Dataset collection and experimental environment
This study employs an online experiment to investigate the impact of eco-positioning strategies on consumer behavior. For participant recruitment, the research team collected data through the SoJump platform. To ensure sample representativeness and experimental validity, random sampling was used to recruit participants from the platform’s user pool. Additionally, the following screening criteria were established to enhance data quality and ensure the accuracy of experimental results. Participants must be at least 18 years old to ensure they possess legal purchasing decision-making capabilities. Participants must be active fashion consumers, having purchased fashion products at least once in the past three months. To effectively test the influence of eco-positioning strategies, participants were required to answer questions about environmental awareness during the preliminary screening phase. Meanwhile, only those with high environmental concern were allowed to proceed with the experiment. During the recruitment process, 123 participants were recruited through platform advertisements and social media promotions, with 102 valid questionnaires ultimately collected, yielding a response rate of 82.9%.
The sample size is determined based on multiple factors. First, referencing prior studies in similar fields, sample sizes between 100 and 150 are common in comparable online experiments to ensure sufficient statistical power for data analysis. According to G*Power analysis, this study assumes the use of regression analysis with an effect size of 0.15 (medium effect), a significance level of 0.05, and a statistical power of 0.80. Based on these parameters, the minimum required sample size is 89. Therefore, the sample size of 102 is sufficient to meet statistical analysis requirements and demonstrates high reliability and representativeness. Furthermore, the sample is drawn from the SoJump platform, where users exhibit diverse socioeconomic backgrounds and consumption behaviors, covering various age groups, professions, and consumption levels. This ensures sample diversity and the generalizability of the research findings. Thus, the 102 valid questionnaires are deemed adequate and reasonable.
In this study, a questionnaire is used to collect relevant data from participants. The questionnaire items are divided into three main sections: brand evaluation, willingness to pay, and sustainable consumption intention. To ensure the validity and reliability of the questionnaire, some items are adapted from previous studies; Others are specifically designed based on the objectives of this study. The brand evaluation section utilized scales from Park et al. (2013) to measure brand attitude and trust61. The willingness-to-pay section referenced scales from Biswas et al. (2016) on willingness to pay for green products62. The sustainable consumption intention section is newly designed based on the specific objectives of this study, developed through discussions with environmental experts. All borrowed scales are appropriately modified to align with this study’s fashion context and objectives. The complete version is provided in the appendix to facilitate readers’ evaluation of the questionnaire’s structure and content. It includes detailed descriptions of each question, the corresponding scale type (Likert scale), and the scoring method for each item.
During the experiment, participants were first asked to complete a questionnaire on their personal information, environmental awareness, and fashion consumption behaviors. They then completed three experimental modules presented in randomized order. Each module included exposure to stimulus materials specific to its manipulation condition, followed by outcome measures relevant to the module. Their responses were based on the content of the stimuli and their perceptions of the brand, product, or sustainability message. The questionnaire utilized a Likert scale to measure consumers’ attitudes and behavioral intentions towards eco-positioning. All experimental data were collected in real-time through the SoJump platform and analyzed using SPSS software. In addition, the personal information of all participants is strictly confidential. The study has been approved by an ethics review committee to ensure compliance with relevant ethical and privacy regulations. The reliability coefficient is 0.81, and the validity is 0.84, indicating the data’s reliability and validity.
The reliability coefficient of 0.81 mentioned here refers to Cronbach’s alpha coefficient of the scale, which is used to assess the internal consistency of the questionnaire. Cronbach’s alpha value is commonly employed to measure the consistency and reliability among items within a scale. The value typically ranges from 0 to 1, with higher values indicating stronger internal consistency of the scale. In this study, Cronbach’s alpha value of 0.81 demonstrates that the scale used exhibits good internal consistency. To evaluate the scale validity, this study employs tests for construct validity and Average Variance Extracted (AVE). Construct validity assesses whether the scale accurately measures the theoretical construct it intends to measure. AVE evaluates the degree of association between individual items and the latent variable, with higher AVE values indicating stronger convergent validity. For discriminant validity, this study ensures the scale’s good discriminant validity by assessing the correlations between latent variables and comparing factor loadings and cross-loadings. Moreover, Confirmatory Factor Analysis (CFA) validates the scale’s validity. In the CFA, all factor loadings exceed 0.6, indicating good construct validity of the scale.
The questionnaire’s descriptive statistical results are depicted in Fig. 2.
Descriptive statistical results of the questionnaire.
The hardware and software environments include Windows 10 operating on a 64-bit system with 8GB RAM and an Intel i7-7500 Central Processing Unit (CPU). Software-wise, the SoJump platform is used for questionnaire design and data collection. Meanwhile, Statistical Product and Service Solutions (SPSS) 26.0 software is employed for reliability and validity testing and descriptive statistical analysis.
Performance evaluation
The data were analyzed across three independent experimental modules. Each module’s manipulation and dependent variables were evaluated separately.
The influence of eco-positioning strategy on brand evaluation results.
The influence of eco-positioning strategies on brand evaluation is shown in Fig. 3. The study results indicate that product-related and process-related eco-positioning substantially influence brand evaluation and purchase intentions. Specifically, for product-related eco-positioning, the regression coefficients for brand evaluation and purchase intentions are 0.4 and 0.3, with p-values of 0.045 and 0.014. For process-related eco-positioning, the regression coefficient for brand evaluation is 0.6 (p-value = 0.003), and for purchase intentions, it is 0.7 (p-value = 0.001). Brand familiarity also plays a significant moderating role, where under high brand familiarity, the regression coefficients for brand evaluation and purchase intentions are 1.2 and 1.3, and the p-values are all 0.001. However, under low brand familiarity, the regression coefficients for brand evaluation and purchase intentions reach 0.5 and 0.4, and their p-values are 0.088 and 0.19. When combined with brand familiarity, the regression coefficient of brand evaluation for product eco-positioning and high brand familiarity is 0.8, with a p-value of 0.007, and the purchase intention’s regression coefficient is 0.9, with a p-value of 0.004; The regression coefficient is 1.1 (p-value = 0.002) for process eco-positioning and high brand familiarity, for purchase intention is 1.0, (p-value = 0.003). This indicates that process-related eco-positioning has a more prominent impact on brand evaluation and purchase intentions, while high brand familiarity enhances the positive effects of eco-positioning strategies.
The regression analysis results are presented in Table 2. The regression coefficients, effect sizes, and related confidence intervals (CI) and VIF for multiple influencing factors are considered in the regression analysis. First, in terms of regression coefficients, both product-related and process-related eco-positioning significantly impact brand evaluation and purchase intention, with strong effects. Their Cohen’s f² values are 0.15 and 0.25, respectively, indicating moderate effect sizes in the model. The regression coefficient for high brand familiarity is 1.2, with an effect size of 0.55, demonstrating a significant and strong influence on purchase intention and brand evaluation. In contrast, the impact of low brand familiarity is smaller, with a regression coefficient of 0.5 and an effect size of only 0.10. This indicates that eco-positioning has a diminished effect on consumer decision-making when brand familiarity is low. Regarding CIs, none of the regression coefficients’ CIs cross zero, further validating their significance and enhancing the credibility of the results. For example, the CI for process-related eco-positioning is [0.28, 0.92], demonstrating a significant and stable positive impact on consumer purchase intention. In terms of effect sizes (Cohen’s f²), all models exhibit moderate to strong effects. For instance, the effect size for high brand familiarity is 0.55, significantly exceeding the threshold of 0.35, highlighting its substantial role in driving consumer purchase intention. In contrast, the effect size for low brand familiarity is only 0.10, reflecting its weaker influence. Multicollinearity analysis reveals that all variables’ VIF values are below 5 (specifically ranging from 2.1 to 2.5), indicating no severe multicollinearity issues in the model. Therefore, the proposed regression model is statistically stable and reliable.
In Module 2 (willingness to pay), participants were exposed to product descriptions containing varying levels of eco-positioning: no eco-related content, moderate eco-positioning (e.g., simple eco-friendly features), and strong eco-positioning (e.g., comprehensive sustainability claims). The dependent variables included willingness to pay and perceived environmental sustainability. The impact of eco-positioning on willingness to pay is suggested in Fig. 4.
The impact of eco-positioning on willingness to pay.
Figure 4 shows that under no eco-positioning conditions, the regression coefficients for willingness to pay and perceived environmental sustainability are 0.2 and 0.3 (p-values = 0.1 and 0.08). Under moderate eco-positioning conditions, the regression coefficients for both variables are 0.5 and 0.6 (p-values = 0.02 and 0.01). Under strong eco-positioning conditions, the two variables’ regression coefficients are 0.8 and 0.9, with p-values of 0.001 and 0.001. Perceived environmental sustainability remarkably mediates the effect of eco-positioning on willingness to pay, with a regression coefficient of 0.7 (p-value = 0.001). The study demonstrates that eco-positioning strategies effectively enhance consumer willingness to pay by increasing perceived environmental sustainability.
In Module 3 (sustainable consumption intention), participants viewed either an eco-themed fashion advertisement highlighting sustainability commitments, a neutral fashion-related message without ecological emphasis, or no message (control). The dependent variables included sustainable fashion consumption intention, with environmental concern and fashion involvement measured as potential moderators. The eco-positioning’s effect on sustainable fashion consumption intention is denoted in Fig. 5.
The effect of eco-positioning on sustainable fashion consumption intention.
Figure 5 reveals that under eco-positioning fashion brand advertisement conditions, the regression coefficients for sustainable fashion consumption intention, environmental concern, and fashion involvement reach 0.6, 0.5, and 0.7; Their corresponding p-values are 0.002, 0.01, and 0.004, respectively. Under non-eco-positioning conditions, the regression coefficients for the above three variables are 0.3, 0.2, and 0.4 (p-values = 0.05, 0.08, and 0.06). In the control group (no advertisement), the three variables’ regression coefficients are 0.1, 0.1, and 0.2, with p-values of 0.2, 0.15, and 0.12. Moreover, environmental concern and fashion involvement markedly moderate the relationship between eco-positioning and sustainable fashion consumption intention. The regression coefficients are 0.5 (p-value = 0.01) for environmental concern and 0.7 (p-value = 0.004) for fashion involvement. These findings illustrate that eco-positioning strategies significantly promote consumers’ sustainable fashion consumption intention, with environmental concern and fashion involvement playing crucial moderating roles.
Discussion
The main findings of this study indicate that eco-positioning strategies have a significant impact on consumers’ brand evaluation, willingness to pay, and sustainable consumption intentions. Among these strategies, process-oriented eco-positioning has a more pronounced effect on enhancing brand evaluation and purchase intention, while brand familiarity plays a crucial moderating role in this process. These findings should be reassessed within the context of existing literature to highlight their theoretical contributions. First, the findings regarding differences in eco-positioning types engage in a dialogue with the study by Majeed et al. (2022)63. Although their research suggested that eco-positioning enhanced purchase intention by improving a brand’s environmental image, it did not differentiate between product-oriented and process-oriented strategies. Through experimental validation, this study demonstrates their distinct psychological effects on consumers, particularly revealing the advantages of process-oriented strategies (e.g., green production processes and third-party certifications) in shaping brand credibility and perceived commitment. This conclusion aligns with the findings of Shanti et al. (2022) on the impact of sustainable practices in the hotel industry on brand equity64. Their study similarly found that consumers’ attention to a company’s full-process environmental responsibility significantly influenced brand perception, indicating cross-industry commonalities. Second, this study’s analysis of the moderating role of brand familiarity further extends the theoretical framework of Kassie et al. (2023). While their research emphasized the fundamental role of brand familiarity in green marketing65, it did not explore the dynamic moderating mechanisms of familiarity under different eco-positioning strategies. This study indicates that within process-oriented strategies, high brand familiarity significantly amplifies consumer trust in environmental commitments, whereas low-familiarity brands rely more on specific certification signals (e.g., third-party endorsements). This finding is consistent with signal theory, which posits that the credibility of information interacts with the recipient’s prior knowledge to influence perception. Furthermore, the findings of this study complement the TPB by demonstrating that process-oriented strategies enhance consumers’ perceived control over brand environmental actions (e.g., transparent production processes), thereby indirectly strengthening subjective norms and perceived behavioral control. This mechanism explains why process-oriented strategies have a stronger effect compared to product-oriented strategies. Notably, previous studies have identified the direct impact of eco-labels on willingness to pay. However, this study employs SEM to reveal the mediating role of perceived environmental sustainability, providing a more granular explanation for the application of signal theory in fashion marketing. Overall, by integrating multiple theoretical perspectives with experimental design, this study validates key arguments in the existing literature (e.g., the relationship between brand familiarity and environmental commitments). Meanwhile, it addresses critical gaps in the typological comparison of eco-positioning strategies, dynamic moderating effects, and psychological mechanism pathways. These contributions establish a scalable analytical framework for future research.
This study also has certain limitations. While it identifies the significant impact of eco-positioning on consumer behavior, the sample composition and experimental design present constraints on generalizability. The current experimental parameters preclude thorough exploration of potential moderating factors beyond those explicitly measured, such as consumers’ cultural backgrounds, socioeconomic status, and prior environmental knowledge. Additionally, the controlled laboratory-like environment of online experiments may not fully capture the complexity of real-world purchasing contexts where multiple competing factors simultaneously influence consumer decisions. Future research could validate these findings in broader samples, especially across diverse cultural contexts, where consumer responses to eco-positioning strategies may vary. Theoretically, this study provides a multidimensional understanding of how eco-positioning strategies influence consumer decision-making processes by integrating TPB, ELM, and Signaling Theory. TPB explains how consumers form purchase intentions through attitudes, subjective norms, and perceived behavioral control when encountering eco-positioning. ELM helps understand how consumers may process ecological information through central or peripheral routes depending on the level of information provided. Signaling Theory offers a framework for how brands convey environmental signals through eco-positioning. By combining these theories, this study enriches the theoretical foundation of eco-marketing. Meanwhile, it provides more systematic theoretical guidance for fashion brands in developing sustainable marketing strategies.
This study delves into the impact of eco-positioning in the fashion marketing context on consumer perceptions and behaviors. It enriches the theoretical literature on sustainable fashion marketing, green advertising, and consumer behavior. It also enhances the theoretical understanding of how different eco-positioning strategies shape consumer responses and how brand familiarity moderates these effects. The study reveals that perceived environmental sustainability mediates the relationship between eco-positioning and consumer willingness to pay. This finding offers new perspectives on the application of Signaling Theory, ELM, Schema Theory, and TPB in sustainable fashion marketing. In practical management, the study provides valuable recommendations for fashion brands to develop effective eco-positioning strategies to communicate environmental commitments, enhance brand awareness, and promote sustainable consumption. It emphasizes the need for strategies to fully consider brand familiarity and consumer characteristics, such as environmental awareness and fashion involvement, to maximize market effectiveness.
link
