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 T

waste classification and recycling behaviors
⁎ ⁎
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
b
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

A R T I C LE I N FO A B S T R A C T

Keywords: Residents’ participation in classification and recycling of urban household solid waste (HSW) is a critical factor
Household solid waste for the success of municipal solid waste management. The aim of this study is to investigate the decision-making
Separation and recycling behavior mechanism of residents’ HSW disposal behaviors by merging the theory of planned behavior and the Attitude-
Structural equation model Behavior-Condition theory. In this study, based on the survey data of 709 residents in Suzhou, China and
Social survey
structural equation modeling method, the main factors that affect residents’ HSW disposal behaviors and their
Situational factors
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 classi-
fication 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 This requires a clear understanding of the main influencing factors and
decision-making mechanisms of residents’ HSW disposal behaviors
With the development of the economy and the improvement of re- (RWDB). Understanding RWDB could enable decision makers and local
sidents’ living standards, urban household solid waste is increasing governments to design more-effective policies for improving waste se-
rapidly in many countries all over the world. At present more and more paration and recycling.
governments regard the principles of HSW’s decrement, recycle and As the world's largest developing country, China has already entered
harmless as the goals of urban municipal solid waste management. a period of rapid urbanization. From 1998–2017, urbanization in China
Nothing wrong with that waste source separation and recycling has has increased at an average annual growth rate of more than 1%. As of
major potential benefits to an effective management of waste by ad- 2017, 58.52% of China’s total population lives in urban areas or cities
dressing the problem of landfill shortage and resource savings. (National Bureau of Statistics of China, 2018). In the meantime, the
However, there is a considerable distance towards achieving its full rapid industrial development has consumed massive resources and
potential in practice. One of the main reasons is the weak residents’ given rise to urban household solid waste (HSW). At present, two thirds
engagement in waste management policies (e.g. classification, re- of China’s big and medium cities are engulfed in waste, with more than
cycling). Since residents' participation in HSW separation and recycling 500 million square meters of land nationwide encroached due to the
is the key to affecting urban solid waste classification management, dumping of household solid waste (Fei et al., 2016). There is no doubt
then, the question is, how to increase the participation rate of residents? that the classification and recycling of HSW is strategically important

⁎
Corresponding authors.
E-mail addresses: mxy3023@126.com (X. Meng), wenzg@tsinghua.edu.cn (Z. Wen), yt289@cantab.net (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.
X. Meng et al. Resources, Conservation & Recycling 140 (2019) 224–234

for alleviating resource and environmental restrictions. As early as conditions (Tucker et al., 1998).
2000, China started to carry out pilot programs for waste separation However, much of the research up to now has separately studied
and recycling in eight cities including Beijing, Guangzhou, and individuals’ waste prevention behavior (Bortoleto et al., 2012), waste
Shanghai (Meng et al., 2018). Since 2016, waste classification has been source separation behavior (Zhang and Wen, 2014), domestic recycl-
elevated to an unprecedented high level. President Xi Jinping specifi- able resource recycling behavior (Fei et al., 2016), etc. Few published
cally pointed out that it is necessary to introduce the waste classifica- studies have been able to draw on any systematic research into HSW
tion system to more areas when he presided over the Central Leading disposal behaviors including all possible waste disposal methods si-
Group on Financial and Economic Affairs in December 2016. The Na- multaneously. There has been no known research on a behavioral de-
tional Development and Reform Commission and the Ministry of cision-making mechanism which considers both residents’ participation
Housing and Urban-Rural Development released the “Implementation in source classification and resource recycling. In addition, previous
Plan on the Household Solid Waste Classification System” on March 18, research on residents’ HSW disposal behavior mainly focuses on will-
2017, requiring that 46 cities nationwide take the lead to implement ingness to participate in source classification and impact factors, how-
mandatory classification of household solid waste and that the re- ever, studies that consider both internal subjective factors and external
cycling rate of household solid waste exceed 35% by 2020. However, situational factors that influence individuals’ waste management be-
the present management of household solid waste classification and haviors are still rare. In this study, in order to comprehensively un-
recycling is not satisfactory. In pilot cities for household solid waste derstanding the decision-making mechanism of residents’ HSW disposal
classification, the participation rate of household solid waste classifi- behaviors, explore the factors have a significant effect on RWDB and the
cation is still low, and there is no substantial progress made in waste degree of their influence, we divided RWDB into three kinds according
reduction (Yan, 2018). It is a pressing problem to cultivate residents’ to different waste disposal ways selected (non-classification, classifi-
habits of source classification and resource recycling, and improve their cation deposition and selling recyclables after classification), and de-
participation in waste classification and recycling in the comprehensive veloped an hypothetical model by merging the TPB theory and A-B-C
management of urban household solid waste. Therefore, it is theoreti- theory. The proposed model can consider both individual subjective
cally and instructively significant to tap into the main impact factor and factors and external situational factors that may influence the residents'
decision-making mechanism and formulate targeted policies to improve HSW classification and recycling behaviors. Next a questionnaire was
residents’ participation in household solid waste classification and re- designed and the field survey was conducted in the five administrative
cycling. districts of Suzhou, China. Then the initial hypothetical model was
There have been some research conducted to explain why in- tested using structural equation modeling (SEM) based on the ques-
dividuals may or may not engage in waste management policies, such tionnaire data, and the significant paths and better indexes of fit were
as waste prevention, source separation and littering (Bortoleto et al., obtained through model evaluation and correction. Finally, the main
2012; Abdelradi, 2018; Wang et al., 2018). In summary, there are two influencing factors of RWDB and the sensitivity coefficient corre-
classes of theoretical methods for research on residents’ environmental sponding to each factor were explored, moreover, the decision-making
behavior and choices at home and abroad (Jackson, 2005), one is the mechanism of residents and policy suggestions were discussed.
research method based on environmental sociology, and the other is the The results of this study could provide a theoretical support for
research method based on environmental psychology. The first method policy formulation on urban household solid waste classification and
starts from the interaction between micro-individuals and socio-en- recycling in China and other countries in the word. The initial research
vironmental systems and considers that an individual’s ideas and be- hypotheses and conceptual models are organized in Section 2 based on
havioral choices are determined by the process and status of social and previous literature reviews. Section 3 introduces the research metho-
technological development (Singh et al., 2018). The second method dology for data collection and structural equation modeling. Section 4
mainly considers the effect of irrational factors on individual behavior. presents data analysis, model testing and results. The discussions and
The most common theory of planned behavior (TPB) falls into the first policy recommendations are conducted in Sections 5, and 6 covers the
category. Ajzen (1985) developed TPB based on the theory of reasoned conclusions.
action, which emphasizes that an individual’s behavior is influenced by
attitude, subjective norm, and perceived behavioral control (Ajzen, 2. Initial research hypotheses and conceptual models
1985). Many scholars have studied waste classification and recycling
behavior with the TPB theory (Botetzagias et al., 2015; Gao et al., 2017; In order to propose reasonable hypotheses of the initial measure-
Lizina et al., 2017). For example, Nguyen et al. (2015) find that per- ment model, we searched and summarized a large number of prior
sonal ethics are a significant impact factor in promoting residents’ be- research on the impact factors of residents’ waste disposal behavior. In
havioral intention to participate in waste classification and recycling; recent years, domestic and foreign scholars have conducted relevant
the study by Park and Ha (2014) indicates that residents are en- research on it (Boonrod et al., 2015; Borthakur and Govind, 2017; Guo
couraged and affected when they see their neighbors or friends classify et al., 2016). Based on literature reviews, this paper summarizes factors
and recycle waste. of frequent occurrence and with a significant effect. Previous studies
Although the TPB theory inspires studies on residents’ recycling have shown that residents’ HSW disposal behaviors may be affected
behavior, its model framework has strong limitations. The TPB theory indirectly by four main aspects, namely, environmental attitudes, social
mainly considers intrinsic factors; however, other factors also affect the norms, environmental knowledge, publicity and education, environ-
process when behavioral intentions turn into behavior (Boldero, 1995). mental facilities and services (situational factors). Further, by com-
Stern and Oskamp (1987) constructed a complex environmental beha- bining the Theory of Planned Behavior and A-B-C Theory, this study
vior model, proposing that environmental behavior is the result of re- puts forward an initial conceptual and measurement model on the de-
lated external factors and intrinsic factors working together. Based on cision-making of RWDB, together with its indicated hypotheses (see
this, Guagnano et al. (1995) proposed the Attitude-Behavior-Condition Fig. 1). The definitions of individual hypotheses regarding RWDB are as
(A-B-C) theory, which states that individuals’ behavior (Behavior, B) follows.
results from the combined effect of residents’ attitudes (Attitude, A) and In the initial model, the study assume that the four latent variables
external conditions (Condition C), and considers that external condi- of “environmental attitudes (EA)”, “social norms (SN)”, “environmental
tions are crucial factors in determining whether residents perform knowledge, publicity and education (EP)” and “environmental facilities
waste recycling behavior. Tucker further refined the model and pro- and services (EF)” have path effect on RWDB (H1,H2, H3 and H4), and
posed a research model in which residents’ HSW disposal behavior is each observed variable is reflected by several observed variables. The
determined by attitude, subjective norms, social norms, and external paper has set up 15 possible observed variables based on the literature

225
X. Meng et al. Resources, Conservation & Recycling 140 (2019) 224–234

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

investigation. are internalized in the individual’s own value system, behaviors that
meet or oppose the norm can lead to self-esteem or guilt. Previous
2.1. Environmental attitudes (EA) studies show that social norms have a significant correlation with re-
sidents' waste recycling behavior (Chu et al., 2013；Botetzagias et al.,
Studies have shown that there is a significant correlation between 2015). Existing research focuses on four major aspects: (1) The first is
environmental attitudes (EA) and residents’ HSW disposal behavior citizens’ social responsibility. Nguyen found that an individual’s sense
(Ajzen, 1985; Begum et al., 2009; Song et al., 2012). The performance of social responsibility and ethics, that is, residents think that waste
of environmental behaviors can be affected directly by attitudes toward recycling is a good thing for the public and themselves, is an important
particular actions (Singh et al., 2018; Bortoleto et al., 2012; Tadesse, factor affecting their participation in recycling (Nguyen et al., 2015).
2009). At present, there is no clear definition of environmental atti- (2) The second is constraints of laws and regulations. Wan and other
tudes. This paper defines environmental attitudes as the general and scholars believe that enactment of corresponding laws and regulations
stable perception or position held by major residents on household solid has a positive effect on the residents’ environmental behavior, and it
waste classification and recovery. In this investigation, we use the fol- helps increase residents' participation in waste separation and recycling
lowing four items to reflect residents’ environmental attitudes: re- (Wan et al., 2014；Timlett and Williams, 2008). A social survey in
sidents’ environmental literacy (Ajzen, 1985), willingness to participate Hong Kong (Wan et al., 2015) shows that the government implements
(Nguyen et al., 2015), awareness of resource conservation and en- “carrot and stick” policy measures, that is, incentives and penalties
vironmental protection (Wan et al., 2013), and recognition of the ne- related to waste separation and recycling, and residents will better
cessity of classification and recycling behavior (Li et al., 2015). perceive the binding of policies, which helps to promote waste se-
Therefore, on the basis of the literature review above, the study assume paration and recycling. (3) Herd behavior effect (the influence from
that the observed variables“environmental literacy (EA1)”, “willingness family and neighbors). Park and Ha’s research shows that when re-
to participate in classification (EA2)”, “environmental awareness sidents see their neighbors or peer group classify and recycle waste,
(EA3)” and “behavioral attitudes (personal recognition of the necessity they are often driven and affected (Park and Ha, 2014). (4) Social re-
of classification and recycling behavior, EA4)” can mainly reflect and cognition. A survey in Michigan shows that non-economic returns such
have the positive path effect on the latent variable EA. The corre- as social recognition, satisfaction from participating in waste recycling
sponding path hypothesis are H1a, H1b, H1c and H1d, respectively. and charity are important factors in promoting residents’ participation
in recycling.
Therefore, this paper defines social norms as the tendency for re-
2.2. Social norms (SN)
sidents to adopt certain HSW disposal behavior due to notable peer and
social pressure (Deng et al., 2013). We assume that the observed vari-
Social norms refer to the pressure of others and the society that
ables “social recognition (SN1)”, “laws and regulations (SN2)”, “citi-
exert an important influence on the households’ behaviors, it take the
zens’ social responsibility (SN3)”, and “influence of neighbors (herd
form of approval or disapproval of others, as well as associated feelings
psychological effect, SN4)” can form and mainly reflect the latent
of pride or shame (Lindbeck, 1997). In addition, once the social norms

226
X. Meng et al. Resources, Conservation & Recycling 140 (2019) 224–234

variable SN, and SN1, SN2, SN3 and SN4 have the positive path effect sociological characteristics are not taken as direct observed variables in
on SN (H2a, H2b, H2c and H2d). this study.

2.3. Environmental knowledge, publicity and education (EP) 3. Research methodology

Several studies have revealed that there is a strong positive corre- 3.1. Questionnaire design and data collection
lation between publicity efforts and residents’ participation in waste
recycling (Grazhdani, 2016; Wang et al., 2018; Xiao et al., 2017). Reddi On the basis of the initial conceptual model proposed (Fig. 1), this
et al.’s study shows that residents’ environmental knowledge and in- study used the five-level Likert scale to design a preliminary ques-
formation is significantly related to their environmental behavior, the tionnaire to investigate the RWDB and their influencing factors. The
lack of related knowledge and information will hinder residents' par- questionnaire consisted of four parts: (1) background of the investiga-
ticipation in waste separation and recycling (Reddi et al., 2013). tion, including a brief introduction of this survey; (2) the current si-
Through public education, residents can better understand the new tuation of households’ waste disposal, including disposal ways, time
recycling policy and improve the ability of classification and recycling spent, economic income and methods of selling recyclable waste, etc. In
(Wan et al., 2013; Izagirre-Olaizola et al., 2015). This paper defines this study, we classified residents’ HSW disposal behaviors into three
environmental knowledge as the knowledge, skills, and information kinds according to different waste disposal ways selected: (1) non-
necessary for residents to carry out waste classification and recycling, classification (mixed disposal), (2) classification deposition (classified
such as classification methods, recycling channels, locations of re- and delivered into trash cans according to the method of quartering),
cycling sites, and recycling hotlines; it defines publicity and education (3) selling recyclables after classification (separate recyclables and sell
as what residents receive through the media, advertisements, education them to recyclable material collectors, and dispose of the rest into trash
at school and other methods on waste classification and recycling. cans); (3) items measuring the initial research hypotheses and the
Based on the literature review above, we assumed that the latent conceptual model, all measures were reported on 1 to 5 point scale from
variable “environmental knowledge, publicity and education” was “Strongly Agree”, “Agree”, “Moderately”, “Disagree” to “Strongly Dis-
mainly reflected by two observed variables: “knowledge and informa- agree”; (4) the demographic and social attribute information of the
tion on classification and recycling (EP1)”, and “publicity and educa- respondents, including gender, age, education level, family monthly
tion (EP2)”. Moreover, EP1 and EP2 have positive and direct influence income and habitation areas.
on EP, the corresponding path hypothesis are H3a and H3b. The data were collected by questionnaire survey. We chose Suzhou
as a study area, because there was a favorable foundation for waste
2.4. Environmental facilities and services (EF) classification and recycling in Suzhou, and it had embarked on a pilot
program on household solid waste classification and recycling in some
Some studies have confirmed that external conditions have an im- residential communities since 2000. In 2010, Suzhou put forward
portant impact on hindering or promoting residents’ household solid “rough separation in the short term and fine classification in the long
waste recycling behavior (Matsumoto, 2014; Wu et al., 2017). Bach run”, a new model for household solid waste classification and re-
et al.’s study shows that to some extent the informal recycling market cycling (jswmw.com). In April 2015, Suzhou was chosen to be one of
makes it more convenient for residents to dispose of waste and pro- the 26 national domestic waste sorting collection pilot cities (bunch 1)
motes residents’ HSW classification and recycling behavior; in the by five ministries. In recent years, source classification of household
meantime, the increase in the number of recycling sites for renewable solid waste in Suzhou has achieved steady progress. In 2017, there were
resources helps increase the recovery rate (Bach et al., 2004). A re- more than 400 pilot communities participating in waste classification.
search in the United Kingdom (Abbott et al., 2011) confirmed that more First, in order to verify the rationality of the initial questionnaire de-
recycling sites established near residential areas had resulted in an in- sign, including the structure, questions and options setting, we con-
crease in the frequency of resident recycling. This study also pointed ducted a pretest via WeChat with nearly 200 survey results retrieved.
that when the government does not have the right to charge residents And the preliminary questionnaire was revised and re-designed ac-
enough garbage disposal fees, residents do not actively reduce the cording to the results. For example, because we didn't know the family
waste generated or try their best to participate in waste separation and monthly income structure of residents in the sample in advance, the
recycling, indicating that the economic factors will also affect the re- answer range setting for this question was unreasonable, the total range
sidents' garbage disposal behaviors. Moreover, the time spent on waste was too large and the grouping interval was not suitable, which made
classification and recycling and the storage space used to store house- some group cases more concentrated. Therefore, the answer settings for
hold solid waste at home also affect residents’ participation in waste this question was re-adjusted based on the applicable answer distribu-
classification and recycling (Liu et al., 2018). This paper uses the latent tion characteristics. The final measurement instruments for the latent
“environmental facilities and services (EF)” to summarize objective variables of the hypothetical model are shown in Table 1. Then, the
external conditions (situational factors) that affect HSW classification research team conducted the formal field survey among permanent
and recycling. We assume that EF may be explained and measured by residents in the central areas of Suzhou (including its five adminis-
five observed variables including “Economic cost & benefits (disposal trative districts, i.e. Gusu, Wuzhong, Xiangcheng, Gaoxin District, and
fees and revenues from sales of waste, EF1)”, “Time spent (EF2)”, Industrial Park) through random sampling. The survey was launched in
“accessibility to classification facilities (EF3)”, “occupied storage space the areas where residents gather, including eight shopping malls in
(EF4)” and “accessibility to recycling facilities (EF5)”. In addition, EF1, downtown Suzhou, Tongjing Park, Jinji Lake Plaza, and Guanqian
EF3 and EF5 have a positive effect on EF, the corresponding path hy- Street. It fully considered the population ratio in each district, as well as
pothesis are H4a, H4c and H4e; EF2 and EF4 have a negative effect on EF, the characteristics of distribution of residents of all age groups and with
the corresponding path hypothesis are H4b, and H4d. different occupations. The questionnaire survey collected 709 valid
In addition to the four aspects above, research has also shown that questionnaires out of 759 in total, with a valid response rate of 93.8%.
residents’ knowledge, attitudes, and practices in waste recycling are
influenced by demographic and sociological factors such as age, edu- 3.2. Measurement model
cation level, gender, and occupation (Babaei et al., 2015; Song et al.,
2012). However, since the influence from different demographic and The initial hypothetical model was tested using Structural Equation
sociological characteristics of the population is already incorporated in Modeling (SEM) with the software AMOS 21.0 in this study. SEM was
“environmental attitudes” and “social norms”, demographic and first proposed in the 1970s by Swedish scholars Jöreskog (1970).

227
X. Meng et al. Resources, Conservation & Recycling 140 (2019) 224–234

Table 1
Measurement instruments for the latent variables of the hypothetical model.
Latent variables Measurement items (observed Sources
variables)

Environmental attitudes (EA) EA1 What I care about is survival and life issues, not environmental issues Ajzen (1985)
such as waste recycling.
EA2 In order to save resources and protect the environment, I am willing to Nguyen et al. (2015)
participate in waste separation and recycling.
EA3 Waste classification and recycling are conducive to saving resources and Wan et al. (2015)
turning waste into treasure.
EA4 I produce less recyclable waste, no need for separate and recycling. Li et al. (2015)

Social norms (SN) SN1 I'm satisfied to participate in waste classification and recycling. Young (1990)
SN2 Waste separation and recycling laws and regulations can play a Wan et al. (2014, 2015)
constraining role for me.
SN3 Waste classification and recycling are the responsibility of the Timlett and Williams
government and enterprises and have nothing to do with residents. (2008)
SN4 Seeing my neighbors and friends to participate in sorting, I’ll do the Park and Ha (2014)
same.

Environmental knowledge, publicity and EP1 I have mastered the waste classification method and I know the location Reddi et al. (2013)
education (EP) of the nearby waste recycling sites.
EP2 I have been exposed to enough publicity and education on waste sorting Grazhdani (2016)
and recycling in my daily life.

Environmental facilities and services EF1 I sell scrap in order to obtain economic benefits. Abbott et al. (2011)
(EF) EF2 Waste classification and recycling waste of time. Bach et al. (2004)
EF3 There are classified garbage bins and garbage kiosks in the community, Guagnano et al. (1995)
with clear identification and close distances.
EF4 Sorting and collecting recyclable waste takes up a lot of storage space in Matsumoto (2014)
my house.
EF5 Waste recycling sites are close to home, and the service of recyclers is Izagirre-Olaizola et al.
good. (2015)

Structural equation models are a family of multivariate statistical 4. Data analysis and results
models that allow the analyst to estimate the effect and relationships
between multiple variables (DellöOlio et al., 2018). The model analyzes 4.1. Data inspection
the relationship between variables based on the covariance matrix of
the variables and is therefore also called Covariance Structure Modeling The statistical analysis was performed on the data of 709 valid
(CSM). In general, SEM combines the advantages of statistical methods questionnaires collected. And the distribution of the characteristics of
such as factor analysis, path analysis, and multiple regression (Bollen the samples collected are shown in Table 2. On the whole, the dis-
and Long, 1993; Jöreskog et al., 1979). SEM have been applied in a tribution of the socio-demographic characteristics of the valid samples
wide range of fields such as sociology, psychology, biological sciences, is comparable with the total population distribution of Suzhou City,
political science, market research, etc (Chou et al., 2014; Yang et al., indicating that this survey has a good representation.
2012; Lee et al., 2017). In the environmental field, this analytical tool is This paper employed SPSS 20.0 to analyze the reliability and va-
firmly established and is frequently used in the municipal solid waste lidity of the survey data and tested the data reliability by calculating the
management (Bortoleto et al., 2012), Contaminated land restoration Cronbach’s Alpha of all observed variables (Tenenhaus et al., 2005).
and air pollutant transmission (Kim and Lee, 2011), etc. The results show that the Cronbach’s Alpha coefficient is 0.643. Ac-
The general form of structural equation model is shown in Formula cording to Wu’s research conclusion: reliability is good when the re-
(1). liability coefficient is greater than 0.7 (Wu, 2003), which means the
overall reliability of the current data is mediocre. According to the
⎧ η = Bη + Γξ + ζ result that “measurements are taken when items are deleted”, the
Y = Δy η + ε “Cronbach’s Alpha value of the deleted items” of EF4 (occupied storage
⎨
space) is greater than the current overall reliability coefficient, so the
⎩ X = Δx ξ + δ (1)
observed variable “occupied storage space” is dropped from the scale.
Among them, η is an endogenous variable, which refers to a variable An overall reliability analysis was performed on the remaining 14 ob-
that is affected by any other variable in the model; ξ is an exogenous served variables. The results are shown in Table 3. The Cronbach's
variable, meaning that the model is not affected by any other variable, Alpha coefficients of all variables are greater than 0.7, and the Cron-
but can affect other variables. Variable ζ is the random term of the bach's Alpha value of the total scale is 0.891, indicating that the overall
system; η can be explained by the observation variable Y, ξ can be reliability of the adjusted data is quite good.
explained by the observation variable X, and δ and ε are respectively The KMO and Bartlett’s test was performed on the questionnaire
the measurement errors of the exogenous variable and the endogenous sample data. The results show that the KMO value is 0.873, which is
variable. We use the software AMOS to evaluate total effects of each greater than 0.6, and the P value for statistical significance of the
predictor variable on the endogenous variables, and conduct the model Bartlett’s test of Sphericity is 0.000. P < 0.001 indicates that the data
calculation using Maximum Likelihood Estimation (MLE). of factor analysis has good validity and is suitable for factor analysis.
In brief, the structural equation modeling in this study consisted of Then, exploratory factor analysis was performed on the sample data
four steps: (1) reliability and validity test of the survey data; (2) con- of 14 observed variables with SPSS 20.0. This paper employed the
firmatory factor analysis (CFA) to evaluate the validity of the constructs principal component analysis approach of Oblimin rotation to factor
of the measuring; (3) model evaluation; (4) model correction. out a total of four common factors. The factor loading matrix after
orthogonal rotation is shown in Table 4. The factor loading matrix can

228
X. Meng et al. Resources, Conservation & Recycling 140 (2019) 224–234

Table 2 Table 4
Distribution of the socio-demographic characteristics of the samples. Factor loading matrix by orthogonal rotation.
Social attribute Samples The Observed variables Extracted common factors
characteristics proportion of
Frequency Proportion (%) total Factor 1 Factor 2 Factor 3 Factor 4
population in
Suzhou EA2 0.548
(2016) (%)a SN3 0.732
EA3 0.693
Gender Male 326 46 49a EA4 0.782
Female 383 54 51 EA1
EP2 0.742
Ageb 18-40 298 42 38
SN1 0.680
40-60 276 39 29
EF1 0.575
61 and above 135 19 22
EF2 0.698
Education Junior high 156 22 34.51 EF3 0.787
levelb school and EF5 0.863
below SN2 0.618
High school, 199 28 30.49 SN4 0.722
secondary EP1 0.693
school
University 262 37 24.53
specialties,
4.2. Structural equation model testing
undergraduate
Graduate and 92 13 0.47
above Based on the above analysis and adjustment result, this paper con-
Family monthly 3999 and Below 128 18 /
structed an initial model for structural equation analysis for residents’
incomec 4000-8000 291 41 / HSW disposal behavior (RWDB) with AMOS 21.0. It used maximum
8000-15000 191 27 / likelihood estimation method to estimate the parameters of the model.
15000-20000 85 12 / The initial evaluation results are shown in Fig. 2.
20001 and 14 2 /
The concomitant P of the statistical test of the CR (Critical Ratio)
above
value was used to test the significance of model path coefficients. The
Habitation Gusu 192 27 23
significance of the standardized path coefficient estimates is shown in
areas Wuzhong 171 24 27
Xiangcheng 135 19 17
Table 5.
Gaoxin District 88 13 14 The results of the goodness-of-fit values of the initial model show
Industrial Park 123 17 19 that the Chi-square of the initial model is 365.2 (p = .000) and the
Sample number 709 degree of freedom is 74, and the values of all commonly used fitting
indexes meet the requirements. However, according to the result of
Note: parameter estimation of the initial model shown in Table 5, the coef-
a
Data Source: Suzhou Statistical Yearbook 2017 (Bureau of Statistics of ficient estimate of the standardized path of the influence of “social
Suzhou, 2017). norms” (SN) on residents’ HSW disposal behavior (RWDB) is only
b
“Age” and “education level” are obtained based on the data of the 6th
0.043, which is very small, and the P value is 0.431, which means that
population census in 2010.
c the path coefficient is not significant at the 0.05 level. In addition, the
The average wage of employees in Suzhou (2016) was 62,722 Yuan.
coefficient for path “Laws and regulations” (SN2) to SN is also not
significant. Therefore, the initial path assumptions of “SN < —RWDB”
test if the latent variable setting, observed variable classification and
and “SN2 < — SN” are not supported. From a practical point of view,
setting, etc. in the conceptual model (Fig. 1) are reasonable. In the
ethical constraints, such as social recognition, have little impact on
initial conceptual model, SN3 (citizens’ social responsibility) is an ob-
residents’ participation in waste classification and recycling in Suzhou
served variable of the latent variable “social norms”, but its factor
city now. Because most of the existing laws and regulations in Suzhou
loading under the common factor “environmental attitudes” is up to
are only encouraging, and lacking mandatory and incentive mechan-
0.732. Therefore, SN3 is adjusted to the observed variable of the latent
isms. Almost all of the respondents' HSW disposal behaviors are not
variable “environmental attitudes”. In addition, it can be seen from
significantly affected by laws and regulations in real life in the current
Table 3 that the load value of EA1 (environmental literacy) is less than
stage. Therefore, based on the model analysis theory above and actual
0.5 regardless of the common factor, so this observed variable is re-
situations, this study considers dropping the path between “social
moved from the initial conceptual model; the load values of the other
norms” (SN) and “residents’ HSW disposal behavior” (RWDB).
observed variables except EA1 are all greater than 0.5, indicating that
Then, the model is extended with the Modification Index (MI). The
these variables can be well explained by corresponding common fac-
MI value between the observed variable “influence of neighbors” (SN4)
tors, and their settings are consistent with the conceptual model.
and the latent variable “environmental knowledge, publicity and edu-
cation” (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

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 4 EA1, EA2, EA3, EA4 0.842 High

Social norms 4 SN1, SN2, SN3, SN4 0.763 Higher
Environmental knowledge & Publicity and education 2 EP1, EP2 0.791 Higher
Environmental facilities and services 4 EF1, EF2, EF3, EF5 0.885 High

229
X. Meng et al. Resources, Conservation & Recycling 140 (2019) 224–234

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

SN4 and EP is added to the model. Moreover, in light of actual situa- 4.3. Result analysis of modified model
tions, residents are easily affected by the behavior of neighbors, family
members, etc., and are able to learn. They absorb knowledge and in- The test results of the structural equation model indicate that 13
formation on classification and recycle and form a habit of classification path assumptions (H1, H3, H4, H1a, H1c, H1d, H2a, H3a, H3b, H4a, H4b,
and recovery when encouraged by people around them. Therefore, H4c, H4e) out of the 19 basic path assumptions of the initial conceptual
based on the model analysis theory above and actual situations, this model of urban residents’ HSW disposal behaviors are established
study considers adding a path between the observed variable “influence through examination. A newly added relation path is also established
of neighbors” (SN4) and the latent variable “environmental knowledge, through examination, that is, the “influence of neighbors” has a positive
publicity and education” (EP). The parameter path coefficients of the effect on residents’ “environmental knowledge, publicity and educa-
modified model are estimated and the results are shown in Fig. 3. tion”. The supported hypotheses in the modified model and standar-
All estimated values of the path coefficients in the modified model dized path coefficient estimates are shown in Fig. 4.
are significant at the 0.05 level, and most of the parameters are sig- The size of the left-sided standardized path coefficient in Fig. 4 re-
nificant at the 0.01 level, indicating good significance and that the presents the degree of direct influence of the observed variables on the
model is credible at the 95% confidence level and up to standard after latent variables, and the size of the three right-side standardized path
modification. Comparing the index evaluation results of the initial hy- coefficients represents the direct influence of the latent variables on the
pothetical model and the modified model (Table 6), it can be seen that: target variable (RWDB). The product represents the degree of indirect
the result of chi-square (χ2) test has dropped from 365.2 to 229.8, and effect of the observed variables on the target variables. For instance:
at the same time, each fitting index is better than that before mod-
ification; the RMSEA is less than 0.08, indicating an acceptable model (1) The degree of influence of “willingness to participate in HSW dis-
fit, and the GFI, CFI, NFI and IFI are all greater than 0.90, showing that posal behavior” on “HSW disposal behavior” is:
the modified model of RWDB has a good fit (Chen, 2016; Bortoleto 0.72 × 0.20 = 0.144;
et al., 2012). (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 0.605 – –

EA2 < — EA 0.371 7.251 *** (Strong Support)
EA3 < — EA 0.690 11.328 *** (Strong Support)
EA4 < — EA 0.725 11.483 *** (Strong Support)
EP1 < — EP 0.643 – –
EP2 < — EP 0.543 5.972 *** (Strong Support)
EF3 < — EF 0.550 – –
EF5 < — EF 0.683 5.921 *** (Strong Support)
EA < — RWDB 0.188 3.925 *** (Strong Support)
EP < — RWDB 0.158 1.974 0.048 (Support)
EF1 < — EF 0.443 6.566 *** (Strong Support)
SN1 < — SN 0.498 – –
SN2 < — SN 0.510 1.884 0.060 (No Support)
SN < — RWDB 0.043 0.787 0.431 (No Support)
EF2 < — EF 0.480 8.954 *** (Strong Support)
SN4 < — SN 0.574 5.846 *** (Strong Support)
EF < — RWDB 0.119 2.246 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
X. Meng et al. Resources, Conservation & Recycling 140 (2019) 224–234

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

“HSW disposal behavior” is: 0.68 × 0.23 = 0.156; 5. Discussions and policy recommendations
(3) The degree of influence of “publicity and education” on “HSW
disposal behavior” is: 0.81 × 0.23 = 0.186. 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'
Similarly, the impact of the remaining eight observed variables on HSW disposal behavior. Most of the hypotheses of the model were
HSW disposal behavior can be calculated: the overall impact of “en- supported. On the basis of the analysis results of SEM modeling, "en-
vironmental awareness”, “citizens’ social responsibility” and “beha- vironmental facilities and services" has the most comprehensive effect,
vioral attitudes” on “HSW disposal behavior” is 0.138, 0.12 and 0.074, while “publicity and education”, “accessibility to recycling facilities”,
respectively, which means that for each additional unit of residents’ “accessibility to classification facilities”, “willingness to participation of
environmental awareness, social responsibility, and behavioral atti- classification” and “environmental awareness of residents” are the five
tudes, utility of their HSW disposal behavior increases by 0.138, 0.12, most significant factors. On the whole, residents’ behavioral selection of
and 0.074 units respectively; “accessibility to classification facilities”, HSW disposal is mainly under the joint action of four intrinsic factors
“time spent”, and “economic cost & benefits” on “HSW disposal beha- and seven external factors. In addition, the combined effect of external
vior” is 0.127, 0.110 and 0.101. Therefore, for example, if the con- factors on residents’ HSW disposal behavior is 0.865, which is nearly
venience of classification facilities increases by 1 unit, the utility of twice that of the combined effects of intrinsic factors (0.476). Further,
HSW disposal behavior will increase by 0.127 units. For each additional based on the above analysis and discussions of the research results, we
unit of time spent on waste management, behavioral utility reduces by proposed some recommendations on urban solid waste classification
0.110 units; the impact of “knowledge and information on classification and recycling management.
and recycling” and “influence of neighbors” on “HSW disposal beha- First of all, it is suggested to strengthen the planning and con-
vior” are 0.094 and 0.090, respectively. struction of urban household solid waste classification and recycling
Through the above fitting analysis by structural equation model, the facilities. According to the research results, “environmental facilities
main influencing factors of residents' HSW disposal behavior and the and services” has the most comprehensive effect on residents’ behavior,
sensitivity coefficient corresponding to each factor are obtained. The and accessibility to classification and recycling facilities are key factors
sensitivity coefficients represent the degree of importance of factors to affecting residents’ HSW disposal behavior. However, at present, the
residents' behavior, and the fitting results and the importance ranking construction of a back-end classification, recycling and transportation
of the factors are shown in Table 7. system in most Chinese cities is lagging behind; the existing construc-
tion plans only set principles for the construction of a waste classifi-
cation, recycling and transportation system. An incomplete and in-
efficient back-end classification, recycling and transportation facilities

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

Observed value Comment Observed value Comment

χ (Chi-square)
2
365.2(P = 0.000) The model is rejected 229.757(P = 0.021) The model is accepted
GFI 0.853 Acceptable model fit 0.914 Good model fit
CFI 0.856 Acceptable model fit 0.933 Good model fit
RMSEA 0.082 Unacceptable model fit 0.056 Acceptable model fit
NFI 0.914 Good model fit 0.952 Good model fit
IFI 0.857 Acceptable model fit 0.926 Good model fit
AIC 427.227 – 281.757 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
X. Meng et al. Resources, Conservation & Recycling 140 (2019) 224–234

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

Table 7 mobile client and mobile Internet; promote knowledge and information
Fitting results of the structural equation model for residents’ HSW disposal on waste classification and recycling, and encourage green and civilized
behavior. lifestyles; and organize training at working units, schools, and com-
Observed variables (factors) Sensitivity coefficients Importance munities to enhance residents’ environmental awareness and will-
(Standardized estimate) ranking ingness to participate, and promote their classification and recycling
knowledge.
Willingness to participate in 0.144 3
Last but not least, it is suggested to improve laws and regulations.
classification
Environmental awareness 0.138 4
According to the result of parameter estimation of the initial model in
Behavioral attitudes 0.074 11 Section 4.2, the coefficient for path “Laws and regulations” (SN2) to“-
Citizens’ social responsibility 0.12 6 social norms” (SN) is not significant, thus the initial path hypothesis
Influence of neighbors 0.090 10 that SN2 have a positive effect on SN perceived by residents is not
Knowledge and information on 0.094 9
supported. The actual reason is that there was no mandatory laws and
Classification and recycling
Publicity and education 0.186 1 regulations and incentive mechanisms on urban HSW classification and
Accessibility to classification 0.127 5 recycling in Suzhou city in that period, and almost all of the re-
facilities spondents' HSW disposal behaviors, regardless of whether they parti-
Accessibility to recycling 0.156 2
cipated in HSW recycling, are not significantly affected by laws and
facilities
Time spent 0.110 7
regulations. This result and situation are quite different from other
Economic cost & benefits 0.101 8 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
will seriously damp front-end residents’ enthusiasm of classification. countries or regions have established a sound system of supporting laws
Therefore, it is recommended that all cities in China incorporate de- and regulations on HSW classification and recycling, and the incentives
tailed waste separation and recycling facilities and system plans during and penalties are effective at promoting waste recycling and reducing
the planning and construction, and accelerate the establishment of a contamination.
complete waste classification and management system for “classifica- However, there is still a big gap and room for improvement in this
tion & disposal, classification & recycling, classification & transporta- aspect in mainland China. At present, China mainland area has not
tion, and classification & management”. At the same time, the con- established enough effective laws and regulations, incentive mechan-
struction of recovery sites for renewable resources and a standardized isms, or mandatory restraint policies on waste classification and re-
resource recovery system should be accelerated to make recycling fa- cycling management. For example, the "Compulsory Recycling List"
cilities or services more accessible. proposed in the "Solid Waste Pollution Prevention Law" and the
Then, it is very necessary to carry out extensive publicity and "Circular Economy Promotion Law" have not been formally established.
education activities on waste classification and recycling among re- And residents as waste producers, are not subject to laws and regula-
sidents through various channels. Research shows that publicity and tions, and their participation in classification and recovery is virtually
education is the most significant factor affecting residents’ HSW dis- voluntary and out of self-discipline. Therefore, it is required that the
posal behavior. When residents are exposed to such publicity and governments strengthen the top-level design of urban household solid
education more frequently through more channels, they participate waste classification and recycling, formulate laws and regulations with
more in waste classification and recycling. Therefore, it is re- “teeth”. It is suggested with the opportunity of amending "The Law of
commended that governments and other administrative departments the People's Republic of China on the Prevention and Control of
strengthen the publicity and education on waste classification and re- Environmental Pollution by Solid Wastes" and the "Circular Economy
cycling through channels such as media, education at school, and ad- Promotion Law", to set up a special chapter to stipulate the basic
vertisements; make full use of media such as television, billboards, principles of waste sorting and collection, the rights and obligations,

232
X. Meng et al. Resources, Conservation & Recycling 140 (2019) 224–234

core systems, and legal responsibilities. Meanwhile, it is recommended issue, some other methodology, such as simulation methods based on
that the State Council introduce special regulations on waste separation complex adaptive system (CAS) theory, can be tried in future research.
management to further clarify the supervision system, residents’ re-
sponsibilities and obligations system, supporting mechanism of reward Acknowledgments
and punishment and credit system.
I would like to extend my gratitude to the general program of China
6. Conclusions Postdoctoral Science Foundation (2018M631585), the Strategic Pilot
Project of “the Research on the Bottleneck Problems of Resource and
The present study proposed the behavioral decision-making me- Environment for the 100-Year Construction of a Strong Country”
chanism which considers both residents’ participation in source classi- (Y8X0771601) launched by the Chinese Academy of Sciences for their
fication and resource recycling, obtained the main factors have a sig- grant support, and General Programs of the National Natural Science
nificant effect on residents’ HSW disposal behaviors and their degree of Foundation of China (71774099). I also would like to thank the staff of
influence. On the basis of the results of our study, the main conclusions the Suzhou Environmental Sanitation and Administration Agency and
are as follows. other organizations for their assistance in the research and the 10 stu-
First, on the whole, residents’ behavioral selection of HSW disposal dents from Suzhou University of Science and Technology who helped
is mainly under the joint action of four intrinsic factors (willingness to me conduct the questionnaire survey.
participate, environmental awareness, social responsibility and beha-
vioral attitudes) and seven external factors (publicity and education References
received, accessibility to recycling facilities, influence of neighbors,
accessibility to classification facilities, time spent, economic cost & Abbott, A., Nandeibam, S., O’Shea, L., 2011. Explaining the variation in household re-
benefits, and knowledge on classification and recycling). The combined cycling rates across the UK. Ecol. Econ. 70 (11), 2214–2223.
Abdelradi, F., 2018. Food waste behaviour at the household level: a conceptual frame-
effect of external factors on residents’ HSW disposal behavior is nearly work. Waste Manag. 71, 485–493.
twice that of the combined effects of intrinsic factors. Ajzen, I., 1985. From intentions to actions: a theory of planned behavior. Action-Control:
Second, for the three latent variable facets, “environmental facilities Cogn. Behav. 1–63.
Babaei, A.A., Alavi, N., Goudarzi, G., et al., 2015. Household recycling knowledge, atti-
and services” has the greatest combined impact on residents’ HSW tudes and practices towards solid waste management. Resour. Conserv. Recycl. 102,
disposal behavior, followed by “environmental attitudes”. 94–100.
“Environmental knowledge, publicity and education” comes last. In 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.
terms of single factors, the most prominent five factors affecting re- Recycl. 41 (1), 65–73.
sidents’ HSW disposal behavior are: “publicity and education”, “ac- Begum, R.A., Siwar, C., Pereira, J.J., Jaafar, A.H., 2009. Attitude and behavioral factors in
cessibility to recycling facilities”, “willingness to participate”, “aware- waste management in the construction industry of Malaysia. Resour. Conserv. Recycl.
53 (6), 321–328.
ness of environmental protection” and “accessibility to classification
Bilitewski, B., 2008. From traditional to modern fee systems. Waste Manag. 28 (12),
facilities”. However, the impact of economic cost & benefits and time 2760–2766.
spent on the residents’ participation in waste classification and re- Boldero, J., 1995. The prediction of household recycling of newspapers: the role of atti-
cycling is relatively smaller. tudes, 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.
Third, the impact of laws and regulations on residents’ behavioral Methodol. 69 (39), 66–67.
selection of household solid waste management does not reach a sta- Boonrod, K., Towprayoon, S., Bonnet, S., et al., 2015. Enhancing organic waste separation
tistically significant level, as current laws and regulations on waste at the source behavior: a case study of the application of motivation mechanisms in
communities in Thailand. Resour. Conserv. Recycl. 95, 77–90.
classification and recycling in Suzhou are mostly incentive and in- Borthakur, A., Govind, M., 2017. Emerging trends in consumers’ E-waste disposal beha-
structive, with no mandatory restraint policies issued yet. That is to say, viour and awareness: a worldwide overview with special focus on India. Resour.
the impact of policies, laws, and regulations on residents surveyed in Conserv. Recycl. 117, 102–113.
Bortoleto, A.P., Kurisu, K.H., Hanaki, K., 2012. Model development for household waste
real life is generally much limited regarding whether or not residents prevention behaviour. Waste Manag. 32, 2195–2207.
participate in waste classification and recycling. The assumption path Botetzagias, I., Dima, A.F., Malesios, C., 2015. Extending the theory of planned behavior
that storage space occupied has a negative impact on residents’ external in the context of recycling: the role of moral norms and of demographic predictors.
Resour. Conserv. Recycl. 95, 58–67.
environmental factor, “environmental facilities and services”, is not Bureau of Statistics of Suzhou, 2017. Suzhou Statistical Yearbook 2017. China Statistics
established. Press, Beijing.
Finally, the government should strengthen the planning and con- Charuvichaipong, C., Sajor, E., 2006. Promoting waste separation for recycling and local
governance in Thailand. Habitat Int. 30, 579–594.
struction of urban household solid waste classification and recycling
Chen, M.F., 2016. Extending the theory of planned behavior model to explain people’s
facilities, carry out extensive publicity and education activities on waste energy savings and carbon reduction behavioral intentions to mitigate climate
classification and recycling among residents through various channels, change in Taiwan-moral obligation matters. J. Clean. Prod. 112, 1746–1753.
and improve laws and regulations in urban comprehensive manage- Chou, P.F., Lu, C.S., Chang, Y.H., 2014. Effects of service quality and customer satisfac-
tion on customer loyalty in high-speed rail services in Taiwan. Transp. Metr. A-
ment in future. Transp. Sci. 10 (10), 917–945.
Of course, the current research still includes some limitations which Chu, Z., Xi, B., Song, Y., et al., 2013. Taking out the trash: household preferences over
need further improvement. First, the samples of the study primarily 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.
select the residents of the central urban area in Suzhou with developed Public Transportation Quality of Service, pp. 141–154.
economy and favorable foundation, the results may not exactly re- Deng, J., Xu, W.Y., Zhou, C.B., 2013. Investigation of waste classification and collection
present entire China scenario. In the future, studies on residents in the actual effect and the study of long acting management in the community of Beijing.
Environ. Sci. 34 (1), 395–400.
middle-developed, underdeveloped areas and suburbs should be in- Fei, F., Qu, L.L., Wen, Z.G., et al., 2016. How to integrate the informal recycling system
creased, and the decision-making mechanism of RWDB may be dif- into municipal solid waste management in developing countries: based on a China’s
ferent, so as to formulate relevant management measures according to 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
local conditions. Second, as there is no clear definition of the concept of behavior to understand individual’s energy saving behavior in workplaces. Resour.
environmental attitudes at present, the selection of possible observation Conserv. Recycl. 127, 107–113.
variables and the design of measurement items in the questionnaire to Grazhdani, D., 2016. Assessing the variables affecting on the rate of solid waste gen-
eration and recycling: an empirical analysis in Prespa Park. Waste Manag. 48, 3–13.
reflect it may not be very accurate. This could be a topic for further
Guagnano, G.A., Stern, P.C., Dietz, T., 1995. Influences on attitude-behavior relation-
investigation. In addition, the study of individual’s environmental be- ships: a natural experiment with curbside recycling. Environ. Behav. 27 (5), 699–718.
haviors based on statistical analysis of questionnaire survey data has Guo, H., Hobbs, B.F., Lasater, M.E., et al., 2016. System dynamics-based evaluation of
certain methodological limitations and some subjectivity. To solve this interventions to promote appropriate waste disposal behaviors in low-income urban

233
X. Meng et al. Resources, Conservation & Recycling 140 (2019) 224–234

areas: a Baltimore case study. Waste Manag. 56, 547–560. India. Resour. Conserv. Recycl. 135, 313–322.
Izagirre-Olaizola, J., Fernández-Sainz, A., Vicente-Molina, M.A., 2015. Internal determi- Song, Q., Wang, Z., Li, J., 2012. Residents’ behaviors, attitudes, and willingness to pay for
nants of recycling behaviour by university students: a cross-country comparative recycling e-waste in Macau. J. Environ. Manage. 106, 8–16.
analysis. Int. J. Consum. Stud. 39 (1), 25–34. Stern, P.C., Oskamp, S., 1987. Managing scarce environmental resources. Handbook of
Jackson, T., 2005. Motivating Sustainable Consumption: A Review of Evidence on Environmental Psychology. Wiley, New York.
Consumer Behaviour and Behavioural Change. pp. 43–61. Tadesse, T., 2009. Environmental concern and its implication to household waste se-
Jöreskog, K.G., 1970. A general method for estimating a linear structural equation paration and disposal: evidence from Mekelle, Ethiopia. Resour. Conserv. Recycl. 53
system. Res. Bull. 2, 1–41. (4), 183–191.
Jöreskog, K.G., Sörbom, D., Magidson, J., 1979. Advances in Factor Analysis and Tenenhaus, M., Vinzi, V.E., Chatelin, Y.M., et al., 2005. PLS path modeling. Comput. Stat.
Structural Equation Models. Abt Books, pp. 24–26. Data Anal. 48 (1), 159–205.
Kim, Y.K., Lee, H.R., 2011. Customer satisfaction using low cost carriers. Tour. Manag. 32 Timlett, R.E., Williams, I.D., 2008. Public participation and recycling performance in
(2), 235–243. England: a comparison of tools for behaviour change. Resour. Conserv. Recycl. 52
Lee, P., Mahoney, K.T., Lee, S., 2017. An application of the exploratory structural (4), 622–634.
equation modeling framework to the study of personality faking. Pers. Individ. Differ. Tucker, P., Murney, G., Lamont, J., 1998. Predicting recycling scheme performance: a
119, 220–226. process simulation approach. J. Environ. Manage. 53 (1), 31–48.
Li, J., Tam, V.W.Y., Zuo, J., et al., 2015. Designers’ attitude and behaviour towards Wan, A.W.A.K.G., Rusli, I.F., Dayang, R., et al., 2013. An application of the theory of
construction waste minimization by design: a study in Shenzhen, China. Resour. planned behaviour to study the influencing factors of participation in source se-
Conserv. Recycl. 105, 29–35. paration of food waste. Waste Manag. 33 (5), 1276–1281.
Lindbeck, A., 1997. Incentives and social norms in household behavior. Am. Econ. Rev. Wan, C., Shen, G.Q., Yu, A., 2014. The role of perceived effectiveness of policy measures
87 (2), 370–377. in predicting recycling behaviour in Hong Kong. Resour. Conserv. Recycl. 83 (83),
Liu, Y., Sun, C., Xia, B., et al., 2018. Impact of community engagement on public ac- 141–151.
ceptance towards waste-to-energy incineration projects: empirical evidence from Wan, C., Shen, G.Q., Yu, A., 2015. Key determinants of willingness to support policy
China. Waste Manag. 76, 431–442. measures on recycling: a case study in Hong Kong. Environ. Sci. Policy 54, 409–418.
Lizina, S., Dael, M.V., Passel, S.V., 2017. Battery pack recycling: behaviour change in- Wang, Z., Guo, D., Wang, X., et al., 2018. How does information publicity influence re-
terventions derived from an integrative theory of planned behaviour study. Resour. sidents’ behaviour intentions around e-waste recycling? Resour. Conserv. Recycl.
Conserv. Recycl. 122, 66–82. 133, 1–9.
Matsumoto, S., 2014. The opportunity cost of pro-environmental activities: spending time Wu, M., 2003. SPSS Statistical Application Practice – Questionnaire Analysis and
to promote the environment. J. Fam. Econ. Issues 35 (1), 119–130. Application Statistics. Science Press, Beijing.
Meng, X., Wen, Z., Qian, Y., 2018. Multi-agent based simulation for household solid waste Wu, Z., Yu, A.T.W., Shen, L., 2017. Investigating the determinants of contractor’s con-
recycling behavior. Resour. Conserv. Recycl. 128, 535–545. struction and demolition waste management behavior in Mainland China. Waste
National Bureau of Statistics of China, 2018. Statistical Bulletin of the People’s Republic Manag. 60, 290–300.
of China on National Economic and Social Development in 2017. Xiao, L., Zhang, G., Zhu, Y., et al., 2017. Promoting public participation in household
Nguyen, T.P., Zhu, D., Le, N.P., 2015. Factors influencing waste separation intention of waste management: a survey based method and case study in Xiamen city, China. J.
residential households in a developing country: evidence from Hanoi, Vietnam. Clean. Prod. 144, 313–322.
Habitat Int. 48, 169–176. Yan, H., 2018. Mandatory Classification May be the Opportunity for Changes [N/OL].
Park, J., Ha, S., 2014. Understanding consumer recycling behavior: combining the theory 2016-09-21. China Environment News. http://news.cenews.com.cn/html/2016-09/
of planned behavior and the norm activation model. Fam. Consum. Sci. Res. J. 42 (3), 21/content_50274.htm.
278–291. Yang, K.C., Hsieh, T.C., Li, H., et al., 2012. Assessing how service quality, airline image
Reddi, K.R., Li, W., Wang, B., et al., 2013. System dynamics modelling of hybrid re- and customer value affect the intentions of passengers regarding low cost carriers. J.
newable energy systems and combined heating and power generator. Int. J. Sustain. Air Transp. Manag. 20, 52–53.
Eng. 6 (1), 31–47. Young, R.D., 1990. Recycling as appropriate behavior: a review of survey data from se-
Sakai, S., Ikematsu, T., Hirai, Y., et al., 2008. Unit-charging programs for municipal solid leeted recycling edueation programs in Michigan. Resour. Conserv. Recycl. 3 (4),
waste in Japan. Waste Manag. 28 (12), 2815–2825. 253–266.
Singh, M.P., Chakraborty, A., Roy, M., 2018. Developing an extended theory of planned Zhang, H., Wen, Z., 2014. Residents’ household solid waste (HSW) source separation
behavior model to explore circular economy readiness in manufacturing MSEMEs, activities: a case study in Suzhou, China. Sustainability 6, 6446–6466.

234