The impact of perceived risk of online takeout packaging and the moderating role of educational level

The study examines whether and how consumers’ perceived packaging pollution risk (PPRP) from online takeout affects their purchase intention. Addressing two research gaps, it first tests if PPRP negatively influences intention, and second, investigates the mechanism via the Theory of Planned Behavior (TPB)—attitude, subjective norms, and perceived behavioral control (PBC)—and the moderating role of education level. In the context of rapid growth in online takeout and heightened post-COVID environmental and health concerns about plastic packaging, the authors integrate the Concept/Theory of Perceived Risk (CPR/TPR) with TPB to model the trade-off consumers make between convenience and environmental/health risks. The study’s contributions include extending risk perception to online takeout packaging, applying and extending TPB in this context, and clarifying mediation (attitude, subjective norms, PBC) and moderation (education level) mechanisms linking PPRP to purchase intention.

The literature review covers: (1) Concept of perceived risk (CPR): originating in psychology and adapted to consumer behavior, perceived risk comprises uncertainty and consequence severity, and can include financial, physical, social, functional, and psychological risks. In online contexts, risks shift toward product quality, authenticity, and privacy; for online takeout, consumers focus less on payment/privacy and more on product quality and environmental/health risks from packaging. (2) Pollution risk of takeout packaging: Plastic packaging dominates takeout for cost and performance advantages but generates significant environmental pollution and health hazards (microplastics, plasticizers like PAEs and bisphenols, especially under heat). Migration into food and bioaccumulation raise toxicity and health concerns; regulations (e.g., BPA limits) exist but waste volumes are increasing. The paper defines packaging pollution risk perception (PPRP) as perceived environmental and health risks from online takeout packaging. (3) Theory of Planned Behavior (TPB): Attitude, subjective norms, and PBC determine behavioral intention and have been validated across contexts. The review argues for integrating perceived risk as a negative antecedent to TPB constructs in online takeout. (4) Theoretical framework and hypotheses: The model posits that attitude, subjective norms, and PBC positively influence purchase intention (H1–H3); PPRP negatively affects attitude, subjective norms, PBC, and purchase intention (H4–H7); attitude, subjective norms, and PBC mediate the PPRP–intention link (H8–H10); and education level moderates key paths between PPRP, TPB components, and intention (H11a–H11g).

Design and measures: A multi-step scale development was used. TPB items (attitude, subjective norms, PBC, purchase intention) were adapted from prior studies and contextualized for online takeout. PPRP items were developed via grounded-theory content analysis of 20 in-depth interviews with Chinese consumers who ordered takeout at least weekly for three months; environmental and health concerns were primary themes. Experts reviewed items for content/face validity. All items used 5-point Likert scales. Back-translation procedures ensured linguistic validity. Pilot and reliability/validity: A pilot survey (n=116) assessed reliability (Cronbach’s alpha across constructs >0.70) and validity (EFA/CFA). Final constructs retained parsimonious 3–4 items each, meeting recommended indicators per construct. Common method bias (CMB) was addressed via procedural remedies (clear items, expert review, anonymity) and tested with CFA comparing a five-factor model to a single-factor model, with the five-factor model fitting significantly better in both pilot and main samples. Sampling and data collection: The main online survey (WJX.cn) produced 336 valid responses from six major Chinese cities—Beijing (18%), Shanghai (16.5%), Guangzhou (17%), Chengdu (16.8%), Wuhan (15%), Xi’an (16.7%)—with balanced gender quotas. Inclusion required online takeout purchase within the last week. Demographics collected included sex, age, occupation, and education. Analysis: Confirmatory factor analysis (CFA) confirmed model fit, composite reliability (CR > 0.77), and convergent/discriminant validity (AVE > 0.50; inter-construct correlations < 0.7). Structural equation modeling (SEM, AMOS 24, maximum likelihood) tested direct (H1–H7) and indirect effects (H8–H10) with bias-corrected bootstrapping (5,000 resamples, 95% CI). Multi-group SEM assessed education-level moderation (H11) across three groups: EL1 (senior high/secondary and below; n=114), EL2 (junior college; n=101), EL3 (bachelor’s and above; n=121). Measurement invariance (configural, metric) was established; partial structural invariance testing identified moderated paths.

- Model fit and explanatory power: SEM fit indices were acceptable (e.g., χ²/df≈1.95, CFI=0.953, RMSEA=0.053). Exogenous variables explained 49.7% of purchase intention variance (R²=0.497); TPB alone explained 47.1% (ΔR²=+2.6% with PPRP). PPRP explained 17.0% of attitude variance, 19.9% of subjective norm variance, and 21.8% of PBC variance. - Direct effects (supporting H1–H7): • Attitude → Purchase intention: β=0.288, p<0.001 (H1 supported) • Subjective norms → Purchase intention: β=0.256, p<0.001 (H2 supported) • PBC → Purchase intention: β=0.202, p<0.01 (H3 supported) • PPRP → Attitude: β=−0.412, p<0.001 (H4 supported) • PPRP → Subjective norms: β=−0.446, p<0.001 (H5 supported) • PPRP → PBC: β=−0.467, p<0.001 (H6 supported) • PPRP → Purchase intention: β=−0.259, p<0.01 (H7 supported) - Mediation (H8–H10): Bootstrapping showed significant partial mediation of PPRP’s effect on purchase intention through TPB components. • Total effect (PPRP → PI): TE=−0.421, 95% CI [−0.519, −0.330] • Direct effect: DE=−0.186, 95% CI [−0.301, −0.078] • Indirect effect: IE=−0.235, 95% CI [−0.338, −0.163] • Specific indirect effects: via Attitude IE≈−0.085 (36.2% of total IE); via Subjective norms IE≈−0.082 (34.9%); via PBC IE≈−0.068 (28.9%). All CIs excluded zero, confirming partial mediation (H8–H10 supported). - Moderation by education level (H11): Multi-group SEM indicated education level significantly moderated several paths: • PBC → Purchase intention differed between EL1 and EL3 (path positive and significant for both; EL1 β≈0.206 vs. EL3 β≈0.201; H11c supported for group difference). • PPRP → Attitude differed across EL1 vs. EL2 and EL2 vs. EL3 (e.g., EL1 β≈−0.420; EL2 β≈−0.398; EL3 β≈−0.456; H11d supported). • PPRP → Subjective norms differed between EL1 and EL2 (EL1 β≈−0.460; EL2 β≈−0.407; H11e supported). • PPRP → PBC differed between EL1 and EL2 (EL1 β≈−0.487; EL2 β≈−0.432; H11f supported). • No significant moderation was found for Attitude → PI (H11a), Subjective norms → PI (H11b), or PPRP → PI (H11g).

Findings confirm that TPB effectively explains online takeout purchase intention in China, with attitude being the strongest positive determinant, followed by subjective norms and PBC. Incorporating PPRP improves explanatory power, demonstrating that perceived packaging-related environmental and health risks undermine the TPB antecedents and directly reduce purchase intention. PPRP most strongly depresses PBC, suggesting that when risks are salient, perceived ease and control of using takeout platforms become less influential than concerns over health and environment. The mediation analysis shows that much of PPRP’s impact on intention operates through reduced attitude, norms, and PBC, highlighting these as leverage points for interventions. Education level conditions several relationships: lower-educated consumers exhibit stronger negative sensitivity of TPB components to PPRP, whereas higher-educated consumers’ PBC–intention link differs from lower-educated consumers, implying nuanced educational effects on risk processing and control perceptions. These results address the research questions by mapping both the pathways (mediation) and boundary conditions (education-level moderation) of how PPRP influences online takeout intentions, with implications for sustainable consumption and platform strategy under post-pandemic conditions.

The study integrates perceived risk with the Theory of Planned Behavior to explain online takeout purchase intention under heightened environmental and health concerns about packaging. PPRP significantly reduces attitude, subjective norms, PBC, and purchase intention, with partial mediation through TPB variables and moderation by education level on several paths. The work extends risk perception and TPB to the online takeout context, clarifies inhibitory mechanisms of risk on planned behavior, and offers actionable insights for policymakers and industry to promote sustainable packaging and consumption. Future research should: (1) assess cross-cultural generalizability beyond China given cultural and development differences; (2) incorporate additional risk dimensions and higher-order constructs to enhance predictive power; and (3) employ longitudinal and behavioral (transactional) data to examine intention–behavior gaps (e.g., actual choices of eco-friendly packaging options).

- Generalizability: Findings are based on Chinese consumers; cultural and developmental differences may limit transferability to other countries. - Scope of risk factors: PPRP focuses on environmental and health risks; other risk dimensions and higher-order constructs were not modeled. - Cross-sectional design: Results reflect intentions at one point in time; longitudinal or behavioral data are needed to validate causal inferences and intention–behavior relationships.