-
Int. J. Environ. Res., 6(1):323-334, Winter 2012ISSN:
1735-6865
Received 7 March 2011; Revised 1 July 2011; Accepted 10 July
2011
*Corresponding author E-mail: [email protected]
323
Structural Equation Model for Environmentally Conscious
Purchasing Behavior
Arslan, T., Yilmaz, V. and Aksoy, H. K.*
Department of Statistics, Eskişehir Osmangazi University,
Eskişehir, 26480 Turkey
ABSTRACT: In this study, the effect of environmental concern,
attitudes and behaviors of the universitystudents on the
environmentally conscious purchasing behavior was investigated with
the help of StructuralEquation Model (SEM). SEM is an effective
data analysis tool that expresses the complicated
causativerelations between the latent variables. In this study, we
collected the data from 400 university students bymeans of a
survey. The results of this research showed that environmental
attitudes, environmental behaviorand green product awareness have a
positive effect on environmentally conscious purchasing behavior.
Thedeveloped model showed that, environmental concern explains 48%
of the variation in the environmentalattitudes and 28% of the
product recovery awareness; environmental attitudes and product
recovery awarenessexplain 55% of green product awareness;
environmental attitudes explains 44% of the variation in
theenvironmental behavior; environmental attitudes, green product
awareness and environmental behavior explain74% of the variation in
the environmentally conscious purchasing behavior.
Key words: Structural Equation Model, Latent variable,
Environmental consciousness, Green product, Product recovery,
Consumers’ behavior
INTRODUCTIONPublic consciousness on environmental issues and
recent more strict environmental legislations have
beencompelling manufacturers from various industries andconsumers
to produce and dispose of used productsin an environmentally
conscious manner. The rapidexhaustion of raw materials and growing
amount ofsolid waste also fuel this trend. Gungor and Gupta
(1999)reviewed the literature and emphasized the two
majorobjectives of the environmentally consciousmanufacturing and
product recovery. These objectivesare environmentally friendly
product generation anddevelopment of product recovery and
wastemanagement technologies. The objective of productrecovery
management, as stated by Thierry et al. (1995),is ‘to recover as
much of the economic (and ecological)value as reasonably as
possible, thereby reducing theultimate quantities of waste’. In a
subsequent paper,Ilgin and Gupta (2010) expand the
environmentallyconscious manufacturing and product
recoveryliterature and examined it under four major
categories;environmentally conscious product design, reverse
andclosed-loop supply chains, remanufacturing, anddisassembly.
Environmental problems and theaccelerating changes in living
conditions have becomea fundamental part of the world in general
andmetropolises in particular. Earlier, environmental
problems have been considered as technical andeconomic problems;
while in the recent decades thesocial dimensions of environmental
problems such aspublic attention and people’s attitudes
towardsenvironment have became one of the areas ofenvironmental
sociology and environmentalpsychology (Kalantari and Asadi,
2010).
Environmentally conscious manufacturing andproduct recovery
efforts are directly related with theconsumer’s awareness and
involvement into thisprocess. Taking different factors into
consideration,the researchers develop many environmental
attitudeand environmental behavior models. Hini et al.
(1995)examine the relationship between environmentalattitudes and
behaviours. Environmental behaviorincludes actions which contr
ibute towardsenvironmental preservation and/or conservation
(forinstance, energy conservation, water conservation,consumerism,
etc). On the other hand the object ofone’s environmental attitude
is either the naturalenvironment itself or conservation behavior
(Axelrodand Lehman (1993), Kaiser et al. (1999) and (2007)).
The literature in the area of environmental attitudeand behavior
involves various models which arearising from the perspective of
the researchers towardsthe issue. Some of the researchers used
environmental
-
324
Arslan, T. et al.
attitude as the estimator of the environmental behavior(Chan,
1999; Kaiser et al., 1999; Fraj and Martinez, 2007;Steg and Vlek,
2009). Alternatively, both environmentalattitude and environmental
behavior are alsoconsidered simultaneously as the estimator of
theenvironmentally conscious purchasing behavior(ECPB) (Mostafa,
2007; Tilikidou and Delistavrou,2008).
The roots of environmental problems lie in humanbehavior, so the
solution could lie in changing thebehavior of organizations and
groups and so incultural and lifestyle changes, i.e.,
environmentalawareness is needed from the point of view of
bothsupply and demand (Mondéjar-Jiménez et al., 2011).The main
focus of this research is individuals’ ECPB,we considered
environmental behavior (EB) as anexogenous latent variable of the
ECPB. The majorreasons that we have considered the ECPB in the
modelas follows;(1) ECPB together with environmentalist behavior
canbe seen as significant actions measuring naturefriendliness and
protection.(2) Recent consumer’s trend towards green productsmay
direct the enterprises to change their entire wayof production
process in an environmentally friendlymanner.3) Companies which
fail to put into serviceenvironmentally responsible operation
procedures willencounter a loss of competitiveness or loose
marketshare. Hereby consumers accompanied by
producers/manufacturers will have more effective roles inprotecting
the environment.
In this study, we developed an analytical model toexamine the
effect of environmental concern, attitudeand behavior of the
consumers on theirenvironmentally conscious purchasing
behavior(ECPB). We employed the Structural Equation Model(SEM) to
investigate the complex correspondence ofabovementioned factors.
SEM describes thecomplicated causative relations between the
latentvariables. For this purpose, firstly, environmentalconcern
and environmental attitude of the students asthe estimator of the
environmental behavior were used.Secondly, environmental concern,
environmentalattitude and product recovery awareness were usedas
the estimator of the green product awareness.Finally, consumer ’s
environmental attitude,environmental behavior and green product
awarenesswere used as the estimator of the environmentallyconscious
purchasing behavior of the universitystudents.
In the literature, the relationship amongenvironmental concern,
environmental attitude,
environmental behavior and environmentallyconscious purchasing
behavior were analyzedseparately. On the other hand, the most
significantdistinction of this work is SEM comprehensivelycomprises
all of the factors in the model, viz.environmental concern,
environmental attitude,product recovery awareness, green product
awareness,environmental behavior and environmentallyconscious
purchasing behavior of individuals’simultaneously.
The paper is organized as follows. Section 2presents mainly
relevant literature in the area. Section3 briefly presents
Structural Equation Model (SEM)and its’ analysis. Section 4
presents the results of theanalysis. Finally, Section 5 provides
conclusion andSection 6 discusses the findings of this study.
Kaiser and Shimoda (1999) indicated thatenvironmental concern
could be used as the estimatorsof the environmental behaviors.
Kaiser et al. (1999)indicated that environmental attitudes could be
usedas the estimator of the environmental behavior.Nordlund and
Garvill (2002) determined that personalnorm mediated the effects
from general values,environmental values and problem awareness
onproenvironmental behavior.
Mannetti et al. (2004), searched the attitudes ofthe Italians
towards recycling efforts by means of theSEM. Authors found that,
personal judgments have asignificant effect on explaining the
product recoveryawareness. Bolaane (2006) determined that
educationalstatus considerably affects the contribution to
productrecovery efforts. Sidique et al. (2010) pointed out
thatdemographic factors like age, educational status,income and
number of the individuals in a householdhave an influence on the
recycling efforts. Tilikidouand Delistavrou (2006) found that there
is a negativecorrelation between the environmental behavior
andenvironmental insensitivities among Greek consumers.Also,
authors emphasized that females with highereducation contribute the
environmental activities more.Mostafa (2007) determined that
environmental anxietiesare effective on the attitudes of the people
on theconsumption of green products and he found that thereis a
weak relationship between the consumers’ greenproduct awareness and
their actual green productconsumption behavior. Fraj and Martinez
(2007)depicted that environmental attitudes are the estimatorsof
the environmental behavior. Tilikidou (2007) foundthat while
environmentally concerned consumption hasa positive correlation
with environmental awareness;it has a negative correlation with
environmentalunconcern. Tilikidou and Delistavrou (2008)
determinedthat the consumers, who behave in accordance with
-
Int. J. Environ. Res., 6(1):323-334, Winter 2012
325
environment and who are interested in productrecovery and
participate in environment-priorityactivities, are high-educated
people. Birgelen et al.(2009) found that env-friendly purchase and
disposaldecisions for beverages are related to theenvironmental
awareness of consumer and their env-friendly attitude.
Dono et al. (2009) found that there is a significantrelationship
between environmental attitudes andenvironmental behaviors. Steg
and Vlek (2009)indicated that when the environmental behaviors
ofindividuals are to be explained, environmental attitudesrelated
to them shall primarily be examined and in casethose environmental
attitudes are to be changed, thiswould be reflected on the
behavior, as well.
MATERIALS & METHODSThe following notation is used through
the paper;
AGFI Adjusted Goodness of Fit Index
CFI Comparative Fit Index EA Environmental Attitude
EB Environmental Behavior EC Environmental Concern ECPB
Environmentally Conscious Purchasing
Behavior GFI Goodness of F it Index GPA Green Product Awareness
NFI Normed Fit Index NNFI Non-Normed Fit Index PRA Product Recovery
Awareness RMSEA Root Mean Square Error of
Approximation RMR Root Mean Square Error SECPBM Students’
Environmentally Conscious
Purchasing Behavior Model SEM Structural Equation Model
Structural Equation ModelingSEM is a comprehensive statistical
method used
in testing hypotheses about causal relationshipsamong observed
and unobserved (latent) variables hasproved to be useful in solving
problems and informulating theoretical constructions (Reisinger
andTurner, 1999). SEM also can expand the explanatoryability and
statistical efficiency for model testing witha single comprehensive
method (Pang, 1996).Steenkamp and Baumgartner (2000) reflect on the
roleof SEM in marketing modeling and managerial decisionmaking, and
discuss some of its benefits. Authorsunderlined that although SEM
has potential fordecision support modeling, it is probably most
usefulfor theory testing, which is a key phase in developingmodels.
Applied to data on attitudes, perceptions,
stated behavioral intentions, and actual behavior, SEMcan be
used to specify and test alternative causalhypotheses (for SEM and
LISREL see Byrne, 1998;Cheng, 2001; Cudeck et al. 2000; Hayduk,
1987;Jöreskog and Sörbom, 2001).
MeasurementThe measurement tool used in this study was
adapted from Kaiser and Wilson (2000), Fraj andMartinez (2007),
Tilikidou and Delistavrou (2008). Inthe measurement tool, the items
of attitude andbehavior were measured by 5 point likert scale.
Theattitudinal questions were labeled either stronglydisagree to
strongly agree. The behavioral questionswere labeled never to
always. There are 26 items in themeasurement tool that include
various attitudes andbehaviors (fifteen of them are measures
attitude andeleven of them are measures behavior). The values
ofCronbach Alpha (α), which is the reliability criteriarelated to
the factors in the model, are given in Table 1.The Cronbach Alpha
(α) value between 0.50 and 0.60means a “close to reliability” and
that they are between0.60 and 0.80 means that it is “reliable”.
Table 1 showsthat, five of the calculated Cronbach Alpha (α)
valuesare between 0.60 and 0.80; one of them is between 0.50and
0.60.
Data collectionIn this research, since the general proportion
of
the attitude and behavior expressions within the frameof
research was not known, we couldn’t prepared theresearch frame by
means of the contingent samplingtechnique. However, to determine
the sample size weutilized the acceptable error level method under
thenormality assumption for the sample statistic. Thesample size
was calculated as 384; on 0.05 relevancelevel, z=1.96 d
(sensitivity) =0.05 or p and q values,being 0.5 (for the details of
sample size calculationplease see; Kish, 1965).
The survey was applied by talking face to face to400 students,
who were chosen randomly within theuniversity campus. Fourty-seven
of the appliedsurveys were soon realized that erroneous
andinconsistent and they were not included within theanalyses.
Theoretical FrameworkThe theoretical premise of this study is
based on
the theory of planned behavior. The theory of plannedbehavior
was formulated by Ajzen (1985) after thedevelopment of the theory
of reasoned action (Ajzenand Fishbein, 1980). According to the
theory of plannedbehavior, human behavior is under the influence
ofcertain factors with certain underlying causes, andexhibits
itself in a planned manner. Initially, an
-
326
Structural Model for Environmentally Conscious Behavior
Tabl
e1. I
tem
s with
in th
e Fac
tors
and
thei
r Mea
nsFa
ctor
s / C
ronb
ach
Alp
ha(α
) / It
ems
M
ale
Fem
ale
Envi
ronm
enta
l Con
cern
(EC
)/ C
ronb
ach
Alp
ha(α
)= 0
.75
a1
. Th
e id
ea th
at h
uman
, as
the
mas
ter
of t
he e
arth
, ha
s th
e rig
ht t
o di
strib
ute
the
natu
ral s
ourc
es i
n w
hate
ver
way
she
/he
wis
hes
to,
frustr
ates
me.
4.
21
4.06
4.
31
a2. T
he fa
ct th
at fa
ctor
ies p
erfo
rm p
rodu
ctio
n w
ithou
t che
ckin
g w
heth
er it
is h
arm
ful t
o na
ture
or n
ot, s
care
me.
4.
37
4.03
4.
58
a3. T
he id
ea o
f not
leav
ing
a cl
ean
wor
ld to
nex
t gen
erat
ions
wor
ries m
e.
4.35
4.
05
4.53
a4
. The
fact
that
peo
ple
hunt
ani
mal
s fo
r the
ir fu
rs a
ggra
vate
s m
e.
4.02
3.
76
4.29
a5
a . I d
on’t
thin
k th
at th
e pr
oble
m o
f env
ironm
enta
l pol
lutio
n is
exa
gger
ated
. 3.
99
3.79
4.
12
Envi
ronm
enta
l Atti
tude
(EA)
/ C
ronb
ach
Alp
ha(α
)= 0
.53
b1
. If
it pr
ovid
ed a
n ad
vant
age
for
the
envi
ronm
ent,
the
prod
ucts
whi
ch a
re p
rodu
ced,
pro
cess
ed a
nd p
acka
ged
in a
frie
ndly
way
to
envi
ronm
ent,
coul
d be
pai
d m
ore.
3.
53
3.26
3.
70
b2. I
f the
re is
to b
e a
choi
ce b
etw
een
two
prod
ucts,
the
prod
uct t
hat c
ause
s th
e le
ast h
arm
to p
eopl
e an
d en
viro
nmen
t sha
ll be
pur
chas
ed.
3.98
3.
80
4.09
b3. P
rodu
cts,
who
se b
asin
s an
d bo
xes c
ould
be
used
for o
ther
pur
pose
s sh
all b
e pr
efer
red.
. 3.
73
3.56
3.
83
b4. R
eact
iona
l act
iviti
es, p
erfo
rmed
in b
ehal
f of p
rote
ctin
g th
e env
ironm
ent,
prov
ide
bene
fit fo
r the
envi
ronm
ent.
2.83
2.
68
2.92
Pr
oduc
t Rec
over
y A
war
enes
s (PR
A) /
Cro
nbac
h A
lpha
(α)=
0.7
7
d1. T
he c
ontri
butio
n, a
cqui
red
as a
resu
lt of
pro
duct
reco
very
, gai
ns fa
vor f
or th
e so
ciet
y.
4.08
3.
84
4.23
d2
. Pro
duct
reco
very
hel
ps to
pro
tect
the
natu
ral s
ourc
es.
4.21
4.
03
4.33
d3
. The
pac
kage
of t
he p
urch
ased
pro
duct
sha
ll be
mad
e of r
ecyc
labl
e m
ater
ial.
4.
16
3.97
4.
27
Gre
en P
rodu
ct A
ware
ness
(GPA
) / C
ronb
ach
Alph
a(α)
= 0.
65
c1
. I b
elie
ve th
at co
nsum
ing
gree
n pr
oduc
ts is
bet
ter f
or h
uman
hea
lth.
3.97
3.
85
4.04
c2
. I th
ink
that
gre
en p
rodu
cts a
re c
ompl
etel
y fr
iend
ly to
nat
ure.
3.
45
3.27
3.
60
c3. T
he w
aste
s of g
reen
pro
duct
s are
har
mle
ss, s
ince
they
can
be
anni
hila
ted
durin
g th
e nat
ural
pro
cess
. 3.
69
3.56
3.
77
Envi
ronm
enta
l Beh
avio
r(E
B) /
Cro
nbac
h A
lpha
(α)=
0.7
4
e1. I
dis
cuss
abo
ut th
e en
viro
nmen
tal p
robl
ems
durin
g th
e fe
llow
con
vers
atio
ns.
2.76
2.
62
2.84
e2
. I th
row
the
was
tes b
y se
para
ting.
3.
20
2.94
3.
37
e3. I
do
not p
urch
ase
the
prod
ucts
of th
e fir
ms t
hat d
amag
e th
e en
viro
nmen
t. 2.
90
2.74
3.
01
e4. I
try
to c
onvi
nce
my
fam
ily m
embe
rs a
nd fr
iend
s no
t to
buy
the
prod
ucts
that
dam
age
the
envi
ronm
ent.
3.
10
2.85
3.
25
e5. I
hav
e ch
ange
d m
y lif
esty
le in
beh
alf o
f pro
tect
ing
the n
atur
e.
2.47
2.
37
2.53
e6
. I d
ecre
ase t
he c
onsu
mpt
ion
of e
lect
ricity
, wat
er a
nd fu
el in
ord
er to
pro
tect
the
natu
re.
3.31
3.
12
3.42
e7
. I d
o at
tend
to p
lant
ing
tree.
3.
05
3.08
3.
03
Envi
ronm
enta
lly C
onsc
ious
Pur
chas
ing
Beh
avio
r (E
CPB
) / C
ronb
ach
Alp
ha(α
)= 0
.77
f1
. Con
side
ring
thei
r dam
ages
, I e
xcha
nge
the
prod
ucts
I ow
n, w
ith e
colo
gica
l pro
duct
s.
2.59
2.
56
2.61
f2
. Whi
le p
urch
asin
g a
prod
uct,
I loo
k w
heth
er th
ere
is a
n ec
olog
ical
pro
duct
labe
l on
the
pack
age
or n
ot.
2.81
2.
58
2.95
f3
. I p
refe
r eco
logi
cal p
rodu
cts s
ince
they
do
not d
istu
rb th
e na
tura
l bal
ance
. 3.
52
3.27
3.
68
f4. I
pur
chas
e eco
logi
cal p
rodu
cts a
lthou
gh th
ey a
re m
ore
expe
nsiv
e th
an th
e si
mila
r pro
duct
s.
2.98
2.
65
3.18
a I
nver
ted
item
s.
-
327
Int. J. Environ. Res., 6(1):323-334, Winter 2012
“Intention” has to be developed in order that a personperforms
the behaviour. Factors affecting “Intention”can be listed as
“Attitude towards the Behavior”,“Subjective Norm”, and “Perceived
BehavioralControl”. According to the planned behavior
theory,“Behavior” is directly under the influence of“Intention”
(Ajzen, 1991; Ajzen, 2005; Ajzen andFishbein, 2000). This paper is
based on Ajzen’s (1985)theory of planned behavior and Kaiser et
al.’s (1999)environmental attitude and ecological behaviour
model.
Research model and hypothesesIn this research model,
environmental concern was
used as the estimator of the environmental attitudeand product
recovery awareness. Environmentalattitude and product recovery
awareness were usedas the estimator of the green product
awareness.Environmental attitude, environmental behavior andgreen
products awareness were used as the estimatorof the environmentally
conscious purchasing behavior(Fig. 1).
Here, we explained the research hypothesis andtheir foundations.
These hypotheses are depicted inTable 2. Individuals, who encounter
with environmentalproblems, are expected to react to the problem
thataffects her/him. This expected reaction constitutes theconcept
of concern about environmental problems.Kaiser and Shimoda (1999)
indicated that as theconcern of individuals about the
environmentincreases, their environmental attitudes increases
inparallel to it. In order to investigate the relationshipbetween
environmental concern and environmentalattitude, an H1 hypothesis
was developed.
Fig. 1. Research Model
Product recovery is being used as an effectiveway in protecting
the environment. Environmentallyconcerned enterprises shape their
productionprocesses in an environmentally conscious manner.Bolaane
(2006), Sidique et al. (2010) stated thatconsumers with
environmental concern preferrecyclable products. In order to search
the relationshipbetween environmental concern and product
recoveryawareness, H2 hypothesis was introduced.
Green products are known to be the products thatplay an
effective role in protecting the environment. Itcould be assumed
that, product recovery awarenessmay affect green product awareness.
Both factors arealso having a positive affect on protecting
theenvironment. In order to investigate the relationshipbetween
product recovery awareness, which is effectivein protecting the
nature, and green products awareness,H4 hypothesis was developed.
On the other hand,environmental concern was used as the estimator
ofproduct recovery awareness and product recoveryawareness was used
as the estimator of green productawareness. H2 and H4 hypothesis,
which examine theserelationships, were adapted from Mannetti et al.
(2004),Bolaane (2006) and Sidique et al. (2010).
In order to reveal the existence of the relationshipbetween
environmental attitude and behavior, H3hypothesis was introduced.
It was assumed thatindividuals with environmental attitudes would
havegreen product awareness by converting this awarenessinto
environmental behavior. Environmental attitudesand behaviors, on
the other hand, would turn intoenvironmentally conscious purchasing
behavior. Inorder to determine the existence of the relationship
of
-
328
Arslan, T. et al.
environmental attitude and environmental behaviorwith green
product awareness and environmentallyconscious purchasing behavior,
H5, H6 ve H7hypothesis were developed respectively. With
theassumption that environmentally consciouspurchasing behavior
would develop by converting thegreen product awareness, H8
hypothesis wasintroduced. The relationship between
environmentalconcern and environmental attitudes was investigatedby
Kaiser and Shimoda (1999). The relationshipbetween the green
product awareness andenvironmentally conscious purchasing behavior
wasexamined in the studies of Mostafa (2007) and Tilikidouand
Delistavrou (2008).
H1, H3, H5, H6, H7 and H8 hypothesis, whichwere examined in this
study, were adapted from ofKaiser and Shimoda (1999), Mostafa
(2007) andTilikidou and Delistavrou (2008) and Fraj and
Martinez(2007). Kaiser and Shimoda (1999) examined therelationship
between environmental concern andenvironmental attitude; Tilikidou
and Delistavrou(2006) examined the relationship
betweenenvironmental concern and environmental behavior;Kaiser et
al. (1999); Fraj and Martinez (2007) examinedthe relationship
between environmental attitude and
Table 2. Research Hypothesis
H yp ot hesis
H 1 As the environme ntal c once rn incr eases, e nvironmenta l
attitude incre ase s, as well.
H 2 As the environme ntal c once rn incr eases, product recovery
a wa reness incr eases, a s well.
H 3 As the environme ntal a ttitude increases, environmental
behavior incre ases, a s we ll.
H 4 As the produc t recovery awar eness incre ase s, gree n pr
oduct aware ne ss increa se, as we ll.
H 5 As the envir onm ental attitude inc reases, envir onme
ntally conscious purchasing behavior inc reases, as
well.
H 6 As the e nvironm enta l beha vior incre ase s, environm
entally consc ious purc ha sing behavior inc reases,
a s we ll.
H 7 As the environme ntal a ttitude increases, green produc ts a
wa reness incr ease, as well.
H 8 As the green pr oducts aware ne ss incre ase s, e
nvironmenta lly c onscious purchasing behavior incr eases,
a s we ll.
environmental behavior. In this research, in additionto the
existing literature a much more comprehensivemodel was developed
and hypotheses were tested.The hypotheses, examined within the
research, arepresented in Table 2.
RESULTS & DISCUSSIONDescriptive statistics
The analysis and averages related to the scale thatwas used in
the studied items within the factors aregiven in Table 1.
Examination of the items within thefactors show that the statement
“The fact that factoriesperform production without checking whether
it isharmful to nature or not, scares me.” within the “EC”factor,
has the highest average (4.37). The statementof “If there is to be
a choice between two products, theproduct that causes the least
harm to people andenvironment shall be purchased.” within the
“EA”factor, has the highest average (3.98). The statementof
“Recycling helps to protect the natural sources.”within the “PRA”
factor, has the highest average (4.21).The statement of “I believe
that consuming ecologicalproducts is better for human health.”
within the “GPA”factor, has the highest average (3.97). The
statementof “I decrease the consumption of electricity, water
Table 3. T-test Results for the Comparison of Males and Females
with Respect to theFactors Considered in the Study
Factors t s ig . M e an D if fere nce EC -5 .072 p< 0.001 -0
.422 EA -4 .069 p< 0.001 -0 .312
PR A 2 .039 0 .043 -0 .136 G PA -2 .821 0 .005 -0 .236 EB -3 .41
0 .001 -0 .275
ECP B -4 .423 p< 0.001 -0 .437
-
329
Int. J. Environ. Res., 6(1):323-334, Winter 2012
and fuel in order to protect the nature.” within the “EB”factor,
has the highest average (3.31). The statement of“I prefer
ecological products since they do not disturbthe natural balance.”
within the “ECPB” factor, has thehighest average (3.52). Checking
the averages relatedto factors, it is seen that the highest average
belongsto the “EC” (4.19) factor, which is respectively followedby
“PRA” (4.15), “GPA” (3.70), “EA” (3.51), “EB” (2.97)and “ECPB”
(2.97).
Examining the green product awareness, it is seenthat females is
more concerned about this subject andconsequently, their positive
attitudes are higher.Besides, the frequency of environmentally
consciouspurchasing behavior is higher in females, being parallelto
green product awareness. Examining the study ingeneral, it could be
said that females are more concernedand conscious about the issues
of environmentalattitude and behavior than males (Table 3).
Structural model resultsThere are more than one goodness of fit
index for
Structural Equation Model. When the literature issearched, the
most commonly used test statistics inSEM are likelihood ratio
chi-square statistics (χ2), rootmean square error of
approximation(RMSEA), goodnessof fit index statistics (GFI) and
adjusted goodness offit index (AGFI). If the value of {χ2 /d.f.} is
less than 3 itmeans that there is an acceptable fit. If the RMSEA
isless than 0.05 it shows the perfect fit, 0,05< RMSEA 0,1 means
poor fit.GFI is used similar to the statistics of coefficient
ofdetermination (R2) in Regression Analysis. AGFI isused similar to
the statistics of adjusted coefficient ofdetermination in
Regression Analysis. AGFI and GFIhave value between 0 and 1 and
generally having thevalue of close to 1 means that the model fits
well.(Raykov & Marcoulides, 2006).
LISREL 8.74 program was used in the analysis ofthe data. The
goodness-of-fit indexes of the model are
calculated as; 2χ = 487.77; df. =291; 2χ / df. =1.68;
RMSEA=0.047; NFI=0.92; NNFI=0.97; CFI=0.97;RMR=0.068; GFI=0.89;
AGFI=0.87. The goodness-of-fit indexes show that the model is
within acceptablelimits. Goodness-of-fit indexes acceptable limits
arepresented in Table 4 (for more information aboutgoodness-of-fit
indexes, see Schermelleh-Engel et al.(2003); Byrne (1998); Hayduk
(1987); Jöreskog andSörbom (2001)).
Structural equations, the results related tohypothesis and
standardized parameter estimationvalues are given in Table 5.
Looking at the Table 5; it isseen that an increase of one unit in
the factor of“environmental concern (EC)” causes an increase of0.69
units in “environmental attitude (EA)” and anincrease of 0.53 units
in “product recovery awareness(PRA)”. Besides, while an increase of
one unit in thefactor of “environmental attitude (EA)” causes
anincrease of 0.67 units in “environmental behavior (EB)”and an
increase of 0.64 unit in “green productawareness (GPA)”, it could
be said that an increase ofone unit in the factor of “product
recovery awareness(PRA)” causes an increase of 0.21 unit in
“greenproduct awareness (GPA)”. Additionally, it is revealedthat an
increase of one unit in the factor of “greenproduct awareness
(GPA)” causes an increase of 0.24unit in “environmentally conscious
purchasingbehavior (ECPB)” and an increase of one unit in thefactor
of “environmental behavior (EB)” causes anincrease of 0.70 unit in
“environmentally consciouspurchasing behavior (ECPB)” (Table 5).
Besides, sincethe relationship of “EA” “ECPB” was notstatistically
found significant, H5 could not beconfirmed. R2 value, related to
the “Model of Students’Environmentally Conscious Purchasing
Behavior(SECPBM)”, was found as 0.74. This R2 value showsthat the
factors explain 74% of the change within “EPB”and 26% of it could
be explained through the factors,which do not exist within the
model.
It could be said that the increase of environmentalconcern and
environmental attitudes of universitystudents display an amplifier
effect on their
Table3. T-test Results for the Comparison of Males and Females
with Respect to theFactors Considered in the Study
Fact or s t sig. Mean Differe nce
EC -5.072 p
-
330
Table 5. Standardized parameter estimation values, t values and
hypothesis
Hypothesis Path Std. parameter estimation values t -value
Result
H1 (EC) (EA) 0.69 6.82 Confirmed
H2 (EC) (PRA) 0.53 6.91 Confirmed H3 (EA) (EB) 0.67 5.52
Confirmed
H5 (EA) (ECPB) 0.02 0.16 Not Confirmed
H7 (EA) (GPA) 0,64 5.38 Confirmed
H4 (PRA (GPA) 0.21 2.66 Confirmed H8 (GPA) (ECPB) 0.24 2.15
Confirmed
H6 (EB) (ECPB) 0.70 5.62 Confirmed
Structural Equations
EA = 0.69*EC R2=0.48 PRA = 0.53*EC R2=0.28
GPA = 0.64*EA + 0.21*PRA R2=0.55
EB = 0.67*EA R2=0.44
ECPB = 0.02*EA + 0.24*GPA + 0.70*EB R2=0.74
EA PRA GPA EB ECPB EC EA 1.00 PRA 0.36 1.00 GPA 0.71 0.44 1.00
EB 0.67 0.24 0.48 1.00 ECPB 0.66 0.28 0.59 0.83 1.00 EC 0.69 0.53
0.55 0.46 0.47 1.00
Table 6. Correlation matrix related to the factors
Table 4. Goodness-of-Fit Indexes Acceptable Limits
Fit Measures Good Fit Acceptable Fit Developed Model
RMSEA 0
-
331
Int. J. Environ. Res., 6(1):323-334, Winter 2012
Fig. 2. Students’ Environmentally Conscious Purchasing Behavior
Model (SECPBM) LISREL 8.74 Output
-
332
Arslan, T. et al.
environmental behaviors. Besides, in the developedstructural
equation model, it was seen thatenvironmental concern,
environmental attitude andproduct recovery awareness play an
amplifier role forthe student’s green products awareness. Finally,
it wasrevealed that environmental attitudes, environmentalbehavior
and green products awareness display apositive effect on
environmentally consciouspurchasing behaviors. Additionally
correlation matrix,related to the factors is given in Table 6.
Fig. 2 shows a structural equation model forstudents’
environmentally conscious purchasingbehavior. While environmental
concern explains 48%of the change in the environmental attitude and
28%of the product recovery awareness, environmentalattitudes and
product recovery awareness explain 55%of the green product
awareness. While environmentalattitudes explain 44% of the change
in theenvironmental behavior; environmental attitudes,green product
awareness and environmental behaviorexplain 74% of the change in
the environmentallyconscious purchasing behavior (Fig. 2).
CONCLUSIONThe results of the study revealed that
environmental
attitudes could be used as the estimator of theenvironmental
behavior (EA EB). This findingcoincides with the findings, acquired
through the studieswhich were performed in different cultures in
theliterature (Kaiser and Shimoda, 1999; Dono et al.
2009).Considering the fact that the study was conducted withthe
university students, it is seen that it has similaritieswith the
results of the study conducted by Kaiser andWilson (2000) with the
students in California, as well.Besides, the significant
relationship betweenenvironmental attitude and environmental
behavior(EA EB), which is obtained from the study, depicts
asimilarity with the findings of the study done by Donoet al.
(2009). No statistically significant relationship wasfound between
environmental attitude andenvironmentally conscious purchasing
behavior(EA ECPB) in this study. Mostafa (2007) determinedthe
existence of a weak relationship between the interestof the
individuals in green consumption and their realconsumption
behaviors. The fact that no statisticallysignificant relationship
between environmental attitudeand environmentally conscious
purchasing behavior(EA ECPB) could be found in the structural
equationmodel, which was developed in this study, confirms
thefindings in the study of Mostafa (2007). In addition tothe
findings of Mannetti et al. (2004), Bolaane (2006)and Sidique et
al. (2010), who investigated the productrecovery awareness, it was
seen in the findings of thisstudy that environmental concern
effects the productrecovery awareness (EC PRA) positively.
While Bolaane (2006) and Sidique et al. (2010) claimthat
educational status influences the product recoveryawareness in
their studies, it was additionally determinedthat gender, as well,
has also a significant effect onproduct recovery awareness in this
study. This resultcould be seen in the result of the t test (Table
3).
The findings of this study show that compared tomales, females
have much more environmental concernand attitudes and they also
support the view of Tilikidouand Delistavrou (2006), which claims
that female withhigh education take more participation in
environmentallyactivities than others. As a result of this study,
it wasrevealed that there is a statistically significant
differencebetween the environmental attitudes of female and
maleuniversity students (Table 4).As a result of the study,two
situations were emerged to be discussed. The firstof them is the
indirect relationship of“EA” ”EB” ”ECPB” ( 46.0=β )
statisticallysignificant, whereas no direct
cause-and-effectrelationship was found between environmental
attitudeand environmentally conscious purchasing behavior(“EA”
”ECPB”, 02.0=β ). From the structure ofthis relationship (“EA” ”EB”
”ECPB”), it isunderstood that among the ones who
developedenvironmental attitude, only the ones who convertedthese
attitudes into environmental behavior, displayenvironmentally
conscious purchasing behavior. In thesecond situation, while the
relationship of“EA” ”ECPB” ( 02.0=β ) was not foundstatistically
significant, the indirect relationship of“EA” ”GPA” ”ECPB” ( 15.0=β
) was foundsignificant. From the structure of this
relationship(“EA” ”GPA” ”ECPB”), it is understood that amongthe
ones who developed environmental attitude, onlythe ones who
supported these attitudes by greenproduct awareness, directed
towards environmentallyconscious purchasing behavior. Considering
the firstsituation, it might be conceived that there is
theconditional dependence structure between greenproduct awareness
and environmentally consciouspurchasing behavior, upon realising
the environmentalbehavior. In the second situation, on the other
hand, itis conceived that there might be the conditionaldependence
structure between environmental attitudeand environmentally
conscious purchasing behavior,upon realising the green product
awareness.
The following could be said for these twosituations; there are
actually conditional relationshipsbetween latent structures. On
condition that theindividual displays environmental behavior,
theremight be a cause-and-effect relationship betweenenvironmental
attitude and environmentally conscious
-
333
Int. J. Environ. Res., 6(1):323-334, Winter 2012
purchasing behavior. In order to reveal theseconditional
dependence structures, an additional studyis being planned with the
help of graphical models. Inthe future studies, the research
of(“EA” ”EB” ”ECPB”) and(“EA” ”EPA” ”ECPB”) conditioned
dependencestructures and results of their conditional
dependencecoefficients are being planned.According to
thedescriptive results of the study (Table 1 and Table 3),it could
be said that females are more environmentalistcompared to males.
Many psychological reasons couldbe depicted among the reasons of
this condition;however, probably one of the most important of
thesereasons might be the sense of protection, which comesfrom
maternal instinct of females.
As a brief summary; in the literature environmentalconcern,
environmental attitude, environmentalbehavior, product recovery
awareness, green productawareness and environmentally conscious
purchasingbehavior concepts analyzed independently from eachother.
Whereas these mentioned concepts arecorrelated and in this research
we consider their jointeffect by means of the SEM. This research
can beextended to evaluate the effect of cultural differenceson the
environmentally conscious purchasingbehavior. Although the
developed model is statisticallysuggestible for intercultural
assessments, still it shouldbe examined with larger samples and
different culturesfor the validation purpose.
REFERENCESAjzen, I. and Fishbein, M. (1980). Understanding
attitudesand predicting social behavior. Englewood Cliffs,
NJ,Prentice-Hall.
Ajzen, I . (1985). From intention to actions: A theory ofplanned
behavior. In J. Kuhl and J. Beckman (eds), Action-control: From
cognition to behavior, 11-39. Heidelberg:Springer.
Ajzen, I. (1991). The theory of planned behavior.Organizational
Behavior and Human Decision Processes,50, 179-211.
Ajzen, I. and Fishbein, M. (2000). Attitudes and the
attitude-behavior relation: Reasoned and automatic process. In
W.Stroebe and M. Hewstone (eds), Europen Review of
SocialPsychology, 11, 1-33. Chichester: Wiley.
Ajzen, I. (2005). Attitudes, personality and behavior. NewYork:
Open University Press.
Axelrod, L. J. and Lehman, D. R. (1993). Responding
toenvironmental concern: what factors guide individual
action?Journal of Environmental Psychology, 13, 149-159.
Birgelen, M. V., Semeijin, J. and Keicher, M. (2009).Packaging
and pro-environmental consumption behavior.Environment and
Behavior, 41 (1), 125-146.
Bolaane, B. (2006). Constraints to promoting peoplecentered
approaches in recycling. Habitat International, 30,731–740.
Byrne, M. B. (1998) Structural equation modeling withLISREL,
PRELIS and SIMPLIS: Basic concepts,applications, and programming.
New Jersey: LawrenceErlbaum.
Chan, R. Y. (1999). Environmental attitudes and behavior
ofconsumers in China. Journal of International ConsumerMarketing,
11 (4), 25-52.
Cheng, E. W. L. (2001). SEM being more effective thanmultiple
regression in parsimonious model testing formanagement development
research. Journal of ManagementDevelopment, 20 (7), 650-667.
Cudeck, R., Toit, D. S. and Sörbom, D. (2000).
Structuralequation modeling: Present and future. Chicago:
ScientificSoftware International Inc.
Dono, J., Webb, J. and Richardson, B. (2010). Therelationship
between environmental activism, pro-environmental behavior and
social identity. Journal ofEnvironmental Psychology, 30 (2),
178-186.
Fraj, E. and Martinez, E. (2007). Ecological consumerbehavior:
An emprical analysis. International Journal ofConsumer Studies, 31,
26-33.
Gungor, A. and Gupta, S. M. (1999). Issues inenvironmentally
conscious manufacturing and productrecovery: A survey. Computers
and Industrial Engineering,36, 811-853.
Hayduk, L. A. (1987). Structural equation modeling withLISREL:
Essentials and Advances. Baltimore: John Hopkins.Hini, D., Gendall,
P. and Kearns Z. (1995). The link betweenenvironmental attitudes
and behavior. Marketing Bulletin,6, 22-31.
Ilgin, M. A. and Gupta, S. M. (2010). Environmentallyconscious
manufacturing and product recovery (ECMPRO):A review of the state
of the art. Journal of EnvironmentalManagement, 91 (3),
563-591.
Jöreskog, K. and Sörbom, D. (2001). LISREL 8: User’sreference
guide. Chicago: Scientific Software InternationalInc.
Kaiser, F. G. and Shimoda, T. A. (1999). Responsibility as
apredictor of ecological behavior. Journal of
EnvironmentalPsychology, 19, 243-253.
Kaiser, F. G., Wölfing, S. and Fuhrer, U. (1999).Environmental
attitude and ecological behavior. Journal ofEnvironmental
Psychology, 19, 1-19.
Kaiser, F. G. and Wilson, M. (2000). Assessing people’sgeneral
ecological behavior: A cross-cultural measure. Journalof Applied
Social Psychology, 30 (5), 952-978.
Kaiser, F. G., Oerke, B. and Bogner, B. X. (2007).
Behavior-based environmental attitude: Development of an
instrumentfor adolescents. Journal of Environmental Psychology,
27,2442-251.
-
334
Kalantari, Kh. and Asadi, A. (2010). Designing a structuralmodel
for explaining environmental attitude and behaviourof urban
residents (Case of Tehran). International Journal ofEnvironmental
Research, 4 (2), 309-320.
Kish, L. (1965). Survey sampling. New York: John Wiley &Sons
Inc.
Loudon, D. L. and Della Bitta, A. J. (1993). Consumerbehavior:
Concept and applications (4th edition). McGrawHill: Auckland.
Mannetti, L., Pierro, A. and Livi, S. (2004). Recycling:Planned
and self-expressive behavior. Journal ofEnvironmental Psychology,
24, 227–236.
Mondéjar-Jiménez, J . A., Cordente-Rodriguéz,
M.,Meseguer-Santamaría, M. L., and Gázquez-Abad, J. C.(2011).
Environmental behavior and water saving in Spanishhousing.
International Journal of Environmental Research,5 (1), 1-10.
Mostafa, M. M. (2007). A hierarchical analysis of the
greenconsciousness of the Egyptian consumer. Psychology
&Marketing, 24 (5), 445-473.
Nordlund, A. M. and Garvill, J. (2002). Value structuresbehind
pro-environmental behavior. Environment andBehavior, 34 (6),
740-757.
Pang, N. S. K. (1996). School values and teachers’ feelings:a
LISREL model. Journal of Educational Administration, 34(2),
64-83.
Raykov, T. and Marcoulides, G. A. (2006). A first course
instructural equation modeling. London: Lawrence
ErlbaumAssociates.
Reisinger, Y. and Turner, L. (1999). Structural equationmodeling
with LISREL: Application in tourism. TourismManagement, 20,
71-88.
Schermelleh-Engel, K., Moosbrugger, H. and Müller, H.(2003).
Evaluating the fit of structural equation models: Testsof
significance and descriptive goodness-of-fit measures.Methods of
Psychological Research Online, 8 (2), 23-74.
Sidique, S. F., Lupi, F. and Joshi, S. V. (2010). The effects
ofbehavior and attitudes on drop-off recycling
activities.Resources, Conservation and Recycling, 54, 163–170.
Steenkamp, B. E. M. and Baumgartner, H. (2000). On theuse of
structural equation models for marketing modelling.International
Journal of Research in Marketing, 17, 195-202.
Steg, L. and Vlek, C. (2009). Encouraging
pro-environmentalbehavior: An integrative review and research
agenda . Journalof Environmental Psychology, 29, 309-317.
Thierry, M., Salomon, M., van Nunen, J. and vanWassenhove, L.
(1995). Strategic issues in product recoverymanagement, California
Management Review, 37 (2), 114–135.
Tilikidou, I. and Delistavrou, A. (2006). Consumers’ecological
activities and their correlates. Retrieved July11,2009, from
http://www.ctw-congress.de/ifsam/download/track_9/pap00169.pdf
Tilikidou, I. (2007). The effects of knowledge and attitudesupon
Greeks’ pro-environmental purchasing behavior.Corporate Social
Responsibility and EnvironmentalManagement, 14, 121–134.
Tilikidou, I. and Delistavrou, A. (2008). Types and
influentialfactors of consumers’ non-purchasing ecological
behaviors.Business Strategy and the Environment, 18, 61-76.
Structural Model for Environmentally Conscious Behavior