Does Internet Access Contribute to Residents’ Pro-Environmental Behavior? Evidence from China

Abstract

Residents' pro-environmental behavior (PEB) is vital for solving environmental pollution. This paper uses a micro sample of 10993 respondents from the Chinese General Social Survey. Propensity score matching was used to assess the effect of Internet access on residents' PEB, and the study finds a significant positive impact on private-sphere PEB and public-sphere PEB. This fundamental conclusion still holds after utilizing the instrumental variable method, changing the dependent variable, core explanatory variable, and the econometric model. Heterogeneous analysis finds that Internet access has a more significant positive impact on private-sphere PEB for elderly, women, low educated and rural residents. Internet access has a more significant positive impact on public-sphere PEB for the middle-aged, women, higher educated, and urban residents. The mediating test suggests that Internet access promotes residents' PEB and raises the overall level of residents' environmental awareness through the mediating factor of residents' PEB. This paper provides the theoretical and empirical basis that Internet technology can promote PEB at the microlevel, which emphasizes the positive impact of information and communications technology on eco-environmental improvement.

Estimating heterogeneous effects of internet use on environmental knowledge: Taking population heterogeneity into consideration

“…The swift proliferation of ICTs has resulted in a significant transformation in the media landscape and an increase in the adoption of internet- and social media-based interventions for environmental education [ 26 , 27 ]. Owing to their cost-free nature and relative efficacy in reaching target audiences, digital media, particularly the internet, have become increasingly integrated into our daily routines [ 28 ] and are viewed as the most cost-effective channels for environment communication and education [ 1 , 20 , 26 , 29 , 30 ]. Furthermore, the internet’s on-demand and crowd-sourcing features offer up-to-date environmental news, enable real-time interactions, and foster engagement and social participation [ 29 , 31 ], empowering audiences to exercise greater agency in environmental learning [ 20 ].…”

“…Variability across units of analysis is inherent in all social phenomena, which also applies to the investigation of digital media effects [ 53 ] on sustainability outcomes. In recent years, an emerging body of research has focused on exploring the heterogeneous effects of internet use/access on pro-environmental behavior [ 30 , 31 ]. Furthermore, scholars have employed quantile regression models to examine the influence of ICTs adoption on reducing agrochemical inputs [ 52 ] and enhancing economic and production performance in sustainable agriculture practices [ 54 ] across various quantiles.…”

“…Extensive empirical research has investigated the impact of internet access and use on attitudinal [33][34][35][36] and behavioral [30,31,37,38] aspects of sustainability outcomes. A common finding from these studies is that environmental knowledge, as a cognition-based sustainability outcome, plays an important intermediary role between internet use/access and both environmental attitude and pro-environmental behavior [e.g., 30,37,38].…”

“…This predisposition could lead individuals to pay more attention to negative information about environmental degradation and pollution [46], making people more conscious of the environmental issues and enhancing their environmental awareness [34,38]. Substantial evidence supports a significant and positive relationship between internet use and environmental awareness [30,31,33,47]. Scholars suggest that this heightened awareness subsequently contributes to enhanced environmental knowledge [24,39,48,49].…”

“…Based on the counterfactual framework [72][73][74] and estimated propensity scores, PSM identifies a counterfactual control group similar to the treatment group via various matching algorithms [30,75]. In this study, the 4-nearest neighbor matching method is used to generate pairs of treatment-controlled samples sharing similar characteristics.…”