sustainability

Article
Internal and External Influential Factors on Waste
Disposal Behavior in Public Open Spaces in Phnom
Penh, Cambodia
Pagnarith Srun and Kiyo Kurisu *
Department of Urban Engineering, The University of Tokyo, Tokyo 113-8654, Japan; prith.srun@gmail.com
* Correspondence: kiyo@env.t.u-tokyo.ac.jp; Tel.: +81-090-9009-2638




Received: 12 February 2019; Accepted: 6 March 2019; Published: 13 March 2019


Abstract: Over the last decade, municipal solid waste generation in Phnom Penh has increased
noticeably; however, the waste management system is far from satisfactory. Considerable amount
of waste is left uncollected, as well as intentionally disposed of in public open spaces. External and
internal factors can trigger these problems. Possible external factors are low collection frequency,
low cleaning services, and insufficient facilities such as small dumpsters. Possible internal factors,
which also play an important role in this issue, include low awareness, insufficient knowledge,
and low responsibility for personal waste. To examine the influences of these internal and external
factors on people’s waste disposal behaviors, we selected and conducted a questionnaire survey at
four sites in Phnom Penh that differ in waste collection frequency and population density. A total of
413 valid responses were obtained. We developed a structural equation model to explain people’s
intentions not to dispose of waste in public open spaces. The results showed that personal and social
norms, such as perception of social pressure from friends and family and from the government,
had significant influences on intention, whereas the influence of external factors was much smaller.

Keywords: waste disposal in public open space; waste collection service; psychological factor;
questionnaire survey; path analysis; hypothetical model

1. Introduction
Phnom Penh, the capital city of Cambodia, has faced a significant increase in waste generation.
Waste management systems have not been developed well and urgent improvement is considered
necessary [1,2]. In Phnom Penh, waste is frequently disposed of and seen to remain in public open
spaces. For example, many bags containing waste are disposed of right beside signs proclaiming
“no waste dumping”. Residents also leave their bags of waste in front of their houses, even when
collection trucks are not expected. This harms human health and the city landscape, and the dumped
and scattered waste clogs drains and causes flooding [3]. Frequent disposal of waste in public open
spaces may be caused by poor waste management systems, such as irregular collection, insufficient
containers, poor cleaning services, and so on. Another potential reason is low awareness about waste
disposal among residents. Therefore, to address this issue, external conditions and internal awareness
should be considered together. Thondhlana and Hlatshwayo (2018) also suggested that understanding
people’s internal conditions can be one of the effective methods to achieve environmental sustainability
as well as taking into consideration the external conditions [4].
Many studies have focused on one or the other aspect—external or internal—with respect to
waste-related behaviors, especially recycling behavior. Nguyen et al. (2019) showed that laws and
regulations were more influential than other determinants as well as the inconvenience and cost related
to recycling intention [5]. Xu et al. (2017) focused on the influence of external conditions on people’s

Sustainability 2019, 11, 1518; doi:10.3390/su11061518 www.mdpi.com/journal/sustainability

Sustainability 2019, 11, 1518 2 of 14

recycling behaviors in China and showed that government interventions, such as the improvement
of segregation infrastructure and the provision of recycling bins, had positive influences on recycling
intention [6]. Conke (2018) focused on the influence of collection service on recycling behavior in
Brazil. He compared two cities in which collection schedules were regular and irregular, respectively.
The results showed that the city with regular collection reached 90% of recycling participation,
in contrast to 45% recycling participation in the irregular-collection city [7]. With respect to internal
influences, many studies have focused on psychological factors for determining recycling behaviors
and intentions [8–13]. Strydom (2018) showed that subjective norms had the largest substantial positive
effects on the recycling intention of the residents in South Africa [14]. Hornik et al. (1995) noted that
in addition to internal facilitators, external facilitators had some influence on recycling behavior [15].
Barr (2003) proposed “curbside bins” as a factor to explain recycling and waste minimization behaviors,
in addition to psychological factors, and it showed significant influence on recycling behavior [16].
Stoeva and Alriksson (2017) evaluated local waste management service conditions, such as collection
and waste bins, under the term “satisfaction with the local service”. They showed that the perception
of external conditions had an important role in predicting waste recycling intentions, in addition
to other psychological factors, and suggested that satisfactory service conditions could increase the
recycling rate [17].
Few studies have focused specifically on waste disposal behaviors in open spaces. Some have
focused on the related behaviors of “littering”. Cialdini et al. (1990) focused on the influence of
social norms, such as injunctive and descriptive norms, on littering behavior [18]. Schultz et al. (2013)
investigated littering behaviors in the United States and showed that the availability of existing litter
and trash receptacles had significant influence on the behavior [19]. They also revealed the significant
influence of age and smoking behavior. Al-Khatib et al. (2009) also focused on the sociodemographic
effects on littering behavior in the Nablus district of the West Bank, demonstrating that religious
convictions had a clear influence on littering habits [20].
Waste disposal behavior in open spaces is not the same as littering and recycling behaviors, but are
similar to some extent. This behavior can be influenced by norms and waste management conditions,
such as collection frequency and waste bin provision. Therefore, in this study, we focused on the
influences of external conditions and internal psychological factors on residents’ intentions not to
dispose of waste in public open spaces, and revealed the factors with significant influence in reducing
open-space waste disposal in Phnom Penh. In previous studies, the influences from internal factors
and external factors were separately discussed and a comparative evaluation was not well conducted.
This study shows the influential pathways and their strength quantitatively by using model analysis,
and gives insight into the influential factors on waste disposal in public open spaces.

2. Methods

2.1. Study Area

Phnom Penh consists of 9 districts and 36 communes, and the provision of waste collection
services varies by location. To determine the study sites, we sought a holistic understanding of current
waste management service conditions, such as collection frequency and facility provision. Cintri,
a private company in charge of collection services for the whole city, provided information on waste
collection in each district in the city [21]. These data were compiled into a digital map by a geographic
information system (GIS; ArcGIS Desktop ver. 21, ESRI, Tokyo, Japan), shown in Figure 1. As Figure 1
illustrates, the central part of the city receives daily waste collection services; in contrast, rural areas
receive collection services only two or three days a week. Two types of dumpsters with 4 ton capacity
(4T) and 10 ton capacity (10T) are provided, and the latter ones are located mainly in the central area.
Sustainability 2019, 11, 1518 3 of 14
Sustainability 2019, 11 FOR PEER REVIEW 3

Figure
Figure 1.
1. Waste
Waste service
service data
data map
map of
of Phnom
Phnom Penh.
Penh.

We assembled
We assembled the the data
data and
and calculated
calculated population
population density
density by
by population
population [22]
[22] and
and area
area
(calculated by GIS). The index “collection frequency/population density” was used to select
(calculated by GIS). The index “collection frequency/population density” was used to select the study the
study sites. As shown in Table 1, four sites were selected, HH, LH, HL, and LL, which
sites. As shown in Table 1, four sites were selected, HH, LH, HL, and LL, which represent varying represent
varying
levels of levels of collection
collection frequency frequency
(H: high,(H: high,and
L: low) L: low) and population
population density
density (H: high,(H: high, L: low).
L: low).

Table 1.
Table Study sites.
1. Study sites.

Collection
Collection Population
Population
Commune Frequency Density Streets Surveyed
Commune Frequency Density Streets Surveyed
(Times/Week) (pop/km2 )
(Times/Week) (pop/km ) 2
HH Phsar Depou Ti Pir High 7 High 53,700 St. 134, 138, 146, 221, 225, 233
HH Phsar
LH Depou Ti Pir
Tuol Sangka High
Low 37 High
High 53,700
19,060 St. 103,
St. 134,105,
138,273,
146,518,
221,
590225, 233
LH HL TuolToek
Sangka
Laak I Low
High 73 High
Low 19,060
14,858 St. 132,
St. 103,138,
105,261,
273,
265518, 590
HL LL Toek Laak
Steung MeanI Chey High
Low 37 Low
Low 14,858
7864 St. 38,
St. 23, 132,62,138,
217,261,
371 265
LL Steung Mean Chey Low 3 Low 7864 St. 23, 38, 62, 217, 371
2.2. Theoretical Framework and Research Hypotheses
2.2. Theoretical Framework and Research Hypotheses
A number of theories have attempted to explain why people do or do not engage in a certain
A number
behavior. One ofofthe theories have attempted
most applauded was the to theory
explainofwhy people
reasoned do or(TRA).
action do notTRA,
engage
firstin a certain
introduced
behavior. One of the most applauded was the theory of reasoned action
by Ajzen and Fishbein, proposed that “behavior” was determined by “intention,” which was in turn(TRA). TRA, first introduced
by Ajzen andby
determined Fishbein,
“attitude proposed
toward that “behavior”and
the behavior” was“subjective
determinednorms” by “intention,” whichrefers
[23]. Attitude was intoturn
the
determined by “attitude toward the behavior” and “subjective norms”
degree that people value the behavior positively or negatively. Subjective norms are the perceived[23]. Attitude refers to the
degree that people
social pressure fromvaluethose the behavior
a person positively
is close to, suchor as negatively. Subjective
family and friends, norms whether
regarding are the perceived
to engage
social
in the pressure
behavior or from not.those
This amodel
person is close
implied to,ifsuch
that people as valued
family aand friends,
behavior regarding
positively andwhether
perceivedto
engage in the behavior or not. This model implied that if people valued
that their friends and family expected them to perform the behavior, they would intend to perform a behavior positively and
perceived
the behavior. that their
However, friends and family
although manyexpected
research them to perform
findings agreethe behavior,
with they (e.g.,
this theory wouldParkintend to
et al.,
perform
2009 [24]),theitbehavior. However,a although
has also received great dealmany research
of criticism forfindings agree with
its assumption thatthis theorybehavior
people’s (e.g., Parkis
et al., 2009 [24]), it has also received a great deal of criticism for its assumption
determined solely by their intention. Such criticism noted that the model applies only when the that people’s behavior
is determined
behavior solely
is under by theircontrol.
volitional intention.
Thus,Such
Ajzencriticism
(1991) noted that the
introduced model
another applies
theory only when
to predict the
people’s
behavior is under volitional control. Thus, Ajzen (1991) introduced another theory
behavior: the theory of planned behavior (TPB) [25]. This theory, a revision of the controversial TRA, to predict people’s
behavior:
introduced the theory
one moreofpredictor,
planned behavior
perceived(TPB) [25]. This
behavioral theory,
control a revision
(PBC), which of the controversial
determines intentionTRA,
and
introduced
behavior. PBC onerepresents
more predictor, perceived
how much behavioral
the target behavior control
can be(PBC), whichbydetermines
controlled intention and
an actor himself/herself.
behavior. PBC represents how much the target behavior can be controlled by an actor himself/herself.
In other words, PBC refers to the degree to which people perceive the behavior as easy or difficult to
Sustainability 2019, 11, 1518 4 of 14

In other words, PBC refers to the degree to which people perceive the behavior as easy or difficult
Sustainability 2019, 11 FOR PEER REVIEW 4
to perform. TPB has become the most popular theory to predict people’s behaviors and it has been
perform.
widely TPBto
applied has become the behaviors
waste-related most popular suchtheory to predict
as recycling people’s
[26–28] behaviors and
and prevention it has been
[29,30].
widely applied to waste-related behaviors such as recycling [26–28] and prevention [29,30].of personal
Schwartz introduced the norm activation model, which explained that the activation
norms Schwartz introduced
is an important the norm
process activation model,
in determining which
altruistic explained
behavior that thePersonal
[31,32]. activationnorms
of personal
are the
norms is an people
“expectations important hold process in determining
for themselves” altruistic
regarding behavior
whether they[31,32].
should Personal
or shouldnorms
notare the
perform
“expectations
a certain behavior people
[31].hold
Thisformodel
themselves” regarding whether
also explained they should
that personal normsorare should not through
created perform athe
certain behavior
internalization [31]. This
of social norms model alsoBortoleto
[32,33]. explainedetthat personal
al. (2012) norms the
showed aresignificant
created through the of
influence
internalization
personal norms on ofwaste
social prevention
norms [32,33]. Bortoleto
behaviors [29].et al. (2012) showed the significant influence of
personal normstoonthe
In addition waste
TPBprevention
variables andbehaviors
personal [29].norms, other variables that predict waste-related
In addition to the TPB variables
behaviors have been reported. Hornik et al. (1995) and personal norms, other variables
showed that besidesthat attitude
predict waste-related
and intention,
behaviors have been reported. Hornik et al. (1995) showed that
laziness and ignorance also played a crucial role in predicting people’s recycling behavior besides attitude and intention,
[15]. Davies
et laziness
al. (2002) andreported
ignorance alsothe
that played a crucialofrole
evaluation in predictingwas
effectiveness people’s recycling behavior
an important predictor[15]. Davies
of recycling
et al. (2002) reported that the evaluation of effectiveness was an important predictor of recycling
behavior [11]. Knowledge, concern, and obligation to environmental issues have also been identified
behavior [11]. Knowledge, concern, and obligation to environmental issues have also been identified
in determining recycling and waste prevention intentions [30,34]. Chiang et al. (2019) suggested
in determining recycling and waste prevention intentions [30,34]. Chiang et al. (2019) suggested that
that locus of control is a key determinant to promote pro-environmental behavior [35]. For waste
locus of control is a key determinant to promote pro-environmental behavior [35]. For waste littering,
littering, locus of control has also been proposed as an influential determinant as well as altruism and
locus of control has also been proposed as an influential determinant as well as altruism and self-
self-efficacy (e.g., Ojedokun, 2011; Ojedokun and Balogun, 2011) [3,36]. Self-efficacy is a belief in one’s
efficacy (e.g., Ojedokun, 2011; Ojedokun and Balogun, 2011) [3,36]. Self-efficacy is a belief in one’s
ability to to
ability accomplish
accomplisha atask taskor orsucceed
succeedin in specific situations.Locus
specific situations. Locusofofcontrol
controlrefers
refers
toto whether
whether people
people
believe that they or external powers are in control
believe that they or external powers are in control of their life. of their life.
Basedononthese
Based thesetheoretical
theoreticalframeworks,
frameworks, the the present
present study
study developed
developeda ahypothetical
hypothetical model
model
involving
involving keykey internal
internal andexternal
and externalfactors
factorsfor forwaste
wastedisposal
disposal behavior
behavior inin public,
public, as
as shown
shownin inFigure
Figure 2.
This2. This
model model set “(a)
set “(a) intention
intention notnot
to to disposeofofwaste
dispose waste inin public”
public” (Int_nd)
(Int_nd)asasthe thetarget
targetdependent
dependent
variable.InInaddition,
variable. addition,the the variable
variable “(b)
“(b) intention
intention to to keep
keephouse
houseandandoutdoor
outdoorsurroundings
surroundings clean”
clean”
(Int_kc)
(Int_kc) was
was assumedtotohave
assumed havean aninfluence
influenceon on Int_nd
Int_nd (H1).
(H1).

c) Attitude toward waste H2

disposal behavior (Att) b) Intention to
keep outside
H12 house and
g) Personal norms (Pn)
H11 outdoor
H9 surroundings
H10 H4 clean (Int_kc)
d) Subjective norms (Sb)
H3 H1
f) Social pressure (Sp) H8
H7
H6 a) Intention not
e) Perceived behavioral to dispose of
control (Pbc) H5 waste in public
(Int_nd)
i) Laziness (Lz), H14
Ignorance (Ig) and H16
E1) Collection H20 H20 Locus of control (Lc) H13
H22 H15
frequency (Cf)
E3) Population j) Satisfaction with h) Past waste disposal
density (Pd) current services (Sat) H17 experience (Exp)
E2) Facility (bin) H19
H23 provision (Fp) H21
k) Concern about waste
management (Con) H18

Figure2.2.Hypothetical
Figure Hypothetical model
model for public
publicwaste
wastedisposal
disposalbehaviors.
behaviors.

The
The TPB
TPB modelwas
model wasadopted
adoptedwith
with respect
respect to internal
internal factors
factorsdetermining
determiningintentions.
intentions.This model
This model
postulates three conceptual determinants, such as “(c) attitude toward the behavior” (Att), “(d)
postulates three conceptual determinants, such as “(c) attitude toward the behavior” (Att), “(d)
subjective
subjective norms”(Sb),
norms” (Sb),and
and“(e)
“(e)perceived
perceived behavioral
behavioral control”
control”(Pbc),
(Pbc),that
thatpredict
predictpeople’s
people’sintentions
intentions
and
and behavior.
behavior. Ourmodel
Our modelassumed
assumedthat
that positive
positive Att
Att has
has aapositive
positiveinfluence
influenceononInt_nd (H2),
Int_nd and
(H2), that
and that
Sb and Pbc have influences on both Int_nd (H3, H5) and Int_kc (H4, H6). Our model
Sb and Pbc have influences on both Int_nd (H3, H5) and Int_kc (H4, H6). Our model also assumed also assumed
some
some additionalinternal
additional internal factors,
factors, namely,
namely,“(f)
“(f)social
socialpressure”
pressure”(Sp) andand
(Sp) “(g)“(g)
personal norms”
personal (Pn). In
norms” (Pn).
addition to perception of social pressure from friends and family, perception of institutional and
Sustainability 2019, 11, 1518 5 of 14

In addition to perception of social pressure from friends and family, perception of institutional and
community-level social pressures can influence people’s intentions. Therefore, we involved Sp as well
as Sb and assumed that Sp has positive influences on Int_nd (H7) and Int_kc (H8). The model also
assumed that Pn has positive influences on Int_nd (H11) and Int_kc (H12), and that social norms such
as Sb and Sp are internalized into Pn (H9, H10).
Past experiences and behaviors have been reported to influence current intentions and behaviors.
For example, Tonglet et al. (2004) showed that the main determinants of recycling behavior were
recycling attitude, previous recycling experience, concern for the community, and consequences of
recycling [37]. Thus, our model assumed the influence of “(h) past experience of waste disposal in
public” (Exp) on Int_nd (H13).
Ojedokun (2011) found that personal attributes, such as altruism and locus of control, had impacts
on people’s attitude toward littering and responsible environmental behaviors [3]. Hornik et al. (1995)
suggested that the basic problems of recycling were internal barriers such as ignorance, laziness,
and inconvenience [15]. McCarty and Shrum (1994) showed that the inconvenience of recycling
strongly influenced people’s recycling behavior [38]. Based on these previous studies, we assumed
that “(i) laziness, ignorance, and locus of control” (Lz, Ig, and Lc) have influences on Int_nd (H14) and
Exp (H15).
In considering influences from external factors, two possible aspects were involved. First was
the objective condition of waste management, indicated by “(E1) collection frequency” (Cf) and “(E2)
facility (bin) provision” (Fp); second was people’s perception of waste management based on current
external conditions, indicated by “(j) satisfaction with current waste management services” (Sat)
and “(k) concern about waste management” (Con). We assumed that “(E3) population density” (Pd)
determines Cf and Fp (H22, H23), Cf and Fp affect Sat (H20, H21), Sat influences Con and Int_nd (H19,
H16), and Sat and Con influence Exp (H17, H18).

2.3. Questionnaire Design

The questionnaire consisted of three parts. The first part (I_1–9) measured demographic
information: age (I_1), gender (I_2), address (I_3), family income (I_4), education level (I_7), family size
(I_8), and house type (I_9). To validate the appropriateness of the answer for family income, car (I_5)
and motorbike (I_6) ownership were also recorded.
The second part (II_1–14) included questions about current waste collection services and how
people managed the waste they generated: awareness of current waste management (II_1, 2), collection
and disposal frequency (II_3, 8), person responsible for waste disposal (II_4), usual number of waste
bags disposed (II_5), disposal location (II_6, 9), method of disposal for recyclable wastes (II_7),
satisfaction with current services (II_10), any experience of disposing of waste in public open spaces
(II_11) and the reason for doing so (II_12), perception of the sufficiency of facility provision (II_13),
and free opinions on current waste collection services (II_14).
The third part (III_1–22) included questions concerning the main internal variables (a–g, i):
Int_nd (III_16), Int_kc (III_20), Att (III_1, 21), Sb (III_2), Pbc (III_3, 9, 13, 18), Sp (III_4, 7, 10, 11, 17,
19), Pn (III_8), Lz (III_12, 22), Ig (III_6, 15), and Lc (III_5, 9). For each of the 22 statements shown,
respondents expressed their opinions using a 6-point scale ranging from 1 “strongly disagree” to 6
“strongly agree”.

2.4. Questionnaire Survey

Target streets were selected for each study site as shown in Table 1, and respondents were
selected by fixed intervals of blocks along the street. For structural equation modeling with more
than 10 variables, it is recommended that the sample size be 100 or greater, as a smaller sample could
produce unstable and insignificant results [29,39]. The total population of Phnom Penh is 1.6 million;
according to Yamane’s (1967) formula, 400 samples are needed for this population to obtain a 95%
Sustainability 2019, 11, 1518 6 of 14

confidence level with significance at the level of 5% [40]. Thus, a sample size of 100 was used for each
of the four sites, resulting in a total sample size of 400.
Waste disposal is a sensitive issue; therefore, face-to-face interviews were considered to be an
appropriate method for the survey, allowing an interviewer to give supplemental explanations if the
interviewee could not understand a question’s meaning. Nine Cambodian university students were
employed to help conduct the interview survey. Prior to the interview campaign, two training sessions
were conducted from 10 January to 11 January 2018. In the first session, all questions were explained to
ensure that the interviewers understood all materials clearly. In the second session, a mock interview
survey was conducted to determine whether the interviewers understood all questions well and
how flexibly they could handle a real situation. After these training sessions, the main survey was
conducted from 12 January to 19 January 2018, collecting a total of 425 samples.

2.5. Analytical Method

All obtained data sheets were coded and input into Excel (Microsoft, Albuquerque, NM, USA).
For statistical and path analysis, SPSS ver. 21 and SPSS Amos ver. 22 (IBM Co., Armonk, NY, USA) were
used, respectively. The obtained data did not follow a normal distribution; therefore, nonparametric
tests including the Kruskal–Wallis and Mann–Whitney U tests were used to compare the average scores
among groups of three or more and groups of two, respectively. For the post hoc (paired comparison)
test, the Dunn–Bonferroni method was used, as provided by SPSS.

3. Results

3.1. Sample Distribution

A total of 413 valid responses were obtained out of the 425 samples. Table 2 shows the
sociodemographic data for the samples with the newest available information about Cambodia from
the national census [22,41]. The sample data show a higher ratio of women than do the census data; this
might be because the survey was conducted during daytime hours on weekdays. With respect to age
distribution, the majority of respondents were in their 20s (31.2%) and 30s (26.5%). In comparison to the
census data, the sample data show a relatively smaller proportion of people over 60. Among the four
survey sites, HL shows a relatively higher ratio of women (67.3%) and LL shows a larger proportion
of people over 50 compared with other regions. According to the national census, 20.7% of people
above age 25 have no or lower education, and only 4.7% of people currently attend university or hold a
degree. In comparison, whereas the ratio of female respondents was relatively high, the educational
level of the respondents is seen to be much higher than that of the census data. Of the four survey
sites, respondents in HH, LH, and HL were more educated than those in LL. With respect to monthly
family income, similar to the census data (about 400 USD/month per family), 43.1% of people reported
a family income ranging from 200 to 500 USD/month. In HH, LH, HL, and LL, 38.4%, 33.0%, 27.9%,
and 19.2% of respondents reported a monthly income of more than 500 USD, respectively. This order
among the four survey sites corresponds to the reported educational level of each region. Additionally,
answers regarding car and motorbike ownership, which were obtained to check the validity of reported
income data, also supported the income conditions of the four sites. When discussing the average
scores of the total area or each site, careful consideration is necessary based on the sociodemographic
characteristics of the respondents and the differences among the four sites.
Sustainability 2019, 11, 1518 7 of 14

Table 2. Sociodemographic data for respondents.

Sample (%)

Category Urban Rural Cambodia Average

Item Total
(n = 413) HH* LH* HL* LL*
(n = 101) (n = 104) (n = 104) (n = 104)
Male 39.7 40.6 46.2 32.7 39.4 48.64 [41]
Gender
Female 60.3 59.4 53.8 67.3 60.6 51.36 [41]
<20 5.3 6.9 7.7 4.8 1.9 34.8 [22]
20s 31.2 23.8 50.0 35.6 15.4 19.4 [22]
30s 26.5 36.6 14.4 26.9 27.9 11.5 [22]
Age
40s 15.8 19.8 13.5 12.5 17.3 10.3 [22]
50s 13.8 9.9 9.6 8.7 26.9 6.6 [22]
≥60 7.5 3.0 4.8 11.5 10.6 17.4 [22]
Uneducated 6.0 5.9 3.8 3.8 10.6 20.7 [41]
Primary school 19.8 18.8 16.3 11.5 32.7 57.9 [41]
Secondary school 18.8 7.9 15.4 24.0 27.9 11.4 [41]
Education level High school 22.5 21.8 11.5 34.6 22.1 5.3 [41]
Bachelor’s degree 31.8 43.6 51.9 25.0 6.7
Master’s degree 0.8 1.0 1.0 1.0 0.0 4.7 [41]
Doctorate 0.3 1.0 0.0 0.0 0.0
<200 USD 26.4 19.8 37.5 18.3 29.8
Monthly family 200–500 USD 43.1 37.6 29.8 53.8 51.0
About 400 USD [41]
income 500–800 USD 19.9 27.7 18.3 19.2 14.4
>800 USD 10.7 14.9 14.4 8.7 4.8
None 80.1 70.3 84.6 78.8 86.5
1 16.0 21.8 11.5 19.2 11.5
Cars owned None: 96% [41]
2 3.7 7.9 3.8 1.0 1.9
3 or more 0.3 0.0 0.0 1.0 0.0
None 7.3 9.9 5.8 4.8 8.7
Motorbikes 1 36.3 36.6 28.8 38.5 41.3
None: 36% [41]
owned 2 30.1 31.7 31.7 28.8 28.3
3 or more 23.0 21.8 33.7 27.9 8.7

3.2. Summary of Obtained Data

Answers regarding internal variables were given as a score from 1 (“strongly disagree”) to
6 (“strongly agree”). Valid question items were extracted on the basis of the single correlation
coefficients with Int_nd and Int_kc and the consistency of the items, checked by Cronbach’s α > 0.70.
Sp was divided into two variables: “social pressure from the government” (Sp_g; III_11) and “social
pressure from surrounding people” (Sp_s; III_7, 10, 17, 19; α = 0.704). The average value of items for
Sp_s are described below. For Pbc and Lc, only one item showed significant correlation with intentions;
therefore, the answers for those items (Pbc: III_13, Lc: III_5) were selected. The items for Att, Ig, Lz,
and Sat did not show significant correlation with intentions.
Figure 3 shows the average scores of the valid internal variables involved in the hypothetical
model. For Int_nd, the total average score was 5.23, and 95% of respondents answered between
“strongly agree” and “slightly agree” regarding avoidance of disposing of waste in public open spaces.
Of the four sites, HL showed the highest (5.35) and LL the lowest (5.13) scores. Significant difference
was observed between LL and HL on using the post hoc test (p = 0.00). For Int_kc, the total average
score was 5.52, slightly higher than that of Int_nd. HL showed a slightly higher score than the other
sites, but the differences among the four sites were not significant (p = 0.11). The average Sb score was
5.01; the difference among the four sites was quite small and not significant (p = 0.49). The average score
for Sp_s was 4.72, the lowest among the three types of social norms (Sb, Sp_s, and Sp_g). The lowest
score was seen in LL, and significant differences were observed with the other three areas: HL (p = 0.00),
LH (p = 0.00), and HH (p = 0.00). The total average score for Sp_g was 5.15, the highest among the three
social norms. The difference among the four sites was not significant (p = 0.45). Pn showed the highest
average score (5.30) among the norms, and the difference among the four regions was not significant
Sustainability 2019, 11, 1518 8 of 14

(p = 0.94). The average score for Pbc was 4.50. LH showed the highest score (4.68), and a significant
difference was observed between LL and LH (p = 0.00). The average score for Lc was 3.87, the lowest
among the variables shown in Figure 3. HL showed the lowest score (3.62), but the difference among
Sustainability
Sustainability
the 2019, 11
2019,
four regions 11 FORnot
FOR
was PEER
PEER REVIEW (p = 0.10).
REVIEW
significant 88

6.00
6.00

5.00
5.00

4.00
4.00

3.00
3.00

2.00
2.00

1.00
1.00
HL

HL

HL

HL

HL

HL

HL

HL
HH

HH

HH

HH

HH

HH

HH

HH
LH

LH

LH

LH

LH

LH

LH

LH
TT

TT

TT

TT

TT

TT

TT

TT
LL

LL

LL

LL

LL

LL

LL

LL
HL

HL

HL

HL

HL

HL

HL

HL
HH

HH

HH

HH

HH

HH

HH

HH
LH

LH

LH

LH

LH

LH

LH

LH
LL

LL

LL

LL

LL

LL

LL

LL
Int_nd
Int_nd Int_kc
Int_kc Sb
Sb Sp_s
Sp_s Sp_g
Sp_g Pn
Pn Pbc
Pbc Lc
Lc

Figure 3.
Figure
Figure 3. Average
3. Average scores of
Average scores
scores of variables
of variables in
variables in the
in the hypothetical
the hypothetical model
hypothetical model by
model by site.
by site. Error
site. Error bars
Error bars represent
bars represent the
represent the
the
standard error
standard error values.
errorvalues. The
values.The
The letter
letter
letter pairs
pairs
pairs represent
represent
represent each
eacheach site’s
site’ssite’s collection
collection
collection frequency
frequency
frequency and population
and population
and population density
density
density
(H: (H:
high,(H: high, L:
L: high,
low). L:represents
T low). TT represents
low). represents the
the totalthe total average
total
average average score.
score. score.

Figure 444 shows

Figure
Figure shows respondents’
shows respondents’ concerns
respondents’ concerns about
concerns about different
about different environmental
different environmental issues,
environmental issues, such
issues, such as
such as water
as water
water
pollution,
pollution, air
air pollution,
pollution, global
global warming,
warming, and
and waste
waste management.
management. Answers
Answers
pollution, air pollution, global warming, and waste management. Answers were given as a score were
were given
given as
as a
a score
score
from 11 (“not
from
from 1 (“not concerned
(“not concerned at
concerned at all”)
atall”) to
all”)to 44 (“highly
to 4(“highly concerned”).
(“highlyconcerned”). The
concerned”). The average
average scores
The scores of
average of the
scores the total
of total sample
the sample (n
total sample(n ==
413)
(n = for
413) water
for pollution,
water air
pollution, pollution,
air pollution,global warming,
global warming,and waste
and management
413) for water pollution, air pollution, global warming, and waste management were 3.33, 3.10, 2.84,
waste management were 3.33,
were 3.10,
3.33, 2.84,
3.10,
and 3.00,
and
2.84, 3.00, respectively.
respectively.
and 3.00, Respondents
Respondents
respectively. Respondents indicated
indicated thethe
the
indicated highest
highest
highestconcern
concern
concernabout
about water
aboutwater pollution
waterpollution
pollutionandand
and thethe
the
lowest about
lowest
lowest about global
about global warming.
global warming. The
warming. The residents
The residents of
residents of LH,
of LH, who
LH, who receive
who receive low
receive low collection
low collection frequency
collection frequency in
frequency in aaa more
in more
more
densely
densely populated
populated area,
area, indicated
indicated thethe highest
highest concern
concern for for
eacheach issue.
issue. Waste
densely populated area, indicated the highest concern for each issue. Waste management wasthird
Waste management
management was was
the the
the
third highest concern in all regions expect for HL, in which it was
third highest concern in all regions expect for HL, in which it was second highest.
highest concern in all regions expect for HL, in which it was second second highest.
highest.

4.00
4.00

3.00
3.00

2.00
2.00

1.00
1.00
HL
LL

HL
LL

HL
LL

HL
LL
HH

HH

HH

HH
LH

LH

LH

LH
TT

TT

TT

TT
HL
LL

HL
LL

HL
LL

HL
LL
HH

HH

HH

HH
LH

LH

LH

LH

Wp
Wp Ap
Ap Gw
Gw Wm
Wm

Figure 4. Average
Figure Average scores of of environmental concern
concern by site.
site. Error bars
bars represent thethe standard error
error
Figure 4.
4. Average scores
scores of environmental
environmental concern by by site. Error
Error bars represent
represent the standard
standard error
values. Wp,
values. Wp, Ap,
Ap, Gw,
Gw, and
and Wm
Wm represent
represent the
the four
four target
target issues:
issues: water
water pollution,
pollution, air
air pollution,
pollution, global
global
values. Wp, Ap, Gw, and Wm represent the four target issues: water pollution, air pollution, global
warming, and
warming,
warming, and waste
and waste management.
waste management. The
management. The letter
The letter pairs represent
letter pairs
pairs represent each
represent each site’s
each site’s collection
site’s collection frequency
collection frequency and
frequency and
and
population
population density
population density (H:
density (H: high,
(H: high, L:
high, L: low).
low).
L: low).

Figures55 and
Figures and66 show
showaverage
average scores
scoresbybygender
gender and and age,
age,respectively.
respectively. As Asseen
seenininFigure
Figure 5,
5, women
women
showed higher scores for all the variables compared with men. Statistically
showed higher scores for all the variables compared with men. Statistically significant differences significant differences
were observed
were observed for
for Int_nd
Int_nd (p
(p == 0.05),
0.05), Int_kc
Int_kc (p
(p == 0.00),
0.00), Sp_g
Sp_g (p (p == 0.04),
0.04), and
and PbcPbc (p(p == 0.00);
0.00); differences
differences
were not
were not significant
significant for
for Sb
Sb (p
(p == 0.33),
0.33), Sp_s
Sp_s (p
(p == 0.06),
0.06), Pn
Pn (p
(p == 0.24),
0.24), or
or Lc
Lc (p
(p == 0.06).
0.06).
Sustainability 2019, 11, 1518 9 of 14

Figures 5 and 6 show average scores by gender and age, respectively. As seen in Figure 5, women
showed higher scores for all the variables compared with men. Statistically significant differences
were observed for Int_nd (p = 0.05), Int_kc (p = 0.00), Sp_g (p = 0.04), and Pbc (p = 0.00); differences
Sustainability 2019, 11 FOR PEER REVIEW 9
Sustainability 2019, 11 FOR
were not significant PEER
for Sb REVIEW
(p = 0.33), Sp_s (p = 0.06), Pn (p = 0.24), or Lc (p = 0.06). 9

6.00
6.00

5.00
5.00

4.00
4.00

3.00
3.00

2.00
2.00

1.00
1.00
M F M F M F M F M F M F M F M F
M F M F M F M F M F M F M F M F
Int_nd Int_kc Sb Sp_s Sp_g Pn Pbc Lc
Int_nd Int_kc Sb Sp_s Sp_g Pn Pbc Lc

Figure 5.
5. Average
Average scores
scores of variables
variables in the
the hypothetical model
model by
by gender.
gender. Error
Error bars
bars represent
represent the
the
Figure
Figure 5. Average scores of
of variables in
in the hypothetical
hypothetical model by gender. Error bars represent the
standard error
standard error values.
standard error values.
values.

As shown
As shownininFigure
Figure6,6,respondents
respondents in in their
their 30s30s showed
showed relatively
relatively lower lower
scoresscores for Int_nd,
for Int_nd, Sb,
Sb, Sp_g,
As shown in Figure 6, respondents in their 30s showed relatively lower scores for Int_nd, Sb,
Sp_g,
and Pn.and Pn. Significant
Significant differences
differences were observed
were observed for Int_nd
for Int_nd (p = Pn
(p = 0.04), 0.04),
(p =Pn0.05),
(p = and
0.05),Pbc
and(pPbc (p =
= 0.03);
Sp_g, and Pn. Significant differences were observed for Int_nd (p = 0.04), Pn (p = 0.05), and Pbc (p =
0.03); differences
differences were were not significant
not significant for Int_kc
for Int_kc (p = (p = 0.72),
0.72), Sb=(p0.14),
Sb (p = 0.14),
Sp_sSp_s(p(p
= =0.63),
0.63),Sp_g
Sp_g(p (p == 0.09),
0.09),
0.03); differences were not significant for Int_kc (p = 0.72), Sb (p = 0.14), Sp_s (p = 0.63), Sp_g (p = 0.09),
or Lc
or Lc (p
(p =
= 0.86).
0.86). For
For Int_nd,
Int_nd, the
the youngest (<20) and
youngest (<20) and the
the oldest
oldest ((≥60)
≥60)groups
groupsshowed
showedthe thehighest
highest scores.
scores.
or Lc (p = 0.86). For Int_nd, the youngest (<20) and the oldest (≥60) groups showed the highest scores.
The post
The post hoc
hoc test
test found
found significant
significant differences
differences between
between <20 <20 and
and 30s
30s (p
(p =
= 0.01),
0.01), 20s
20s and
and 30s
30s (p
(p = = 0.02),
0.02),
The post hoc test found significant differences between <20 and 30s (p = 0.01), 20s and 30s (p = 0.02),
and <20 and 50s (p = 0.02). For Pbc, the highest and lowest scores were found in the youngest (<20)
and <20 and 50s (p = 0.02). For Pbc, the highest and lowest scores were found in the youngest (<20) and
and <20 and 50s (p = 0.02). For Pbc, the highest and lowest scores were found in the youngest (<20)
and oldest
oldest (≥60)(≥60) groups,
groups, respectively.
respectively. The The
postpost
hoc hoc
test test
found found significant
significant difference
difference between
between <20 <20
and and
30s
and oldest (≥60) groups, respectively. The post hoc test found significant difference between <20 and
30s= (p
(p = 0.01),
0.01), 40s
40s (p (p = 0.03),
= 0.03), 50s (p 50s (p = 0.04),
= 0.04), and ≥ and ≥60= (p
60 (p = 0.00)
0.00) age groups.
age groups. Significant
Significant difference
difference was
was also
30s (p = 0.01), 40s (p = 0.03), 50s (p = 0.04), and ≥60 (p = 0.00) age groups. Significant difference was
also found
found betweenbetween
20s and20s30sand(p30s (p =and
= 0.05) 0.05)≥60
and(p≥60 (p =age
= 0.03) 0.03) age groups.
groups. For Pn,For Pn, relatively
relatively higher
higher values
also found between 20s and 30s (p = 0.05) and ≥60 (p = 0.03) age groups. For Pn, relatively higher
valuesseen
were were seenyoungest
in the in the youngest
(<20) and(<20) and the
the oldest (≥60)oldest (≥60)The
groups. groups.
post hocThetestpost hoc test
showed showed
significant
values were seen in the youngest (<20) and the oldest (≥60) groups. The post hoc test showed
significant between
difference difference thebetween
youngest thegroup
youngest
and group and(20s,
all others all others (20s,30s,
p = 0.02; p =p0.02; 30s,40s,
= 0.00; p = p0.00; 40s,50s,
= 0.01; p=
significant difference between the youngest group and all others (20s, p = 0.02; 30s, p = 0.00; 40s, p =
p0.01; 50s,except
= 0.02) p = 0.02)
forexcept for the
the oldest oldest group.
group.
0.01; 50s, p = 0.02) except for the oldest group.

Figure 6.
Figure 6. Average scores of
Average scores of variables
variables in
in the
the hypothetical
hypothetical model
model by
by age.
age. Error
Error bars
bars represent
represent the
the
Figure 6. Average scores of variables in the hypothetical model by age. Error bars represent the
standard error values.
standard error values.
standard error values.

3.3. Model Estimation

3.3. Model Estimation
The hypothetical model was modified and estimated by path analysis. For internal variables, the
The hypothetical model was modified and estimated by path analysis. For internal variables, the
variables shown in the previous sections were used. For Exp, answers were given on a scale from 1
variables shown in the previous sections were used. For Exp, answers were given on a scale from 1
(“never”) to 5 (“always”). For external variables such as Pd, Cf, and Fp, the objective data prepared
(“never”) to 5 (“always”). For external variables such as Pd, Cf, and Fp, the objective data prepared
in Section 2.1 were used. For Fp, the indicator “total facility volume (L)/area (km2)” was used. Total
in Section 2.1 were used. For Fp, the indicator “total facility volume (L)/area (km2)” was used. Total
 Sustainability 2019, 11 FOR PEER REVIEW 10

was excluded
Sustainability from
2019, 11, 1518 the model, and perceived external conditions were newly added. Dis_per10
referred
of 14
to the perceived distance of the disposal site (III_12) and Fp_per to the perceived sufficiency of facility
provision (II_14).
3.3. Model Estimation
Figure 7 depicts the estimated model to explain intention not to dispose of waste in public. Only
theThesignificant (p < 0.01)
hypothetical modelpathswasare shown. The
modified and model
estimatedfit indices
by path were acceptable;
analysis. the comparative
For internal variables,fit
theindex (CFI) shown
variables was 0.909 (>0.90),
in the and the
previous root mean
sections weresquare
used. error
For Exp,of approximation
answers were(RMSEA) given on was 0.071
a scale
from(<0.080). The explained
1 (“never”) variance (R2)
to 5 (“always”). of Int_ndvariables
For external was 0.65,suchas seenas to
Pd,the
Cf,upper-right
and Fp, the ofobjective
Int_nd in data
Figure
7, whichinwas
prepared high2.1
Section enough
were for model
used. For acceptance.
Fp, the indicator “total facility volume (L)/area (km2 )” was
used. Total Int_kc had volume
facility a substantial positive effect
was calculated (0.37)the
by totaling on volume
Int_nd. of Among the constituents
each waste container in of theTPB, two
target
types
street. ofwas
Sat social norms from
excluded (Sb and the Sp_g)
model,showed higher external
and perceived direct influences
conditions (0.22
wereand 0.20,added.
newly respectively)
Dis_peron
Int_nd,towhereas
referred Att had
the perceived no significant
distance influence.
of the disposal siteIn addition,
(III_12) andpersonal
Fp_per tonorms (Pn) andsufficiency
the perceived Pbc showed
of significant direct influences on Int_nd (Pn: 0.12; Pbc: 0.11). The internalization effects of social norms
facility provision (II_14).
(SbFigure
and Sp) on personal
7 depicts norms (Pn)
the estimated wereto
model observed
explain as significantly
intention not tolarge pathof
dispose coefficients between
waste in public.
Onlythesethe variables
significant(Sb→→Pn:
(p < 0.01) paths0.36;are
Sp_s→Pn:
shown. The 0.14; Sp_g→Pn:
model fit indices0.23).
wereLz, Ig, and the
acceptable; Lc comparative
did not show
fit significant
index (CFI)influences
was 0.909on Int_ndand
(>0.90), or the
Int_kc,
rootand
mean were
squareexcluded
error of from the model. Other
approximation (RMSEA) influential
was
variables
0.071 (<0.080). on The
Int_nd included
explained Exp (−0.10).
variance (R2) of Int_nd was 0.65, as seen to the upper-right of Int_nd in
Figure 7, People’s
which was perception
high enoughof theforexisting waste service provision also had significant influences
model acceptance.
(Dis_per→Exp: 0.13; Fp_per→Con:
Int_kc had a substantial 0.07).(0.37)
positive effect External factor Cf
on Int_nd. showed
Among thea constituents
negative influence
of TPB,on twoperceived
types
of distance
social norms of the(Sbdisposal
and Sp_g) site showed
(Cf→Dis_per:
higher−0.13).
direct influences (0.22 and 0.20, respectively) on Int_nd,
whereasTable 3 shows
Att had the total effect
no significant of each
influence. variable on
In addition, Int_ndnorms
personal and Int_kc, involving
(Pn) and indirect
Pbc showed and direct
significant
effects.
direct With respect
influences to total
on Int_nd (Pn: effects, social
0.12; Pbc: norms
0.11). showed high influences
The internalization at thenorms
effects of social same level of direct
(Sb and Sp)
oninfluence
personal normsof Int_kc(Pn)(0.37)
were on Int_ndas(Sb:
observed 0.38; Sp_g:
significantly 0.36).
large pathPncoefficients
and Pbc showed betweensignificantly
these variableslarge
(Sbinfluences
→→Pn: 0.36; on Int_nd
Sp_s→Pn: (Pn:0.14;
0.20;Sp_g
Pbc:→0.18). Compared
Pn: 0.23). Lz, Ig, with
and Lc the
dideffects of the
not show internal variable,
significant influencesthe
oninfluences on Int_nd
Int_nd or Int_kc, and and
wereInt_kc of external
excluded from thevariables such as
model. Other Cf and Fp,
influential as wellonasInt_nd
variables the influences
included of
Exp their perception, such as Dis_per and Fp_per, were small.
( − 0.10).

Figure
Figure 7. Model
7. Model estimation
estimation results.
results. Model
Model fit fit indices
indices were
were CFICFI = 0.909
= 0.909 andand RMSEA
RMSEA = 0.071.
= 0.071. TheThe lines
lines
represent
represent significant
significant (<0.01)
(<0.01) paths.The
paths. Thevalue
valueshown
shownbeside
beside each
each path
path is is
thethe standardized
standardized path
path
coefficient
coefficient value.
value.
Sustainability 2019, 11, 1518 11 of 14

People’s perception of the existing waste service provision also had significant influences
(Dis_per→Exp: 0.13; Fp_per→Con: 0.07). External factor Cf showed a negative influence on perceived
distance of the disposal site (Cf→Dis_per: −0.13).
Table 3 shows the total effect of each variable on Int_nd and Int_kc, involving indirect and
direct effects. With respect to total effects, social norms showed high influences at the same level of
direct influence of Int_kc (0.37) on Int_nd (Sb: 0.38; Sp_g: 0.36). Pn and Pbc showed significantly
large influences on Int_nd (Pn: 0.20; Pbc: 0.18). Compared with the effects of the internal variable,
the influences on Int_nd and Int_kc of external variables such as Cf and Fp, as well as the influences of
their perception, such as Dis_per and Fp_per, were small.

Table 3. Direct and total effects on intentions.

Dependent Variables
Independent Variables Int_nd Int_kc
Direct Indirect Total Direct Indirect Total
(b) Int_kc 0.37 - 0.37
(d) Sb 0.22 0.16 0.38 0.15 0.08 0.23
(e) Pbc 0.11 0.07 0.18 0.20 - 0.20
(f-1) Sp_s - 0.03 0.03 - 0.03 0.03
(f-2) Sp_g 0.20 0.16 0.36 0.26 0.05 0.31
(g) Pn 0.12 0.08 0.20 0.23 - 0.23
(h) Exp −0.10 - −0.10 - - -
(k) Con 0.04 −0.02 0.02 - - -
(l-1) Dis_per −0.01 −0.03 −0.04 - −0.07 −0.07
(l-2) Fp_per - 0.01 0.01 - 0.01 0.01
(E1) Fp - −0.01 −0.01 - - -
(E2) Cf - 0.01 0.01 0. 0.01 0.01
(E3) Pd - −0.01 −0.01 - −0.01 −0.01

4. Discussion
Our findings suggest that TPB is a good framework for predicting intention not to dispose of
waste in public. Social norms, such as Sb and Sp_g, and Pbc were found to have significant effects on
behavioral intention, which is supported by the TPB framework. In contrast, Att did not demonstrate
any significant influences on intention. Attitude here refers to the degree to which people value
avoiding public waste disposal positively or negatively. This indicates that such perceptions do not
connect with actual behavioral intentions. More normative right-or-wrong perceptions, like personal
norms, had a significant effect on intention. Previous studies (e.g., Bortoleto et al., 2012 and Barr et al.,
2001) demonstrated that environmental attitude did not always have a direct effect on environmental
behavior, but also had indirect effects on behavioral intention [29,34]. Bortoleto et al. (2012) found
no significant direct influence of attitude on intention to prevent waste, but did find a relationship
between attitude and PBC. This leads to the conclusion that attitude has a potential indirect effect on
intention [29].
The substantial positive influences of norms coincided with the findings of many other studies,
which showed the important role of personal and subjective norms in environmental behaviors [42–45].
Chan and Bishop (2013) found that moral and subjective norms had influential effects on the intent
to recycle [26]. Matthies et al. (2012) concluded that subjective norms not only had a direct effect on
recycling and re-use behavior, but also had an additional indirect effect through personal norms [42].
They also noted that this would be reasonable because the past behaviors of parents and other close
friends or family could impact the development of personal norms. The results of our model also
found significant indirect effects of Sb through Pn. Bortoleto et al. (2012) also indicated that subjective
norms interacted with personal norms to create a large influential effect on prevention behavior, rather
than having a direct effect [29].
Sustainability 2019, 11, 1518 12 of 14

In addition to subjective norms, perceived social pressure from the government (Sp_g) showed
significant direct and indirect influences on intentions. This social norm can be categorized as an
injunctive norm. The influence of injunctive norms has been well investigated and has been found
to have substantial influence on recycling behavior [42]. Thøgersen (2006) found that descriptive,
injunctive, subjective, and personal norms have a statistically significant positive correlation with
environmentally responsible behavior [33]. The significant influence of this type of norm indicates that
governmental regulations can enhance people’s intentions not to dispose of waste in public.
Pbc was also found to have a direct effect on Int_nd. Pbc here referred to people’s perceptions
of the ease or difficulty of avoiding public waste disposal. This result was consistent with the TPB
framework and other studies about environmental behaviors, especially waste recycling and waste
prevention (e.g., [16,29,46,47]). However, although significant influence from PBC was demonstrated
by many previous studies, the size of the influence has been controversial. Some previous studies
showed significantly large influence of PBC on intention—for example, Chan and Bishop (2013) found
the influence of PBC on recycling behavior to be 0.27 [26]—whereas numerous others showed a
relatively smaller relationship between PBC and intention.
Influence of past experience on Int_nd was also found. This indicates that people who do not have
any experience disposing of waste in public open spaces will highly intend to continue to perform this
good behavior. Past experience was mainly determined by awareness of waste management, as seen
in the higher path coefficient from Con to Exp (−0.45). This indicates that people with lower concern
about waste management have greater experience disposing of waste in public open spaces, and this
connects with future disposal intentions. Therefore, increase of awareness about waste management is
considered important to avoid waste disposal behaviors in public open spaces.

5. Conclusions
Focusing on external and internal factors determining people’s intention not to dispose of waste
in public open spaces, we conducted a face-to-face questionnaire survey in Phnom Penh, Cambodia.
A total of 413 valid samples were gained from the four targeted sites, which differed in collection
frequency and population density. A model based on a modification of TPB was developed and
estimated using the obtained data. The model estimation results indicated that the influence of internal
factors was much larger than that of external factors in predicting the target behavioral intention.
Among the internal factors, influences from social and personal norms were largest. Social pressure
from friends and family and from the government showed significantly large influences on the target
intention. Additionally, the personal norms internalized from these social norms showed significant
influence on the target intention. For the policy makers, top-down interventions, such as regulations
and laws, have been usually considered effective. The higher path strength from the social pressure
from the government can fit with this kind of intervention. On the other hand, higher path strengths
from personal norms and subjective norms indicate that bottom-up interventions stimulating residents’
norms can also have the potential to reduce waste disposal in public open spaces. Comparison of the
effectiveness of different interventions can give more insights into reducing waste disposal in public
open spaces as future research.

Author Contributions: All authors contributed equally to this paper.

Funding: This research was partially supported by the Environment Research and Technology Development
Fund (S-16) of the Ministry of Environment, Japan.
Conflicts of Interest: The authors declare no conflict of interest.
Sustainability 2019, 11, 1518 13 of 14

References
1. Kum, V.; Sharp, A.; Harnpornchai, N. Improving the solid waste management in Phnom Penh city: A strategic
approach. Waste Manag. 2005, 25, 101–109. [CrossRef] [PubMed]
2. Seng, B.; Kaneko, H.; Hirayama, K.; Katayama-Hirayama, K. Municipal solid waste management in Phnom
Penh, capital city of Cambodia. Waste Manag. 2010, 29, 491–500. [CrossRef] [PubMed]
3. Ojedokun, O. Attitude towards littering as a mediator of the relationship between personality attributes and
responsible environmental behavior. Waste Manag. 2011, 31, 2601–2611. [CrossRef]
4. Thondhlana, G.; Hlatshwayo, T.N. Pro-environmental behavior in student residences at Rhodes University,
South Africa. Sustainability 2018, 10, 2746. [CrossRef]
5. Nguyen, H.T.T.; Hung, R.J.; Lee, C.H.; Nguyen, H.T.T. Determinants of residents’ E-waste recycling
behavioral intention: A case study from Vietnam. Sustainability 2019, 11, 164. [CrossRef]
6. Xu, L.; Ling, M.; Lu, Y.; Shen, M. External influences on forming residents’ waste separation behaviour:
Evidence from households in Hangzhou, China. Habitat Int. 2017, 63, 21–33. [CrossRef]
7. Conke, L.S. Barriers to waste recycling development: Evidence from Brazil. Resour. Conserv. Recycl. 2018,
134, 129–135. [CrossRef]
8. Oskamp, S.; Harrington, M.J.; Edwards, T.C.; Sherwood, D.L.; Okuda, S.M.; Swanson, D.C. Factors
influencing household recycling behavior. Environ. Behav. 1991, 23, 494–519. [CrossRef]
9. Hopper, J.R.; Nielsen, J.M. Recycling as altruistic behavior: Normative and behavioral strategies to expand
participation in a community recycling program. Environ. Behav. 1991, 23, 195–220. [CrossRef]
10. Ewing, G. Altruistic, egoistic, and normative effects on curbside recycling. Environ. Behav. 2001, 33, 733–764.
[CrossRef]
11. Davies, J.; Foxall, G.R.; Pallister, J. Beyond the intention–behaviour mythology: An integrated model of
recycling. Mark. Theory 2002, 2, 29–113. [CrossRef]
12. Tabernero, C.; Hernández, B.; Cuadrado, E.; Luque, B.; Pereira, C.R. A multilevel perspective to explain
recycling behaviour in communities. J. Environ. Manag. 2015, 159, 192–201. [CrossRef] [PubMed]
13. Botetzagias, I.; Dima, A.-F.; Malesios, C. Extending the theory of planned behavior in the context of recycling:
The role of moral norms and of demographic predictors. Resour. Conserv. Recycl. 2015, 95, 58–67. [CrossRef]
14. Strydom, W. Applying the theory of planned behavior to recycling behavior in South Africa. Recycling 3.3
2018, 3, 43. [CrossRef]
15. Hornik, J.; Cherian, J.; Madansky, M.; Narayana, C. Determinants of recycling behavior: A synthesis of
research results. J. Socio Econ. 1995, 24, 105–127. [CrossRef]
16. Barr, S. Strategies for sustainability: Citizens and responsible environmental behaviour. Area 2003, 35,
227–240. [CrossRef]
17. Stoeva, K.; Alriksson, S. Influence of recycling programmes on waste separation behaviour. Waste Manag.
2017, 68, 732–741. [CrossRef] [PubMed]
18. Cialdini, R.B.; Reno, R.R.; Kallgren, C.A. A focus theory of normative conduct: Recycling the concept of
norms to reduce littering in public places. J. Pers. Soc. Psychol. 1990, 58, 1015–1026. [CrossRef]
19. Schultz, P.W.; Bator, R.J.; Large, L.B.; Bruni, C.M.; Tabanico, J.J. Littering in context: Personal and
environmental predictors of littering behavior. Environ. Behav. 2013, 45, 35–59. [CrossRef]
20. Al-Khatib, I.A.; Arafat, H.A.; Daoud, R.; Shwahneh, H. Enhanced solid waste management by understanding
the effects of gender, income, marital status, and religious convictions on attitudes and practices related to
street littering in Nablus—Palestinian territory. Waste Manag. 2009, 29, 449–455. [CrossRef]
21. Cintri. Waste Collection Reports for Khan Dongkor, Khan Prampir Meakkakra, Khan Chbar Amporv,
Khan Chhroy Chongva, Khan Chomkar Mom, Khan Duon Penh, Khan Mean Chey, Khan Pousenchey,
Khan Prek Pnov, Khan Reussey Keo District, Khan Sensok, and Khan Tuol Kork. 2016. Available online:
http://www.cintri.com.kh/view.aspx?LID=KH&mID=H11 (accessed on 10 March 2019). (In Khmer)
22. National Institute of Statistics of Cambodia (NIS). General Population Census of Cambodia 2008; National
Institute of Statistics, Ministry of Planning: Phnom Penh, Cambodia, 2009.
23. Ajzen, I.; Fishbein, M. Attitude-behavior relations: A theoretical analysis and review of empirical research.
Psychol. Bull. 1975, 84, 888–918. [CrossRef]
24. Park, H.S.; Levine, T.R.; Sharkey, W.F. The theory of reasoned action and self-construals: Understanding
recycling in Hawaii. Commun. Stud. 2009, 49, 196–208. [CrossRef]
Sustainability 2019, 11, 1518 14 of 14

25. Ajzen, I. The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 1991, 50, 179–211. [CrossRef]
26. Chan, L.; Bishop, B. A moral basis for recycling: Extending the theory of planned behaviour. J. Environ.
Psychol. 2013, 36, 96–102. [CrossRef]
27. Arı, E.; Yılmaz, V. A proposed structural model for housewives’ recycling behavior: A case study from
Turkey. Ecol. Econ. 2016, 129, 132–142. [CrossRef]
28. Wang, Z.; Guo, D.; Wang, X. Determinants of residents’ e-waste recycling behaviour intentions: Evidence
from China. J. Clean. Prod. 2016, 137, 850–860. [CrossRef]
29. Bortoleto, A.P.; Kurisu, K.H.; Hanaki, K. Model development for household waste prevention behaviour.
Waste Manag. 2012, 32, 2195–2207. [CrossRef] [PubMed]
30. Tucker, P.; Douglas, P. WR0112: Understanding Household Waste Prevention Behaviour; Technical Report 2007
No. 3: A Conceptual and Operational Model of Household Waste Prevention Behaviour; Defra: London,
UK, 2018. Available online: http://randd.defra.org.uk (accessed on 21 August 2018).
31. Schwartz, S.H. Normative explanations for helping behavior: A critique, proposal, and empirical test. J. Exp.
Soc. Psychol. 1973, 9, 349–364. [CrossRef]
32. Schwartz, S.H. Normative influences on altruism. In Advances in Experimental Social Psychology; Berkowitz, L.,
Ed.; Academic Press: New York, NY, USA, 1977; pp. 221–279.
33. Thøgersen, J. Norms for environmentally responsible behaviour: An extended taxonomy. J. Environ. Psychol.
2006, 26, 247–261. [CrossRef]
34. Barr, S.; Gilg, A.W.; Ford, N.J. A conceptual framework for understanding and analysing attitudes towards
household-waste management. Environ. Plan. A 2001, 33, 2025–2048. [CrossRef]
35. Chiang, Y.T.; Fang, W.T.; Kaplan, U.; Ng, E. Locus of control: The mediation effect between emotional
stability and pro-environmental behavior. Sustainability 2019, 11, 820. [CrossRef]
36. Ojedokun, O.; Balogun, S.K. Psycho-sociocultural analysis of attitude towards littering in a Nigerian urban
city. Ethiopian J. Environ. Stud. Manag. 2011, 4, 68–80. [CrossRef]
37. Tonglet, M.; Phillips, P.S.; Read, A.D. Using the theory of planned behaviour to investigate the determinants of
recycling behaviour: A case study from Brixworth, UK. Resour. Conserv. Recycl. 2004, 41, 191–214. [CrossRef]
38. McCarty, J.A.; Shrum, L.J. The recycling of solid wastes: Personal values, value orientations, and attitudes
about recycling as antecedents of recycling behavior. J. Bus. Res. 1994, 30, 53–62. [CrossRef]
39. Loehlin, J.C. Latent Variables Models: An Introduction to Factor, Path, and Structural Analysis; Lawrence Erlbaum
Associates: Mahwah, NJ, USA, 1998.
40. Yamane, T. Statistics, an Introductory Analysis, 2nd ed.; Harper and Row: New York, NY, USA, 1967.
41. National Institute of Statistics of Cambodia (NIS). National Socio-Economic Census; National Institute of
Statistics, Ministry of Planning: Phnom Penh, Cambodia, 2015.
42. Matthies, E.; Selge, S.; Klöckner, C.A. The role of parental behaviour for the development of behaviour
specific environmental norms—The example of recycling and re-use behaviour. J. Environ. Psychol. 2012, 32,
277–284. [CrossRef]
43. Bratt, C. The impact of norms and assumed consequences on recycling behavior. Environ. Behav. 1999, 31,
630–656. [CrossRef]
44. Harland, P.; Staats, H.; Wilke, H.A.M. Explaining proenvironmental intention and behavior by personal
norms and the theory of planned behavior. J. Appl. Soc. Psychol. 1999, 29, 2505–2528. [CrossRef]
45. Thøgersen, J. A model of recycling behaviour, with evidence from Danish source separation programs.
Int. J. Res. Mark. 1994, 11, 145–163. [CrossRef]
46. White, K.M.; Smith, J.R.; Terry, D.J.; Greenslade, J.H.; McKimmie, B.M. Social influence in the theory of
planned behaviour: The role of descriptive, injunctive, and in-group norms. Br. J. Soc. Psychol. 2009, 48,
135–158. [CrossRef] [PubMed]
47. Ioannou, T.; Zampetakis, L.A.; Lasaridi, K. Psychological determinants of household recycling intention in
the context of the theory of planned behaviour. Fresen. Environ. Bull. 2013, 22, 2035–2041.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).