Modeling the intention and adoption of food waste prevention practices among Chinese households

The study addresses household food waste as a major contributor to global food loss with significant environmental, social, and economic impacts. Despite global recognition, most behavioral research on reducing food waste has focused on high-income countries. China, due to its population size, generates substantial household food waste, making the issue particularly salient. The research question explores how cognitive (perceived value of sustainability), environmental (awareness of consequences and ascription of responsibility), social (social norms), and emotional (anticipated guilt) factors influence Chinese households’ intentions to reduce food waste and how these intentions translate into actual waste-reduction behaviors. Grounded in the Theory of Interpersonal Behavior, the study aims to identify determinants of intention and behavior and to test whether intention mediates the relationship between these determinants and behavior, thereby informing interventions to curb household food waste in China.

The literature highlights the scale and impacts of household food waste globally and in China, noting per capita estimates and higher waste in developed Chinese cities. Prior research has often quantified waste and its impacts, with limited focus on household-level participation in prevention. Studies leveraging the theory of planned behavior underscore attitudes and knowledge but have inconsistently accounted for emotions and non-cognitive factors. Recent work integrating TIB indicates that cognitive, environmental, social, and emotional factors shape waste-reduction behaviors, yet gaps remain—particularly regarding perceived values of sustainability and ascription of responsibility as predictors. In the Chinese context, few empirical, theory-driven studies have examined determinants of food waste reduction intentions and behavior. This study extends TIB by including perceived values on sustainability (cognitive), awareness of consequences and ascription of responsibility (environmental), social norms (social), and anticipated guilt (emotional) to explain intentions and subsequent behavior, and by testing the mediating role of intention.

Design: Quantitative, cross-sectional survey targeting Chinese households with at least one adult (18+). Data collection was conducted via WJX (online platform) from March 15 to June 22, 2022, using screening questions. Sample: Power analysis (G*Power 3.1) indicated a minimum of 153 respondents (effect size 0.15, power 0.95, seven predictors). The study obtained 1,090 valid responses. Descriptives: 52.4% female; largest age group 31–40 years (23.0%); 69.1% employed; 56.4% ate out 0–3 times/week. Measures: Section A captured demographics. Section B measured constructs with five items each from validated scales: perceived values on sustainability (Oviedo-García et al.; Kim et al.; Han et al.), awareness of consequences and anticipated guilt (Attiq et al.), ascription of responsibility (Kim et al.), social norms (Kim et al.), food waste reduction intention (Aktas et al.), and food waste reduction behavior (Attiq et al.). Items used a 5-point Likert scale (1=strongly disagree to 5=strongly agree). Behavior items were tailored to household food waste reduction; intention items included both household and non-household contexts for comparative insight. Bias and distribution checks: Common method bias assessed via Harman’s single-factor test (largest factor 42.996% < 50%) and full collinearity VIFs (<3.3). Multivariate normality was violated (WebPower), supporting use of PLS-SEM. Analysis: Partial least squares structural equation modeling (PLS-SEM) was used to evaluate measurement and structural models. Reliability and validity were assessed via Cronbach’s alpha, composite reliability, Dijkstra-Henseler’s rho, AVE, Fornell-Larcker, HTMT, and cross-loadings. Model quality was evaluated with R². Measurement invariance of composite models (MICOM) and multi-group analyses (by age, gender, education, and eating-out frequency) were conducted. VIFs within the measurement model were below 5 (1.000–1.630).

- Measurement model: High reliability across constructs (Cronbach’s alpha ~0.906–0.943; composite reliability ~0.930–0.943; rho_A ~0.910–0.934). Convergent validity supported (AVE 0.728–0.767). Discriminant validity supported by Fornell-Larcker and HTMT (<0.9). Measurement model VIFs 1.000–1.630. - Structural model: R² for intention (FWI) = 0.424; R² for behavior (FWB) = 0.365, indicating moderate explanatory power. - Direct effects on intention (all positive and significant, p<0.05): • Perceived values on sustainability → intention: β=0.198 (CI 0.140–0.260; t=5.198) • Awareness of consequences → intention: β=0.232 (CI 0.175–0.294; t=6.404) • Ascription of responsibility → intention: β=0.073 (CI 0.007–0.136; t=1.903) • Social norms → intention: β=0.128 (CI 0.068–0.194; t=3.349) • Anticipated guilt → intention: β=0.231 (CI 0.170–0.291; t=6.192) - Intention → behavior: β=0.604 (CI 0.557–0.648; t=22.004; p<0.001), indicating that stronger intentions predict greater food waste reduction behavior. - Mediation: Intention significantly mediated the effects of all antecedents on behavior: • PV → FWI → FWB: β=0.120 (CI 0.083–0.159) • AC → FWI → FWB: β=0.140 (CI 0.102–0.179) • AR → FWI → FWB: β=0.044 (CI 0.004–0.082) • SN → FWI → FWB: β=0.077 (CI 0.040–0.118) • AG → FWI → FWB: β=0.139 (CI 0.102–0.184) - Multi-group analysis and MICOM: Partial measurement invariance generally supported across groups (age, gender, education, eating-out frequency). One significant difference identified related to the relationship between awareness of consequences and intention when comparing groups by eating-out frequency. - Descriptive highlights: Majority ate out ≤3 times/week (56.4%); spending per meal commonly RMB51–70 (30.0%) or RMB30–50 (30.8%).

Findings confirm that cognitive, environmental, social, and emotional determinants from TIB jointly shape Chinese households’ intentions to reduce food waste, and that intention strongly translates into actual waste-reduction behavior. Perceived values of sustainability and awareness of consequences emerged as robust predictors of intention, underscoring the importance of value-driven cognition and knowledge of impacts. Ascription of responsibility, though smaller in effect, significantly contributes to forming pro-environmental intentions by reinforcing personal accountability. Social norms positively influence intentions, highlighting the role of perceived societal expectations and group identity in shaping waste-reduction motivations. Anticipated guilt is a strong emotional driver, indicating that negative anticipatory emotions can motivate corrective actions to avoid moral dissonance. The mediating role of intention across all antecedents validates TIB’s mechanism, showing that interventions targeting these determinants can increase intentions, which in turn enhance actual behaviors. The results extend prior research by integrating underexplored predictors (perceived sustainability values and ascription of responsibility) within TIB and by demonstrating their indirect effects on behavior via intention in a developing-country context. Practically, the findings suggest that policies and campaigns that elevate sustainability values, increase awareness of environmental and social consequences, foster personal responsibility, leverage social norms, and ethically engage emotions (e.g., guilt-aversion) can effectively reduce household food waste.

This study extends the Theory of Interpersonal Behavior to the context of Chinese households by integrating perceived values on sustainability, awareness of consequences, ascription of responsibility, social norms, and anticipated guilt as predictors of food waste reduction intention and behavior. All factors positively influenced intention, and intention strongly predicted behavior; intention also mediated each antecedent’s effect on behavior. These results offer a comprehensive, multidimensional explanation of household food waste prevention and provide empirically grounded guidance for designing interventions. Future research should: (1) incorporate direct observation and weighing of household food waste to complement self-reports; (2) examine generalizability across different cultural and national contexts; (3) explore additional TIB elements such as habits and facilitating conditions; and (4) assess intervention efficacy (e.g., educational, normative, and emotional appeals) over time using longitudinal or experimental designs.

- Self-reported measures of behavior may be subject to bias; alignment with observed or weighed waste was not feasible in this large-scale study. - Sample limited to China, potentially constraining generalizability to other cultural contexts. - Direct observation/weighing of food waste was not conducted due to practical constraints (time, resources, scale). Future studies should include objective measures and cross-cultural samples.