Investigation on decision-making mechanism of residents’ household solid waste classification and recycling behaviors

Resources, Conservation & Recycling 140 (2019) 224–234

Contents lists available at ScienceDirect

Resources, Conservation & Recycling journal homepage: www.elsevier.com/locate/resconrec

Full length article

Investigation on decision-making mechanism of residents’ household solid waste classification and recycling behaviors ⁎

T

⁎

Xiaoyan Menga, Xianchun Tana,d, Yi Wanga,d, Zongguo Wenb, , Yuan Taoc, , Yi Qianb a

Institutes of Science and Development, Chinese Academy of Sciences, Beijing 100190, China State Key Joint Laboratory of Environment Simulation and Pollution Control (SKLESPC), School of Environment, Tsinghua University, Beijing 100084, China c Institute for Manufacturing, Department of Engineering, University of Cambridge, Cambridge CB3 0FS, United Kingdom d School of Public Policy and Management, University of Chinese Academy of Sciences, Beijing 100049, China b

A R T I C LE I N FO

A B S T R A C T

Keywords: Household solid waste Separation and recycling behavior Structural equation model Social survey Situational factors

Residents’ participation in classification and recycling of urban household solid waste (HSW) is a critical factor for the success of municipal solid waste management. The aim of this study is to investigate the decision-making mechanism of residents’ HSW disposal behaviors by merging the theory of planned behavior and the AttitudeBehavior-Condition theory. In this study, based on the survey data of 709 residents in Suzhou, China and structural equation modeling method, the main factors that affect residents’ HSW disposal behaviors and their degree of influence were analyzed, followed by discussion on decision-making mechanisms. The findings show that residents’ behavioral selection has been significantly related to four intrinsic subjective factors and seven external objective factors, and the combined effect of the latter ones is nearly twice of that of the former ones. Moreover, the convenient of environmental facilities and services are most effective at promoting residents’ participation in HSW classification and recycling. Specifically, the observed variables of publicity and education, accessibility to recycling facilities, accessibility to classification facilities, willingness to participation of classification and residents' environmental awareness are the five most significant factors. The impact of laws and regulations is not significant; however, this may be because that there was no mandatory laws, regulations and incentive mechanisms on HSW classification and recycling in Suzhou in this period, and there is still a big gap and room for improvement in this aspect in mainland China. Finally, the study put forward relevant policy recommendations for the comprehensive management of urban HSW classification and recycling.

1. Introduction With the development of the economy and the improvement of residents’ living standards, urban household solid waste is increasing rapidly in many countries all over the world. At present more and more governments regard the principles of HSW’s decrement, recycle and harmless as the goals of urban municipal solid waste management. Nothing wrong with that waste source separation and recycling has major potential benefits to an effective management of waste by addressing the problem of landfill shortage and resource savings. However, there is a considerable distance towards achieving its full potential in practice. One of the main reasons is the weak residents’ engagement in waste management policies (e.g. classification, recycling). Since residents' participation in HSW separation and recycling is the key to affecting urban solid waste classification management, then, the question is, how to increase the participation rate of residents?

⁎

This requires a clear understanding of the main influencing factors and decision-making mechanisms of residents’ HSW disposal behaviors (RWDB). Understanding RWDB could enable decision makers and local governments to design more-effective policies for improving waste separation and recycling. As the world's largest developing country, China has already entered a period of rapid urbanization. From 1998–2017, urbanization in China has increased at an average annual growth rate of more than 1%. As of 2017, 58.52% of China’s total population lives in urban areas or cities (National Bureau of Statistics of China, 2018). In the meantime, the rapid industrial development has consumed massive resources and given rise to urban household solid waste (HSW). At present, two thirds of China’s big and medium cities are engulfed in waste, with more than 500 million square meters of land nationwide encroached due to the dumping of household solid waste (Fei et al., 2016). There is no doubt that the classification and recycling of HSW is strategically important

Corresponding authors. E-mail addresses: [email protected] (X. Meng), [email protected] (Z. Wen), [email protected] (Y. Tao).

https://doi.org/10.1016/j.resconrec.2018.09.021 Received 30 June 2018; Received in revised form 17 September 2018; Accepted 19 September 2018 0921-3449/ © 2018 Elsevier B.V. All rights reserved.

Resources, Conservation & Recycling 140 (2019) 224–234

X. Meng et al.

conditions (Tucker et al., 1998). However, much of the research up to now has separately studied individuals’ waste prevention behavior (Bortoleto et al., 2012), waste source separation behavior (Zhang and Wen, 2014), domestic recyclable resource recycling behavior (Fei et al., 2016), etc. Few published studies have been able to draw on any systematic research into HSW disposal behaviors including all possible waste disposal methods simultaneously. There has been no known research on a behavioral decision-making mechanism which considers both residents’ participation in source classification and resource recycling. In addition, previous research on residents’ HSW disposal behavior mainly focuses on willingness to participate in source classification and impact factors, however, studies that consider both internal subjective factors and external situational factors that influence individuals’ waste management behaviors are still rare. In this study, in order to comprehensively understanding the decision-making mechanism of residents’ HSW disposal behaviors, explore the factors have a significant effect on RWDB and the degree of their influence, we divided RWDB into three kinds according to different waste disposal ways selected (non-classification, classification deposition and selling recyclables after classification), and developed an hypothetical model by merging the TPB theory and A-B-C theory. The proposed model can consider both individual subjective factors and external situational factors that may influence the residents' HSW classification and recycling behaviors. Next a questionnaire was designed and the field survey was conducted in the five administrative districts of Suzhou, China. Then the initial hypothetical model was tested using structural equation modeling (SEM) based on the questionnaire data, and the significant paths and better indexes of fit were obtained through model evaluation and correction. Finally, the main influencing factors of RWDB and the sensitivity coefficient corresponding to each factor were explored, moreover, the decision-making mechanism of residents and policy suggestions were discussed. The results of this study could provide a theoretical support for policy formulation on urban household solid waste classification and recycling in China and other countries in the word. The initial research hypotheses and conceptual models are organized in Section 2 based on previous literature reviews. Section 3 introduces the research methodology for data collection and structural equation modeling. Section 4 presents data analysis, model testing and results. The discussions and policy recommendations are conducted in Sections 5, and 6 covers the conclusions.

for alleviating resource and environmental restrictions. As early as 2000, China started to carry out pilot programs for waste separation and recycling in eight cities including Beijing, Guangzhou, and Shanghai (Meng et al., 2018). Since 2016, waste classification has been elevated to an unprecedented high level. President Xi Jinping specifically pointed out that it is necessary to introduce the waste classification system to more areas when he presided over the Central Leading Group on Financial and Economic Affairs in December 2016. The National Development and Reform Commission and the Ministry of Housing and Urban-Rural Development released the “Implementation Plan on the Household Solid Waste Classification System” on March 18, 2017, requiring that 46 cities nationwide take the lead to implement mandatory classification of household solid waste and that the recycling rate of household solid waste exceed 35% by 2020. However, the present management of household solid waste classification and recycling is not satisfactory. In pilot cities for household solid waste classification, the participation rate of household solid waste classification is still low, and there is no substantial progress made in waste reduction (Yan, 2018). It is a pressing problem to cultivate residents’ habits of source classification and resource recycling, and improve their participation in waste classification and recycling in the comprehensive management of urban household solid waste. Therefore, it is theoretically and instructively significant to tap into the main impact factor and decision-making mechanism and formulate targeted policies to improve residents’ participation in household solid waste classification and recycling. There have been some research conducted to explain why individuals may or may not engage in waste management policies, such as waste prevention, source separation and littering (Bortoleto et al., 2012; Abdelradi, 2018; Wang et al., 2018). In summary, there are two classes of theoretical methods for research on residents’ environmental behavior and choices at home and abroad (Jackson, 2005), one is the research method based on environmental sociology, and the other is the research method based on environmental psychology. The first method starts from the interaction between micro-individuals and socio-environmental systems and considers that an individual’s ideas and behavioral choices are determined by the process and status of social and technological development (Singh et al., 2018). The second method mainly considers the effect of irrational factors on individual behavior. The most common theory of planned behavior (TPB) falls into the first category. Ajzen (1985) developed TPB based on the theory of reasoned action, which emphasizes that an individual’s behavior is influenced by attitude, subjective norm, and perceived behavioral control (Ajzen, 1985). Many scholars have studied waste classification and recycling behavior with the TPB theory (Botetzagias et al., 2015; Gao et al., 2017; Lizina et al., 2017). For example, Nguyen et al. (2015) find that personal ethics are a significant impact factor in promoting residents’ behavioral intention to participate in waste classification and recycling; the study by Park and Ha (2014) indicates that residents are encouraged and affected when they see their neighbors or friends classify and recycle waste. Although the TPB theory inspires studies on residents’ recycling behavior, its model framework has strong limitations. The TPB theory mainly considers intrinsic factors; however, other factors also affect the process when behavioral intentions turn into behavior (Boldero, 1995). Stern and Oskamp (1987) constructed a complex environmental behavior model, proposing that environmental behavior is the result of related external factors and intrinsic factors working together. Based on this, Guagnano et al. (1995) proposed the Attitude-Behavior-Condition (A-B-C) theory, which states that individuals’ behavior (Behavior, B) results from the combined effect of residents’ attitudes (Attitude, A) and external conditions (Condition C), and considers that external conditions are crucial factors in determining whether residents perform waste recycling behavior. Tucker further refined the model and proposed a research model in which residents’ HSW disposal behavior is determined by attitude, subjective norms, social norms, and external

2. Initial research hypotheses and conceptual models In order to propose reasonable hypotheses of the initial measurement model, we searched and summarized a large number of prior research on the impact factors of residents’ waste disposal behavior. In recent years, domestic and foreign scholars have conducted relevant research on it (Boonrod et al., 2015; Borthakur and Govind, 2017; Guo et al., 2016). Based on literature reviews, this paper summarizes factors of frequent occurrence and with a significant effect. Previous studies have shown that residents’ HSW disposal behaviors may be affected indirectly by four main aspects, namely, environmental attitudes, social norms, environmental knowledge, publicity and education, environmental facilities and services (situational factors). Further, by combining the Theory of Planned Behavior and A-B-C Theory, this study puts forward an initial conceptual and measurement model on the decision-making of RWDB, together with its indicated hypotheses (see Fig. 1). The definitions of individual hypotheses regarding RWDB are as follows. In the initial model, the study assume that the four latent variables of “environmental attitudes (EA)”, “social norms (SN)”, “environmental knowledge, publicity and education (EP)” and “environmental facilities and services (EF)” have path effect on RWDB (H1,H2, H3 and H4), and each observed variable is reflected by several observed variables. The paper has set up 15 possible observed variables based on the literature 225

Resources, Conservation & Recycling 140 (2019) 224–234

X. Meng et al.

Fig. 1. The initial conceptual and measurement model of urban residents’ HSW disposal behaviors.

are internalized in the individual’s own value system, behaviors that meet or oppose the norm can lead to self-esteem or guilt. Previous studies show that social norms have a significant correlation with residents' waste recycling behavior (Chu et al., 2013；Botetzagias et al., 2015). Existing research focuses on four major aspects: (1) The first is citizens’ social responsibility. Nguyen found that an individual’s sense of social responsibility and ethics, that is, residents think that waste recycling is a good thing for the public and themselves, is an important factor affecting their participation in recycling (Nguyen et al., 2015). (2) The second is constraints of laws and regulations. Wan and other scholars believe that enactment of corresponding laws and regulations has a positive effect on the residents’ environmental behavior, and it helps increase residents' participation in waste separation and recycling (Wan et al., 2014；Timlett and Williams, 2008). A social survey in Hong Kong (Wan et al., 2015) shows that the government implements “carrot and stick” policy measures, that is, incentives and penalties related to waste separation and recycling, and residents will better perceive the binding of policies, which helps to promote waste separation and recycling. (3) Herd behavior effect (the influence from family and neighbors). Park and Ha’s research shows that when residents see their neighbors or peer group classify and recycle waste, they are often driven and affected (Park and Ha, 2014). (4) Social recognition. A survey in Michigan shows that non-economic returns such as social recognition, satisfaction from participating in waste recycling and charity are important factors in promoting residents’ participation in recycling. Therefore, this paper defines social norms as the tendency for residents to adopt certain HSW disposal behavior due to notable peer and social pressure (Deng et al., 2013). We assume that the observed variables “social recognition (SN1)”, “laws and regulations (SN2)”, “citizens’ social responsibility (SN3)”, and “influence of neighbors (herd psychological effect, SN4)” can form and mainly reflect the latent

investigation. 2.1. Environmental attitudes (EA) Studies have shown that there is a significant correlation between environmental attitudes (EA) and residents’ HSW disposal behavior (Ajzen, 1985; Begum et al., 2009; Song et al., 2012). The performance of environmental behaviors can be affected directly by attitudes toward particular actions (Singh et al., 2018; Bortoleto et al., 2012; Tadesse, 2009). At present, there is no clear definition of environmental attitudes. This paper defines environmental attitudes as the general and stable perception or position held by major residents on household solid waste classification and recovery. In this investigation, we use the following four items to reflect residents’ environmental attitudes: residents’ environmental literacy (Ajzen, 1985), willingness to participate (Nguyen et al., 2015), awareness of resource conservation and environmental protection (Wan et al., 2013), and recognition of the necessity of classification and recycling behavior (Li et al., 2015). Therefore, on the basis of the literature review above, the study assume that the observed variables“environmental literacy (EA1)”, “willingness to participate in classification (EA2)”, “environmental awareness (EA3)” and “behavioral attitudes (personal recognition of the necessity of classification and recycling behavior, EA4)” can mainly reflect and have the positive path effect on the latent variable EA. The corresponding path hypothesis are H1a, H1b, H1c and H1d, respectively. 2.2. Social norms (SN) Social norms refer to the pressure of others and the society that exert an important influence on the households’ behaviors, it take the form of approval or disapproval of others, as well as associated feelings of pride or shame (Lindbeck, 1997). In addition, once the social norms 226

Resources, Conservation & Recycling 140 (2019) 224–234

X. Meng et al.

variable SN, and SN1, SN2, SN3 and SN4 have the positive path effect on SN (H2a, H2b, H2c and H2d).

sociological characteristics are not taken as direct observed variables in this study.

2.3. Environmental knowledge, publicity and education (EP)

3. Research methodology

Several studies have revealed that there is a strong positive correlation between publicity efforts and residents’ participation in waste recycling (Grazhdani, 2016; Wang et al., 2018; Xiao et al., 2017). Reddi et al.’s study shows that residents’ environmental knowledge and information is significantly related to their environmental behavior, the lack of related knowledge and information will hinder residents' participation in waste separation and recycling (Reddi et al., 2013). Through public education, residents can better understand the new recycling policy and improve the ability of classification and recycling (Wan et al., 2013; Izagirre-Olaizola et al., 2015). This paper defines environmental knowledge as the knowledge, skills, and information necessary for residents to carry out waste classification and recycling, such as classification methods, recycling channels, locations of recycling sites, and recycling hotlines; it defines publicity and education as what residents receive through the media, advertisements, education at school and other methods on waste classification and recycling. Based on the literature review above, we assumed that the latent variable “environmental knowledge, publicity and education” was mainly reflected by two observed variables: “knowledge and information on classification and recycling (EP1)”, and “publicity and education (EP2)”. Moreover, EP1 and EP2 have positive and direct influence on EP, the corresponding path hypothesis are H3a and H3b.

3.1. Questionnaire design and data collection On the basis of the initial conceptual model proposed (Fig. 1), this study used the five-level Likert scale to design a preliminary questionnaire to investigate the RWDB and their influencing factors. The questionnaire consisted of four parts: (1) background of the investigation, including a brief introduction of this survey; (2) the current situation of households’ waste disposal, including disposal ways, time spent, economic income and methods of selling recyclable waste, etc. In this study, we classified residents’ HSW disposal behaviors into three kinds according to different waste disposal ways selected: (1) nonclassification (mixed disposal), (2) classification deposition (classified and delivered into trash cans according to the method of quartering), (3) selling recyclables after classification (separate recyclables and sell them to recyclable material collectors, and dispose of the rest into trash cans); (3) items measuring the initial research hypotheses and the conceptual model, all measures were reported on 1 to 5 point scale from “Strongly Agree”, “Agree”, “Moderately”, “Disagree” to “Strongly Disagree”; (4) the demographic and social attribute information of the respondents, including gender, age, education level, family monthly income and habitation areas. The data were collected by questionnaire survey. We chose Suzhou as a study area, because there was a favorable foundation for waste classification and recycling in Suzhou, and it had embarked on a pilot program on household solid waste classification and recycling in some residential communities since 2000. In 2010, Suzhou put forward “rough separation in the short term and fine classification in the long run”, a new model for household solid waste classification and recycling (jswmw.com). In April 2015, Suzhou was chosen to be one of the 26 national domestic waste sorting collection pilot cities (bunch 1) by five ministries. In recent years, source classification of household solid waste in Suzhou has achieved steady progress. In 2017, there were more than 400 pilot communities participating in waste classification. First, in order to verify the rationality of the initial questionnaire design, including the structure, questions and options setting, we conducted a pretest via WeChat with nearly 200 survey results retrieved. And the preliminary questionnaire was revised and re-designed according to the results. For example, because we didn't know the family monthly income structure of residents in the sample in advance, the answer range setting for this question was unreasonable, the total range was too large and the grouping interval was not suitable, which made some group cases more concentrated. Therefore, the answer settings for this question was re-adjusted based on the applicable answer distribution characteristics. The final measurement instruments for the latent variables of the hypothetical model are shown in Table 1. Then, the research team conducted the formal field survey among permanent residents in the central areas of Suzhou (including its five administrative districts, i.e. Gusu, Wuzhong, Xiangcheng, Gaoxin District, and Industrial Park) through random sampling. The survey was launched in the areas where residents gather, including eight shopping malls in downtown Suzhou, Tongjing Park, Jinji Lake Plaza, and Guanqian Street. It fully considered the population ratio in each district, as well as the characteristics of distribution of residents of all age groups and with different occupations. The questionnaire survey collected 709 valid questionnaires out of 759 in total, with a valid response rate of 93.8%.

2.4. Environmental facilities and services (EF) Some studies have confirmed that external conditions have an important impact on hindering or promoting residents’ household solid waste recycling behavior (Matsumoto, 2014; Wu et al., 2017). Bach et al.’s study shows that to some extent the informal recycling market makes it more convenient for residents to dispose of waste and promotes residents’ HSW classification and recycling behavior; in the meantime, the increase in the number of recycling sites for renewable resources helps increase the recovery rate (Bach et al., 2004). A research in the United Kingdom (Abbott et al., 2011) confirmed that more recycling sites established near residential areas had resulted in an increase in the frequency of resident recycling. This study also pointed that when the government does not have the right to charge residents enough garbage disposal fees, residents do not actively reduce the waste generated or try their best to participate in waste separation and recycling, indicating that the economic factors will also affect the residents' garbage disposal behaviors. Moreover, the time spent on waste classification and recycling and the storage space used to store household solid waste at home also affect residents’ participation in waste classification and recycling (Liu et al., 2018). This paper uses the latent “environmental facilities and services (EF)” to summarize objective external conditions (situational factors) that affect HSW classification and recycling. We assume that EF may be explained and measured by five observed variables including “Economic cost & benefits (disposal fees and revenues from sales of waste, EF1)”, “Time spent (EF2)”, “accessibility to classification facilities (EF3)”, “occupied storage space (EF4)” and “accessibility to recycling facilities (EF5)”. In addition, EF1, EF3 and EF5 have a positive effect on EF, the corresponding path hypothesis are H4a, H4c and H4e; EF2 and EF4 have a negative effect on EF, the corresponding path hypothesis are H4b, and H4d. In addition to the four aspects above, research has also shown that residents’ knowledge, attitudes, and practices in waste recycling are influenced by demographic and sociological factors such as age, education level, gender, and occupation (Babaei et al., 2015; Song et al., 2012). However, since the influence from different demographic and sociological characteristics of the population is already incorporated in “environmental attitudes” and “social norms”, demographic and

3.2. Measurement model The initial hypothetical model was tested using Structural Equation Modeling (SEM) with the software AMOS 21.0 in this study. SEM was first proposed in the 1970s by Swedish scholars Jöreskog (1970). 227

Resources, Conservation & Recycling 140 (2019) 224–234

X. Meng et al.

Table 1 Measurement instruments for the latent variables of the hypothetical model. Latent variables

Measurement items (observed variables)

Sources

Environmental attitudes (EA)

EA1

What I care about is survival and life issues, not environmental issues such as waste recycling. In order to save resources and protect the environment, I am willing to participate in waste separation and recycling. Waste classification and recycling are conducive to saving resources and turning waste into treasure. I produce less recyclable waste, no need for separate and recycling.

EA2 EA3 EA4 Social norms (SN)

SN1 SN2 SN3 SN4

Environmental knowledge, publicity and education (EP)

EP1 EP2

Environmental facilities and services (EF)

EF1 EF2 EF3 EF4 EF5

I'm satisfied to participate in waste classification and recycling. Waste separation and recycling laws and regulations can play a constraining role for me. Waste classification and recycling are the responsibility of the government and enterprises and have nothing to do with residents. Seeing my neighbors and friends to participate in sorting, I’ll do the same.

Nguyen et al. (2015) Wan et al. (2015) Li et al. (2015) Young (1990) Wan et al. (2014, 2015) Timlett and Williams (2008) Park and Ha (2014)

I have mastered the waste classification method and I know the location of the nearby waste recycling sites. I have been exposed to enough publicity and education on waste sorting and recycling in my daily life.

Reddi et al. (2013)

I sell scrap in order to obtain economic benefits. Waste classification and recycling waste of time. There are classified garbage bins and garbage kiosks in the community, with clear identification and close distances. Sorting and collecting recyclable waste takes up a lot of storage space in my house. Waste recycling sites are close to home, and the service of recyclers is good.

Abbott et al. (2011) Bach et al. (2004) Guagnano et al. (1995)

Grazhdani (2016)

Matsumoto (2014) Izagirre-Olaizola et al. (2015)

4. Data analysis and results

Structural equation models are a family of multivariate statistical models that allow the analyst to estimate the effect and relationships between multiple variables (DellöOlio et al., 2018). The model analyzes the relationship between variables based on the covariance matrix of the variables and is therefore also called Covariance Structure Modeling (CSM). In general, SEM combines the advantages of statistical methods such as factor analysis, path analysis, and multiple regression (Bollen and Long, 1993; Jöreskog et al., 1979). SEM have been applied in a wide range of fields such as sociology, psychology, biological sciences, political science, market research, etc (Chou et al., 2014; Yang et al., 2012; Lee et al., 2017). In the environmental field, this analytical tool is firmly established and is frequently used in the municipal solid waste management (Bortoleto et al., 2012), Contaminated land restoration and air pollutant transmission (Kim and Lee, 2011), etc. The general form of structural equation model is shown in Formula (1).

⎧ η = Bη + Γξ + ζ Y = Δy η + ε ⎨ ⎩ X = Δx ξ + δ

Ajzen (1985)

4.1. Data inspection The statistical analysis was performed on the data of 709 valid questionnaires collected. And the distribution of the characteristics of the samples collected are shown in Table 2. On the whole, the distribution of the socio-demographic characteristics of the valid samples is comparable with the total population distribution of Suzhou City, indicating that this survey has a good representation. This paper employed SPSS 20.0 to analyze the reliability and validity of the survey data and tested the data reliability by calculating the Cronbach’s Alpha of all observed variables (Tenenhaus et al., 2005). The results show that the Cronbach’s Alpha coefficient is 0.643. According to Wu’s research conclusion: reliability is good when the reliability coefficient is greater than 0.7 (Wu, 2003), which means the overall reliability of the current data is mediocre. According to the result that “measurements are taken when items are deleted”, the “Cronbach’s Alpha value of the deleted items” of EF4 (occupied storage space) is greater than the current overall reliability coefficient, so the observed variable “occupied storage space” is dropped from the scale. An overall reliability analysis was performed on the remaining 14 observed variables. The results are shown in Table 3. The Cronbach's Alpha coefficients of all variables are greater than 0.7, and the Cronbach's Alpha value of the total scale is 0.891, indicating that the overall reliability of the adjusted data is quite good. The KMO and Bartlett’s test was performed on the questionnaire sample data. The results show that the KMO value is 0.873, which is greater than 0.6, and the P value for statistical significance of the Bartlett’s test of Sphericity is 0.000. P < 0.001 indicates that the data of factor analysis has good validity and is suitable for factor analysis. Then, exploratory factor analysis was performed on the sample data of 14 observed variables with SPSS 20.0. This paper employed the principal component analysis approach of Oblimin rotation to factor out a total of four common factors. The factor loading matrix after orthogonal rotation is shown in Table 4. The factor loading matrix can

(1)

Among them, η is an endogenous variable, which refers to a variable that is affected by any other variable in the model; ξ is an exogenous variable, meaning that the model is not affected by any other variable, but can affect other variables. Variable ζ is the random term of the system; η can be explained by the observation variable Y, ξ can be explained by the observation variable X, and δ and ε are respectively the measurement errors of the exogenous variable and the endogenous variable. We use the software AMOS to evaluate total effects of each predictor variable on the endogenous variables, and conduct the model calculation using Maximum Likelihood Estimation (MLE). In brief, the structural equation modeling in this study consisted of four steps: (1) reliability and validity test of the survey data; (2) confirmatory factor analysis (CFA) to evaluate the validity of the constructs of the measuring; (3) model evaluation; (4) model correction.

228

Resources, Conservation & Recycling 140 (2019) 224–234

X. Meng et al.

Table 2 Distribution of the socio-demographic characteristics of the samples. Social attribute characteristics

Samples Frequency

Proportion (%)

The proportion of total population in Suzhou (2016) (%)a

Gender

Male Female

326 383

46 54

49a 51

Ageb

18-40 40-60 61 and above

298 276 135

42 39 19

38 29 22

Junior high school and below High school, secondary school University specialties, undergraduate Graduate and above

156

22

34.51

199

28

30.49

92

13

0.47

Family monthly incomec

3999 and Below 4000-8000 8000-15000 15000-20000 20001 and above

128 291 191 85 14

18 41 27 12 2

/ / / / /

Habitation areas

Gusu Wuzhong Xiangcheng Gaoxin District Industrial Park

192 171 135 88 123

27 24 19 13 17

23 27 17 14 19

Education levelb

Sample number

262

37

Table 4 Factor loading matrix by orthogonal rotation. Observed variables

Extracted common factors Factor 1

EA2 SN3 EA3 EA4 EA1 EP2 SN1 EF1 EF2 EF3 EF5 SN2 SN4 EP1

Factor 2

Factor 3

Factor 4

0.548 0.732 0.693 0.782 0.742 0.680 0.575 0.698 0.787 0.863 0.618 0.722 0.693

24.53

4.2. Structural equation model testing Based on the above analysis and adjustment result, this paper constructed an initial model for structural equation analysis for residents’ HSW disposal behavior (RWDB) with AMOS 21.0. It used maximum likelihood estimation method to estimate the parameters of the model. The initial evaluation results are shown in Fig. 2. The concomitant P of the statistical test of the CR (Critical Ratio) value was used to test the significance of model path coefficients. The significance of the standardized path coefficient estimates is shown in Table 5. The results of the goodness-of-fit values of the initial model show that the Chi-square of the initial model is 365.2 (p = .000) and the degree of freedom is 74, and the values of all commonly used fitting indexes meet the requirements. However, according to the result of parameter estimation of the initial model shown in Table 5, the coefficient estimate of the standardized path of the influence of “social norms” (SN) on residents’ HSW disposal behavior (RWDB) is only 0.043, which is very small, and the P value is 0.431, which means that the path coefficient is not significant at the 0.05 level. In addition, the coefficient for path “Laws and regulations” (SN2) to SN is also not significant. Therefore, the initial path assumptions of “SN < —RWDB” and “SN2 < — SN” are not supported. From a practical point of view, ethical constraints, such as social recognition, have little impact on residents’ participation in waste classification and recycling in Suzhou city now. Because most of the existing laws and regulations in Suzhou are only encouraging, and lacking mandatory and incentive mechanisms. Almost all of the respondents' HSW disposal behaviors are not significantly affected by laws and regulations in real life in the current stage. Therefore, based on the model analysis theory above and actual situations, this study considers dropping the path between “social norms” (SN) and “residents’ HSW disposal behavior” (RWDB). Then, the model is extended with the Modification Index (MI). The MI value between the observed variable “influence of neighbors” (SN4) and the latent variable “environmental knowledge, publicity and education” (EP) is very high at 45.363. It means that the chi-square of the modified model will be reduced by 45.363 at least if a path between

709

Note: a Data Source: Suzhou Statistical Yearbook 2017 (Bureau of Statistics of Suzhou, 2017). b “Age” and “education level” are obtained based on the data of the 6th population census in 2010. c The average wage of employees in Suzhou (2016) was 62,722 Yuan.

test if the latent variable setting, observed variable classification and setting, etc. in the conceptual model (Fig. 1) are reasonable. In the initial conceptual model, SN3 (citizens’ social responsibility) is an observed variable of the latent variable “social norms”, but its factor loading under the common factor “environmental attitudes” is up to 0.732. Therefore, SN3 is adjusted to the observed variable of the latent variable “environmental attitudes”. In addition, it can be seen from Table 3 that the load value of EA1 (environmental literacy) is less than 0.5 regardless of the common factor, so this observed variable is removed from the initial conceptual model; the load values of the other observed variables except EA1 are all greater than 0.5, indicating that these variables can be well explained by corresponding common factors, and their settings are consistent with the conceptual model.

Table 3 Test results of the reliability of all latent variables. Latent variables

Number of observed variables

Observed variables

Cronbach's Alpha

Reliability

Environmental attitudes Social norms Environmental knowledge & Publicity and education Environmental facilities and services

4 4 2 4

EA1, EA2, EA3, EA4 SN1, SN2, SN3, SN4 EP1, EP2 EF1, EF2, EF3, EF5

0.842 0.763 0.791 0.885

High Higher Higher High

229

Resources, Conservation & Recycling 140 (2019) 224–234

X. Meng et al.

Fig. 2. Evaluation results of the parameter estimates of the initial structural equation model.

4.3. Result analysis of modified model

SN4 and EP is added to the model. Moreover, in light of actual situations, residents are easily affected by the behavior of neighbors, family members, etc., and are able to learn. They absorb knowledge and information on classification and recycle and form a habit of classification and recovery when encouraged by people around them. Therefore, based on the model analysis theory above and actual situations, this study considers adding a path between the observed variable “influence of neighbors” (SN4) and the latent variable “environmental knowledge, publicity and education” (EP). The parameter path coefficients of the modified model are estimated and the results are shown in Fig. 3. All estimated values of the path coefficients in the modified model are significant at the 0.05 level, and most of the parameters are significant at the 0.01 level, indicating good significance and that the model is credible at the 95% confidence level and up to standard after modification. Comparing the index evaluation results of the initial hypothetical model and the modified model (Table 6), it can be seen that: the result of chi-square (χ2) test has dropped from 365.2 to 229.8, and at the same time, each fitting index is better than that before modification; the RMSEA is less than 0.08, indicating an acceptable model fit, and the GFI, CFI, NFI and IFI are all greater than 0.90, showing that the modified model of RWDB has a good fit (Chen, 2016; Bortoleto et al., 2012).

The test results of the structural equation model indicate that 13 path assumptions (H1, H3, H4, H1a, H1c, H1d, H2a, H3a, H3b, H4a, H4b, H4c, H4e) out of the 19 basic path assumptions of the initial conceptual model of urban residents’ HSW disposal behaviors are established through examination. A newly added relation path is also established through examination, that is, the “influence of neighbors” has a positive effect on residents’ “environmental knowledge, publicity and education”. The supported hypotheses in the modified model and standardized path coefficient estimates are shown in Fig. 4. The size of the left-sided standardized path coefficient in Fig. 4 represents the degree of direct influence of the observed variables on the latent variables, and the size of the three right-side standardized path coefficients represents the direct influence of the latent variables on the target variable (RWDB). The product represents the degree of indirect effect of the observed variables on the target variables. For instance: (1) The degree of influence of “willingness to participate in HSW disposal behavior” on “HSW disposal behavior” is: 0.72 × 0.20 = 0.144; (2) The degree of influence of “accessibility to recycling facilities” on

Table 5 Result of parameter estimation of the initial model. Hypothesized relationship paths

Standardized regression weight estimates

Statistic test parameter

P (Strength of support)

SN3 < — EA EA2 < — EA EA3 < — EA EA4 < — EA EP1 < — EP EP2 < — EP EF3 < — EF EF5 < — EF EA < — RWDB EP < — RWDB EF1 < — EF SN1 < — SN SN2 < — SN SN < — RWDB EF2 < — EF SN4 < — SN EF < — RWDB

0.605 0.371 0.690 0.725 0.643 0.543 0.550 0.683 0.188 0.158 0.443 0.498 0.510 0.043 0.480 0.574 0.119

– 7.251 11.328 11.483 – 5.972 – 5.921 3.925 1.974 6.566 – 1.884 0.787 8.954 5.846 2.246

– *** (Strong Support) *** (Strong Support) *** (Strong Support) – *** (Strong Support) – *** (Strong Support) *** (Strong Support) 0.048 (Support) *** (Strong Support) – 0.060 (No Support) 0.431 (No Support) *** (Strong Support) *** (Strong Support) 0.025 (Support)

Note: “***” indicates significant at the 0.001 level. This study takes a 95% confidence interval, that is, P < 0.05 means that it is significant at the 0.05 level. In this case, the path coefficient is considered to be significant. 230

Resources, Conservation & Recycling 140 (2019) 224–234

X. Meng et al.

Fig. 3. Parameter estimation results of the modified structural equation model.

“HSW disposal behavior” is: 0.68 × 0.23 = 0.156; (3) The degree of influence of “publicity and education” on “HSW disposal behavior” is: 0.81 × 0.23 = 0.186.

5. Discussions and policy recommendations A preliminary finding of this research is that combining TPB theory with A-B-C theory is a good starting point for the modeling of residents' HSW disposal behavior. Most of the hypotheses of the model were supported. On the basis of the analysis results of SEM modeling, "environmental facilities and services" has the most comprehensive effect, while “publicity and education”, “accessibility to recycling facilities”, “accessibility to classification facilities”, “willingness to participation of classification” and “environmental awareness of residents” are the five most significant factors. On the whole, residents’ behavioral selection of HSW disposal is mainly under the joint action of four intrinsic factors and seven external factors. In addition, the combined effect of external factors on residents’ HSW disposal behavior is 0.865, which is nearly twice that of the combined effects of intrinsic factors (0.476). Further, based on the above analysis and discussions of the research results, we proposed some recommendations on urban solid waste classification and recycling management. First of all, it is suggested to strengthen the planning and construction of urban household solid waste classification and recycling facilities. According to the research results, “environmental facilities and services” has the most comprehensive effect on residents’ behavior, and accessibility to classification and recycling facilities are key factors affecting residents’ HSW disposal behavior. However, at present, the construction of a back-end classification, recycling and transportation system in most Chinese cities is lagging behind; the existing construction plans only set principles for the construction of a waste classification, recycling and transportation system. An incomplete and inefficient back-end classification, recycling and transportation facilities

Similarly, the impact of the remaining eight observed variables on HSW disposal behavior can be calculated: the overall impact of “environmental awareness”, “citizens’ social responsibility” and “behavioral attitudes” on “HSW disposal behavior” is 0.138, 0.12 and 0.074, respectively, which means that for each additional unit of residents’ environmental awareness, social responsibility, and behavioral attitudes, utility of their HSW disposal behavior increases by 0.138, 0.12, and 0.074 units respectively; “accessibility to classification facilities”, “time spent”, and “economic cost & benefits” on “HSW disposal behavior” is 0.127, 0.110 and 0.101. Therefore, for example, if the convenience of classification facilities increases by 1 unit, the utility of HSW disposal behavior will increase by 0.127 units. For each additional unit of time spent on waste management, behavioral utility reduces by 0.110 units; the impact of “knowledge and information on classification and recycling” and “influence of neighbors” on “HSW disposal behavior” are 0.094 and 0.090, respectively. Through the above fitting analysis by structural equation model, the main influencing factors of residents' HSW disposal behavior and the sensitivity coefficient corresponding to each factor are obtained. The sensitivity coefficients represent the degree of importance of factors to residents' behavior, and the fitting results and the importance ranking of the factors are shown in Table 7.

Table 6 Evaluation of the fit of the overall SEM model. Model fit criterion

χ (Chi-square) GFI CFI RMSEA NFI IFI AIC 2

Initial hypothetical model

Modified model

Observed value

Comment

Observed value

Comment

365.2(P = 0.000) 0.853 0.856 0.082 0.914 0.857 427.227

The model is rejected Acceptable model fit Acceptable model fit Unacceptable model fit Good model fit Acceptable model fit –

229.757(P = 0.021) 0.914 0.933 0.056 0.952 0.926 281.757

The model is accepted Good model fit Good model fit Acceptable model fit Good model fit Good model fit Better model fit

Note: χ2:Chi-square test, GFI: goodness-of-fit index, CFI: comparative fit index, RMSEA: root-mean-square error of approximation, NFI: normed fit index; IFI: incremental fit index, AIC: Akaike information criterion. 231

Resources, Conservation & Recycling 140 (2019) 224–234

X. Meng et al.

Fig. 4. Supported hypotheses in the modified model and standardized path coefficient estimates (*P < 0.05, ** P < 0.01, ***P < 0.001).

mobile client and mobile Internet; promote knowledge and information on waste classification and recycling, and encourage green and civilized lifestyles; and organize training at working units, schools, and communities to enhance residents’ environmental awareness and willingness to participate, and promote their classification and recycling knowledge. Last but not least, it is suggested to improve laws and regulations. According to the result of parameter estimation of the initial model in Section 4.2, the coefficient for path “Laws and regulations” (SN2) to“social norms” (SN) is not significant, thus the initial path hypothesis that SN2 have a positive effect on SN perceived by residents is not supported. The actual reason is that there was no mandatory laws and regulations and incentive mechanisms on urban HSW classification and recycling in Suzhou city in that period, and almost all of the respondents' HSW disposal behaviors, regardless of whether they participated in HSW recycling, are not significantly affected by laws and regulations. This result and situation are quite different from other regions like the U.S. (Timlett and Williams, 2008), Germany (Bilitewski, 2008), Janpan, Taipei (Charuvichaipong and Sajor, 2006) and Hong Kong, China (Wan et al., 2015; Sakai et al., 2008). Because these countries or regions have established a sound system of supporting laws and regulations on HSW classification and recycling, and the incentives and penalties are effective at promoting waste recycling and reducing contamination. However, there is still a big gap and room for improvement in this aspect in mainland China. At present, China mainland area has not established enough effective laws and regulations, incentive mechanisms, or mandatory restraint policies on waste classification and recycling management. For example, the "Compulsory Recycling List" proposed in the "Solid Waste Pollution Prevention Law" and the "Circular Economy Promotion Law" have not been formally established. And residents as waste producers, are not subject to laws and regulations, and their participation in classification and recovery is virtually voluntary and out of self-discipline. Therefore, it is required that the governments strengthen the top-level design of urban household solid waste classification and recycling, formulate laws and regulations with “teeth”. It is suggested with the opportunity of amending "The Law of the People's Republic of China on the Prevention and Control of Environmental Pollution by Solid Wastes" and the "Circular Economy Promotion Law", to set up a special chapter to stipulate the basic principles of waste sorting and collection, the rights and obligations,

Table 7 Fitting results of the structural equation model for residents’ HSW disposal behavior. Observed variables (factors)

Sensitivity coefficients (Standardized estimate)

Importance ranking

Willingness to participate in classification Environmental awareness Behavioral attitudes Citizens’ social responsibility Influence of neighbors Knowledge and information on Classification and recycling Publicity and education Accessibility to classification facilities Accessibility to recycling facilities Time spent Economic cost & benefits

0.144

3

0.138 0.074 0.12 0.090 0.094

4 11 6 10 9

0.186 0.127

1 5

0.156

2

0.110 0.101

7 8

will seriously damp front-end residents’ enthusiasm of classification. Therefore, it is recommended that all cities in China incorporate detailed waste separation and recycling facilities and system plans during the planning and construction, and accelerate the establishment of a complete waste classification and management system for “classification & disposal, classification & recycling, classification & transportation, and classification & management”. At the same time, the construction of recovery sites for renewable resources and a standardized resource recovery system should be accelerated to make recycling facilities or services more accessible. Then, it is very necessary to carry out extensive publicity and education activities on waste classification and recycling among residents through various channels. Research shows that publicity and education is the most significant factor affecting residents’ HSW disposal behavior. When residents are exposed to such publicity and education more frequently through more channels, they participate more in waste classification and recycling. Therefore, it is recommended that governments and other administrative departments strengthen the publicity and education on waste classification and recycling through channels such as media, education at school, and advertisements; make full use of media such as television, billboards, 232

Resources, Conservation & Recycling 140 (2019) 224–234

X. Meng et al.

issue, some other methodology, such as simulation methods based on complex adaptive system (CAS) theory, can be tried in future research.

core systems, and legal responsibilities. Meanwhile, it is recommended that the State Council introduce special regulations on waste separation management to further clarify the supervision system, residents’ responsibilities and obligations system, supporting mechanism of reward and punishment and credit system.

Acknowledgments I would like to extend my gratitude to the general program of China Postdoctoral Science Foundation (2018M631585), the Strategic Pilot Project of “the Research on the Bottleneck Problems of Resource and Environment for the 100-Year Construction of a Strong Country” (Y8X0771601) launched by the Chinese Academy of Sciences for their grant support, and General Programs of the National Natural Science Foundation of China (71774099). I also would like to thank the staff of the Suzhou Environmental Sanitation and Administration Agency and other organizations for their assistance in the research and the 10 students from Suzhou University of Science and Technology who helped me conduct the questionnaire survey.

6. Conclusions The present study proposed the behavioral decision-making mechanism which considers both residents’ participation in source classification and resource recycling, obtained the main factors have a significant effect on residents’ HSW disposal behaviors and their degree of influence. On the basis of the results of our study, the main conclusions are as follows. First, on the whole, residents’ behavioral selection of HSW disposal is mainly under the joint action of four intrinsic factors (willingness to participate, environmental awareness, social responsibility and behavioral attitudes) and seven external factors (publicity and education received, accessibility to recycling facilities, influence of neighbors, accessibility to classification facilities, time spent, economic cost & benefits, and knowledge on classification and recycling). The combined effect of external factors on residents’ HSW disposal behavior is nearly twice that of the combined effects of intrinsic factors. Second, for the three latent variable facets, “environmental facilities and services” has the greatest combined impact on residents’ HSW disposal behavior, followed by “environmental attitudes”. “Environmental knowledge, publicity and education” comes last. In terms of single factors, the most prominent five factors affecting residents’ HSW disposal behavior are: “publicity and education”, “accessibility to recycling facilities”, “willingness to participate”, “awareness of environmental protection” and “accessibility to classification facilities”. However, the impact of economic cost & benefits and time spent on the residents’ participation in waste classification and recycling is relatively smaller. Third, the impact of laws and regulations on residents’ behavioral selection of household solid waste management does not reach a statistically significant level, as current laws and regulations on waste classification and recycling in Suzhou are mostly incentive and instructive, with no mandatory restraint policies issued yet. That is to say, the impact of policies, laws, and regulations on residents surveyed in real life is generally much limited regarding whether or not residents participate in waste classification and recycling. The assumption path that storage space occupied has a negative impact on residents’ external environmental factor, “environmental facilities and services”, is not established. Finally, the government should strengthen the planning and construction of urban household solid waste classification and recycling facilities, carry out extensive publicity and education activities on waste classification and recycling among residents through various channels, and improve laws and regulations in urban comprehensive management in future. Of course, the current research still includes some limitations which need further improvement. First, the samples of the study primarily select the residents of the central urban area in Suzhou with developed economy and favorable foundation, the results may not exactly represent entire China scenario. In the future, studies on residents in the middle-developed, underdeveloped areas and suburbs should be increased, and the decision-making mechanism of RWDB may be different, so as to formulate relevant management measures according to local conditions. Second, as there is no clear definition of the concept of environmental attitudes at present, the selection of possible observation variables and the design of measurement items in the questionnaire to reflect it may not be very accurate. This could be a topic for further investigation. In addition, the study of individual’s environmental behaviors based on statistical analysis of questionnaire survey data has certain methodological limitations and some subjectivity. To solve this

References Abbott, A., Nandeibam, S., O’Shea, L., 2011. Explaining the variation in household recycling rates across the UK. Ecol. Econ. 70 (11), 2214–2223. Abdelradi, F., 2018. Food waste behaviour at the household level: a conceptual framework. Waste Manag. 71, 485–493. Ajzen, I., 1985. From intentions to actions: a theory of planned behavior. Action-Control: Cogn. Behav. 1–63. Babaei, A.A., Alavi, N., Goudarzi, G., et al., 2015. Household recycling knowledge, attitudes and practices towards solid waste management. Resour. Conserv. Recycl. 102, 94–100. Bach, H., Mild, A., Natter, M., et al., 2004. Combining socio-demographic and logistic factors to explain the generation and collection of waste paper. Resour. Conserv. Recycl. 41 (1), 65–73. Begum, R.A., Siwar, C., Pereira, J.J., Jaafar, A.H., 2009. Attitude and behavioral factors in waste management in the construction industry of Malaysia. Resour. Conserv. Recycl. 53 (6), 321–328. Bilitewski, B., 2008. From traditional to modern fee systems. Waste Manag. 28 (12), 2760–2766. Boldero, J., 1995. The prediction of household recycling of newspapers: the role of attitudes, intentions, and situational factors. J. Appl. Soc. Psychol. 25 (5), 440–462. Bollen, K.A., Long, J.S., 1993. Testing structural equation models. BMS Bull. Sociol. Methodol. 69 (39), 66–67. Boonrod, K., Towprayoon, S., Bonnet, S., et al., 2015. Enhancing organic waste separation at the source behavior: a case study of the application of motivation mechanisms in communities in Thailand. Resour. Conserv. Recycl. 95, 77–90. Borthakur, A., Govind, M., 2017. Emerging trends in consumers’ E-waste disposal behaviour and awareness: a worldwide overview with special focus on India. Resour. Conserv. Recycl. 117, 102–113. Bortoleto, A.P., Kurisu, K.H., Hanaki, K., 2012. Model development for household waste prevention behaviour. Waste Manag. 32, 2195–2207. Botetzagias, I., Dima, A.F., Malesios, C., 2015. Extending the theory of planned behavior in the context of recycling: the role of moral norms and of demographic predictors. Resour. Conserv. Recycl. 95, 58–67. Bureau of Statistics of Suzhou, 2017. Suzhou Statistical Yearbook 2017. China Statistics Press, Beijing. Charuvichaipong, C., Sajor, E., 2006. Promoting waste separation for recycling and local governance in Thailand. Habitat Int. 30, 579–594. Chen, M.F., 2016. Extending the theory of planned behavior model to explain people’s energy savings and carbon reduction behavioral intentions to mitigate climate change in Taiwan-moral obligation matters. J. Clean. Prod. 112, 1746–1753. Chou, P.F., Lu, C.S., Chang, Y.H., 2014. Effects of service quality and customer satisfaction on customer loyalty in high-speed rail services in Taiwan. Transp. Metr. ATransp. Sci. 10 (10), 917–945. Chu, Z., Xi, B., Song, Y., et al., 2013. Taking out the trash: household preferences over municipal solid waste collection in Harbin, China. Habitat Int. 40 (7), 194–200. Dell’Olio, L., Ibeas, A., Oña, J., et al., 2018. Structural Equation Models. Chapter 8. Public Transportation Quality of Service, pp. 141–154. Deng, J., Xu, W.Y., Zhou, C.B., 2013. Investigation of waste classification and collection actual effect and the study of long acting management in the community of Beijing. Environ. Sci. 34 (1), 395–400. Fei, F., Qu, L.L., Wen, Z.G., et al., 2016. How to integrate the informal recycling system into municipal solid waste management in developing countries: based on a China’s case in Suzhou urban area. Resour. Conserv. Recycl. 110, 74–86. Gao, L., Wang, S., Li, J., et al., 2017. Application of the extended theory of planned behavior to understand individual’s energy saving behavior in workplaces. Resour. Conserv. Recycl. 127, 107–113. Grazhdani, D., 2016. Assessing the variables affecting on the rate of solid waste generation and recycling: an empirical analysis in Prespa Park. Waste Manag. 48, 3–13. Guagnano, G.A., Stern, P.C., Dietz, T., 1995. Influences on attitude-behavior relationships: a natural experiment with curbside recycling. Environ. Behav. 27 (5), 699–718. Guo, H., Hobbs, B.F., Lasater, M.E., et al., 2016. System dynamics-based evaluation of interventions to promote appropriate waste disposal behaviors in low-income urban

233

Resources, Conservation & Recycling 140 (2019) 224–234

X. Meng et al.

India. Resour. Conserv. Recycl. 135, 313–322. Song, Q., Wang, Z., Li, J., 2012. Residents’ behaviors, attitudes, and willingness to pay for recycling e-waste in Macau. J. Environ. Manage. 106, 8–16. Stern, P.C., Oskamp, S., 1987. Managing scarce environmental resources. Handbook of Environmental Psychology. Wiley, New York. Tadesse, T., 2009. Environmental concern and its implication to household waste separation and disposal: evidence from Mekelle, Ethiopia. Resour. Conserv. Recycl. 53 (4), 183–191. Tenenhaus, M., Vinzi, V.E., Chatelin, Y.M., et al., 2005. PLS path modeling. Comput. Stat. Data Anal. 48 (1), 159–205. Timlett, R.E., Williams, I.D., 2008. Public participation and recycling performance in England: a comparison of tools for behaviour change. Resour. Conserv. Recycl. 52 (4), 622–634. Tucker, P., Murney, G., Lamont, J., 1998. Predicting recycling scheme performance: a process simulation approach. J. Environ. Manage. 53 (1), 31–48. Wan, A.W.A.K.G., Rusli, I.F., Dayang, R., et al., 2013. An application of the theory of planned behaviour to study the influencing factors of participation in source separation of food waste. Waste Manag. 33 (5), 1276–1281. Wan, C., Shen, G.Q., Yu, A., 2014. The role of perceived effectiveness of policy measures in predicting recycling behaviour in Hong Kong. Resour. Conserv. Recycl. 83 (83), 141–151. Wan, C., Shen, G.Q., Yu, A., 2015. Key determinants of willingness to support policy measures on recycling: a case study in Hong Kong. Environ. Sci. Policy 54, 409–418. Wang, Z., Guo, D., Wang, X., et al., 2018. How does information publicity influence residents’ behaviour intentions around e-waste recycling? Resour. Conserv. Recycl. 133, 1–9. Wu, M., 2003. SPSS Statistical Application Practice – Questionnaire Analysis and Application Statistics. Science Press, Beijing. Wu, Z., Yu, A.T.W., Shen, L., 2017. Investigating the determinants of contractor’s construction and demolition waste management behavior in Mainland China. Waste Manag. 60, 290–300. Xiao, L., Zhang, G., Zhu, Y., et al., 2017. Promoting public participation in household waste management: a survey based method and case study in Xiamen city, China. J. Clean. Prod. 144, 313–322. Yan, H., 2018. Mandatory Classification May be the Opportunity for Changes [N/OL]. 2016-09-21. China Environment News. http://news.cenews.com.cn/html/2016-09/ 21/content_50274.htm. Yang, K.C., Hsieh, T.C., Li, H., et al., 2012. Assessing how service quality, airline image and customer value affect the intentions of passengers regarding low cost carriers. J. Air Transp. Manag. 20, 52–53. Young, R.D., 1990. Recycling as appropriate behavior: a review of survey data from seleeted recycling edueation programs in Michigan. Resour. Conserv. Recycl. 3 (4), 253–266. Zhang, H., Wen, Z., 2014. Residents’ household solid waste (HSW) source separation activities: a case study in Suzhou, China. Sustainability 6, 6446–6466.

areas: a Baltimore case study. Waste Manag. 56, 547–560. Izagirre-Olaizola, J., Fernández-Sainz, A., Vicente-Molina, M.A., 2015. Internal determinants of recycling behaviour by university students: a cross-country comparative analysis. Int. J. Consum. Stud. 39 (1), 25–34. Jackson, T., 2005. Motivating Sustainable Consumption: A Review of Evidence on Consumer Behaviour and Behavioural Change. pp. 43–61. Jöreskog, K.G., 1970. A general method for estimating a linear structural equation system. Res. Bull. 2, 1–41. Jöreskog, K.G., Sörbom, D., Magidson, J., 1979. Advances in Factor Analysis and Structural Equation Models. Abt Books, pp. 24–26. Kim, Y.K., Lee, H.R., 2011. Customer satisfaction using low cost carriers. Tour. Manag. 32 (2), 235–243. Lee, P., Mahoney, K.T., Lee, S., 2017. An application of the exploratory structural equation modeling framework to the study of personality faking. Pers. Individ. Differ. 119, 220–226. Li, J., Tam, V.W.Y., Zuo, J., et al., 2015. Designers’ attitude and behaviour towards construction waste minimization by design: a study in Shenzhen, China. Resour. Conserv. Recycl. 105, 29–35. Lindbeck, A., 1997. Incentives and social norms in household behavior. Am. Econ. Rev. 87 (2), 370–377. Liu, Y., Sun, C., Xia, B., et al., 2018. Impact of community engagement on public acceptance towards waste-to-energy incineration projects: empirical evidence from China. Waste Manag. 76, 431–442. Lizina, S., Dael, M.V., Passel, S.V., 2017. Battery pack recycling: behaviour change interventions derived from an integrative theory of planned behaviour study. Resour. Conserv. Recycl. 122, 66–82. Matsumoto, S., 2014. The opportunity cost of pro-environmental activities: spending time to promote the environment. J. Fam. Econ. Issues 35 (1), 119–130. Meng, X., Wen, Z., Qian, Y., 2018. Multi-agent based simulation for household solid waste recycling behavior. Resour. Conserv. Recycl. 128, 535–545. National Bureau of Statistics of China, 2018. Statistical Bulletin of the People’s Republic of China on National Economic and Social Development in 2017. Nguyen, T.P., Zhu, D., Le, N.P., 2015. Factors influencing waste separation intention of residential households in a developing country: evidence from Hanoi, Vietnam. Habitat Int. 48, 169–176. Park, J., Ha, S., 2014. Understanding consumer recycling behavior: combining the theory of planned behavior and the norm activation model. Fam. Consum. Sci. Res. J. 42 (3), 278–291. Reddi, K.R., Li, W., Wang, B., et al., 2013. System dynamics modelling of hybrid renewable energy systems and combined heating and power generator. Int. J. Sustain. Eng. 6 (1), 31–47. Sakai, S., Ikematsu, T., Hirai, Y., et al., 2008. Unit-charging programs for municipal solid waste in Japan. Waste Manag. 28 (12), 2815–2825. Singh, M.P., Chakraborty, A., Roy, M., 2018. Developing an extended theory of planned behavior model to explore circular economy readiness in manufacturing MSEMEs,

234