The Role of Two Social Marketing Strategies and Communication Design in Chinese Households’ Waste-Sorting Intentions and Behavior: A Theory of Planned Behavior Approach

Abstract

As part of the 14th Five-Year Plan, China aims to establish waste-sorting facilities in 436 cities by 2035. The success of waste sorting is dependent on individual-level changes as well as macro-level changes. Guided by the theory of planned behavior, the present analysis aimed to answer two questions: (a) What variables (e.g., attitudes and norms) were related to Chinese residents’ waste-sorting intentions, and (b) what social marketing strategies and design were related to the participants’ waste-sorting intentions? A cross-sectional survey of 459 Chinese participants was conducted in March 2021. Participants’ attitudes toward waste sorting, perceived norms, self-efficacy, exposure to social marketing strategies, evaluation of communication design, and several other variables were measured using questionnaire items. The participants’ attitudes toward waste sorting, perceived norms, self-efficacy, and hope predicted their intentions to sort waste. More importantly, two main social marketing strategies emerged from the analysis: Benefits and ways of waste sorting and the consequences of noncompliance. Promoting the benefits of and ways of waste sorting and communication design predicted intentions: These relationships were mediated by attitudes, norms, self-efficacy, and hope. On the other hand, strategies related to the consequences of noncompliance were a weak predictor of perceived norms and self-efficacy. Our results were significant because they provide guidance for future waste-sorting programs on the social marketing strategies to use and the need for more professional and well-designed promotional materials.

1. Introduction

As a result of rapid economic development and urbanization, the amount of domestic solid waste production in China has increased from 118 million tons in 2000 to 235 million tons in 2020 [1,2]. If not appropriately processed, domestic solid waste can negatively impact the living and natural environments and people’s health [3]. For example, openly dumped wastes can contaminate soil and water, plastic bags and facemasks can pollute the oceans and threaten marine lives, and waste management and landfills emit greenhouse gasses and contribute to climate change. Household waste sorting refers to residents sorting household waste into different categories (e.g., recycles, food waste, batteries, and other garbage) before placing it into bins. Waste sorting is one vital way to help manage domestic waste and carries several benefits, for example, helping protect the living environment and foster green and sustainable development [3]. Such measures can help preserve a sustainable environment for future generations.

At the macro-level, the United Nations Department of Economic and Social Development calls for “sustainable cities and human settlements” [4]. One goal is to promote “the integrated provision of environmental infrastructure: water, sanitation, drainage and solid waste management” [4]. Consistent with the United Nations’ call, China piloted and established a complete waste-sorting and recycling system in 46 major cities before 2020 (e.g., Beijing, Shanghai, Tianjin, and Hangzhou) [5]. It also aims to establish additional waste-sorting programs and systems in 337 regional-level cities by 2025 [6]. Clearly, without such facilities, it is not possible to process and separate large quantities of reusables from landfills.

At the micro-level, the United Nations calls on local authorities to mobilize different stakeholders, facilitate “‘bottom-up’ and inclusive processes,” and form different partnerships [4]. Admittedly, the effectiveness of waste-sorting programs is often the result of the micro-level changes in the citizens’ willingness to cooperate and sort waste before disposal, in addition to the macro-level infrastructure changes to facilitate waste sorting and processing. As such, China has heavily relied on social marketing campaigns to educate residents about waste sorting and gain cooperation.

Then, it is essential to share the Chinese experience and lessons learned with the sustainability community within and outside China: What factors are associated with Chinese residents’ willingness to sort garbage before disposal, and how do the social marketing efforts affect their willingness and behavior to do so? Unlike much of the previous research focusing on the psychological aspects of environmental behaviors, the present research will examine the effectiveness of two social marketing strategies (i.e., emphasizing the benefits and ways of waste sorting, and the consequences of noncompliance) as well as the role of communication design (i.e., how well were the promotional materials designed). Successful strategies should be adopted in the future, whereas less successful ones should be improved upon. To this end, the present research adopts the most recent version of the theory of planned behavior (TPB) [7], coupled with the literature on social marketing strategies and communication design (i.e., attitudes toward ad/promotional materials) [8].

1.1. Basic Version of the TPB

Health and environment research often applies the TPB to identify the factors that motivate individual behaviors. The TPB assumes that many human behaviors reasonably follow an individual’s beliefs and attitudes [7]. In its basic form, the TPB (Figure 1) states that attitudes toward the behavior, perceived norms, and self-efficacy (or perceived behavioral control) predict intentions that predict behavior. Attitudes commonly refer to individuals’ general evaluation of a behavior favorably or unfavorably. Perceived norms are the perceived approval or disapproval from significant others (e.g., family members or friends) or the frequency of a behavior by others [7]. Finally, self-efficacy refers to the individual’s confidence in successfully performing a behavior, which considers the internal locus of control and external barriers [7]. Armitage and Conner’s meta-analysis [9] found that attitudes and self-efficacy were generally stronger predictors of intentions than subjective norms.

These TPB variables are related to pro-social behaviors, including waste sorting [10,11]. Those who hold more favorable attitudes toward an environmental behavior (i.e., it is good to perform a pro-environmental behavior, or it carries good benefits) are more likely to perform the pro-environmental behavior. Those with higher self-efficacy (e.g., having more confidence, knowing more about the procedure, and having easy access to facilities) are more likely to participate in such behaviors. Individuals may have low intentions of sorting waste if they do not have confidence in doing so or cannot access nearby facilities. Lastly, subjective norms are perceived approval from significant others (e.g., family and friends) in interpersonal behaviors. However, in pro-environmental behaviors, subjective norms can be extended to the expectations in one’s neighborhood or local governments.

Regarding recycling behavior, previous research has found that attitudes, self-efficacy, and personal norms directly predicted recycling intentions in several countries, including China [12], India [12,13], Iran [14], Malaysia [15], and the United States [16]. S. Wang et al. found that attitudes toward waste sorting, norms, and self-efficacy all predicted intentions to sort garbage among a group of Chinese participants [17]. Govindan et al. surveyed residents in Shanghai and found that attitudes, norms, and self-efficacy predicted waste-sorting intentions [18]. Based on a survey, Zheng et al. found that individuals who have more positive attitudes and higher efficacy toward waste sorting were more likely to sort garbage [19]. Furthermore, the people that individuals are surrounded by influence their intention to sort waste [19]. Haj-Salem and Al-Hawari surveyed participants in the United Arab Emirates. They found that subjective norms, perceived efficacy, and recycling knowledge predicted their recycling intentions [20]. Jain et al. surveyed participants in India [21]. They found that attitudes, norms, and self-efficacy predicted intentions to recycle waste even after additional variables (e.g., regulatory pressure) were statistically controlled for. Regulatory pressure resembles subjective norms in the TPB. In contrast, Y. Wang et al. found that subjective norms did not significantly predict waste-sorting intentions in China [22].

The research discussed in the preceding has shown some limited inconsistencies, likely due to regional differences, the inclusion of additional variables, and differences in measurements. However, the overall patterns are clear. Furthermore, a meta-analysis of the theory in environmental research has confirmed that, in general, attitudes are the strongest predictor, followed by self-efficacy and then norms [10]: The pooled effects of attitudes, self-efficacy, and subjective norms on behavioral intentions were 0.30, 0.30, and 0.15, respectively. These three variables explained 55% of the variance in intentions [10]. Greaves et al. found that the TPB constructs explained 46% to 61% of the variance in intentions to perform three environmental behaviors [23]. That is, collectively, the TPB is useful in predicting various social and personal behaviors, including recycling behaviors. Thus, the following hypothesis is proposed:

Hypothesis 1 (H1). 

Chinese residents’ (a) attitudes toward waste sorting, (b) perceived norms, and (c) self-efficacy predicted their intentions to sort domestic waste.

1.2. Extended Version of the TPB: Antecedent Variables

Fishbein and Ajzen proposed an extended version of the theoretical framework [7]. It specifies that distal variables such as media or marketing techniques can predict attitudes, norms, and self-efficacy and indirectly predict behavioral intentions and behaviors (Figure 1). As such, social policies and marketing strategies’ role in facilitating intent to sort garbage is indirect and mediated by attitudes, subjective norms, and self-efficacy. However, the exact relationships between the distal variables and the core TPB variables were not specified by Fishbein and Ajzen because there are many such variables. As a result, this research will draw on the literature to specify such relationships (explained below).

1.2.1. Social Marketing Strategies: Benefits/Ways of Waste Sorting and Consequences of Noncompliance

Social marketing materials in China, such as signage or pamphlets, emphasize two aspects: The benefits and ways of sorting waste and the consequences of noncompliance [24,25]. These materials reflect the nature of waste sorting and government policies. First, waste sorting is associated with social good, including positive environmental outcomes. Second, a government’s policies, such as mandates and tax incentives, can influence individuals’ behaviors [26]. Many cities in China enforce a fine for noncompliance ranging from RMB50 [27] to RMB2000 [28] depending on the locality and the severity of the noncompliance. Such mandates (e.g., fines) enforce that it is also individual responsibility (i.e., norms) to sort waste before disposal. For example, Zheng et al. found that penalties for noncompliance were significantly associated with waste sorting in Chinese households [19]. Y. Wang et al. found that self-assessed effects of environmental regulations were associated with Chinese residents’ intention to sort garbage [22]. Thirdly and importantly, promotional materials guide how to sort waste [29]. For example, videos or signages clearly show the types of refuse to be accepted by different types of bins, and community volunteers and monitors assist residents with their waste sorting.

That is, social marketing strategies entail the beneficial consequences of waste sorting, the consequences of noncompliance, and how to sort garbage. According to Fishbein and Ajzen [7], these strategies can influence individuals’ waste-sorting intentions via participants’ attitudes toward waste sorting (i.e., the outcome of waste sorting: Benefits and punishments), norms (i.e., what is expected), and self-efficacy (e.g., how to sort and how to sort household waste; that is, perceived confidence in performing a behavior). Z. Wang et al. found that information publicity was indirectly associated with intentions via perceived norms and recycling attitudes [30]. The same research group found that government stimulus (e.g., publicity) was associated with perceptions of accessibility of facilities, hence the ease of sorting and disposing of garbage (i.e., self-efficacy) [31]. Thus, we propose hypothesis 2:

Hypothesis 2 (H2). 

Social marketing strategies in promotional campaigns are positively associated with participants’ (a) attitudes toward waste sorting, (b) perceived norms, and (c) self-efficacy.

1.2.2. Communication Design

One stream of advertising research has shown that communication design (i.e., how well a message is designed) is positively associated with persuasion outcomes (i.e., attitude or behavioral change). Similarly, how well social marketing materials are designed can be associated with persuasion outcomes. Unlike strategies focusing on the reasons and ways to sort waste, communication design focuses on the design aspect. One stream of research on the effects of persuasive messaging focuses on A_ad. A_ad refers to “a predisposition to respond in a favorable or unfavorable manner to a particular advertising stimulus during a particular exposure occasion” [32]. That is, A_ad is one’s evaluation of a message, whether a message is good or bad. It was often measured by items such as this ad/message is “favorable,” “interesting,” “credible,” “convincing,” and “informative” [33,34].

The idea that communication design (and hence the evaluation of it or A_ad) is associated with persuasion is not new. Research in this area has shown that A_ad can be positively associated with brand attitudes (i.e., whether a brand or a product is good) and purchase intentions. Similarly, such an evaluation is relevant in social marketing [34]. Previous research theorizes and provides evidence for the pathways between the two: A_ad was positively associated with attitudes toward a brand (or an issue/behavior in social marketing), which was, in turn, associated with behavioral intent [33]. The elaboration likelihood model provides a theoretical explanation for the role of A_ad or communication design [35]. The model states that individuals are persuaded via two routes: The central route (e.g., the content of an argument) and the peripheral route (e.g., the design of a message or the presenter). Those with abilities and time (vs. not) are more likely to be persuaded via a central route. In contrast, those that do not have the abilities or time are more likely to be persuaded via a peripheral route. That is, the affective evaluation of a message can predict persuasion outcomes. Although not related to environmental communication, Huang et al. found that evaluative reactions toward a viral ad were positively associated with cognitive beliefs and were then associated with attitudes toward a brand and behavioral intentions (i.e., sharing a viral video) [36].

For persuasive messaging (e.g., signage or posters) in waste sorting, individuals may form attitudes about recycling, norms, and self-efficacy based on the content in the promotional materials (e.g., communication strategies) via a central route. However, it is also possible that individuals may not always engage in the central processing route: For example, they may read a pamphlet carefully or come across signage in their neighborhood when passing by. That is, individuals will be persuaded by their evaluative reactions to the design of the social marketing materials (e.g., signage or pamphlets). As such, the following hypothesis is proposed.

Hypothesis 3 (H3). 

If participants perceive the messaging (e.g., signage and posters) well designed, they are more likely to develop (a) more favorable attitudes toward waste sorting, (c) perceived norms, and (c) self-efficacy.

1.3. The Addition of Hope

Fishbein and Ajzen state that the TPB is open to additional variables if chosen carefully [7]. Relatively recent theorizing states that hope can facilitate individuals’ behavioral intent and behaviors [37,38] and is linked to environmental behaviors [39,40]. Hope is defined as the emotion that an individual experiences when believing that the future will be better than the present. Those who experience hope feel optimistic, upbeat, and confident about a recommended action [38]. Such uplifting emotions can motivate individuals to form behavioral intentions and perform a behavior. In the case of environmental behaviors (e.g., recycling), Ojala found that hope predicted pro-environmental behavior among a sample of Swedish high school students and young adults [40]. Similarly, Kerret et al. found that Israeli students in grades 5–6 in the environmental hope programs reported a higher level of pro-environmental behavior [41]. Maartensson and Loi reported a similar finding between hope and pro-environmental behavior [42]. Based on such evidence, H4 is proposed.

Hypothesis 4 (H4). 

Participants’ hope feelings are positively associated with intentions to sort garbage.

In terms of hope appraisal, hope is a goal-congruent emotion whereby (a) the expected outcome will be congruent with the preset goal and (b) the expected outcome is possible but has not been realized [38]. Because individuals want to obtain rewards and avoid punishment, perceived benefits associated with behavior will bring hope [43]. In contrast, perceived punishment may be negatively associated with hope. For example, Wu et al. analyzed social media posts using a data mining approach. They found that penalties were associated with negative sentiments toward waste sorting on social media [44]. In addition, individuals experience hope when confident in performing a behavior. Therefore, based on the theorizing on hope, the positive content characteristics (e.g., benefits) and the negative content characteristics or policies have differential relationships with hope.

Hypothesis 5 (H5). 

(a) Benefits of waste sorting and (b) communication design are positively associated with hope.

Hypothesis 6 (H6). 

Consequences of noncompliance are negatively associated with hope.

Furthermore, this present analysis will explore whether the above communication variables, the core TPB variables, and hope predict the actual waste-sorting behavior.

2. Method

2.1. Procedure and Sample

The data were collected through an online survey via Wen Juan Xing (WJX.com), a Chinese online survey platform and a survey sample panel provider, on 15 and 16 March 2021. WJX states that its online panels consist of 6.2 million members recruited from those who previously completed surveys posted on its website. Panel members were 18 years and older. This was a non-probability opt-in panel similar to most online panels. The present research did not specifically ask the head of household to complete the questionnaire because all adults are responsible for sorting waste.

The final sample consisted of 459 cases, excluding those who did not complete the survey or failed WJX’s attention-checking questions. The final response rate was 28.7%. The present research used Fritz and MacKinnon’s table to determine the required sample size for percentile-based bootstrapping mediation analysis [45]. A small effect size (0.14) of a relationship between an antecedent variable (e.g., communication strategy) and a mediating variable (e.g., attitudes) and a small-medium effect size between a mediating variable to the behavioral intention required a sample size of approximately 412.

Table 1. Socio-demographic characteristics of the sample.

Variable	Statistic
Average age (SD)	31.7 (8.3)
Average annual income (SD)	RMB100,070 (RMB65,110)
Average year of formal education (SD)	13.8 (4.5)
Gender
Female	55.3%
Male	44.7%
Ethnicities (%)
Han	98.0%
Other ethnicities	2.0%
Employment status or occupation
Students	10.9%
Management personnel	15.7%
Administrative/support personnel	10.5%
Research & development	14.6%
Sales	10.5%
Accounting or finance	6.3%
Other	4.6%
Not employed	1.1%

Note. N = 459.

presents the sample characteristics.

2.2. Questionnaire

The present research and analysis used part of the data collected via the survey. Participants completed the questionnaires in Chinese. Responses ranged from 1 (strongly disagree) to 7 (strongly agree) unless noted otherwise. The items below were submitted to confirmatory factor analysis. Results showed that the data fit the final model well; that is, the factor structure was satisfactory: Satorra–Bentler χ²(278, N = 459) = 759.4, p < 0.001, root mean square error of approximation (RMSEA) = 0.061, 90% CI of RMSEA [0.056–0.067], and comparative fit index = 0.98, standardized root mean square residue = 0.055. Standardized factor loadings for the indicators ranged from 0.58 to 0.91, providing support for the construct validity of the measures.

Social marketing strategies (i.e., benefits and ways of waste sorting) were measured by the following items constructed based on a pilot analysis of signage, promotional materials, and waste-sorting policies. Factor analysis showed that items used to measure social marketing strategies loaded on two factors: Consequences of noncompliance and benefits of waste sorting/how to sort garbage.

The benefits and ways of waste sorting formed one scale: “The published materials showed the environmental benefits of waste sorting”, “the published materials showed how to sort waste”, and “the signage showed the environmental benefits of waste sorting”, and “the signage showed how to sort waste”. The alpha coefficient was 0.82.

The consequences of noncompliance were measured by two items: “The published materials showed the individual consequences of noncompliance (e.g., a fine),” and “the signage showed the individual consequences of noncompliance (e.g., a fine)”. The alpha coefficient was 0.82.

Attitudes were measured by five items adapted from Fishbein and Ajzen [7]: “Sorting waste before disposal is important/wise/beneficial/convenient/pleasant”. The alpha coefficient was 0.74.

Norms were based on four items constructed based on Fishbein and Ajzen’s definitions of both subjective and descriptive norms [7]: “Regarding sorting waste before disposal, my family/people in my neighborhood expect me to do so”, and “regarding sorting waste before disposal, my family/people in my neighborhood do so”. These items loaded on one factor. The alpha coefficient was 0.90.

Self-efficacy was measured based on four items reflecting both individuals’ confidence in performing a behavior and external factors that help facilitate the behavior [7]: “Regarding sorting waste before disposal, I know how to sort waste correctly,” “for me, it is easy to sort waste”, “there are convenient locations for me to sort and dispose of waste”, and “there are clear signs near waste disposal”. The alpha coefficient was 0.79.

Hope was measured by four items [43]. “Thinking waste sorting, if the majority of the people do so, I would feel hopeful”, “…, I would feel optimistic”, “…, I would think of the positives”, and “…, I would feel upbeat and better”. The alpha coefficient was 0.72.

Intentions to sort waste were measured by three items adapted from Conner and Sparks [46]: “I will sort waste before disposal”, “I plan to sort waste before disposal”, and “I intend to sort waste before disposal”. The alpha coefficient was 0.77.

Waste-sorting behavior was calculated by obtaining the percentage of times a participant sorted waste in the previous two weeks divided by the total number of times they did and did not sort waste before disposal. The use of percentages signifies behavioral adherence and consistency and is an important behavioral measure. In this case, the use of percentages was also more meaningful than the absolute number of correctly sorting waste or how often an individual correctly sorted waste. Note that a large family needs to dispose of waste more often than a small family. It is possible that the large family inconsistently sorts waste yet registers higher on the absolute number of times of correctly sorting waste than a small family that sorts waste all the time.

Finally, variables related to socio-demographics, political philosophy, and location (e.g., pilot cities or not) were measured and controlled for.

3. Results

3.1. Main Analysis

Sample characteristics are presented in Table 1. Means, standard deviations, and Pearson correlations are presented in the Supplementary Materials (Table S1). Hayes’ PROCESS MACRO (model 4) and the percentile-based bootstrapping method with 5000 samples were adopted for the mediation analysis. Model 4 allows for the examination of mediating relationships (A -> B -> C) and the inclusion of multiple antecedent variables (A) and up to 10 parallel mediating variables (B). The analysis controlled for sex, education, annual income, and political philosophy. In addition, social marketing strategies (i.e., consequences of noncompliance and benefits/ways to sort waste) and communication design were used as antecedent variables, attitudes, norms, self-efficacy, and hope as mediators, and intentions to sort waste before disposal as the dependent variable. Unstandardized coefficients, along with 95% confidence intervals, are reported below. Additional information is presented in

Table 2. Predicting attitudes Toward waste sorting, norms, self-efficacy, hope, and waste sorting intentions: Direct relationships.

	Attitudes toward Waste Sorting			Norms			Self-Efficacy			Hope			Intention
Predictor	B	SE	β	B	SE	β	B	SE	β	B	SE	β	B	SE	β
Control variable
Sex (1 = male, 2 = female)	0.02	0.06	0.01	0.01	0.08	0.01	−0.01	0.08	−0.01	0.06	0.06	0.04	0.10	0.05	0.06 *
Age	0.00	0.00	−0.04	0.00	0.01	0.00	0.00	0.00	0.01	0.00	0.00	−0.05	0.00	0.00	0.00
Income (1 = RMB10,000)	0.00	0.00	−0.01	0.02	0.01	0.12 **	0.02	0.01	0.13 **	0.01	0.00	0.08 *	0.00	0.00	0.04
Education (year)	0.00	0.01	0.02	0.00	0.01	0.01	0.00	0.01	0.01	0.00	0.01	0.03	0.00	0.01	0.02
Personal philosophy (1 = conservative, 7 = liberal)	0.04	0.03	0.05	0.04	0.04	0.04	0.09	0.04	0.10 *	0.02	0.03	0.04	0.01	0.02	0.01
Antecedent
Consequences of non-compliance	0.03	0.02	0.06	0.16	0.03	0.21 ***	0.07	0.03	0.10 *	0.01	0.02	0.01	−0.01	0.02	−0.02
Benefits and ways of waste sorting	0.19	0.03	0.26 ***	0.18	0.05	0.17 ***	0.15	0.05	0.16 ***	0.20	0.03	0.30 ***	0.05	0.03	0.06 +
Communication design	0.28	0.04	0.35 ***	0.37	0.05	0.32 ***	0.34	0.05	0.32 ***	0.23	0.04	0.30 ***	0.00	0.03	0.00
Mediator
Attitudes toward waste sorting													0.31	0.05	0.27 ***
Perceived norms													0.28	0.03	0.35 ***
Self-efficacy													0.18	0.04	0.20 ***
Hope													0.14	0.05	0.11 **

Note. N = 459. Attitudes: R² = 0.348, F(8, 450) = 3.1, p < 0.001; Norms: R² = 0.378, F(8, 450) = 34.2, p < 0.001; Self-efficacy: R² = 0.319, F(8, 450) = 26.3, p < 0.001; hope: R² = 0.324, F(8, 450) = 26.9, p < 0.001; Intentions: R² = 0.678, F(12, 448) = 78.3, p < 0.001. + p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001.

and

Table 3. Predicting waste-sorting intentions in China: Indirect and total relationships.

	Indirect Relationship via												Total Relationship
	Attitudes			Norms			Self-Efficacy			Hope			Intention
Predictor	B	SE	β	B	SE	β	B	SE	β	B	SE	β	B	SE	β
Consequences of non-compliance	0.01	0.01	0.02	0.05	0.01	0.07 ***	0.01	0.01	0.02	0.00	0.00	0.00	0.06	0.03	0.09 *
Benefits and ways of waste sorting	0.06	0.02	0.07 **	0.05	0.02	0.06 *	0.03	0.01	0.03 **	0.03	0.01	0.03 **	0.22	0.04	0.26 ***
Communication design	0.09	0.03	0.09 **	0.10	0.02	0.11 ***	0.06	0.02	0.06 **	0.03	0.02	0.03	0.28	0.05	0.30 ***

Note. N = 459. Socio-demographic variables (i.e., gender, age, annual income, year of education, and political philosophy) were controlled for. Only annual income had a significant total relationship with intentions to sort garbage (B = 0.01, β = 0.11, p = 0.008). * p < 0.05, ** p < 0.01, *** p < 0.001.

and Figure 2.

For H1, Chinese residents’ intentions to sort garbage were positively predicted by their attitudes toward waste sorting (B = 0.31, p < 0.001, 95% CI [0.21–0.41]), their perceived norms (B = 0.28, p < 0.001, 95% CI [0.22–0.35]), and their perceived self-efficacy (B = 0.18, p < 0.001, 95% CI [0.11–0.25]).

For H2, the consequences of noncompliance positively predicted perceived norms (B = 0.16, p < 0.001, 95% CI [0.10–0.22]) and self-efficacy (B = 0.07, p = 0.026, 95% CI [0.01–0.13]), but not attitudes toward waste sorting.

Benefits/how to sort garbage was positively associated with attitudes toward waste sorting (B = 0.19, p < 0.001, 95% CI [0.12–0.26]), perceived norms (B = 0.18, p < 0.001, 95% CI [0.08–0.27]), and self-efficacy (B = 0.15, p < 0.001, 95% CI [0.06–0.25]).

For H3, communication design positively predicted attitudes toward waste sorting (B = 0.28, p < 0.001, 95% CI [0.20–0.36]), perceived norms (B = 0.37, p < 0.001, 95% CI [0.26–0.47]), and self-efficacy (B = 0.34, p < 0.001, 95% CI [0.23–0.44]).

For H4, hope positively predicted Chinese residents’ intentions to sort garbage (B = 0.14, p = 0.003, 95% CI [0.05–0.24]).

For H5 and H6, communicating benefits and ways to sort waste (B = 0.20, p < 0.001, 95% CI [0.13–0.26]) and communication design (B = 0.23, p < 0.001, 95% CI [0.15–0.30]) positively predicted hope, whereas communicating the consequences of noncompliance was not associated with the intention to sort garbage (B = 0.01, p = 0.745, 95% CI [−0.04–0.05]).

Based on Hayes’ PROCESS for mediation analysis, the results showed that the total effects of consequences of noncompliance on behavioral intentions to sort garbage was 0.06 (p = 0.037, 95% CI [0.00–0.11]). The effects were mediated by norms (B = 0.05, p < 0.001, 95% CI [0.02–0.08]) and self-efficacy (B = 0.01, p = 0.061, 95% CI [0.00–0.03]).

The total effects of the benefits of sorting garbage and how to sort garbage on behavioral intentions to sort garbage were 0.22 (p < 0.001, 95% CI [0.14–0.30]). The effects were mediated by attitudes (B = 0.06, p = 0.001, 95% CI [0.03–0.10]), norms (B = 0.05, p = 0.013, 95% CI [0.02–0.10]), self-efficacy (B = 0.03, p = 0.042, 95% CI [0.01–0.06]), and hope (B = 0.03, p = 0.041, 95% CI [0.00–0.06]).

The total effect of communication design on behavioral intentions to sort garbage was 0.28 (p < 0.001, 95% CI [0.19–0.37]). The effects were mediated by norms (B = 0.10, p < 0.001, 95% CI [0.06–0.16]) and attitudes (B = 0.09, p = 0.002, 95% CI [0.04–0.15]), followed by self-efficacy (B = 0.06, p = 0.003, 95% CI [0.03–0.10]), and hope (B = 0.03, p = 0.035, 95% CI [0.01–0.07]).

3.2. Additional Analysis of Waste-Sorting Behavior

The relationship between intent and the percentage of times that participants sorted waste (divided by the total number of times of garbage disposal) was 0.52.

Table 4. Predicting waste-sorting behaviors in China: Indirect and total relationships.

	Percentage of Waste Sorting
	Direct Relationship			Total Relationship
	B	SE	β	B	SE	β
Control variable
Gender (1 = male, 2 = female)	0.00	0.02	0.00	0.00	0.03	−0.01
Age	0.00	0.00	0.05	0.00	0.00	0.05
Annual income (1 = RMB10K)	0.01	0.00	0.11 **	0.01	0.00	0.19 ***
Year of education	0.00	0.00	0.03	0.00	0.00	0.04
Personal Philosophy	−0.01	0.01	−0.03	0.00	0.01	0.01
City (1 = pilot city, 0 = others)	−0.02	0.02	−0.04	−0.03	0.03	−0.06
Number of times corrected by trash monitors	−0.02	0.01	−0.08	−0.01	0.01	−0.04
Antecedent
Consequences of noncompliance	0.01	0.01	0.03	0.03	0.01	0.13 *
Benefits and ways of waste sorting	0.01	0.01	0.03	0.03	0.02	0.12 *
Communication design	−0.01	0.02	−0.04	0.04	0.02	0.14 *
Mediator
Attitudes	0.00	0.02	0.01
Norms	0.11	0.02	0.40 ***
Self-efficacy	0.08	0.02	0.28 ***
Hope	−0.05	0.02	−0.11 *

Note. N = 459. R² = 0.388, F(14, 444) = 19.3, p < 0.001. * p < 0.05, ** p < 0.01, *** p < 0.001.

shows that norms and self-efficacy were the most important predictors of the percentage of waste-sorting behaviors. Regarding the total effects of demographic and antecedent variables, annual income was predictive of the percentage of times of waste sorting before disposal. The benefits of waste sorting, the consequences of noncompliance, and communication design were predictive of the percentage of times of waste sorting.

4. Discussion

The present research examined the factors that predicted Chinese residents’ intentions to sort garbage before disposal and their garbage disposal behavior. Furthermore, it examined the role of social marketing strategies (i.e., benefits of waste sorting and consequences of noncompliance) and communication design in Chinese residents’ waste-sorting intentions before disposal. This research specifies how social marketing strategies and communication design relate to participants’ behavioral intentions.

Chinese residents’ intentions to sort waste were primarily predicted by their attitudes toward sorting garbage, subjective norms, and self-efficacy. The results were generally consistent with the research discussed in the literature review [10]. The main difference is that personal norms were a stronger predictor of waste-sorting intentions, and self-efficacy was a less strong predictor than those in Klöckner [10]. The reasons can be attributed to the collectivist cultural values in China, whereby certain public behaviors are expected to conform to the norms. Self-efficacy is less important due to the availability of disposal bins in the neighborhood.

Comparatively speaking, hope was a less strong predictor of the participants’ intentions to sort garbage before disposal. First, regarding total effects in predicting intentions, communication design and communication strategies for promoting benefits and ways to sort garbage were stronger predictors than communication strategies for promoting the consequences of noncompliance. Second, the Chinese residents’ waste-sorting behaviors were primarily predicted by perceived norms and self-efficacy. Furthermore, the two social marketing strategies and communication design were equally strong (and weak) predictors.

4.1. Theoretical Discussion

First, the present research extended the theoretical framework by examining the role of distal variables and mediating relationships. Communicating the consequences of noncompliance was related to behavioral intentions via perceived norms and less so by self-efficacy. The relationship between communicating noncompliance and perceived norms was reasonable, and the consequences of noncompliance were part of normative expectations. The relationship between the consequences of noncompliance and self-efficacy is weak and tricky, and no prior research has revealed this finding. Such a relationship may also be due to statistical artifacts and should be confirmed or disconfirmed in future research.

On the other hand, communicating the benefits and ways to sort garbage is more important in predicting intentions than communicating noncompliance. The indirect relationships with intentions were mediated via attitudes, perceived norms, and self-efficacy, whereas the indirect relationship with behavior was mediated via perceived norms and self-efficacy. The paths in Figure 2 and the total relationships were of relatively strong effect sizes. The results provided evidence for the need to communicate the positives and ways to sort garbage, in addition to the relatively modest role of communicating the consequences of noncompliance to facilitate behavior among the target audience.

Second, rooted in the literature of A_ad [32], better communication design is associated with stronger behavioral intent and behavior. A_ad specifies the path to brand attitudes and purchase intention. Although not identical, attitudes toward waste sorting are conceptualized and measured similarly to attitudes toward a brand or a product: Both focus on the participants’ general attitudes toward an attitude object (i.e., a behavior or a brand). Those with a more favorable evaluation of a message may like a brand or product better because of the transfer of positive feelings toward an associated product or behavior. Unlike previous research and theorizing in A_ad, the present research further found paths from communication design to perceived norms and self-efficacy. There are two possible reasons: First, better-designed communication messages can communicate expectations/norms and ways to sort garbage, which leads to stronger norms and self-efficacy. Second, those in a low-involvement situation (e.g., either lack of time or ability to process messages) are more likely to be persuaded by a peripheral cue (e.g., a well-designed message).

Thirdly, the present research aimed to extend the reasoned action theoretical framework by adding hope. The use of hierarchical multiple regression analysis, where hope was entered in the last step, showed that hope added a relatively modest amount (0.01%) of the variance in intentions (p = 0.009). The results showed that hope is weakly associated with behavioral intentions. Thus, the results indicate that for predicting waste-sorting intentions, the traditional reasoned action theoretical variables are more important than hope. However, although a less preferred statistic compared to multiple regression, the Pearson correlation between hope and intentions was 0.58, a large effect size. Given that hope was a moderately strong predictor of other behaviors [43], it might be premature to dismiss the role of hope. Future research should further examine the role of hope in environmental psychology and communication.

Lastly, an additional exploratory analysis examined the post-prediction of waste-sorting behaviors in the previous two weeks. The role of norms and self-efficacy showed similar predictive power of waste-sorting intentions. On the other hand, communication strategies and design showed weaker predictive power. The effects of antecedent variables on the downstream variables (e.g., behavior) would be less strong. On the other hand, annual income predicted the percentage of times that the Chinese participants sorted garbage before disposal, indicating the need to consider some background variables such as income, economic situations, and the types of residential neighborhoods.

4.2. Practical Implications

A subsection of the United Nations’ Sustainable Development Goal 11 calls on local and regional governments to share their experience in promoting sustainability among “a wide range of stakeholders” and facilitating “bottom-up and inclusive processes” [4]. Scholars and researchers should also help evaluate the success and failure of social marketing campaigns in promoting sustainability among many stakeholders and residents, thus providing lessons for future campaigns.

First, this research provides evidence for the role of the TPB variables in promoting effective social marketing campaigns to encourage waste sorting. Social marketing initiatives must first identify why individuals perform or do not perform a behavior (e.g., waste sorting). This research helps identify the variables necessary to promote waste sorting in China. Suppose further research confirms the findings in this research. In that case, social marketing campaigns should enhance the residents’ perceived norms, attitudes, and self-efficacy to enhance their behavioral intentions and behaviors. The means for norms, attitudes, and self-efficacy were 5.56 (SD = 1.09), 5.85 (SD = 0.76), and 5.53 (SD = 0.99), indicating that there is room for change. Of note, because perceived norms and self-efficacy were relatively strong or moderate predictors of intentions and behaviors, they should be the focus for promotion. It is tricky that although hope positively predicted intentions, it negatively predicted percentages of times correctly sorting garbage before disposal. Thus, the effects on behavior were canceled out. Future research should further consider the role of hope in facilitating waste-sorting intentions and behaviors.

More importantly, the present research has showcased the importance of two social marketing strategies and communication design in predicting attitudes, norms, and self-efficacy. First, social marketing campaigns should consider communicating the benefits and ways of waste sorting (strategy 1) in mediated promotional materials and interpersonal communication. Such a strategy was positively associated with intentions to sort garbage via attitudes, norms, and efficacy. The present research examined such a strategy in both forms of communication, whether mediated or interpersonal. Second, communicating the consequences of noncompliance did not result in negative attitudes toward waste sorting. Instead, it primarily resulted in higher perceived norms about what should be performed. Furthermore, communicating the consequences of noncompliance (strategy 2) appears to be equally important as communicating benefits in predicting behavior, although less important in predicting intentions. Lastly, social marketing campaigns should strive to create more professionally designed and clear messages. Such messages are easier to understand and can facilitate the transfer of good feelings toward the messages to intentions and behavior to sort garbage. As such, future campaigns should distribute well-designed and -written messages to residents, including info-graphic and visually appealing posters. Such messages can communicate the content better and translate positive feelings toward the messages into attitudes and behaviors.

Observing the associations between communication strategies/design and waste-sorting intentions and behaviors is promising. However, other implications should be noted. First, waste sorting and many other environmental behaviors can be guided by other beliefs and values, such as environmental concerns and moral obligations [14]. Such values are generally not open to change within a short time. Second, decades of research on the diffusion of innovations have shown that the media are important in raising awareness. In contrast, interpersonal communication via opinion leaders is often more effective in facilitating adoption. Taken together, social marketing planners should identify individuals with strong environmental beliefs (i.e., opinion leaders) and encourage them to help others and act as role models to show how to sort waste and its associated benefits. Such actions can instill self-efficacy in the target audience (i.e., by showing that waste sorting can be performed easily), normative expectations (e.g., it is the community’s collective action), and positive attitudes toward waste sorting (e.g., the behavior is indeed pleasant and may help connect with other community members). Third, local governments and waste management should make waste sorting easy by providing bins within walking distance. Without easy access to garbage bins, education would not have achieved the desired outcome. Social marketing campaigns should aim to change micro-level thinking and behaviors and macro-level waste-sorting accessibility.

4.3. Limitations

The first limitation is related to the method of data collection. The data were collected via an internet survey. The sample was not presentative of the Chinese population; the participants were more educated and younger and had a higher income. However, waste sorting and recycling are less likely to be enacted in rural areas. Based on the participants’ IP addresses, they were from cities and towns where waste sorting and recycling have been and are likely being enacted. Second, the present research reported results from a cross-sectional survey. Thus, this research cannot establish cause–effect relationships. Additional longitudinal or quasi-experimental designs can be employed. Thirdly, content characteristics were measured by six items and formed two factors based on factor analysis. In particular, the benefits of waste sorting and ways to sort garbage formed one factor. Such results appear reasonable because both were related to goal attainment. However, future research might consider using additional items and examining whether benefits and ways to sort garbage can load on different factors and thus provide a more fine-tuned analysis of the role of self-efficacy and benefits of waste sorting.

5. Conclusions

The present research examined the psychological factors and content characteristics related to waste-sorting intentions and behavior. Such relationships enrich the reasoned action theoretical framework by adding distal variables and providing a more detailed account of message design and behavioral intention than the attitude toward the ad theoretical framework. The results provide a detailed account of the content characteristics of a promotional campaign and the residents’ behavioral intentions. These relationships also pinpoint the potential areas for future environmental campaign design. Such insights are effective and easy to implement. In particular:

• Communicating benefits and ways of waste sorting and consequences of noncompliance can be positively associated with residents’ waste-sorting intentions and behaviors.
• Better and more professionally designed promotional materials can facilitate residents’ waste-sorting intentions and behaviors.