The Sustainable Development Goals (SDGs) are a commitment established by the United Nations (UN) in 2015 as a roadmap to build a better and more sustainable future for all people. This agenda consists of 17 interconnected goals, recognizing that action in one area will affect outcomes in others. The SDGs address a wide range of urgent global challenges, from eradicating poverty and hunger to improving health, education, and gender equality. In addition, the SDGs promote actions to protect the planet, including tackling climate change, preserving oceans and terrestrial ecosystems, and supporting responsible consumption and production.

Three specific SDG goals are closely linked. Unsustainable consumption and production patterns (Goal 12) can lead to ocean pollution (Goal 14) and land degradation (Goal 15). For example, plastic waste from consumer products often ends up in oceans, polluting marine ecosystems and endangering marine life. Deforestation for agriculture and plantations can also cause land degradation and loss of biodiversity. Therefore, efforts to achieve Goals 12, 14, and 15 must be pursued in an integrated manner—and this is the reason why the researcher is interested in conducting this study.

The Stimulus-Organism-Response (SOR) theory, developed by Mehrabian and Russell (1974), explains that external stimuli influence internal conditions (organism), which then trigger actual behavioral responses. In the context of green marketing, stimuli include green advertising, CSR activities, eco-labels, and perceived value, which affect green trust as an internal psychological response and ultimately lead to a purchase decision. This model is relevant for understanding environmentally conscious consumer behavior and has been widely used in various studies to explain the influence of green communication on purchase decisions through the mediation of trust and consumer emotions. This study adopts the SOR model as a conceptual framework to explain the influence of green factors on sustainable bottled water (AMDK) purchase decisions.

Green advertising is a communication strategy that emphasizes sustainability values in line with increasing environmental awareness. Banerjee et al. (1995) define it as advertising that links products with the environment, promotes environmentally friendly lifestyles, and portrays the company as environmentally responsible. Zhu (2013), as cited in Rohmadoni (2017), also highlights the role of green advertising in demonstrating a company's efforts to reduce negative environmental impacts.

Corporate Social Responsibility (CSR) refers to a company’s commitment to being accountable for the social and environmental impacts of its business operations. CSR is not only a legal obligation but also reflects a company’s voluntary contribution to societal welfare and environmental sustainability (Suharto, 2007; Lamo Said, 2018). Antonius (2017) states that CSR encompasses three main pillars: profit, people, and planet.

An eco-label is a marketing communication tool that provides consumers with information about the environmental aspects of a product. D’Souza et al. (2007) define it as a label that indicates the environmental attributes of a product, emphasizing that eco-labels should be credible and standardized. Mufidah et al. (2018) add that eco-labels can enhance consumer trust in environmental claims and demonstrate a brand's commitment to environmental sustainability.

Perceived value reflects the consumer’s subjective assessment of a product’s benefits compared to the sacrifices made to obtain it. Zeithaml (1988) defines it as the evaluation of what is received versus what is given in a transaction. Sweeney and Soutar (2001) further elaborate that perceived value includes dimensions such as quality, emotional, social, and price value, which influence how consumers assess the overall benefits of a product.

Green trust refers to consumers’ confidence in a company’s environmental commitment. Chen and Chang (2012) define it as the belief that a brand’s environmental claims are credible. Scoll et al. (2013) add that green trust is formed through perceptions of credibility, honesty, and reliability in the company’s sustainable business practices. This trust grows when a company consistently demonstrates genuine commitment to environmental protection.

Purchase decision is the final stage in the consumer behavior process, reflecting the decision to buy or not to buy a product. Schiffman and Kanuk (2008) describe it as the process of selecting from among various alternatives, while Peter and Olson (2013) emphasize the role of knowledge and value considerations in making that decision.

H1: Green Advertising has a positive effect on Purchase Decisions

H2: CSR Activity has a positive effect on Green Trust

H3: Eco-Label has a positive effect on Green Trust

H4: Perceived Value has a positive effect on Purchase Decisions

H5: Green Advertising has a positive effect on Purchase Decisions

H6: CSR activity has a positive effect on Purchase Decisions

H7: Eco-Label has a positive effect on Purchase Decisions

H8: Green Trust has a positive effect on Purchase Decisions

H9: Green Trust has a mediating role in the influence of Green Advertising on Purchase Decisions.

H10: Green Trust has a mediating role in the influence of CSR activity on Purchase Decisions.

H11: Green Trust has a mediating role in the influence of Eco-Label on Purchase Decisions.

This study employed quantitative and descriptive causal methods. The population consisted of respondents who had purchased bottled drinking water. The sample size was 133. The sampling method used non-probability sampling with a purposive sampling technique. The criteria were: individuals residing in Jakarta, aged 17 years or older, having purchased bottled drinking water, and willing to complete the questionnaire. The instrument used a Likert scale ranging from 1-5 (Strongly Disagree - Strongly Agree).

Figure 4.1 Initial Path Diagram Output Results

Source: Primary data processed by researchers, 2025

Source: Primary data processed by researchers, 2025

Based on Figure 4.1 and Table 4.1, all statements are declared valid. This is because the loading factor values are above 0.70 (Hair et al., 2021). In addition to using loading factor values, the validity of the data in this study was also analyzed using the Average Variance Extracted (AVE) value. The following presents the results of the validity test based on the AVE value.

Table 4.2 AVE Test Results

Source: Primary data processed by researchers, 2025

Referring to Table 4.2, all variables in this study are declared valid because their AVE values exceed the minimum limit of 0.50.

Table 4.3 Fornell Lacker Criterion Test

Source: Primary data processed by researchers, 2025

Based on the table above, it can be seen that all constructs have square root values of AVE greater than the correlations with other constructs. For instance, the square root of the AVE for the Purchase Decision (PD) construct is 0.826, which is higher than its correlation with other constructs.

"Table 4.4 Cross Loading Values"

Source: Primary data processed by researchers, 2025

Based on Table 4.4, it can be concluded that each indicator has the highest loading value on the construct it measures compared to other constructs. Thus, all indicators are declared valid, and no issues were found in the discriminant validity test.

Table 4.5 Hajil Reliability Test

Source: Primary data processed by researchers, 2025

Table 4.5 shows that all latent variables have a Composite Reliability value of ≥ 0.7, and the Cronbach's Alpha results also show a value of ≥ 0.7 for all latent variables. Based on these results, it can be concluded that each construct in this research model has good reliability. This means that the instruments or questionnaires used in this study have proven reliable and consistent in measuring the intended variables.

Structural Model Evaluation (Inner Model)

Table 4.6 R-Square Test Results

Source: Primary data processed by researchers, 2025

Based on Table 4.6, it is known that the coefficient of determination R Square for the Green Trust (Z) construct is 0.358. This value indicates that 35.8% of the variation in Green Trust can be explained by three Independent variables, namely Green Advertising, CSR Activity, and Eco Label, while the remaining 64.2% is influenced by other factors outside the research model. The R-square value for the Purchase Decision (Y) construct is recorded at 0.710, which indicates that 71% of changes in Purchase Decision can be explained by the variables Green Advertising, Eco Label, CSR Activity, Perceived Value and Green Trust as Intervening variables. Meanwhile, the remaining 29% is explained by other variables not included in this model.

F Square

Source: Primary data processed by researchers, 2025

The analysis results show that most of the relationships between variables in the model have weak effect sizes. Green Advertising, CSR Activity, and Eco Label each provide small contributions to Green Trust and Purchase Decision. However, Perceived Value shows a moderate effect size on Purchase Decision (f² = 0.248), as does Green Trust on Purchase Decision (f² = 0.294). These findings confirm the important role of Green Trust as a mediator that strengthens the influence of Green Advertising, CSR Activity, and Eco Label variables on consumer purchasing decisions.

Table 4.8 Results of Q-square predictive relevance values

Source: Primary data processed by researchers, 2025

Based on Table 4.8, it is known that the Green Trust (Z) variable has a Q- square value of 0.320, which indicates that the model has moderate predictive ability for this construct. This means that the Green Trust construct is quite well predicted by the independent variables in the model, such as Green Advertising, Eco Label, and CSR Activity. Meanwhile, the Purchase Decision (Y) variable shows a Q-square value of 0.595, which is included in the strong category. This indicates that the structural model has excellent predictive ability.

Table 4.9 Model Fit Results

Source: Primary data processed by researchers, 2025

The test results show that the SRMR values for the saturated and estimated models are 0.065 and 0.069, respectively, below the threshold of 0.09, indicating good model fit. Furthermore, the NFI values of 0.773 and 0.775, respectively, indicate that the model has an adequate overall fit.

Table 4. 10 Results of Hypothesis Test Analysis

Source: Primary data processed by researchers, 2025

Based on Table 4.10, the results of the hypothesis test analysis show that the path coefficient values (original sample) range from 0.080 to 0.371. This indicates that all variables in the model have a positive relationship with each other. Furthermore, the results of the hypothesis test are also visualized in the form of a structural model graph, which displays the path coefficients based on the t-statistic significance level and the p-value for each relationship between variables.

The study indicates that green advertising has a positive and significant effect on green trust. In other words, the stronger the environmental message embedded in advertisements, the higher the consumer trust in the product. This finding aligns with Li & Shan (2025), who found that green advertising effectively fosters positive perceptions and consumer trust. It suggests that sustainability messaging in ads plays a relevant role in establishing trust toward bottled water (AMDK) products.

Results show that CSR activities positively and significantly enhance green trust. This means that the more active and environmentally relevant a company’s CSR programs are, the higher the consumer's trust in the company's commitment to the environment. AMDK consumers in Jakarta now evaluate aspects like waste management or packaging reuse education as integral to a company's integrity. This echoes the findings of Hang et al. (2022) and Mo et al. (2022), which concluded that consistent and environmentally focused CSR enhances public trust in organizations. For urban consumers, CSR is viewed not just as image-building but as a genuine reflection of environmental responsibility.

The research confirms that eco-labels have a positive and significant effect on green trust among AMDK consumers. As symbols of sustainability, eco-labels signal compliance with environmental standards, boosting consumers’ perception of brand responsibility. This result mirrors studies by Gorton et al. (2021) and Hossain et al. (2022), which state that eco-labels enhance consumer trust in the brand’s sustainability integrity. In urban markets like Jakarta, eco- labels serve as environmental assurance, reinforcing a brand’s credibility amid ongoing issues related to plastic waste.

Perceived value was found to have a positive and significant influence on AMDK purchase decisions in Jakarta. Consumers evaluate products not only by price but also by quality, convenience, safety, and sustainability contributions. When perceived benefits are equal to or exceed the sacrifices, consumers are more likely to make a purchase. This is consistent with the findings of Chauhan & Goyal (2024) in Santri (2024) and Nurhalisa et al. (2023), which identify perceived value as a strong predictor of sustainable consumption behaviors.

The data reveal that green advertising positively and significantly affects AMDK purchase decisions. As a communication tool that highlights sustainability, green advertising shapes positive consumer perceptions, driving purchase behaviors—especially in urban, environment-conscious segments. This finding corroborates Shimp & Andrews (2013) in Rahimah (2022), who assert that green advertising educates and influences consumer decision-making. Companies that consistently convey environmental messages—such as using recycled packaging or reducing carbon emissions—can build positive brand images and drive purchase preference.

CSR activities were found to positively and significantly influence purchase decisions for AMDK in Jakarta. CSR initiatives such as plastic waste reduction, recycling programs, or community engagement resonate with increasingly socially and environmentally conscious consumers. When companies demonstrate real commitment, CSR builds positive image and drives buying behavior. This aligns with Verma (2024), who reports that CSR significantly impacts purchasing decisions, particularly when it resonates with consumer values.

Eco-labels also showed a positive and significant influence on purchase decisions. As indicators that products meet environmental standards, eco-labels help consumers identify items aligned with their sustainability values, enhancing trust in environmental claims and prompting purchase. This finding is in line with Ferarri et al. (2015) in Hameed & Waris (2018), who note that eco-label credibility influences eco-conscious consumer behavior.

Green trust was found to have a positive and significant effect on AMDK purchase decisions in Jakarta. Trust in a company's environmental commitment drives consumers to choose products from brands perceived as eco-friendly. Green trust develops from positive experiences, transparent communication, and credible certifications. This mirrors Kusuma & Permana (2023) in Puspaningrum et al. (2023), which finds green trust a significant driver of eco-friendly purchasing behavior. It confirms that green trust is a key determinant in building consumer loyalty and preference in eco-conscious markets.

The results show that green trust significantly mediates the relationship between green advertising and AMDK purchase decisions. Green advertising that emphasizes environmental commitment builds trust, which in turn influences buying behavior. This aligns with Sun et al. (2020), who demonstrate that green trust strengthens the effect of green advertising on consumers’ purchasing actions. Thus, the effectiveness of green advertising hinges on a company's ability to build trust through honest and transparent communication.

Green trust also mediates the relationship between CSR activity and purchase decisions. CSR implemented consistently and transparently builds consumer trust in a company's environmental sustainability, ultimately influencing purchases. This finding aligns with Yu et al. (2020), which confirms trust as a mediator between CSR and purchase behavior. Hence, CSR’s effectiveness depends on its credibility and transparency in building green trust. Mediation: Green Trust Between Eco-Label and Purchase Decision

The study finds that green trust significantly mediates the eco-label → purchase decision pathway. A credible eco-label strengthens consumer trust in environmental claims, driving green product purchases. This matches Chen et al. (2015) in Rakhmawati (2023), which suggests environmental labels improve trust and purchase behaviors in eco-conscious consumers, highlighting that eco-labels are effective only when they genuinely build trust.