Addressing the awareness-behavior gap in SDG 12: a participatory arts-based approach to food waste in Indonesia

1 Introduction

Global sustainability is increasingly defined by the urgent need to reconcile human consumption patterns with the finite ecological boundaries of the planet. At the heart of this challenge lies Sustainable Development Goal 12: Responsible Consumption and Production (SDG 12), which remains one of the most elusive objectives of the “2030 Agenda for Sustainable Development” (United Nations General Assembly, 2015). Despite international policy efforts, global material consumption continues to accelerate. Between 2015 and 2022, global domestic material consumption rose by 23.3%, reaching a record 14.2 metric tons per capita (United Nations Department of Economic and Social Affairs, 2025). This acceleration is particularly acute in developing regions, where rapid infrastructure growth has pushed material footprints upward. In Indonesia, this challenge is localized in the critical issue of food waste, which constitutes 41.7% of the total waste stream (Mahmud et al., 2026). This trajectory underscores that achieving responsible consumption demands more than the passive acquisition of knowledge and requires a fundamental behavioral transformation that shifts individual habits away from a “buy-use-dispose” culture toward sustainable resource management.

A critical barrier to this transformation is the awareness-behavior gap, which Kollmuss and Agyeman (2002) define as the persistent discrepancy between a person's environmental knowledge or awareness and their actual display of pro-environmental behavior. Despite global awareness regarding food security, over one billion meals are wasted daily, contributing to approximately 10% of global greenhouse gas emissions (United Nations Environment Programme, 2024). Research shows that while cognitive knowledge serves as a necessary precursor to environmental awareness, it remains insufficient for behavioral change unless mediated by affective engagement, such as empathy or a sense of personal responsibility (Hines et al., 1987; Kals et al., 1999; Carmi et al., 2015). However, addressing this gap at a systemic level requires looking beyond individual cognition and affective states to understand how everyday environments shape action. Building on this, social practice theory (SPT) posits that this gap persists because behaviors such as food waste are deeply ingrained communal routines rather than isolated cognitive choices, necessitating interventions that disrupt established social norms (Shove et al., 2012; Keegan and Breadsell, 2021; Ozanne et al., 2022). Traditional top-down pedagogical models, characterized by rote learning and product-oriented outcomes, have proven insufficient to bridge this gap (Kollmuss and Agyeman, 2002; Stanišić and Maksić, 2014; Redman and Redman, 2014; UNESCO, 2014). Consequently, UNESCO (2017) emphasizes that for education to foster “agents of change,” it must move toward action-oriented, transformative pedagogies that are learner-centered and empower students to become active participants in their own learning process.

In response, participatory arts-based approaches within Education for Sustainable Development (ESD) have gained increasing attention as a means of supporting affective engagement and meaning-making alongside cognitive learning. By engaging learners through sensory, interpretive, and creative processes, the arts can help translate abstract sustainability concepts into culturally meaningful experiences that are situated within everyday contexts (Publicover et al., 2018; Papavasileiou et al., 2020). This study introduces a participatory arts-based learning framework designed to address the gap between cognitive awareness and behavioral change. This approach is grounded in the premise that arts provide the necessary emotional and affective engagement to drive behavioral change (Publicover et al., 2018), while participatory methods ensure student ownership and social learning (Barth, 2014). Rather than positioning students as passive recipients of sustainability knowledge, participatory arts-based learning emphasizes student-led inquiry and collective meaning-making, shifting education from a top-down instructional model toward a more collaborative learning process.

However, as affective engagement is inherently shaped by cultural and moral values, the efficacy of such frameworks may vary significantly across different social contexts. While arts-based ESD is gaining traction globally, extant literature remains heavily skewed toward Western secular contexts (Edwards et al., 2020), leaving limited evidence on how these models can be adapted to local cultural frameworks, such as waste management practices in Islamic educational settings. In the Indonesian context, environmental stewardship among Muslim youth activists has been shown to operate within a distinct moral and ethical landscape in which ecological responsibility is deeply intertwined with religious orthopraxy rather than secular environmental rationales (Nilan, 2021).

Against this backdrop, the present study examined a pilot intervention conducted at SMA Budi Cendekia Islamic Senior High School (BCI) in Depok, Indonesia. The program represented a contextual adaptation of a participatory arts-based educational model originally developed and validated in South Korea (Shin and Park, 2026). While the foundational South Korean model provided a successful non-Western approach to ESD, it operated within a secular paradigm. Adapting the model to address the local challenge of food waste in Indonesia through student-led filmmaking represented a rare instance of intra-Asian pedagogical transfer, moving beyond the secular frameworks that dominate Western ESD literature. By assessing how participatory arts-based learning, implemented through student-led film productions, was associated with changes in SDG awareness and observed food waste behaviors among secondary school students in an Islamic educational context, this study investigated the efficacy of student-led filmmaking as a catalyst for cognitive and behavioral changes in a non-Western context. Specifically, this pilot intervention was guided by the following research questions:

• How can a participatory arts-based ESD framework be pedagogically and culturally adapted to address localized sustainability challenges in a resource-constrained Indonesian setting?
• To what extent does participation in this mentor-led filmmaking framework influence students' aggregate understanding and awareness of SDG 12?
• What observable changes in cafeteria food waste practices occur following the implementation of the participatory arts-based intervention?

The structure of this article is as follows: Section 2 provides a review of the existing literature and establishes the study's conceptual framework, synthesizing transformative learning, social learning, and social practice theories with the specific affordances of arts-based ESD in an Islamic context. Section 3 details the pilot intervention, outlining its contextual adaptation and mentor-led facilitation structure. Section 4 outlines the methodological approach including data collection, analysis, and ethical considerations, while Section 5 presents the results, detailing the quantitative increases in SDG awareness and the observed 13-percentage point reduction in food waste. Finally, Section 6 offers a discussion of the findings, exploring the implications of this model for scalable, culturally responsive sustainability education in the Global South, before concluding with recommendations for future research in Section 7.

2 Literature review

2.1 Conceptualizing the awareness-behavior gap: a social practice perspective

The observed disparity between environmental awareness and actual pro-environmental behavior, commonly referred to as the awareness-behavior gap, is a fundamental challenge in sustainable consumption (Franco and Ghisetti, 2022; Gould et al., 2023). While various theoretical lenses attempt to explain this phenomenon–ranging from cognitive behavioral models that emphasize individual knowledge deficits (e.g., Hines et al., 1987), to Schwartz's theory of basic human values (Schwartz, 1992), and systems-of-provision approaches that emphasize the broader infrastructures and institutional arrangements shaping consumption (Fine and Leopold, 1993)–these frameworks often attribute the gap to individual deficits in knowledge or personal motivation. However, addressing the awareness-behavior gap requires recognizing that cognitive knowledge alone is often insufficient to break automated, everyday habits. Because actions like food consumption and disposal are rarely calculated individual choices, it is necessary to examine the broader, collective routines that dictate daily life. To bridge this conceptual divide, this study adopts social practice theory (SPT) to conceptualize the awareness-behavior gap as a byproduct of entrenched communal practices rather than a failure of individual choices. This focus is appropriate because the present study examines school-level practices through which sustainability knowledge, institutional routines, peer norms, and cafeteria environments interact in students' everyday food consumption and waste behaviors.

Instead of framing food consumption and waste as calculated individual decisions, SPT conceptualizes them as deeply ingrained routines, which are actively structured by social norms, institutional rules, and material environments (Keegan and Breadsell, 2021; Ozanne et al., 2022). When applied to an educational setting, these material infrastructures take the form of cafeteria supply structures. From this perspective, the awareness-behavior gap emerges because cognitive awareness alone (i.e., knowing food waste is environmentally harmful) is insufficient to disrupt automatic, habitual routines (Ozanne et al., 2022).

To bridge this gap, intervention models must move beyond didactic information delivery. This is where participatory arts-based education for sustainable development (ESD) becomes critical. Arts-based ESD acts as a disruptive mechanism that brings unconscious routines into a conscious, reflexive space (Bentz and O'Brien, 2019; Papavasileiou et al., 2020). By prioritizing active engagement over passive information reception, these participatory models allow students to physically and socially enact new behaviors, shifting the reduction of food waste from an abstract global concept into a localized, shared practice.

2.2 Participatory learning in education for sustainable development (ESD)

The transition toward participatory learning represents a fundamental shift in Education for Sustainable Development (ESD), moving away from didactic knowledge-transmission toward transformative, learner-centered approaches. UNESCO (2017) emphasizes that fostering “agents of change” requires a departure from traditional “teacher-as-expert” hierarchies toward educational models based on collaboration, dialogue, and co-construction. This shift reflects the growing recognition that sustainability challenges are complex and context-dependent, necessitating educational approaches that foster agency and critical engagement rather than passive knowledge acquisition (Redman and Redman, 2014; Papavasileiou et al., 2020).

Grounded in transformative learning theory, this participatory paradigm enables learners to actively construct meaning through experiential processes that prompt them to re-evaluate existing beliefs and values. Instead of engaging with sustainability as an abstract or externally imposed topic, learners participate in dialogue, collaborative problem-solving, and action-oriented projects that empower students to take ownership and responsibility (Bentz and O'Brien, 2019; UNESCO, 2017). By shifting students from passive recipients to active investigators, these methods aim to bridge the awareness-behavior gap by prioritizing action-oriented competencies alongside cognitive understanding.

Social learning perspectives, including (Bandura 1977) account of observational learning and Lave and Wenger's (1991) concept of “Communities of Practice,” help explain how behavioral patterns are modeled and reinforced by peers with a shared social environment. However, applying participatory approaches within educational contexts characterized by high power distance may require navigating traditional hierarchical pedagogical structures (Hofstede, 1986). In such environments, schooling structures shaped by transmissive traditions can inadvertently constrain the dialogic, student-centered, and co-constructive pedagogies essential for transformative ESD (Guerra et al., 2024). To effectively foster environmental agency, the learning environment may need to mitigate these hierarchical barriers. Recent work on near-peer mentoring models suggests that shifting from senior, hierarchical instruction toward near-peer or mentor-facilitated structures can reduce psychological distance, offering emotional support and relatable experiences that may be less accessible in traditional authority-based relationships (Al-Thani et al., 2023; Persohn et al., 2024). By cultivating a more egalitarian dynamic, students are afforded the psychological safety required to take interpersonal risks and express creative ideas during the problem-solving process (Reiter-Palmon and Millier, 2023). Ultimately, bridging hierarchical divides (Guerra et al., 2024; Reiter-Palmon and Millier, 2023) is crucial for enhancing collective efficacy, which plays a central role in motivating joint action against complex social and ecological crises (Hamann et al., 2024).

2.3 The role of arts in behavioral change

Parallel to participatory learning, the arts have been increasingly recognized as a vital medium for translating abstract sustainability targets into tangible, culturally relevant, and personally meaningful experiences (Bentz and O'Brien, 2019; Papavasileiou et al., 2020). Artistic practices, such as the creation of artworks and ecological storytelling, allow students to deconstruct and visualize environmental issues, making the problem feel personal rather than theoretical (Egüz et al., 2025). This active way of learning reduces the mental distance between the student and the environment, helping them move past feeling overwhelmed and instead allowing them to imagine and design practical solutions for the future (Hannigan et al., 2025).

The effectiveness of these artistic methods lies in their ability to spark emotional engagement. It is increasingly accepted that simply knowing the facts about a problem like waste or climate change is rarely enough to change behavior, and that knowledge must be linked to emotions like care or concern to trigger a response (Carmi et al., 2015). Artistic practices provide aesthetic experiences that can deepen emotional engagement and support empathetic perception in ESD, helping students develop empathy for the natural world and reflect on future-oriented responsibilities to coming generations (Niederhauser et al., 2024). Drawing on arts-based climate education research that reports emotional, aesthetic, and transformative shifts in how young people understand their place in the world, participatory video production may similarly prompt reflexive engagement with local environmental narratives (Hannigan et al., 2025).

Building on these foundations, participatory video (PV) and related forms of digital storytelling have emerged as practical tools that combine digital literacy with environmental agency. Research on place-based filmmaking demonstrates that participatory digital storytelling can enhance environmental literacy by localizing climate change, increasing personal responsibility, and fostering students' perceived ability to act, thereby supporting agency-oriented approaches to climate education (Littrell et al., 2020). PV involves learners collaboratively scripting, filming, and editing narratives about issues that directly affect them, shifting engagement away from an information-deficit model toward participatory knowledge production. Prior studies suggest that the iterative learning process involved in filmmaking–encompassing inquiry, storytelling, and reflection–can itself be highly transformative, shifting emphasis from the final product to the reflective and dialogic production process. By embedding environmental concepts within personally and socially meaningful narratives, such participatory storytelling approaches have been shown to foster empowerment and a strengthened sense of agency (Davis and Weinshenker, 2012; Rohmah et al., 2024; Weder and Kierans, 2025). By grounding sustainability challenges in local contexts and meaningful places, digital storytelling can help bridge the gap between abstract environmental knowledge and everyday practice.

2.4 The cross-cultural gap in implementing participatory arts-based ESD

Despite the documented benefits of participatory arts-based approaches in ESD, their effective implementation remains constrained by a significant cross-cultural gap in the literature. The majority of existing studies are derived from Western contexts, particularly Europe and North America, resulting in a limited understanding of how ESD models can be transferred, adapted, and operationalized within Asia (Edwards et al., 2020). This geographical imbalance is especially pronounced in intra-Asian transfers, such as adapting pedagogical models developed in secular, industrialized contexts to socio-culturally distinct educational settings.

Implementing ESD in these contexts requires careful navigation of local cultural and religious realities that are often insufficiently addressed in Western-derived curricula. In Indonesia, environmental education for Muslim youth is closely intertwined with Islamic principles, particularly the concept of khalifah fil ardh (stewardship), which frames environmental protection as a sacred religious duty to maintain the balance of the Earth (Rahmania et al., 2025; Suraijiah et al., 2026; Wijaya Mulya and Salvi, 2025). While there are ongoing efforts to integrate Islamic values with environmental education in Indonesia, these initiatives often face the challenge of connecting abstract religious ethics with concrete, everyday sustainability practices. As Nilan (2021) suggests, for environmentalism to be effective, it must be translated into a “moral habitus,” where piously motivated youth view ecological protection not just as a theory, but as a mandatory expression of their faith. This calls for the need to explore how participatory arts-based frameworks can facilitate this translation, joining scientific sustainability knowledge with religious orthopraxy to help youth enact their role as khalifah (stewards) in tangible ways.

2.5 Conceptual Framework

Drawing on specific elements of transformative learning, social learning theory, social practice theory, and arts-based ESD, this study proposes a contextualized conceptual framework to address the awareness-behavior gap in sustainable consumption. Traditional ESD models frequently tend to stall as they rely on cognitive knowledge transmission within hierarchical structures, which is insufficient to drive behavioral change. To overcome this, our framework posits that the transition from cognitive awareness of SDG 12 to observable pro-environmental behavior requires two critical mediating layers: affective engagement–driven by the critical reflection of transformative learning and the digital storytelling of arts-based ESD–and culturally resonant social learning, achieved through peer-to-peer modeling (social learning theory) to disrupt entrenched communal routines (social practice theory).

In this model, participatory video making served as the primary catalyst for affective engagement. By engaging in the creative, hands-on process of filmmaking, students translated abstract environmental data into personally meaningful narratives, fostering empathy and perceived agency. Simultaneously, the social learning environment was restructured to bridge the cross-cultural gap. By utilizing a near-peer mentoring system (Al-Thani et al., 2023; Persohn et al., 2024), the learning environment was structured to reduce hierarchical barriers associated with high power distance teacher-student relations described in cross-cultural education research (Hofstede, 1986), thereby creating more supportive conditions for students to collaborate and negotiate meaning. Crucially, within this egalitarian space, the students' meaning making was anchored in the Islamic ethical concept of khalifah fil ardh (stewardship). Thus, ecological responsibility was framed not as a secular obligation, but as an enactment of religious orthopraxy (Nilan, 2021).

By intertwining the affective affordances of arts-based learning with the localized moral habitus of an Islamic educational setting, this framework hypothesizes that students would develop the collective efficacy necessary to transform cognitive awareness into behavioral reductions in food waste. The theoretical pathways connecting these pedagogical inputs, mediating mechanisms, and final behavioral outcomes are visually synthesized in Figure 1.

Based on this conceptual framework, the study tests the following hypotheses:

• - H1: The participatory arts-based intervention will increase students' aggregate understanding and awareness of SDG 12.
• - H2: The intervention will result in a reduction in observable cafeteria food waste.

3 Intervention

This study aimed to explore the effectiveness of SDGs education, particularly focusing on SDG 12 through participatory arts education methodology in Indonesia. The intervention was conducted at SMA Budi Cendekia Islamic Senior High School (BCI), a private secondary school located in Depok City, West Java. All intervention activities and data collection were completed between August and October 2024. SMA BCI was selected as the pilot site due to its existing interest in sustainability and its capacity to integrate the program into the P5 Curriculum (Project for Strengthening the Profile of Pancasila Student), a national curriculum component emphasizing character development and community values. The intervention specifically targeted Grade 10 students, comprising an initial cohort of 224 participants. This specific grade was selected because their academic schedule formally incorporates the P5 curriculum, allowing for immersive project-based learning without disrupting preparation for upper-secondary national examinations.

The core pedagogical framework was adapted from a successful predecessor participatory arts-based model implemented at Gyeongnam Electronic High School in South Korea (Shin and Park, 2026). While the Korean model utilized music video production to address plastic waste (SDG 12.5), the Indonesian iteration used short films and stop-motion animation to target food waste (SDG 12.3). This contextualization addresses a critical issue in the country, where up to 48 million tons of food waste are generated annually, with 80% of this total originating at the consumer and household levels (Mahmud et al., 2026). This adaptation involved three key shifts to enhance student participation, pedagogical accessibility, and cultural relevance: content localization, a transition in facilitation structure, and activity modification.

Educational materials were developed to address SDG 12 from cognitive, socio-emotional, and behavioral perspectives. These materials were designed to be student-centered, featuring user-friendly interfaces and review quizzes to reinforce knowledge retention. To ensure global applicability and support the international dissemination of arts-based SDG programs, the core materials were also translated into English. By integrating this localized program into the P5 Curriculum at SMA BCI, students could apply theoretical sustainability concepts to real-life scenarios within their specific school environment. Furthermore, the school's existing cafeteria infrastructure provided a highly controlled environment conducive to the intervention's behavioral tracking. Specifically, the daily lunch tray system allowed for the accurate observation, measurement, and tracking of aggregate daily tray waste by categorizing returns directly as either “clean plates” or “food waste” (see Section 4.1 for operational definitions). This binary categorization facilitated the systematic validation of behavioral shifts over the observation periods.

A distinct methodological innovation in this study was the transition from a teacher-centered instruction model to a university-student mentoring system. In many other recent applications of participatory video for the SDGs, such as those described by Rohmah et al. (2024) and Johnston et al. (2026), the curriculum was primarily delivered by schoolteachers and professional arts instructors. Conversely, this intervention established a partnership with Universitas Indonesia (UI), where undergraduate and graduate students from UI's Department of International Relations served as mentors. To ensure highly personalized guidance, one dedicated mentor was assigned to each of the school's nine participating classes. These mentors regularly provided four structured mentoring sessions throughout the intervention period to help students comprehend SDG 12. This mentor-led structure was designed to reduce the hierarchical distance often present in traditional Asian classrooms (Ying and Briffett-Aktaş, 2024), thereby fostering a more “participatory” environment where high school students felt greater agency to express creative ideas. Within this structure, the university mentors facilitated discussions on responsible consumption and guided the ongoing production of the films. To further strengthen students' technical capacities, external subject-matter experts conducted a specialized “Filmography and Videography for SDG Workshop” on September 6, 2024, equipping the broader cohort of participating students with practical expertise in filmography and videography.

While the original Korean framework utilized music video production to address single-use plastics, the Indonesian iteration shifted to short films and stop-motion animation to address food waste. This modification was deliberately chosen to ensure pedagogical accessibility and scalability within a developing country context. In contrast to media production models that rely on specialized equipment, digital storytelling and short filmmaking were implemented using students' ubiquitous personal devices and freely available mobile applications. Such low-tech approaches lower barriers to participation and support equitable, context-sensitive, and scalable implementation in resource-constrained educational settings (Aijaz, 2025; Eliza et al., 2024). Furthermore, this accessible medium operationalized the core mechanisms of participatory video (PV) and supported the action learning required to address cafeteria food waste. By collaboratively scripting, filming, and editing narratives about their own cafeteria, students engaged in an iterative process of inquiry, storytelling, and reflection. Because reducing food waste requires deconstructing entrenched daily habits, this localized meaning-making was designed to shift engagement away from a passive, information-deficit model toward participatory knowledge production, effectively bridging the gap between abstract sustainability concepts and everyday practice.

To align the evaluation process with the local educational context, the project leveraged SMA BCI's existing tradition of collective celebrations. The intervention culminated in a school-wide SDG 12 Film Festival on October 19, 2024, serving as a public exhibition designed to reinforce subjective norms by making the students' commitment to sustainability visible to the entire school community. The festival featured screenings and an awards ceremony across six categories: Filming, Videography, Concept, Substance, Directing, and a Grand Prize (Best Overall). Furthermore, a specific incentive structure was integrated: if SMA BCI achieved a targeted 10% reduction in food waste by the end of the intervention, the Ban Ki-moon Foundation for a Better Future would award the school a “Sustainable Development Model School through Culture and Arts Education” certificate. Methodologically, embedding the final evaluation within a familiar communal event was intended to foster collective efficacy and provide a social anchor for the targeted pro-environmental behaviors.

Table 1 summarizes the intervention timeline and specific activities carried out during the study.

Table 1. Overview of the intervention.

4 Data and methods

4.1 Data

To assess the effectiveness of the adapted model, data were collected through two primary channels:

• Behavioral Monitoring (Food Waste Tracking): To measure actual behavioral change, the school cafeteria monitored “clean plates” vs. “food waste” daily. Specifically, a “clean plate” was strictly operationalized as a tray containing zero solid, edible food waste. Minor, unpreventable residues–such as leftover sauces, liquids, or scattered crumbs–were not counted as waste. Conversely, any tray returned with intact, unconsumed portions of solid food waste was categorized as “food waste.” Data were collected for 15 days prior to the intervention to establish a baseline, and for 15 days post-intervention. The cafeteria's physical food trays, standard portion sizes, and overarching supply operations remained entirely unchanged throughout both observation periods, ensuring that any reduction in waste was strictly the result of student behavior.
• Descriptive Surveys: Pre- and post-intervention surveys were administered to measure shifts in cognitive awareness and knowledge regarding the SDGs. Specifically, general awareness was assessed using a binary (Yes/No) parameter:

“Do you know the term SDGs (Sustainable Development Goals)?”

Familiarity with specific targets was measured on a 5-point Likert scale (1 = lowest to 5 = highest) for the items:

“Target 12.3: Halve global per capita food waste” and “Target 12.5: Substantially reduce waste generation.”

Because individual-level matched data were not available for regression analysis, aggregate data were utilized to determine descriptive shifts in student awareness of SDG terms and specific targets (12.3 and 12.5).

The participant pool consisted of Grade 10 students at SMA BCI. As shown in Table 2, a total of 224 students (48.7% male, 51.3% female; predominantly aged 15) completed the pre-intervention survey. Following the intervention, 153 students completed the post-intervention survey (49.0% male, 51.0% female; predominantly aged 15). This reduction in sample size was attributed to standard student absenteeism and the rigorous exclusion of incomplete survey responses during the final data collection phase. Crucially, the demographic stability between the pre- and post-intervention cohorts indicated that this attrition did not introduce systematic demographic bias. To maintain analytical integrity, aggregate data analysis was conducted using this final, representative sample (N = 153).

Table 2. Participant demographics pre- and post intervention.

4.2 . Analysis methods

This study adopted a quantitative descriptive case study approach to evaluate the contextual adaptation and effectiveness of a participatory arts-based Education for Sustainable Development (ESD) model. While previous implementations of this model in South Korea were validated using quantitative structural equation modeling (SEM) to test the Theory of Planned Behavior (TPB) (Ajzen, 1991; Shin and Park, 2026), the present study was conducted as a pilot intervention project focusing on the implementation process and aggregate behavioral outcomes within a new cultural context. To capture both cognitive shifts and observable actions, the analysis focused on pre- and post-intervention comparisons of awareness change and food waste.

To evaluate the significance of the observed shifts in student awareness and knowledge, categorical survey data were analyzed using Chi-square (χ²) tests of independence. Pre- and post-intervention responses were organized into 2 x 2 contingency tables [Timing (Pre-Intervention vs. Post-Intervention) x Awareness (Aware vs. Unaware)] to test the null hypothesis that the distribution of SDG awareness was independent of the intervention timing. Expected frequencies (E) for each category were calculated based on the marginal totals:

The chi-square statistic was then derived using the following formula, where O represents the observed frequency:

E= row total x column totalgrand total

Significance was assessed at the p < 0.05 level with 1 degree of freedom (df = 1).

χ2=Σ(O-E)2E

Furthermore, because χ² values are sensitive to sample size, Cramér's V was calculated to determine the practical magnitude (effect size) of the intervention. The statistic was defined as:

where n is the total sample size (n = 377) and k is the number of rows or columns (k = 2). Following Cohen (1988), values of 0.10, 0.30, and 0.50 were used as benchmarks for small, medium, and large effects, respectively.

V=χ2n(k-1)

4.3 . Ethical considerations

The study protocol, including the involvement of student participants and the mentoring arrangement, was reviewed and approved by the Universitas Indonesia's Institutional Review Board (IRB) prior to implementation (KET-13/UN2.F9.KEP/PPM.00.02/2024). Informed consent was obtained from the school administration and participating students, and all data were collected and analyzed in accordance with ethical research standards.

5 Results

The participatory arts-based intervention resulted in measurable shifts across both cognitive and behavioral domains. As detailed below, these outcomes were evidenced by two primary data streams: (1) quantitative survey data, which revealed significant aggregate increases in SDG awareness and familiarity with Targets 12.3 and 12.5, and (2) behavioral monitoring data, which recorded a 13-percentage point reduction in cafeteria food waste.

5.1 . Cognitive and awareness shifts

To test H1–which posited that the participatory arts-based intervention would increase students' aggregate understanding and awareness of SDG 12–changes in student awareness and knowledge related to the SDGs were assessed using pre- and post-intervention survey data. The analysis focuses on aggregate shifts in the recognition of the SDGs and self-reported understanding of specific targets related to waste reduction.

General awareness of the SDGs increased markedly over the course of the intervention. While 48% of respondents reported recognizing or being familiar with the term “SDGs” in the pre-survey, this proportion rose to 81% in the post-survey, representing a 33-percentage point increase. As shown in Table 3, a chi-square test of independence confirmed that the distribution of SDG awareness differed significantly between the pre- and post-survey periods, χ² (1, N = 377) = 41.03, p < 0.001. This yielded a medium practical effect size (V = 0.33) and an odds ratio of 4.68 (OR = 4.68). This indicates that following the intervention, students were nearly five times as likely to be aware of the SDGs, suggesting that the shift in awareness represented a substantively meaningful change across the student population.

Table 3. Contingency table and chi-square results for SDG awareness.

χ² = 41.03, df = 1, p < 0.001, V = 0.33.

Knowledge of SDG Target 12.3 (halving global per capita food waste) also demonstrated a statistically significant shift between baseline and endline surveys. In the pre-survey, a significant portion of students reported low familiarity, but post-intervention responses shifted toward higher comprehension categories. A chi-square test of independence indicated that the distribution of responses changed significantly, χ²(4, N = 377) = 15.36, p = 0.004, with a small-to-moderate practical effect (V = 0.20). These findings suggest that students' understanding of the specific sustainability goals underlying their daily cafeteria behaviors improved following participation in the intervention. The pre- and post-intervention distributions are presented in Figure 2.

Similarly, student familiarity with SDG Target 12.5 (waste generation reduction) also showed a statistically meaningful shift over the course of the intervention. In the pre-survey, only 41% of respondents indicated that they knew this target “well” or “very well.” In the post-survey, responses were more heavily concentrated in higher self-reported knowledge categories. A chi-square test validated the statistical significance of this shift, χ²(1, N = 377) = 28.62, p < 0.001, yielding a small-to-medium practical effect size (V = 0.28). These findings indicate that participation in the mentor-led participatory intervention was associated with a substantive increase in students' perceived knowledge of waste-related SDG targets. The pre- and post-intervention distributions are presented in Figure 3.

5.2 . Behavioral impact: school food waste reduction

To test H2–which hypothesized that the intervention would result in a reduction in observable cafeteria food waste–the primary behavioral outcome measured in this study was the incidence of food waste among Grade 10 students at SMA BCI, assessed through tray observations conducted before and after the intervention.

During the pre-intervention period (August 19— September 6, 2024), a total of 3,392 lunch trays were observed. Of these, 623 trays contained leftover food, resulting in a food waste incidence rate of 18%. During the post-intervention period (September 10—October 3, 2024), 3,591 trays were recorded, with 180 trays containing leftovers, corresponding to a food waste incidence rate of 5%. As illustrated in Figure 4, a comparison of the two observation periods indicated a 13-percentage point reduction in food waste incidence following the intervention, exceeding the project's predefined target of a 10% reduction.

6 Discussion

The findings from SMA Budi Cendekia Islamic Senior High School (BCI) demonstrate that a participatory arts-based intervention can effectively bridge the awareness-behavior gap—the persistent discrepancy between environmental awareness and corresponding pro-environmental behavior. While traditional Education for Sustainable Development (ESD) often stalls at didactic knowledge transmission within hierarchical structures, this model successfully synthesized cognitive shifts with observable behavioral changes. The quantitative results reveal a significant dual impact: observed cafeteria food waste was reduced by 13 percentage points—dropping from an 18% baseline to 5% post-intervention—which exceeded the project's predefined target. This behavioral transformation was accompanied by a marked increase in both general SDG awareness and familiarity with SDG 12.3 and 12.5.

The observed 13-percentage point reduction in aggregate waste was driven entirely by a measurable shift in student behavior at the point of disposal—specifically, an increase in students returning ‘clean plates' with no edible leftovers. These physical outcomes align with our conceptual framework regarding the ‘moral habitus,' suggesting that the arts-based intervention provided the necessary context for students to translate their heightened cognitive awareness of SDG 12.3 into concrete action.

Conceptualizing the awareness-behavior gap through social practice theory has direct implications for how these results are interpreted. Because food waste is viewed as a social routine rather than an isolated cognitive choice, effective interventions must alter shared subjective norms, not just individual test scores. Consequently, the observed 13-percentage point physical reduction in food waste is interpreted not merely as a cognitive victory, but as a successful restructuring of the students' communal practices. The participatory arts process, culminating in the public SDG 12 Film Festival, provided the necessary social infrastructure to anchor these new pro-environmental routines, effectively bridging the gap between individual value shifts and collective action.

The dual cognitive and behavioral outcomes observed in this study suggest that the intervention moved beyond rote instruction into the realm of reflexive engagement, where global sustainability targets became deeply situated within the students‘ everyday contexts. By utilizing accessible digital storytelling, the pedagogical framework shifted students from being passive recipients of sustainability discourse to active creators of local environmental narratives. This shift facilitates what Bentz and O'Brien (2019) describe as a “subjective transformation,” wherein students reflect on their individual roles within a larger ecological system. The filmmaking process—encompassing inquiry, narrative design, and technical execution—required students to deconstruct their immediate cafeteria environment, effectively translating abstract concepts into the “personally and socially meaningful narratives” that characterize transformative arts-based ESD. As students navigated these creative challenges, they cultivated a profound sense of collective ownership over this local sustainability challenge. This hands-on problem-solving fostered the action-oriented agency necessary to move from imagining sustainability solutions in a film to executing them in the cafeteria.

Consistent with the causal pathways analyzed in Shin and Park (2026), the effectiveness of this participatory approach likely stems from its influence on internal psychological constructs. Prior research has established that environmental awareness remains insufficient for behavioral change unless mediated by affective engagement, such as a sense of personal responsibility (Kollmuss and Agyeman, 2002; Carmi et al., 2015). The study by Shin and Park (2026) further confirms that arts-based education drives behavior changes by positively shifting attitudes and reinforcing subjective norms. In the Indonesian context, the “participating” aspect of filmmaking—involving collective brainstorming and collaborative production—internalized the social pressure to reduce waste, thereby creating a normative environment that supported responsible consumption. Consequently, the creative process served as both an instructional tool and a behavioral catalyst, enabling students to transcend mere information acquisition and adopt sustainable moral habits.

Beyond the internal psychological shifts, the physical environment of the cafeteria functioned as a “living lab” that provided the direct feedback necessary to reinforce these new habits. This process is best understood through the principle of observability, which Rogers (2003) identifies as a core attribute for the successful adoption of new behaviors. Unlike abstract environmental data, the “clean plate” serves as a highly visible, daily indicator of success that reduces uncertainty and reinforces student motivation through vivid, tangible evidence of collective effectiveness. This visibility facilitates reflexive monitoring in action (RMA), allowing students to monitor the real-time consequences of their actions and transition from merely “believing” in a goal to learning through direct environmental feedback, as RMA posits that learning is most effective when participants can monitor the real-time consequences of their actions and adjust their behavior accordingly (van Mierlo et al., 2010). Furthermore, this transition highlights a form of sensory reflexivity, where the material changes in the cafeteria—literally seeing and feeling the return of empty trays—situates sustainability knowledge in a way that abstract discourse cannot (Pink, 2012). By bridging the gap between the creative narrative of the films and the sensory reality of the cafeteria, this direct observation strengthens the reflexive link between knowledge and action, ensuring that sustainability is not only practiced but visibly reinforced as a new social norm.

A significant methodological innovation that facilitated these outcomes was the transition from a traditional, teacher-centered instruction model to a near-peer mentoring system involving university students from Universitas Indonesia (UI). While conventional Asian educational settings often maintain high “teacher-as-expert” hierarchies that can stifle student-centered pedagogies, the UI mentors actively mediated this power distance by positioning themselves as facilitators rather than authority figures. By utilizing mentors who were closer in age and social status to the participants, the intervention effectively operationalized the psychological safety identified in our framework as a prerequisite for collaborative problem-solving. This egalitarian environment allowed Grade 10 students to move beyond the rote compliance typical of transmissive traditions and instead assume authentic ownership over their filmmaking projects. The reduction in perceived social distance thus served as a structural catalyst, enabling the dialogic engagement necessary for students to negotiate and express their own creative sustainability narratives.

This peer-to-peer dynamic established by the mentors facilitated a functional “Community of Practice” (Lave and Wenger, 1991) that moved beyond theoretical instruction into situated, shared activity. While traditional structures may separate technical skill from conceptual knowledge, the university mentors provided the integrated scaffolding—ranging from narrative development to digital editing—that enabled students to navigate the iterative complexities of filmmaking. These interpersonal connections were further solidified by leveraging SMA BCI's existing cultural tradition of collective school celebrations through the SDG 12 Film Festival. By transitioning a classroom-based project into a public forum, the festival functioned as a critical behavioral anchor that rendered the students' commitment to sustainability visible to the broader school community. This public exhibition transformed private intentions into shared subjective norms, effectively signaling that responsible consumption was a valued institutional identity. Consequently, embedding the project's culmination within a familiar communal tradition provided the social scaffolding necessary to strengthen collective efficacy—the shared belief in a group's capability to organize and execute joint action against complex ecological crises (Bandura, 1977; Hamann et al., 2024).

In the specific Indonesian context, this collective celebration served as the operational bridge between abstract religious ethics and observable daily orthopraxy. Rather than treating environmental stewardship as a secular obligation, the intervention allowed students to frame their work through the Islamic principle of khalifah fil ardh (stewardship) (Nilan, 2021). The filmmaking process and the subsequent festival translated this theological principle into a tangible “moral habitus” (Nilan, 2021), where reducing cafeteria food waste became a piously motivated enactment of religious duty. The institutional impact of this culturally contextualized approach was externally validated by the “Sustainable Development Model School Certificate” awarded by the Ban Ki-moon Foundation for a Better Future. This recognition served as institutional proof that student creative agency, when anchored in local religious values, can drive recognized behavioral shifts. Such outcomes suggest that for ESD to be sustainable, it must be integrated into the existing moral and cultural fabric of the educational setting.

Despite the observed efficacy of the intervention, this study is subject to several methodological limitations. First, the lack of individual-level matched data precluded the use of paired t-tests or more advanced longitudinal modeling. To minimize potential privacy concerns and comply with institutional guidelines, personal identification was not collected during the survey process. Secondly, the reduction in sample size from the pre-intervention cohort (N = 224) to the post-intervention phase (N = 153) was primarily driven by standard student absenteeism and the rigorous exclusion of incomplete survey responses. While the demographic stability between the two cohorts suggests that this attrition did not introduce systematic demographic bias specifically, the potential impact of this attrition on the study's validity should be acknowledged. It is plausible that the students who failed to complete the post-survey systematically differed in non-demographic ways not captured by the data, such as possessing lower baseline motivation or less interest in environmental-related topics to start with. Consequently, the final analytical sample may disproportionately represent the most engaged students, potentially leading to an overestimation of the intervention's aggregate impact on cognitive SDG awareness.

Furthermore, an important limitation of the present study is the reliance on a single-group pre-test/post-test design without a control group. Consequently, it is not possible to definitively rule out the influence of confounding variables on the observed cognitive and behavioral shifts. It is plausible that the reduction in food waste was in part influenced by the Hawthorne effect, wherein students altered their disposal habits simply because they were aware that their cafeteria trays were being actively monitored. Additionally, there is a risk of contamination from external factors during the core intervention period from August to October 2024. Concurrent school activities or informal peer discussions outside the structured monitoring sessions may have independently influenced student awareness. While the 13-percentage point reduction in food waste represents a substantial behavioral shift, acknowledging such potential confounds provides a more cautious interpretation of the findings. To address these limitations, future research should employ more robust methodological designs, such as quasi-experimental setups incorporating matched control groups, to effectively isolate the specific impact of participatory arts-based ESD interventions from broader environmental and social influences.

Additionally, the measurement of student awareness in the descriptive surveys was relatively simple and generic. The assessment was limited to broad recognition of the SDGs and self-reported familiarity with specific targets. Future research could incorporate more specific and comprehensive metrics to evaluate cognitive shifts. For instance, additional survey items could examine students' understanding of concrete actions that can be taken to reduce food waste, their perceived barriers to reducing waste, and other related behavioral dimensions. Measuring these nuanced aspects in future studies would build upon these initial findings, providing deeper insights into the specific cognitive mechanisms driving the observed behavioral shifts.

7 Conclusion

This study demonstrated that a culturally contextualized, participatory arts-based approach effectively catalyzes both cognitive awareness and observable behavioral change within a non-Western, Islamic educational setting. By leveraging reflexive engagement through student-led filmmaking, the intervention successfully bridged the awareness-behavior gap by disrupting entrenched consumption routines, resulting in a substantive 13-percentage point reduction in cafeteria food waste and significant gains in general SDG awareness. These outcomes validate the premise that when environmental stewardship is anchored in localized ethical frameworks—such as the principle of khalifah fil ardh (stewardship)—sustainability education moves beyond theoretical instruction to become a tangible, shared social practice and internalized moral habitus.

The successful adaptation of this framework offers broader implications for the global scaling of Education for Sustainable Development (ESD). To ensure efficacy across resource-constrained and diverse international contexts, international policy and Official Development Assistance (ODA) should prioritize the transfer of adaptable pedagogical “software” rather than relying on rigid, equipment-heavy models. This includes flexible, low-tech models—such as near-peer mentoring and student-led filmmaking—that can be tailored to diverse socio-cultural and resource-constrained realities. Such a shift in ODA strategy can empower local institutions to lead their own sustainability transitions.

While the immediate results are promising, it is essential that future research employs longitudinal designs to determine the long-term durability of these behavioral transformations. Specifically, tracking whether the 5% post-intervention food waste rate is sustained following the departure of university mentors will provide critical insight into whether these newly formed habits have become permanent or remain contingent on external facilitation. Investigating the longitudinal impact of participatory arts-based interventions is essential to verify their capacity to foster the enduring ecological stewardship required to achieve the goals of the 2030 Agenda for Sustainable Development.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/frsus.2026.1837816/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Diana Gregory-Smith, Newcastle University, United Kingdom

Reviewed by: Endang Sri Maruti, University of PGRI Madiun, Indonesia

Liza Diniarizky Putri, Serang Raya University, Indonesia