior through socially reinforced and digitally mediated pathways. Theoretically, these findings engage with the existing literature in a productive tension. They confirm the central role of intention within a broadly TPB-consistent framework while also demonstrating that the intention–behavior relationship is not fixed but varies with generational context. The results similarly validate the Stimulus–Organism–Response framework’s contention that external stimuli can activate behavior through pathways that bypass or supplement the motivational route, a point especially relevant when internal commitment is limited. Taken together, the two frameworks offer complementary rather than competing explanations, and their integration appears productive for capturing the multi-stage nature of sustainable consumption.

The compensatory mechanism we observe speaks directly to the long-running debate on the intention–behavior gap. The dominant accounts of this gap, from Sheeran and Webb [3] to Carrington and colleagues [5], read it primarily as a translation problem, in which motivation is present but does not convert into action because of perceived control deficits, situational constraints, or competing priorities. The implicit policy lever, on this reading, is to strengthen intention itself, on the assumption that stronger intention will translate into more behavior. Our findings push back against this assumption in a theoretically meaningful way. They suggest that, for Gen Z, the strongest contribution of generational characteristics to behavior shows up precisely where intention is weakest. This sits well with a broader line of work in consumer behavior on contextual compensation, in which situational stimuli substitute for absent motivation rather than amplify present motivation [33,34]. Read this way, the intention–behavior gap is not only a translation problem but also, in part, a substitution opportunity, since the absence of strong intention does not foreclose pro-environmental behavior when the surrounding generational and digital context is activating enough. That is the contribution this study aims to add to the existing theoretical literature on the gap, by positioning Gen Z characteristics not as antecedents competing with intention but as conditional compensators that come into play where intention is most fragile.

The study also responds to a recognized gap in the literature. Prior research has disproportionately relied on intention as a proxy for sustainable engagement and has tended to examine individual predictors in isolation [24,26,43]. By treating Gen Z characteristics as a multidimensional construct and examining intention, behavior, and their interaction simultaneously, the present study offers a more complete account of how generational traits shape sustainable consumption across its different stages. Practically, the findings argue for a more differentiated approach to sustainability-oriented marketing. Social and value-driven messaging appears well-suited to cultivating intentions, particularly among consumers for whom community belonging and environmental identity are already salient. For consumers who have not yet formed strong purchase intentions, however, the activation of behavior may depend more on the quality and visibility of the digital environments they inhabit (visually compelling content, socially circulating endorsements, and accessible peer-generated information). The finding that influencer marketing correlates with intention but not with actual behavior is a particularly pointed observation for practitioners who assume a direct line from influencer exposure to sales. Overall, sustainable consumption among Gen Z is shaped by a combination of motivational, social, and contextual factors that operate differently at different stages of the consumption process. Some findings reinforce existing assumptions while others complicate them. What the study ultimately underscores is the value of moving beyond single-stage, single-predictor models toward a more dynamic and contextually grounded understanding of how sustainable behavior is formed, activated, and ultimately enacted.

6. Conclusions

This study examined the role of Gen Z characteristics in shaping sustainable consumption, with purchase intention and actual purchase behavior treated as distinct yet related outcomes, and the moderating function of intention in bridging the gap between the two as a central analytical concern. The findings establish Gen Z characteristics as a meaningful and multidimensional construct whose influence on sustainable consumption operates through both motivational and contextual channels. Purchase intention remains a primary determinant of behavior, yet generational traits, particularly digital engagement, responsiveness to online stimuli, and social influence, make a complementary contribution that is most pronounced precisely where motivational commitment is weakest. The analysis further reveals that not all dimensions of Gen Z characteristics carry equal explanatory weight. Digitally and visually oriented factors, specifically online reviews, online buzz, and visual elements, emerge as the most consistent predictors of actual purchasing behavior, while the moderation results underscore that these generational traits are particularly consequential in activating behavior among consumers who have not yet formed strong intentions. In this way, the study offers a concrete account of one mechanism through which the intention–behavior gap may be narrowed. More broadly, it contributes to the literature by incorporating generational characteristics into sustainable consumption models and by advancing a more granular understanding of the conditions under which intention converts into action.

The results of this study modify the dominant reading of Gen Z sustainable consumption in three concrete ways. First, they shift the analytical center of gravity away from intention as a near-sufficient proxy for behavior, and toward a model in which generational characteristics act in parallel with, rather than only through, intention. Second, they reframe the intention–behavior gap as a conditional phenomenon: generational traits compensate for weak intention rather than uniformly amplify strong intention, which means that the gap is widest precisely where most prior interventions have been targeted, namely highly motivated consumers. Third, they reveal an asymmetry between the predictors of intention (value, community, and norm-driven) and those of actual behavior (digital, visual, and word-of-mouth driven), implying that sustainable consumption among Gen Z is shaped by two partially distinct ecosystems rather than a single attitudinal pipeline. The practical implication is that sustainability strategies that perform well at the intention-formation stage are not the same strategies that translate intention into purchase, and the segments that most benefit from each are different. Recognizing this dual structure is what the present study contributes to the understanding of the dynamics of this consumer environment.

6.1. Limitations

Two methodological features of the present design merit emphasis up front, because they bound the inferential reach of the analysis throughout. The first is that all variables, including actual purchase behavior, were captured through self-report on a single questionnaire. Self-reported behavior is known to be sensitive to social desirability and recall effects, both of which tend to push pro-environmental responses upward and can inflate the strength of the relationships we observe. The second is that the design is cross-sectional. Even where the moderation pattern is consistent with the compensatory mechanism we propose, the data cannot establish temporal precedence or rule out reverse and reciprocal influences. The associations reported in the results should therefore be read as patterns consistent with the theoretical model rather than as evidence of established causal mechanisms. Longitudinal data and behavioral or transactional measures of purchase would be needed to test the model under more demanding conditions. The remainder of this section sets out the more specific limitations of the present study.

First, the sample consists exclusively of Greek undergraduate students aged 18 or older. While this group falls squarely within the Gen Z bracket, undergraduates are unrepresentative of the wider cohort in important ways: they are more highly educated than the population average, their financial autonomy is constrained, and their digital media diet differs from that of non-student Gen Z consumers. The findings should therefore be read as applicable to this sub-population, and care should be taken before extending them to working, non-student, or younger Gen Z consumers. Second, actual purchase behavior was measured by self-report, which is known to be vulnerable to social desirability bias and recall error, both of which tend to inflate pro-environmental responses; behavioral or transactional data would provide more conservative estimates. Third, the cross-sectional design rules out causal inference since while the moderation pattern is consistent with the proposed compensatory mechanism, longitudinal or experimental designs are needed to establish temporal precedence. Fourth, the sampling procedure, although structured by a fixed positional pattern within an enrollment list, does not constitute systematic probability sampling, and generalizability is therefore restricted. Fifth, the environmental responsibility and influencers sub-scales returned Cronbach’s alpha values below the 0.70 benchmark, and although McDonald’s omega and CR estimates are closer to the threshold, this remains a measurement limitation that warrants expanded item pools in future work. Despite the supporting evidence from omega, CR, AVE, and the Spearman–Brown projection, the alpha values for these two sub-scales remain a real measurement limitation. Findings that hinge specifically on these dimensions should be read with that uncertainty in mind, and any implications drawn from them, whether theoretical or practical, are tentative pending replication with more reliable instruments. Methodologically, the analysis here relied on EFA together with regression-based moderation rather than on a full CFA or PLS-SEM. Because the Gen Z characteristics construct is theory-driven and second-order, a confirmatory measurement model or a fully specified structural equation model would have been the more rigorous route for validating the higher-order configuration, including a formal test of the higher-order factor and of measurement invariance across subsamples. The EFA-plus-regression route fits the moderation focus of our hypotheses and the exploratory integration of dimensions previously studied separately, but it does not carry the same psychometric guarantees as CFA, and future replications should put the construct through a CFA or PLS-SEM framework. Finally, the Gen Z characteristics construct, while broader than prior operationalizations, does not exhaust the relevant generational influences; cultural context, psychological dispositions, and structural market conditions were not modeled. Taken together, these limitations narrow the scope of the conclusions in specific and identifiable ways. The student sample restricts external validity to a relatively homogeneous slice of the Gen Z cohort. The self-report measure of behavior raises the floor on social desirability and recall bias. The cross-sectional design rules out direct causal inference, even though the moderation pattern is consistent with the proposed compensatory mechanism. We do not regard any of these limitations as undermining the central finding, namely that Gen Z characteristics activate sustainable purchasing most strongly where intention is weakest, but they do indicate that the result should be read as a starting point for further work rather than as a definitive account. Replication with non-student samples, with behavioral or transactional data, and with longitudinal or experimental designs is the natural next step.

6.2. Future Research

Future research could address several of these limitations productively. To begin with, probability-based sampling across more diverse Gen Z populations, including those outside university settings, would substantially strengthen the generalizability of findings in this area. Building on this, longitudinal designs would allow researchers to trace how the relationships between generational characteristics, purchase intention, and actual behavior shift over time, providing a more dynamic picture of sustainable consumption. In parallel, behavioral or transactional data could complement or replace self-reported measures, thereby reducing reliance on respondents’ retrospective accounts of their own purchasing. Beyond methodological refinements, there is also room to broaden the construct space by incorporating cultural, psychological, and structural factors that may shape how generational characteristics translate into consumption outcomes. Moreover, cross-generational comparative studies would help establish whether the patterns identified here are specific to Gen Z or reflect more general consumer dynamics, an important step toward building cumulative theoretical knowledge in this field. In addition, future replications should expand the item pools for the environmental responsibility and influencers sub-scales and apply Confirmatory Factor Analysis to obtain a more rigorous estimate of their measurement properties.

Beyond these broader directions, several research avenues follow directly from the results reported here. First, the divergence between intention- and behavior-driving dimensions invites systematic replication across categories of sustainable products (e.g., fashion vs. food vs. consumer electronics vs. mobility services), since the relative weight of value-driven versus stimulus-driven mechanisms is likely to vary by category and by the visibility of consumption. Second, and relatedly, the compensatory role of Gen Z characteristics at low levels of intention should be tested across digital platforms with different affordances (short-form video, peer-review communities, immersive AR/VR commerce, AI-generated content), as the activating power of digital stimuli plausibly depends on platform format. Furthermore, future studies could test the present moderation model against alternative moderators, such as perceived behavioral control, environmental self-identity, or price consciousness, in order to determine whether intention is uniquely compensable or whether the compensatory dynamic generalizes more broadly. In a similar vein, multi-country comparative designs would clarify whether the patterns observed in the Greek context replicate under different economic and regulatory conditions for sustainable products. Additionally, integrating behavioral or transactional data with experimental manipulations of digital stimuli would allow the causal direction of the compensatory effect to be tested directly. Finally, a complementary methodological extension would be to specify a moderated-mediation model in which purchase intention simultaneously mediates the Gen Z characteristics—behavior relationship and is itself moderated by relevant contextual variables. Such a specification could be tested with PROCESS Model 7 or Model 14, or within a full PLS-SEM framework, and would offer a more comprehensive test of how the motivational and stimulus-driven pathways interact.

6.3. Practical and Managerial Implications

The findings carry several concrete implications for practitioners and policymakers working to promote sustainable consumption among Gen Z consumers. The prominence of digitally and visually oriented factors in the analysis points to a clear strategic priority where organizations should invest in visually engaging, socially shareable content when communicating sustainability messages. Campaigns that draw on online reviews, user-generated content, and social visibility are likely to resonate more effectively with this cohort precisely because they align with the digital environments Gen Z consumers already inhabit and trust. The results also complicate some widely held assumptions about influencer marketing. While influencer-related content appears to contribute to intention formation, its relationship to actual purchasing behavior is weaker than practitioners might expect. This suggests that influencer endorsements, deployed in isolation, may not be sufficient to drive behavioral outcomes. A more productive approach likely involves building credibility through transparent communication, consistent sustainability practices, and genuine consumer engagement over time. Authenticity, in other words, appears to be a more reliable behavioral driver than visibility alone. The moderating role of purchase intention adds a further layer of strategic nuance. Consumers at lower levels of motivational commitment represent a segment where activating mechanisms, social proof, digital engagement, and visual storytelling can make a genuine difference in stimulating purchasing behavior. For consumers whose intention is already strong, however, the priority shifts. Continued efforts to reinforce attitudes that are already favorable may yield diminishing returns; what these consumers more likely need is the removal of practical obstacles such as price, limited availability, and inconvenience. For policymakers, the implications follow a similar logic. Interventions that combine informational content with digitally native engagement strategies are likely to outperform those relying on traditional communication channels, particularly when delivered through the platforms and formats that align with how Gen Z actually consumes media. Across both managerial and policy contexts, the overarching message is the same: promoting sustainable consumption among this generation requires a differentiated, context-sensitive approach that accounts for where consumers are in their motivational journey, rather than treating them as a homogeneous group.