Introduction
Urban agriculture is widely advocated as a multifunctional response to contemporary urban challenges, offering pathways to enhance local food security, optimize resource flows, and bolster urban resilience in the face of climatic variability and demographic change. The degree to which urban agriculture can deliver these benefits is contingent not only on land availability and technical inputs but critically on the knowledge base, dispositions, and practical competencies of urban residents who manage productive green spaces. Within this nexus, environmental education emerges as a pivotal enabling strategy—one that aims to build capacity among local actors, promote sustainable, practice-oriented resource use, and catalyze participatory governance of urban green infrastructures. Despite proliferating programs and policy interest, the evidence base linking environmental education interventions to measurable outcomes in participatory urban agriculture has remained fragmented. This meta-analytic review integrates empirical studies published between 1995 and 2024 to quantify the magnitude of educational effects on knowledge, attitudes, behaviors, and ecological outcomes in urban agriculture, and to elucidate the mechanisms and contextual moderators that shape those effects.
Materials and methods
Study design and selection criteria: We undertook a systematic review coupled with meta-analysis to synthesize quantitative and qualitative evidence on the impacts of environmental education interventions on urban-agriculture-related outcomes. Comprehensive searches were executed across international and national bibliographic databases for peer-reviewed articles published in English and Persian through 2024. Eligible studies explicitly evaluated the effects of pedagogical interventions—ranging from school- and community-based curricula to peer-led trainings and extension-style workshops—on at least one of the following outcome domains: environmental or urban-agriculture knowledge, attitudes and participatory intention, observed practices, resource-use efficiency (e.g., water or nutrient use), soil health indicators, input-use patterns (fertilizer, pesticide), biodiversity metrics, or other ecological indicators. Methodological rigor, clear intervention description, and availability of extractable quantitative outcomes were required for inclusion. Seventy studies satisfied these criteria and were included in the quantitative synthesis.
Data extraction and effect-size computation: For each retained study we systematically extracted publication metadata, geographic and socioecological context, sample characteristics, design features (randomization, controls), intervention attributes (delivery mode, content focus, degree of localization and stakeholder co-design), duration and intensity (frequency and follow-up), measured outcomes, and reported statistics or raw data. Where necessary, means, standard deviations, proportions, or test statistics were converted into standardized mean differences (Cohen’s d) or comparable metrics. We pooled effect sizes using random-effects models to accommodate between-study heterogeneity, and we conducted moderator analyses to test the influence of intervention format (interactive/practice-based versus lecture-based), localization and stakeholder participation, duration (short versus sustained), and study quality. Publication-bias diagnostics and sensitivity checks were performed. Complementing quantitative synthesis, a thematic qualitative analysis distilled theorized mechanisms of change and recurring implementation barriers reported across studies.
Results and discussion
Aggregate effects across outcome domains: The pooled analysis of 70 studies demonstrates that environmental education interventions exert statistically significant positive effects across three principal domains relevant to participatory urban agriculture: (1) domain-specific knowledge about environmental processes and urban cultivation; (2) attitudinal shifts and motivational dispositions toward participation and pro-environmental management; and (3) observable practices and on-site performance in urban growing systems. Effect-size estimates reveal a clear gradient: interventions generate moderate-to-large impacts on cognitive and attitudinal outcomes (pooled d ≈ 0.45–0.75), while impacts on behavioral and biophysical outcomes measured in situ—such as adoption of sustainable practices, water-use efficiency, reduced chemical-input application, and soil-health improvements—are smaller but still positive (pooled d ≈ 0.25–0.50). This pattern indicates that educational inputs more readily shift knowledge and attitudes than they immediately transform ecological or agronomic indicators, reflecting the additional temporal, infrastructural, and institutional requisites for behavioral translation into measurable ecological change.

Moderators of intervention effectiveness: Subgroup analyses identify consistent moderators associated with larger and more durable effects. Interventions that prioritized interactive, experiential pedagogies—field exercises, hands-on demonstrations, participatory problem-solving, and peer-to-peer learning—produced larger effect sizes than predominantly lecture-based or one-off informational sessions. Programs that were co-designed with local stakeholders and tailored to place-specific agroecological and socio-cultural contexts also yielded stronger outcomes, as did interventions implemented over sustained timelines with explicit follow-up, monitoring, and technical support. Conversely, short-term, didactic trainings and externally imposed curricula exhibited lower effectiveness in changing practice and sustaining outcomes. Study design quality further moderated effect estimates, with better-controlled evaluations generally reporting more conservative but robust effect sizes.
Environmental and ecosystem-service outcomes: Empirical findings indicate that when environmental education is tightly integrated with participatory urban-agriculture practices, it can lead to measurable environmental benefits. Reported gains include increases in water-use efficiency (frequently in the range of ~15–30% in studies that quantified irrigation practices or volumetric use), reductions in synthetic input dependence (reported declines in fertilizer and pesticide application ranging roughly ~10–40% where measured), and improvements in soil health indicators such as enhanced organic matter content, improved aggregate stability, and ameliorated compaction. Biodiversity-related outcomes—such as increases in pollinator abundance, native flora cover, and functional diversity in multispecies plots—were also documented in a subset of studies that explicitly measured ecological assemblages. Nevertheless, the magnitude and persistence of these environmental gains varied substantially across contexts: larger, longer-lasting impacts were concentrated where enabling conditions—access to inputs and organic amendments, irrigation infrastructure, extension services, secure land tenure, and supportive policy frameworks—were present. Where institutional support or resource access was lacking, educational gains in knowledge and attitude often failed to scale into substantive ecological improvements.
Limitations, heterogeneity, and theoretical implications: The qualitative synthesis highlighted methodological and conceptual limitations pervasive in the literature. Many studies lacked explicit theoretical articulation of behavioral-change mechanisms, limiting causal interpretation of how cognitive and affective shifts translate into sustained practice change. Heterogeneity in outcome measures, short follow-up windows in numerous evaluations, and variable measurement rigor constrained cross-study comparability. Publication-bias diagnostics suggested modest asymmetry consistent with under-reporting of null or small effects in some subfields. Despite these limitations, the aggregated evidence aligns with behaviorally oriented and experiential-learning frameworks: approaches emphasizing active participation, situated practice, social learning, and iterative feedback loops are most plausibly linked to sustained adoption of sustainable urban-agriculture practices. The observed lag between attitudinal gains and ecological outcomes further underscores that behavior-change theories must account for structural enablers and constraints—material, institutional, and socio-economic—that mediate the relationship between knowledge and environmental performance.
Conclusion
Environmental education embedded in participatory urban-agriculture initiatives produces consistent, positive effects on citizens’ domain-specific knowledge and pro-participation attitudes, and yields positive but more context-sensitive improvements in on-site practices and ecological indicators.
Interventions employing interactive, practice-based pedagogies, local adaptation and co-design with stakeholders, and sustained engagement with follow-up and monitoring achieve the largest and most durable outcomes.
When effectively implemented and supported by enabling infrastructure and institutions, integrated education–practice programs can improve water-use efficiency, reduce dependence on chemical inputs, enhance soil quality, and bolster urban biodiversity; however, the degree and longevity of these benefits depend on local conditions and policy support.
Absence of structured, practice-oriented environmental education risks perpetuating maladaptive practices that can exacerbate soil degradation, water contamination, and ecological decline in urban agricultural settings.
Research priorities include longer-term, well-controlled evaluations; explicit theoretical development and testing of behavioral mechanisms linking education to practice and ecological outcomes; comprehensive assessments of socio-economic impacts; and focused analyses of how urban policy and governance arrangements can scale and institutionalize effective education–agriculture models.
Programmatic implications: design participatory environmental-education initiatives that emphasize hands-on training, stakeholder engagement, local contextualization, sustained timelines, and continuous monitoring and adaptive feedback to secure enduring environmental and socio-economic benefits from urban agriculture.