The impact of green accounting on environmental performance: mediating effects of energy efficiency

Abstract

This study aims to investigate the relationship between green accounting and energy e ciency and environmental performance in the context of Bangladeshi pharmaceutical and chemical companies. The study also explores the mediating role of energy e ciency in the relationship between green accounting and environmental performance. A total of 326 responses were collected using a simple random sampling technique from pharmaceutical and chemical companies in Bangladesh. The study employed Partial Least Squares Structural Equation Modeling (PLS-SEM) to analyze the data. The results indicate that green accounting has a signi cant positive impact on both energy e ciency and environmental performance. Moreover, energy e ciency partially mediates the relationship between green accounting and environmental performance. The study also found that economic, environmental, and social practices of green accounting have a positive impact on both energy e ciency and environmental performance, with environmental practices having the highest impact. The ndings of this study provide important insights for managers and policymakers of pharmaceutical and chemical companies in Bangladesh, highlighting the need for green accounting practices that promote environmental sustainability. The study suggests that the integration of green accounting practices can lead to better energy e ciency and environmental performance, which can enhance the reputation and competitive advantage of these companies. This study identi es the mediating role of energy e ciency in the relationship between green accounting and environmental performance, providing a unique perspective on the mechanism behind the relationship.

Factors affecting retailer social responsibility: a PLS-SEM approach in the context of Bangladesh

“…The value is 0.752 and can be regarded as acceptable. Construct validity is a measurement that considers the extent to which a test measures an intended hypothetical construct (Rahman and Islam, 2023). Convergent validity can be defined when indicators or items of a specific construct converge or share a high proportion of variance (Hair et al , 2017).…”

“…Convergent validity can be defined when indicators or items of a specific construct converge or share a high proportion of variance (Hair et al , 2017). We also measured convergent validity under construct validity of the study of three particular parameters for item reliability along with Cronbach’s alpha, construct (composite) reliability and average variance extracted (AVE) from the recommendation of Hair et al (2017) and Rahman and Islam (2023). All the constructs had been satisfactory.…”

“…Two ways were used to assess discriminant validity under the construct validity of the study. First, cross-loading of the indicators used for indicating discriminant validity is by examining the cross-loadings of the indicators where outer loading should be greater than cross-loadings (Rahman and Islam, 2023). In our study, the outer loadings were greater than the cross-loadings of the constructs listed in the table of the rotated component matrix.…”

“…However, chi-square is known to be sensitive to sample size, and other fit indices should also be considered. NFI is the normed fit index, which measures the model’s overall fit (Hair et al , 2021; Rahman and Islam, 2023). In this case, an NFI value of 0.913 indicates that the model fits well.…”

“…The AVE is a tool that measures the quality of the measurement model. The AVE for each construct exceeded the accepted threshold of 0.5 (Hair et al, 2017;Rahman and Islam, 2023). The AVE for each factor was more than 0.5 (ranging from 0.529 to 0.593), further ensuring convergent validity (see Table 3).…”