Children’s Environmental Communicative Agency for Sustainability: A Structural Equation Model Bridging the Knowledge–Action Gap

Abstract

This study investigates the psychological and pedagogical architecture predicting environmental communicative agency among 304 primary school students (grades 5–6) participating in a climate change education programme. Aiming to bridge the persistent “knowledge–action gap” in sustainability education, the research identifies the cognitive, emotional, and instrumental pathways that transform children into active agents of low-carbon, pro-sustainability change in their everyday lives. Employing Structural Equation Modeling (SEM) with a latent climatic knowledge literacy construct, the analysis reveals that the latent construct of General Environmental Knowledge, comprising Conceptual Climate Knowledge and Relational-Systems Climate Knowledge, is the strongest direct predictor of Environmental Communicative Agency. Intrinsic curiosity emerged as a dominant driver of practical competence, while future-oriented tools function as the critical mediator between understanding and social action. Together, these mechanisms outline a school-based pathway through which climate literacy and motivation can be converted into household-level behavioural change and intergenerational climate resilience. The findings advocate for a paradigmatic shift from knowledge transfer to building operative sustainability agency, offering a pedagogical roadmap that fosters “constructive hope” and positions children as “trusted messengers” who catalyse intergenerational learning and reverse socialization toward more sustainable lifestyles.

1. Introduction: From Knowledge Transfer to Agency Building

The deepening climate crisis of the twenty-first century poses not merely a scientific or technological challenge, but above all an unprecedented educational and psychological one. While the educational challenge requires dismantling outdated linear knowledge-transfer frameworks to build systemic literacy, the psychological challenge addresses the profound cognitive disengagement, eco-anxiety, and behavioral barriers that paralyze action in the face of ecological distress. For decades, public education systems have operated under the traditional foundational assumption known as the Knowledge–Attitude–Behavior (KAB) model, according to which the transmission of scientific knowledge leads to attitudinal change, which in turn matures into pro-environmental behavioral change [1]. Although this linear paradigm, often termed the ‘information deficit model’, has been widely discredited in environmental psychology theory, it stubbornly persists as the default architectural blueprint for public education frameworks globally. Recent literature underscores that educational systems continue to operate under the bureaucratic illusion that quantifying and maximizing climate literacy scores will organically catalyze social action. This structural reliance on knowledge accumulation overlooks the deep socio-emotional and institutional barriers that prevent cognitive awareness from translating into systemic behavioral shifts, thereby perpetuating the very ‘knowledge–action gap’ this study seeks to dismantle [2,3]. A substantial body of empirical research, however, points to the persistence of the knowledge-action gap: factual knowledge about global warming and positive dispositions toward the natural world constitute neither a guarantee nor a sufficient condition for the adoption of sustainable lifestyles or for active environmental engagement [4,5,6]. This gap is not merely didactic in nature; it is simultaneously psychological, structural, and normative, and must therefore be addressed across multiple levels concurrently. Psychological barriers, such as conflicting life goals and interpersonal relationship dynamics, have been shown to moderate and often weaken the link between environmental concern and actual action [7]. Moreover, large-scale cross-national studies demonstrate that while climate anxiety can drive pro-environmental behaviors in many contexts, structural and economic barriers often prevent individuals from translating distress into action, highlighting the importance of the socio-political environment in enabling agency [8].

The present study emerges from a recognition of these limitations and from the imperative to bridge the divide between cognitive awareness and practical action. Its starting premise is that in a world saturated with “threatening” information about ecological collapse, learners may develop learned helplessness or disengagement unless they are equipped with concrete tools for action [9]. Accordingly, the central concern of this study is not what children know, but rather how they experience change. The pivotal construct is students’ perceived self-efficacy in promoting behavioral change. Drawing on Bandura’s [10,11] self-efficacy theory, the study contends that the most powerful source of such efficacy beliefs is the mastery experience: a practical, tangible encounter with genuine success in managing a real-world challenge. their own capacity to effect change. Evidence suggests that subjective knowledge is a powerful driver of value-action consistency, as it enhances self-confidence in decision-making processes [12].

To this end, a purpose-built curriculum for primary school students was examined, one grounded in a ten-stage pedagogical sequence integrating creative processing, systems thinking, and engagement with role models, structured around the head–hands–heart framework. The core component of the programme is the experiential learning that unfolds beyond the classroom: a personal challenge in which each student adopts a sustainable behavior, and a family challenge in which they mobilize household members toward the adoption of environmentally responsible habits. For children at this developmental stage, “leading behavioral change” assumes the distinctive form of communicative agency, the capacity to articulate a perspective and persuade within the asymmetrical power structures of family life. The programme thus aspires to transform environmental education from the transmission of information about problems to the cultivation of genuine competence to address them, while simultaneously generating mechanisms of intergenerational learning and reverse socialization [4,13]. In the power-laden intergenerational space, where structural authority typically resides with adults, the most meaningful resource available to the child is precisely this capacity to voice, persuade, and mediate ecological reality to their immediate social environment [13,14].

Through an examination of the interfaces among different levels of knowledge (conceptual, relational, and systems-level), intrinsic motivation (curiosity and challenge), and future-oriented competencies, this study seeks to advance an integrative educational model that positions tools, practical experience, and structured reflection as the essential connective link between understanding and action, thereby repositioning environmental education from an informational encounter with problems to a competence-building encounter with their resolution.

The methodological choice to foreground variables of knowledge, intrinsic motivation, and Action-Oriented Future Competence, rather than the conventionally privileged variable of attitudes, rests on the well-established need to bridge the value-action gap in environmental education research. Classical and contemporary scholarship alike, most notably Kollmuss and Agyeman [1], demonstrates that positive environmental attitudes frequently function as a “normative” variable with insufficient predictive power for actual behavioral adoption, particularly among children who tend to express elevated pro-environmental attitudes as a function of social desirability rather than genuine behavioral intent. Rather than examining what students “feel” toward nature, the present study therefore foregrounds the construction of action competence, grounded in the integration of cognitive understanding (knowledge), internal drive (motivation), and practical capacity to enact change (tools).

A further theoretical rationale for the selection of these variables resides in Bandura’s self-efficacy theory, which holds that belief in one’s capacity to execute a specific task is a far stronger behavioral predictor than a generalized attitude. Students’ perceived future-oriented tools and communicative self-efficacy represent operationally meaningful variables that index readiness to act as agents of change within family and community contexts. The theoretical model developed here posits that knowledge and motivation constitute the “engine,” while Action-Oriented Future Competence provides the mechanism through which this engine drives behavioral change; in the absence of instrumental resources, positive attitudes risk remaining at the level of declarative endorsement, never translating into genuine agency in the face of the climate crisis.

Finally, centering knowledge and practical competence over attitudes enables a more precise assessment of the programme’s pedagogical effectiveness. System-level knowledge and the perceived relevance of action tools are direct products of the pedagogical sequence enacted in the classroom. In employing structural equation modeling, this study seeks to examine the functional pathway through which tools serve as the connective link, thereby proposing an educational model that moves beyond the reinforcement of existing dispositions and instead equips learners with both the competence and the moral authority to lead processes of reverse socialization and intergenerational environmental learning. This focus aligns with strategic priorities in environmental education research, which call for more studies examining intergenerational learning and educational visions that support collective action [15,16].

Building on this measurement logic, the explicit theoretical innovation of this study lies in shifting from descriptive frameworks of the value-action gap to empirical, structural path validation. While prior environmental psychology scholarship has conceptualized concepts like constructive hope or reverse socialization in isolation, this research offers a theoretical breakthrough by operationalizing and statistically testing an integrative model that maps the non-linear pathways between cognitive literacy, intrinsic motivation, and instrumental competence. By explicitly modeling future-oriented tools as a missing structural mediator, this study provides a predictive pedagogical architecture that explains exactly how and under what conditions subjective literacy transforms a child into an active, communicative agent within the family ecosystem, thereby directly addressing current calls in the literature for empirically grounded models of intergenerational environmental learning.

Indeed, Aleksi et al. [16] and Lawson [15] identify intergenerational learning as a key underexplored research priority in the value-action gap literature, noting that most existing studies focus on individual behaviour change rather than on the relational dynamics through which children catalyse household-level transformation. The present model responds directly to this gap by specifying and testing the pathways through which children’s communicative agency and future-oriented tools function within the family context.

2. Theoretical Framework: Building Climate Agency Through Practical Experience, Emotion, and Reverse Socialization

Cross-national evidence from 32 countries confirms that climate anxiety does not consistently translate into environmental action: whereas negative climate emotions predict pro-environmental behaviour in 32 countries, structural and economic barriers prevent this translation in others, underscoring that emotional engagement without actionable tools may heighten distress without generating agency [8]. While these systemic limitations of the linear framework are heavily documented, modern environmental education faces the urgent challenge of architecting operational alternatives. Bridging this divide requires moving beyond a deficit-based focus on what children know to theoretically mapping how they emotionally process crisis information and instrumentally enact change. By shifting the pedagogical focus toward action competence, this framework synthesizes the emotional dimension of climate education, the psychological mechanics of self-efficacy, and the relational dynamics of youth agency within institutional and social systems.

To bridge this gap, this study proposes a shift from focusing on the child’s passive knowledge acquisition to analyzing how children actively experience change through action and emotion. Crucially, navigating the ethical boundaries of this shift requires recognizing that positioning children as change agents within the family unit is not an imposition of systemic adult burdens on minors, but rather a fulfillment of their normative rights to participation and expression, as codified in the UN Convention on the Rights of the Child (UNCRC). From an ethical standpoint, environmental education must balance the cultivation of children’s agency with psychological safety, ensuring that youth are not weaponized against parental authority. By framing this intergenerational dynamic through a relational ethics of care and dialogue, rather than coercion, the child acts not as an intrusive corrector of family habits, but as a co-constructor of shared household resilience. This rights-based framing justifies the child’s agency right here and now, honoring their relational commitment to the living world while respecting the integrity of the family ecosystem [13,17,18] (Chawla, 2022; Furu et al., 2026; Trott, 2022). This is where the work of Maria Ojala [19,20] comes into play, emphasizing the emotional dimension of climate change education.

Ojala argues that exposure to knowledge about the climate crisis without providing coping tools can lead to eco-anxiety or denial. Importantly, research on the directionality of the hope-action relationship suggests that the causal arrow runs in both directions: not only does hope motivate action, but collective engagement and tangible experiences of efficacy serve as generative sources of constructive hope in their own right [21]. She proposes the concept of “Constructive Hope”, a hope not based on naive optimism, but on a sense of Agency and the ability to influence. In the current study, practical experience in adopting personal and family behaviors serves as a tool for producing that constructive hope; it allows the child to process their concern through tangible action, thereby turning emotion into a driver for change rather than a barrier. A three-wave longitudinal study of Finnish adolescents corroborates this claim empirically, demonstrating that climate change distress alone predicts decreases in pro-environmental behaviour over time; however, when meaning-focused coping is present, the same level of distress is associated with significant increases in environmental action [22]. Empirical evidence from ecological restoration programs confirms this dynamic: young participants who engaged in hands-on habitat work demonstrated significant increases in self-efficacy and reported agency, with the social dimension of working alongside adult experts functioning as a critical catalyst [23].

The ability to turn emotion into action rests on building a sense of self-efficacy, as defined by Albert Bandura [10]. According to Bandura, the strongest source of efficacy is “Mastery Experiences.” This study contends that the tasks of dealing with behavioral change at the personal and family levels are, in effect, mastery experiences that shape the child’s identity. Louise Chawla [24], a pioneer in research on “Significant Life Experiences,” emphasizes that adult environmental identity is built from a sequence of positive and active childhood experiences, combined with support from adult figures. Chawla argues that when a child experiences success in changing their immediate reality (such as within the family unit), they develop a “memory of power” that accompanies them as an influential citizen throughout their life.

The dynamics in which the child acts within their family lead us to the concept of “Agency” in the context of Reverse Socialization. Studies highlight that children are not merely passive recipients of education, but “Trusted Messengers” capable of driving change in their parents [13]. Trott explains that when a child receives “tools for action” and implements them at home, they challenge traditional knowledge hierarchies and generate reciprocal intergenerational learning. This process of reverse socialization is particularly critical regarding climate change, where adults sometimes experience difficulty changing old habits, and the child, driven by authentic emotion and efficacy acquired at school, serves as a catalyst for family behavioral change [4]. This reorientation towards active agency rather than passive exposure aligns with a broader theoretical shift in environmental psychology, which distinguishes between children as passive recipients of nature-based benefits and as active agents who exercise capabilities, satisfy fundamental needs for autonomy and competence, and build relational commitments to the living world [17].

The integration of these approaches allows for the understanding of the “tools for action” provided in this study not as a mere school exercise, but as an act of “Active Environmental Citizenship.” According to Dobson [25], such citizenship requires involvement in the public and social sphere. When a student is required to adopt family behavior, they are essentially performing a deconstruction of power relations at home and building “communicative-environmental efficacy.” This ability to express an opinion, persuade, and act together with adults is the bridge pointed to by Kollmuss and Agyeman; it is the bridge that transforms scientific knowledge into social and moral practice. Recent studies in EER confirm that a pedagogy combining the “Head” (knowledge), the “Heart” (hope and emotion according to [19]), and the “Hands” (mastery experience [10,24] according to Bandura and Chawla) is the only way to produce a deep and stable change in the reality of the climate crisis [26,27].

In conclusion, the proposed theoretical framework views practical experience in personal and family behavioral change as the key to breaking “the gap.” Through the lens of Bandura and Chawla, we see how experience builds efficacy and identity; through Ojala’s lens, we understand how action processes emotion into constructive hope; and through the lens of Trott and Kollmuss & Agyeman, we see how Action-Oriented Future Competence transforms the child into a change agent in the family and community systems. This integration emphasizes that environmental education no longer settles for teaching facts but grants students the authority and ability to reshape their world.

3. Research Question and Hypotheses

The central research question examines which factors, across cognitive dimensions (systems-level and relational knowledge), emotional-motivational dimensions (curiosity and challenge), and practical-instrumental dimensions (future-oriented tools), predict the environmental communicative agency of fifth- and sixth-grade students in promoting behavioral change in relation to climate change. Drawing on the empirical literature, the research hypotheses were formulated within an integrative theoretical framework (Figure 1). First (H₁), it is hypothesized that higher levels of climatic knowledge, operationalised through students’ relational-systems climate knowledge and conceptual climate knowledge, will be positively and directly associated with their perceived environmental communicative agency; a nuanced understanding of the interconnected dynamics of the climate crisis constitutes a foundational prerequisite for civic agency. Second (H₂ and H₃), the study anticipates that intrinsic motivation grounded in curiosity and challenge will predict higher levels of communicative agency toward parents and the broader community, in accordance with the model of constructive hope: intrinsic epistemic curiosity (H₂) and academic challenge/resilience (H₃) are expected to show positive direct associations with environmental communicative agency. Third (H_4a–H_4c), higher levels of climatic knowledge (H_4a), intrinsic epistemic curiosity (H_4b), and academic challenge/resilience (H_4c) are expected to be positively associated with students’ perceived action-oriented future competence. Finally (H₅), the core hypothesis posits that students’ perceived future-oriented tools will function as a critical mediating variable: knowledge and motivational engagement will translate into heightened efficacy beliefs and environmental communicative agency primarily when students perceive that the programme has equipped them with practical, applicable tools for everyday life.

4. Materials and Methods

4.1. Research Approach

This study is grounded in a quantitative-explanatory paradigm, designed to examine and validate a theoretical model of climate agency development among children. This developmental path aligns with frameworks that view actionable knowledge as a coevolutionary process between science and society, prioritizing the building of trust and shared problem definition over linear knowledge transfer [16]. The research design follows a correlational-predictive framework, mapping the relational pathways between Climatic Knowledge and motivation on the one hand, and manifestations of active agency on the other. This design enables an assessment of the proposed model’s fit to the psychological and pedagogical realities of the student participants, with particular attention to the predictive power of cognitive and emotional variables on students’ perceived practical competence and communicative agency within social and family contexts.

A central methodological feature of this study is the deliberate decision to shift the research focus away from the conventional variable of attitudes toward operationally meaningful variables of competence and agency. This decision is grounded in the scientific imperative to bridge the knowledge-action gap and in the recognition that general pro-environmental attitudes frequently function as a normative variable shaped by social desirability pressures, one that offers limited predictive validity for actual behavioral adoption among children. In its place, the study foregrounds the construction of action competence and the sources of self-efficacy as theorized by Bandura, premised on the understanding that belief in one’s capacity to perform specific communicative and practical tasks is a substantially stronger predictor of lifestyle change than a declarative, dispositional stance.

Finally, the research approach conceptualizes the child as an active agent of change who exercises communicative agency within family and community contexts. The study examines not only internal cognitive transformation, but the potential for reverse socialization, a process through which the child serves as a mediator of knowledge and action-oriented future Competence for their parents [4,13,14]. This framing aligns with current directions in environmental education research, emphasizing transformative and intergenerational learning [15,28]. Recent scholarship underscores that moving beyond individual behavioral outcomes requires exploring the relational dynamics through which children act as catalyst agents within their household ecosystems, thereby transforming the family into a vital space for climate resilience [15]. Accordingly, both the measurement instruments and the statistical analyses were designed to identify the pathways that predict students’ confidence in leading change, thereby providing an empirical foundation for a pedagogical model that integrates head, heart, and hands as a coherent response to the climate crisis.

4.2. Participants

A total of 304 primary school students in fifth and sixth grade (aged 10–12 years) completed the questionnaire upon finishing the pedagogical programme. The sample comprised 124 fifth-grade students, of whom 62 were boys (50%) and 62 were girls (50%), and 180 sixth-grade students, of whom 108 were boys (60%) and 72 were girls (40%). Across the full sample, 170 participants were boys (55.9%) and 134 were girls (44.1%). All participants engaged in the complete instructional sequence, centred on the development of climate agency and future-oriented practical competence.

4.3. Research Instrument, Validation, and Reliability

The study employed a structured, anonymous, online questionnaire administered via Google Forms (Appendix A). The questionnaire was adapted from a previously validated instrument [29], with modifications made to fit the specific pedagogical context and developmental objectives of the present study. The baseline structural architecture, construct validity, and factor configurations of this measurement scale were formally established during its original development phase via Exploratory Factor Analysis (EFA), which demonstrated high sampling adequacy and clean internal factorization [29]. To confirm that the adapted instrument maintains its psychometric robustness and internal stability within the specific primary school demographic of the current sample (N = 304), reliability was evaluated utilizing Cronbach’s alpha coefficient (α) for each sub-scale. As detailed above, the internal consistency coefficients ranged from 0.78 to 0.90, substantially exceeding the conventional exploratory threshold of 0.70, validating the data’s readiness for subsequent path modeling.

The final instrument consisted of two parts. Part A collected demographic information, including the municipality in which the participant’s school was located, gender, and grade level. Part B, Section A, Environmental Communicative Agency (7 items, α = 0.84) assessed students’ perceived Environmental Communicative Agency, namely their perceived self-efficacy in expressing climate-related opinions and promoting behavioral change regarding climate change mitigation. Responses were recorded on a five-point Likert-type frequency scale ranging from “almost never” to “almost always.” A sample item was: “I can express my opinions about climate change freely at school.” Section B (9 items, α = 0.88) assessed Conceptual Climate Knowledge through students’ self-appraised ability to explain key climate-related concepts. Responses were recorded on a four-point scale ranging from “not at all” to “to a great extent.” A sample item was: “To what extent can you explain the term atmosphere to classmates?” Section C (8 items, α = 0.89) assessed Relational-Systems Climate Knowledge, namely the ability to identify and explain relationships among climate-related concepts and phenomena. The same four-point scale was used. A sample item was: “To what extent can you explain to classmates the connection between veganism and climate-change phenomena?” Section D (7 items, α = 0.90) assessed students’ attitudes regarding climate change and the perceived importance of climate-related learning topics, using a four-point importance scale ranging from “not important at all” to “very important.” A sample item was: “How important is it to learn about consumer culture at school?” Section E (5 items, α = 0.86) assessed Intrinsic Epistemic Curiosity regarding climate change learning, while Section F (5 items, α = 0.78) assessed Academic Challenge/Resilience in climate change studies. Both sections used five-point frequency scales ranging from “almost never” to “almost always.” Sample items included: “I find climate change interesting” and “I think climate studies are too difficult.” Section G (3 items, α = 0.86) assessed Action-Oriented Future Competence, focusing on students’ perceptions regarding the practical relevance of climate change education for future life. Responses were recorded on a five-point agreement scale ranging from “strongly disagree” to “strongly agree.” A sample item was: “Learning about climate change equips children with practical tools for everyday life.” Negatively worded items were reverse-scored prior to analysis where appropriate. The questionnaire was adapted from a previously validated instrument [29], with modifications made to fit the context and objectives of the present study.

4.4. Research Variables

The definition of variables in this study was constructed in a manner that shifts the emphasis from the normative “Attitudes” variable toward operative variables expressing action competence and active agency. The primary dependent variable, based on the first chapter of the questionnaire, is defined as the Environmental Communicative Agency. The items composing this variable examine the student’s confidence to express their opinion, explain content, and lead peers and parents toward behavioral change, thereby reflecting the child’s ability to act as a change agent in the social sphere. It is important to note that the construct of Environmental Communicative Agency, as operationalized in this study, is theoretically grounded in Bandura’s (1982) [10] concept of domain-specific self-efficacy rather than general environmental attitudes. Bandura explicitly argues that efficacy beliefs must be measured at the level of specific behaviors within specific domains. For elementary school children, the most meaningful and developmentally appropriate domain of environmental efficacy is not abstract ‘influencing policy’ but the immediate social ecology of the family unit, the arena in which the child exercises genuine agency. Communicative acts such as explaining climate connections to parents or persuading family members to adopt sustainable behaviors constitute, in Bandura’s terms, performance accomplishments within the child’s proximal zone of influence [11,13,30]. This reframing positions communicative agency not as a proxy for efficacy, but as its most ecologically valid expression for this age group. This focus on fifth and sixth graders is significant as this age group enters a critical developmental window for interest in global issues, facilitated by their burgeoning capacity for abstract thinking and systemic reasoning [31].

The cognitive predictor variables were split into two levels of depth: Conceptual Climate Knowledge, which examines the ability to explain fundamental terms, and Relational-Systems Climate Knowledge, which examines the ability to link human activity, social injustice, and consumer culture with climate phenomena. The emotional-motivational dimension was measured through two variables reflecting transformative learning: Intrinsic Epistemic Curiosity, measuring the level of interest and excitement in learning, and Academic Challenge/Resilience, which examines the perception of learning as a challenging experience that consolidates the ability to cope with difficulties.

Finally, the mediating variable in the model, based on the final chapter of the instrument, is defined as Action-Oriented Future Competence. This variable measures the interface between learning and daily life, as well as the students’ perception of the Action-Oriented Future Competence provided to them for dealing with future challenges. This variable structure enables an examination of the statistical path in which literacy and motivation transform into tangible agency through the mediation of Action-Oriented Future Competence, while avoiding the social desirability biases often associated with the measurement of general attitudes.

4.5. Data Analysis Strategy and Path Analysis (SEM)

The analytical strategy adopted in this study is best characterized as ‘exploratory-confirmatory’: a hybrid design in which theoretical propositions derived from the literature on self-efficacy and climate agency guided the initial structural model, while the exploratory factor analysis served to optimize the operationalization of constructs prior to model testing. Following established practice in educational research, this approach is methodologically transparent in acknowledging that the hypotheses were refined in light of initial analyses, and claims are accordingly framed as theoretical propositions warranting further confirmatory testing in independent samples with pre-registered designs. The model presented here should therefore be understood as an empirically grounded, theory-generating contribution rather than a definitive causal test, a distinction consistently maintained throughout the Section 6.

To mitigate the inherent risk of Common Method Bias (CMB) typical of single-source, cross-sectional data, several procedural design remedies were systematically applied. First, the survey introduction guaranteed absolute participant anonymity to reduce evaluation apprehension and diminish social desirability bias among the children. Second, the questionnaire architecture intentionally varied its configurations and endpoints across sections, structurally alternating between distinct 4-point and 5-point Likert scales anchoring frequency, importance, and agreement.

A central methodological feature of the model is the specification of ‘Climatic Knowledge’ as a latent construct, derived from two highly correlated yet conceptually distinct indicators: conceptual and relational-systems climate knowledge. By separating the measurement model (the latent construct) from the structural model (the pathways), the analysis controls for measurement error, resolves multicollinearity between the knowledge variables, and provides a more parsimonious and theoretically grounded estimation of the pathways leading to communicative agency.

4.6. The Instrument

The measurement instrument was designed to faithfully reflect the “Head, Heart, and Hands” structure presented in the theoretical framework. Section A of the questionnaire examines the perception of efficacy to lead change, focusing on items measuring communicative agency (influence on parents, friends, and community). Sections B and C measure the conceptual and relational-systems climate knowledge components, while Section D examines students’ attitudes toward the importance of the topics. Sections E and F complete the picture by measuring curiosity, challenge, and the perception of the “tools for the future” provided to the students as part of the pedagogical sequence. The link between the questionnaire and the theory is established by identifying these Action-Oriented Future Competences as the bridge allowing cognitive awareness to transform into action confidence in the social sphere.

Statistical analysis was conducted using Structural Equation Modeling (SEM) due to the methodological necessity of examining a complex system of simultaneous relationships rather than simple linear connections. The choice of SEM stemmed from the model’s ability to examine the mediation processes of “Action-Oriented Future Competence” as the connecting link between knowledge and efficacy.

5. Results

This chapter presents the results of the empirical analysis conducted among 304 fifth- and sixth-grade students who participated in the pedagogical programme. The analysis aims to examine the psychological and educational mechanisms predicting the development of the Environmental Communicative Agency in children. The findings are reported in two stages. First, descriptive statistics and Pearson intercorrelations among the study variables are presented, providing a preliminary empirical basis for understanding the relationships between Conceptual and Relational-Systems Climate Knowledge, Intrinsic Epistemic Curiosity, and Action-Oriented Future Competence (

Table 1. Descriptive Statistics and Pearson Intercorrelations Among Study Variables (N = 304).

Variable	M	SD	1	2	3	4	5	6
Environmental Communicative Agency	2.86	0.86	—
Conceptual Climate Knowledge	2.61	0.79	0.592 ***	—
Relational-Systems Climate Knowledge	2.37	0.84	0.113 *	0.167 **	—
Intrinsic Epistemic Curiosity	2.47	0.94	0.637 ***	0.630 ***	0.170 **	—
Academic Challenge/Resilience	2.25	0.92	−0.214 ***	−0.324 ***	−0.063	−0.329 ***	—
Action-Oriented Future Competence	3.27	0.84	−0.010	0.087	0.566 ***	0.052	−0.016	—

* p < 0.05 ** p < 0.01 *** p < 0.001.

). Second, a structural equation model (SEM) is reported, including an evaluation of model fit indices and an examination of path coefficients (Figure 1). This analysis enables the disentanglement of direct from indirect effects and serves to validate the theoretical model, which positions Action-Oriented Future Competence as a mediating mechanism between cognitive climate knowledge and children’s capacity to initiate environmental communicative action within their social environment.

Table 1 presents the means, standard deviations, and Pearson intercorrelations among the six observed variables (N = 304). Note that General Environmental Knowledge is operationalized as a latent variable comprising two observed indicators, Intrinsic Epistemic Curiosity and Academic Challenge/Resilience, and is therefore not represented as a separate variable in the correlation matrix; its constituent indicators are reported individually.

Environmental Communicative Agency demonstrated significant positive associations with Intrinsic Epistemic Curiosity (r = 0.637, p < 0.001), Conceptual Climate Knowledge (r = 0.592, p < 0.001), Action-Oriented Future Competence (r = 0.373, p < 0.001), and Relational-Systems Climate Knowledge (r = 0.113, p < 0.05). In contrast, Academic Challenge/Resilience exhibited a significant negative association with Environmental Communicative Agency (r = −0.214, p < 0.001) and with Intrinsic Epistemic Curiosity (r = −0.329, p < 0.001), yet a positive association with Conceptual Climate Knowledge (r = 0.565, p < 0.001). This pattern suggests that perceiving learning as challenging may support the accumulation of declarative climate knowledge while simultaneously suppressing the intrinsic motivational orientation necessary for the development of environmental communicative agency. Action-Oriented Future Competence showed no significant direct association with Environmental Communicative Agency (r = −0.010, p = ns), yet correlated strongly with Relational-Systems Climate Knowledge (r = 0.566, p < 0.001), consistent with the hypothesized mediated pathway in the structural model.

These bivariate associations provide preliminary support for the theorized structural pathways. The strongest observed correlation, between Intrinsic Epistemic Curiosity and Environmental Communicative Agency (r = 0.637), corroborates the theoretical claim that intrinsic curiosity functions as the primary emotional-motivational engine driving children’s sense of environmental agency. The absence of a direct association between Action-Oriented Future Competence and agency, alongside its robust correlation with Relational-Systems Climate Knowledge, supports the hypothesis that future competence tools operate through systemic knowledge as a mediating mechanism rather than as an independent predictor. It is important to acknowledge, however, that these correlations represent unconditional bivariate relationships; the structural equation model presented in the following section disentangles direct from indirect effects and provides a more rigorous test of the proposed theoretical architecture.

Prior to evaluating the structural model, preliminary analyses were conducted to examine intergroup differences across the research variables. Independent samples t-tests revealed significant developmental differences between grade levels (grade 5 vs. grade 6). Sixth-grade students demonstrated significantly higher levels of Environmental Communicative Agency, Conceptual Climate Knowledge, and Relational-Systems Knowledge (all p < 0.001). In contrast, fifth-grade students reported a significantly higher perception of Academic Challenge (t₍₃₀₂₎ = 4.41, p < 0.001), underscoring the cognitive complexity of climate topics at this earlier stage. Notably, no significant differences were found in perceived Action-Oriented Future Competence between grades, suggesting that the pedagogical tools provided were perceived as equally relevant and applicable across the target age range.

Conversely, no statistically significant differences were observed based on gender for any of the measured variables (p > 0.05 for all scales). To further explore these dynamics, a two-way ANOVA was performed with grade and gender as independent factors. The results confirmed a strong main effect for grade level but yielded no significant main effect for gender and no significant interaction between grade and gender. The absence of interaction effects indicates that the developmental trajectories of Intrinsic Epistemic Curiosity and Agency are consistent for both boys and girls. Collectively, these findings provide a robust methodological justification for utilizing a pooled sample within the Structural Equation Model (SEM) and validate the instrument’s sensitivity to expected developmental shifts in climate literacy and agency.

5.1. Model Fit Assessment

Figure 2 represents the SEM model. To evaluate the degree of correspondence between the proposed structural model and the empirical data, a comprehensive set of fit indices was examined in accordance with prevailing statistical conventions. The findings indicate an excellent fit across all indices assessed. The Chi-square value was 4.854 with 3 degrees of freedom and a significance level of p = 0.183. As this p-value exceeds the 0.05 threshold, no statistically significant discrepancy was detected between the model-implied and observed covariance matrices, a finding consistent with a high degree of model-data correspondence. The Chi-square-to-degrees-of-freedom ratio was 1.618, substantially below the widely accepted threshold of 3.0, indicating a parsimonious and well-fitting model.

Examination of the absolute and incremental fit indices further substantiates the structural adequacy of the model. The Goodness of Fit Index (GFI) yielded a value of 0.995, and the Adjusted Goodness of Fit Index (AGFI) was 0.964. The Root Mean Square Residual (RMR) presented a notably low value of 0.007, indicative of minimal residual discrepancy. The comparative fit indices were equally strong: the Comparative Fit Index (CFI) was 0.998, the Tucker–Lewis Index (TLI) was 0.991, and the Normed Fit Index (NFI) was 0.995. These values substantially exceed the conventional minimum threshold of 0.95, confirming that the model faithfully represents the theoretical structure of climate agency development.

Finally, indices of approximation error and stability were examined. The Root Mean Square Error of Approximation (RMSEA) was 0.045, indicating a low approximation error and a high probability of close model fit in the population. Additionally, the Akaike Information Criterion (AIC) for the research model (40.854) was lower than that of the saturated model (42.000), attesting to the methodological superiority of the proposed model in terms of parsimony and precision. The HOELTER (0.05) index yielded a value of 498, well above the conventional threshold of 200, confirming that the sample size is fully adequate for the estimation of this SEM.

5.2. Path Coefficients

Following confirmation of adequate model fit, the standardized path coefficients (β) were examined to assess the strength of the predictive relationships between the independent variables and students’ environmental communicative agency. The latent variable “Climatic Knowledge”, comprising Conceptual Climate Knowledge (0.89) and relational-systems climate knowledge (0.92), emerged as the strongest, most direct, and most statistically significant predictor of environmental communicative agency (β = 0.43, p < 0.001). This finding corroborates the first research hypothesis (H₁), according to which a deep cognitive understanding of the climate crisis constitutes a necessary foundation for the development of active agency among children.

With respect to the emotional-motivational dimension, intrinsic epistemic curiosity was found to be a positive and significant predictor of communicative agency (β = 0.23, p < 0.001), a finding that supports Hypothesis 2 (H₂). By contrast, the variable “academic challenge and resilience” did not emerge as a significant predictor of communicative agency (β = 0.50, p > 0.05), and Hypothesis 3 (H₃) was therefore not supported in this sample. This null result warrants critical reflection. One plausible interpretation, consistent with Ojala’s [21] distinction between cognitive hope and emotional coping, is that academic challenge as a construct measures effort-investment and perseverance rather than meaning-making. In the existential context of the climate crisis, effort without a sense of meaning or forward-looking competence may amplify rather than reduce disengagement. This finding resonates with cross-national evidence demonstrating that climate distress without coping tools heightens passivity rather than action [8,32], suggesting that pedagogical rigour divorced from emotional relevance fails to generate the agentive disposition required for environmental communicative action.

A central component of the model analysis concerned the examination of Hypothesis 4 (H₄) regarding the mediating role of the variable “action-oriented future competence.” The results indicate that Climatic Knowledge (as a latent variable) significantly predicts the level of practical competence (β = 0.23, p < 0.001). Intrinsic epistemic curiosity emerged as the strongest predictor of practical competence (β = 0.58, p < 0.001), while academic challenge and resilience were associated with this variable in a negative and non-significant direction (β = −0.19, p > 0.05). Action-oriented future competence, in turn, was found to significantly predict environmental communicative agency (β = 0.16, p < 0.01). These findings confirm the existence of an indirect pathway through which knowledge and curiosity predict students’ agency via the mediating role of perceived Action-Oriented Future Competence and the perceived relevance of learning to everyday life.

Empirical support for the integrated model (β = 0.16 ** for the mediation pathway) corroborates Hypothesis 5 (H₅) and indicates that the integrative structure of the “pedagogical value chain” accounts for variance in communicative agency more effectively than any single dimension in isolation. The validated model suggests that the translation of cognitive knowledge into genuine agency among students depends critically on the mediation of Action-Oriented Future Competence that confers a sense of operative competence in confronting the challenges of climate change. Taken together, these findings indicate that systems-level literacy and intrinsic curiosity translate into social influence only when students perceive learning as equipping them with actionable, real-world competencies.

6. Discussion

The central objective of this study was to decode the psychological and pedagogical architecture that characterizes the development of environmental communicative agency in primary school children, specifically addressing the persistent “knowledge-action gap.” The structural equation model confirms that agency emerges as a systemic property, where the transition from literacy to action is a non-linear process mediated by emotional and instrumental factors. By validating an integrative framework that connects cognitive, affective, and practical dimensions, these findings demonstrate how children transform from passive learners into “Trusted Messengers” capable of leading reverse socialization within their family units. Recent evidence from primary education settings confirms that while direct environmental or disaster experiences can significantly enhance factual knowledge among fifth and sixth graders, they do not necessarily translate into more positive environmental attitudes, further reinforcing the theoretical necessity of mediated pathways to build genuine agency [33]. This repositioning of the child from a ‘passive patient’ receiving information to an ‘active agent’ exercising autonomy and building relational commitments is central to a transactional approach to climate resilience [17].

The cognitive foundation of this model, Climatic Knowledge, functions as a primary driver of agency, illustrating how systemic and relational literacy serves as the “memory of power” theorized by Louise Chawla [24]. Environmental identity, as Chawla [23] recently posited, is not built on abstract data but on the “epistemic authority” a child gains when they grasp the complexities of ecological interdependencies. In line with contemporary scholarship [6,34], my results suggest that this deep literacy provides children with the moral authority required to mediate climate reality to adults. This systemic anchor allows children to feel sufficiently empowered to voice their perspectives within the typically asymmetrical power structures of the family, transforming climate information from a source of stress into a tool for social positioning.

The emotional engine of the model is articulated through the dominant role of intrinsic epistemic curiosity, representing an empirical validation of ‘constructive hope’ [19,35]. In this framework, curiosity functions as a resilient, meaning-focused coping strategy that prevents disengagement by transforming potential eco-anxiety into a driver for active engagement [21,22]. This suggests that in the existential context of climate change, interest-driven engagement and cognitive flexibility, top-down processes that allow individuals to override unsustainable habits, are far more critical for agency than conventional academic rigor or effortful persistence [2]. As noted in recent scholarship [5,36], such curiosity-driven pedagogies are essential for maintaining children’s engagement without succumbing to hopelessness. Furthermore, hope serves as a vital relational prerequisite for transmitting action competence, rooted in an ethics of care for children’s futures [18]. From a developmental perspective, this age group (grades 5–6) occupies a critical window for building climate agency, as emotional investment at this stage serves as a foundation for a lifelong civic identity [31].

Integrating practical agency into climate education serves as a critical psychological buffer against the rising levels of eco-anxiety and insecurity documented among primary school students [19]. However, anchoring practical agency as an emotional stabilizer introduces a subtle pedagogical risk: the potential to inadvertently foster an ideological bias that decouples emotional relief from actual ecological mitigation [37]. If green micro-habits within the household are celebrated primarily for their therapeutic value as psychological buffers, education risks institutionalizing a form of moral licensing. In this scenario, families may succumb to behavioral rebound effects, offsetting localized sustainable actions with carbon-intensive compensations elsewhere under the ideological illusion of compliance. To prevent constructive hope from devolving into structural complacency, educational practice must ensure that action competence is consistently tethered to critical systems reasoning, balancing emotional safety with a transparent, scientifically rigorous assessment of material environmental impacts [38].

In this regard, the family challenge component of the current curriculum functions as a form of real-world ‘role-playing,’ a pedagogical strategy that has been shown to reduce epistemological doubt and help learners navigate the inherent uncertainty of climate decision-making. By providing a structured arena for communicative action, the program allows children to process negative emotions, such as the fear or sadness often associated with direct disaster experience, and transform them into a sense of collective empowerment rather than withdrawal. This repositioning suggests that agency is not merely a social outcome but an essential cognitive mechanism for maintaining psychological resilience and fostering constructive hope in the face of ecological crises [33,39]. This is particularly critical as longitudinal data suggests that high-frequency pro-environmental behavior without proper coping strategies can sometimes increase emotional burden; however, meaning-focused coping effectively moderates this relationship, allowing distress to fuel action without undermining well-being [22].

The pivotal functional link in this architecture is established by Albert Bandura’s [10] theory of self-efficacy, specifically through the mechanism of “mastery experiences.” The analysis confirms that knowledge and curiosity translate into agency primarily through the perceived relevance of future-oriented tools. The personal and family challenges integrated into the curriculum served as proximal mastery experiences, allowing children to move from declarative knowledge to operative competence, essentially shifting from “what they know” to “how they can act”. By framing agency as communicative, the study aligns with Bandura’s concept of “proxy agency,” where children lead behavioral change by influencing those with structural authority, their parents. As Trott [13] emphasizes in EER, this reverse socialization allows children to experience collective efficacy, which is a key predictor of lifelong environmental commitment.

It is worth noting, however, that the predictive strength of systemic knowledge (β = 0.43) over Action-Oriented Future Competence alone (β = 0.16) challenges simplistic ‘hands-on first’ pedagogical frameworks. Scholarship warns [1,2] that action-oriented programmes without accompanying conceptual grounding risk producing what they term ‘efficacy without literacy’, a sense of empowerment decoupled from systemic understanding. The present findings empirically substantiate this caution: tools function as mediators, not substitutes, for deep knowledge. Their predictive value emerges only in the presence of prior cognitive scaffolding, suggesting that environmental education programmes that emphasise practical tasks at the expense of conceptual rigour may inadvertently perpetuate the very gap they seek to bridge.

Building upon these mechanisms, the theoretical breakthrough achieved by this study addresses a persistent gap in environmental education literature by moving beyond legacy linear frameworks. Specifically, this theoretical contribution is threefold: First, it introduces Environmental Communicative Agency as an empirically validated construct distinct from general self-efficacy, locating children’s agentive capacity specifically within the asymmetrical power structures of intergenerational family interaction. Second, it establishes intrinsic epistemic curiosity, rather than academic challenge or attitudinal concern, as the emotional engine of climate agency in primary-aged children, resolving a persistent ambiguity in the hope-engagement literature [20]. Third, and most importantly, it empirically validates future-oriented tools as a mediating mechanism, demonstrating that the knowledge-to-agency pathway is neither direct nor attitudinal, but instrumental: it requires a perceived bridge between systemic understanding and the everyday social world of the child.

Consequently, this integrated model demonstrates that dismantling the knowledge-action gap [1] requires a pedagogical synergy between cognitive literacy, Ojala’s constructive hope, and Bandura’s mastery experiences. The study repositions environmental education from an informational encounter with global crises to a competence-building encounter with their local resolution. By equipping children with both the systemic understanding and the Action-Oriented Future Competence to navigate their social environments, the program empowers them to function as active agents of change, transforming the school into a vital hub for intergenerational climate resilience.

7. Conclusions

The current study provides empirical validation for the need for a paradigmatic shift in climate change education, moving from models focused on knowledge transfer and attitude change to the construction of operative agency. The central conclusion arising from the validation of the SEM model is that dismantling the “Knowledge-Action Gap” does not necessarily require the addition of more scientific content, but rather the integration of components that grant the student practical competence and applied tools for daily life. The findings suggest that when students perceive learning as relevant to their future and as equipping them with tangible action capabilities, their confidence to speak out and lead change in their social circles is significantly higher. In doing so, the study offers a pedagogical roadmap that realizes the ideal of “Active Environmental Citizenship” starting in primary school.

At the pedagogical level, the findings indicate that fostering intrinsic epistemic curiosity is the strongest predictor for building students’ sense of practical competence. The applied significance for curriculum designers is the need to design learning environments that arouse personal and emotional interest, emphasizing “constructive hope” rather than messages based on ecological anxiety or threats. The model teaches that enjoyment and curiosity in learning are not merely “add-ons” but a critical mechanism that allows students to process complex information and transform it into an active identity as change agents. A pedagogy focused on challenge-based learning and the provision of Action-Oriented Future Competence enables children to experience authentic mastery experiences, which form the psychological foundation for long-term behavioral change.

Regarding educational policy, the study highlights the importance of the child as a legitimate and influential partner in the process of “Reverse Socialization.” The educational system must recognize that the communicative agency a student develops in the classroom is intended to cross the boundaries of the school institution toward the family and the community. A key policy implication is the need to integrate tasks that actively involve parents and families, as these allow the child to fulfill their role as a “trusted messenger” who mediates environmental knowledge and tools to adults. This approach expands the circles of influence of climate education and transforms the school into a central hub of intergenerational learning, which is essential for addressing the climate crisis in the 21st century.

To operationalize these findings into daily teaching practice, environmental education must move beyond declarative content toward specific, high-leverage practical interventions. First, schools should deploy structured “Household Sustainability Audits,” wherein students utilize digital carbon-footprint toolkits to evaluate real-world household practices alongside their parents, directly strengthening the validated instrumental mediation pathway. Second, pedagogies should integrate localized “Climate Simulation Role-Playing Games,” allowing children to safely practice negotiating resource-allocation dilemmas, which builds communicative self-efficacy while mitigating eco-anxiety through active systems-level reasoning. Finally, educators should scaffold intergenerational learning by establishing community-based “Eco-Storytelling Forums,” where students present empirical observations of local environmental shifts to families and community elders. These targeted interventions ensure that scientific literacy is immediately paired with concrete action-oriented competencies, transforming primary students from passive consumers of crisis data into operational change agents within their immediate social ecology.

8. Research Limitations

Despite its contributions, several methodological limitations define the scope of this study:

First, the cross-sectional design without a control group restricts internal validity. It cannot entirely rule out confounding variables like maturation or concurrent external events. Consequently, the SEM pathways indicate associative-predictive relationships rather than definitive causal determinations regarding program effectiveness.

Second, data collection relied exclusively on student self-reports, introducing potential social desirability bias common among primary school children. Furthermore, the dependent variable measures perceived communicative agency rather than actual, field-observed behavioral changes. Future research should integrate behavioral observations or parental reports to validate these findings.

Third, an unmeasured risk inherent to youth-led family behavioral change is the potential for behavioral rebound effects (or moral licensing). While the program catalyzed immediate agency, isolated green habits can inadvertently trigger compensatory, carbon-intensive actions elsewhere in the household ecosystem if families feel they have already fulfilled their ecological obligations. Future longitudinal research is required to track net household footprint dynamics over time.

Finally, sample characteristics limit generalizability. Focusing solely on grades 5–6 restricts applicability to younger or older age cohorts. Geographically and culturally, the sample reflects a specific educational setting with a relatively high socio-economic and educational parental profile. Families with greater cultural capital and baseline climate awareness may be inherently more receptive to intergenerational learning. In lower socio-economic or highly traditional cultural contexts, where rigid hierarchical hierarchies govern parent-child relations or economic survival overshadows ecological concerns, the pathway from children’s knowledge to household behavioral transformation may face higher resistance. Thus, the validated structural model serves as a proof-of-concept for supportive family ecosystems, and expanding research to more diverse demographic strata is essential.