Xu, Yalin e al.
A icle — Published Ve sion
Can ca bon labels shi consume s owa ds sus ainable
ood? E idence om Chinese consume s
Sus ainable Fu u es
P o ided in Coope a ion wi h:
Leibniz Ins i u e o Ag icul u al De elopmen in T ansi ion Economies (IAMO), Halle (Saale)
Sugges ed Ci a ion: Xu, Yalin e al. (2024) : Can ca bon labels shi consume s owa ds sus ainable
ood? E idence om Chinese consume s, Sus ainable Fu u es, ISSN 2666-1888, Else ie ,
Ams e dam, Vol. 8, pp. 1-13,
h ps://doi.o g/10.1016/j.s .2024.100363 ,
h ps://www.sciencedi ec .com/science/a icle/pii/S2666188824002120
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/306867
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h p://c ea i ecommons.o g/licenses/by-nc/4.0/
Can ca bon labels shi consume s owa ds sus ainable ood? E idence om
Chinese consume s
Yalin Xu
a
, Zhiwen Zhang
a
, Yanjun Ren
a,b
, Rong Yuan
c
, Yanan Wang
a,*
, Rui Li
a
,
Shunan Zhao
a
, Lu Qiu
a
a
College o Economics and Managemen , No hwes A&F Uni e si y, Yangling 712100, China
b
Leibniz Ins i u e o Ag icul u al De elopmen in T ansi ion Economies (IAMO), Theodo -Liese -S . 2, 06120 Halle (Saale), Ge many
c
School o Economics and Business Adminis a ion, Chongqing Uni e si y, Shazhengjie 174, 400040 Chongqing, China
ARTICLE INFO
Keywo ds:
Ca bon-labeled ag icul u al p oduc s
Willingness o pay
Theo y o planned beha io
No m ac i a ion model
China
ABSTRACT
Ca bon labels a e becoming an essen ial ool o policymake s in many coun ies o p omo e low-ca bon con-
sump ion. To assess cus ome s’willingness o pay o i e ca bon-labeled ag icul u al p oduc s (CAP), we used
paymen ca d o conduc a ques ionnai e su ey among 641 esponden s in Shanghai, Nanjing, Wuhan, and
Xi’an, all loca ed in China. This s udy quan i a i ely analyzes he in luencing ac o s and in e ac i e mechanisms
o he publicʼs willingness o pu chase CAP h ough he ex ended Theo y o Planned Beha io and No m Ac i-
a ion Model. The esul s show ha consume s’willingness o pay a p emium o ca bon-labeled milk, co n,
bananas, oma oes, and eggs is 27.50 %, 29.73 %, 26.86 %, 26.51 %, and 24.26 % espec i ely. Pe cei ed
beha io al con ol has he s onges posi i e in luence on pu chase in en ion, ollowed by subjec i e no ms,
a i udes owa d he beha io , and pe sonal no ms. The e is a signi ican media ing e ec be ween awa eness o
consequences and pe sonal no ms, which indi ec ly in luences pe sonal no ms h ough he asc ip ion o e-
sponsibili y and subsequen ly a ec s pu chase in en ion. Addi ionally, he e is a gap be ween pu chase in en ion
and beha io , and isk pe cep ion nega i ely mode a es he ela ionship be ween he wo. Based on he esea ch
indings o his pape , p ac ical and e ec i e policy sugges ions a e p oposed o he go e nmen o p omo e
ca bon labeling policies and educe ca bon emissions.
1. In oduc ion
The e is g owing awa eness o he impac o ood choices on clima e
change. The ood sys em is es ima ed o be esponsible o 26–34 % o
global g eenhouse gas (GHG) emissions [1,2]. Recen modeling in-
dica es ha e en i ossil uel emissions we e o cease immedia ely,
cu en ends in he wo ld’s ood sys em would make i challenging o
achie e he IPCC’s 1.5 ◦C objec i e. By he end o he cen u y, hese
ends would h ea en he 2 ◦C goal [3]. Though ood p oduce s should
ocus on lessening hei en i onmen al impac , consume beha io
changes can also impac p oduc ion sys em imp o emen s [2]. Shi ing
o low-ca bon oo p in die s has he po en ial o signi ican ly dec ease
ca bon emissions and alle ia e he s ain on he en i onmen [4–6].
Acco dingly, ca bon label p oduc s a e a good low-ca bon consump ion
o ien a ion, and cus ome s can educe hei ca bon oo p in by pu -
chasing ood ha is less ha m ul o he en i onmen [7,8].
Among he policy ools used o p omo e he de elopmen o low-
ca bon economy, ca bon labels ha e been implemen ed in se e al
coun ies, such as he Uni ed Kingdom, Ge many, F ance, Sweden, and
he Uni ed S a es [9]. Ca bon labels e e o using quan i a i e measu es
on p oduc labels o indica e he amoun o g eenhouse gas eleased
du ing p oduc ion [10]. Acco ding o Edwa ds-Jones e al. [11], ca bon
labels can help educe g eenhouse gas emissions. Rega ding co po a e
emissions educ ion, Shi [12] poin ed ou ha ca bon labels can b ing
economic bene i s o companies, such as educing emissions and
achie ing cos sa ings. A he consume le el, p o iding in o ma ion o
consume s abou he ca bon con en o p oduc s h ough labels can
assis hem in making in o med decisions abou pu chasing low-ca bon
p oduc s. This, in u n, con ibu es o he o e all educ ion o ca bon
emissions [13]. Despi e he need o a signi ican change in pe sonal
ood consump ion, esea ch indica es ha people a ely conside hei
die a y choices when asked wha hey can do o help he en i onmen
[14] and o en unde es ima e he en i onmen al impac o hei ood
[15,16]. Meanwhile, in some cases, consume s may be eluc an o pay
mo e o ca bon-labeled p oduc s. The gene al lack o awa eness
* Co esponding au ho a : College o Economics and Managemen , No hwes A&F Uni e si y: NO.3 Taicheng Road, Yangling, Shaanxi, China.
E-mail add ess: [email p o ec ed] (Y. Wang).
Con en s lis s a ailable a ScienceDi ec
Sus ainable Fu u es
jou nal homepage: www.sciencedi ec .com/jou nal/sus ainable- u u es
h ps://doi.o g/10.1016/j.s .2024.100363
Recei ed 13 June 2024; Recei ed in e ised o m 30 Augus 2024; Accep ed 1 No embe 2024
Sus ainable Fu u es 8 (2024) 100363
A ailable online 3 No embe 2024
2666-1888/© 2024 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY-NC license (
h p://c ea i ecommons.o g/licenses/by-
nc/4.0/ ).
ega ding how die a y decisions a ec he en i onmen emains a majo
obs acle o educing emissions o g eenhouse gases caused by ood [17].
Globally, he e is a g owing body o esea ch on ca bon labeling,
pa icula ly ega ding how i a ec s consume pu chasing beha io [18,
19]. Se e al s udies ha e been done on consume s om a ious
geog aphic and cul u al backg ounds [20]. Ne e heless, e en i he e
ha e been some p elimina y in es iga ions o esea ch on Chinese
consume s, such as a s udy o ca bon-labeled bee among u ban Chinese
cus ome s [21], addi ional empi ical esea ch is equi ed o ob ain a
deepe unde s anding o he p e e ences and beha io al pa e ns o
a ious consume g oups. Cu en ly, go e nmen s, p oduce s, and
me chan s ha e no been able o ecei e comp ehensi e p omo ional
ad ice o guidance. To achie e he ansi ion o a low-ca bon con-
sump ion pa e n, China should in eg a e ca bon labeling in o a wide
ange o consume p oduc ca ego ies. To add ess he esea ch gaps
men ioned abo e, his pape aims o in es iga e Chinese consume s’
pe cep ions o ca bon-labeled ag icul u al p oduc s (CAP), including
hei willingness o pay and he speci ic amoun hey a e willing o pay
o di e en ypes o CAP. The inno a ions a e as ollows: i s ly, his
s udy c ea es a mo e ho ough explana o y amewo k, in eg a es he
No m Ac i a ion Model (NAM) and he Theo y o Planned Beha io
(TPB), and expands he o iginal model by adding ex e nal a iables
(such as isk pe cep ion, low-ca bon awa eness, and pe sonal knowl-
edge). All hese enhancemen s signi ican ly imp o e he modelʼs ca-
paci y o explain he willingness o he CAP and pu chasing beha io . In
addi ion o examining he di ec e ec s o esponden s’subjec i e
no ms, pe sonal no ms, and pe cei ed beha io al con ol on pu chase
in en ions, he s udy sheds ligh on he media ing a iables hese ac o s
use o a ec pu chase in en ions and beha io . Addi ionally, his s udy
con i ms ha isk pe cep ion mode a es he ela ionship be ween buy
in en ion and ac ion, p o iding a mo e comp ehensi e explana ion o
he inconsis ency be ween consume in en ion and beha io when
pu chasing CAP. Secondly, in o ma ion in e en ion is included in he
expe imen . This s udy uses an expe imen al design o explo e how
p o iding speci ic in o ma ion on ca bon labeling a ec s he publicʼs
WTP o CAP and how he ce ain y ha esponden s will choose o
imp o e he en i onmen changes. Th ough empi ical esea ch in spe-
ci ic cul u al and ma ke backg ounds, we p opose policies o p omo e
he implemen a ion o ca bon labels a all le els, including go e nmen
and en e p ises, o encou age widesp ead pa icipa ion in en i on-
men al p o ec ion and low-ca bon ac i i ies, ul ima ely achie ing “ca -
bon educ ion o all and bene i s o all.”
The emainde o he pape is s uc u ed as ollows: Sec ion 2 p e-
sen s he heo e ical amewo k and li e a u e e iew, Sec ion 3 de-
sc ibes he su ey’s me hodology and da a collec ion, Sec ion 4 p esen s
he model’s esul s, Sec ion 5 discusses he indings, and he inal sec ion
aises policy implica ions.
2. Li e a u e e iew and hypo heses
2.1. Ca bon labeling concep
The e is inc easing esea ch on ca bon labeling due o he u gen
need o ans o m he global low-ca bon economy. Ca bon labels can
success ully and e ec i ely aise consume unde s anding o he emis-
sions o mo e en i onmen ally haza dous oods, which can hen help
consume s choose mo e en i onmen ally iendly ood op ions [17].
Ra ing labels a e mo e e ec i e in guiding consume s o mo e sus ain-
able p oduc s han posi i e and nega i e labels [22]. Edenb and &
Lage k is [23] in es iga ed consume willingness o subs i u e
high-emissions mea p oduc s wi h lowe -emissions p o ein p oduc s,
including blends o mea and ege ables. They ound he a ic ligh
ca bon label a ec s choice beha io . In Canada and A gen ina, younge
consume s in he Ame icas exhibi di e ing pe cep ions o ca bon la-
beling, wi h hose who a e mo e educa ed demons a ing a p e e ence
o he a ic ligh ca bon label [24]. The g ea es impac is obse ed
when ca bon oo p in in o ma ion is exp essed in mone a y uni s and
colo -coded in a manne analogous o he a ic signal sys em [25].
T a ic ligh labels a e mos e ec i e in educing ca bon emissions,
p o iding empi ical suppo o he design o ca bon labels [26,27].
Howe e , he e a e also s udies ha ca bon labels don’ succeed in
a ac ing people’s a en ion when cus ome s a e unguided [28]. Ca bon
labeling has a posi i e bu small impac on sus ainable ood choices
[29]. As consume us in he label g ows, ca bon labeling boos s he
pe cei ed en i onmen al sus ainabili y o animal-based p oduc s bu
no ha o plan -based p oduc s.
Addi ionally, he “halo e ec ”o plan -based oods may lessen he
e ec s o ca bon labeling [30]. To change consume beha io , con-
sume s mus de elop us in labels and he o ganiza ions behind hem
[31]. This us and posi i e a i ude ema kably in luence he WTP and
he posi i e e ec o policy measu es on consume s’WTP [32]. In e -
es ingly, amily and pee in luences a e mo e likely o a ec consume s’
pu chasing decisions han media exposu e. This would appea o
emphasize he impo ance o social dynamics in p omo ing en i on-
men ally iendly consump ion [33,34].
By combining he esul s o hese s udies, we may conclude ha
ca bon labeling may encou age sus ainable consump ion, bu i s e ec s
may be a ec ed by cul u al, social, and psychological ac o s. In addi-
ion o enhancing ou knowledge o he mechanisms unde lying he
e ec s o ca bon labeling, in eg a ing na ional and in e na ional
esea ch o e s a solid scien i ic ounda ion o c ea ing mo e ocused
en i onmen al egula ions and ma ke ing s a egies. Fu u e esea ch
could u he explo e he bes p ac ices o ca bon labeling in di e en
cul u al con ex s and how o combine socio-cul u al ac o s o imp o e
i s accep ance and e ec i eness in he global ma ke .
2.2. Theo y o planned beha io
The TPB model has i s oo s in psychological and sociological he-
o ies and is based on he Theo y o Ra ional Beha io [35]. TPB, p o-
posed by Ajzen [36], is he classic social psychological model o
explaining people’s beha io al in en ions and ac ions. I is a powe ul
and widely used ool o e alua ing, modeling, and in es iga ing peo-
ple’s beha io abou ac i i ies, p oduc s, o se ices [36]. Acco ding o
he Theo y o Planned Beha io , a i udes, subjec i e no ms, and
pe cei ed beha io al con ol impac beha io al in en ions. A i udes
owa d he beha io e e o he deg ee o which a pe son has a a o -
able o un a o able e alua ion o assessmen o he beha io unde
Nomencla u e
Abb e ia ions
ATT A i udes owa d he beha io
AR Asc ip ion o esponsibili y
AC Awa eness o consequences
CAP Ca bon-labeled ag icul u al p oduc s
CVM Con ingen alua ion me hod
LA Low-ca bon awa eness
NAM No m Ac i a ion Model
PBC Pe cei ed beha io al con ol
PN Pe sonal no ms
PK Pe sonal knowledge o CAP
PB Pu chase beha io
PI Pu chase in en ion
RP Risk pe cep ion
SEM S uc u al equa ion modeling
SN Subjec i e no ms
TPB Theo y o Planned Beha io
WTP Willingness o pay
Y. Xu e al. Sus ainable Fu u es 8 (2024) 100363
2
discussion. Subjec i e no ms ep esen he pe cei ed social p essu e o
engage in a beha io o e ain om doing so. The hi d an eceden o
in en ion is he deg ee o pe cei ed beha io al con ol, which, as we’ e
seen, e e s o how easy o di icul he beha io is ega ded o be o
ca y ou . This an eceden is hough o e lec p io expe ience as well
as an icipa ed impedimen s. The Theo y o Planned Beha io sugges s
ha beha io al pe o mance is in luenced by beha io al in en ions,
which a e s onge i he indi idual has a mo e posi i e a i ude, eels
mo e p essu e om ex e nal no ms, o eels ha he o she has mo e
con ol o e his o he beha io [36]. When a pe son can clea ly un-
de s and he objec i e condi ion es ic ions, such as esou ces and
abili ies, ha he o she equi es o conduc a gi en beha io , pe cei ed
beha io al con ol can also di ec ly in luence he occu ence o an in-
di idual’s beha io .
The TPB model has been widely deployed o s udy sus ainable ood
consump ion beha io . Fo example, Ve mei and Ve beke [37] con-
duc ed a s udy wi h young people in Belgium. They ound ha pe sonal
a i udes, pe cei ed social impac s, pe cei ed consume e ec s, and
pe cei ed a ailabili y we e key ac o s in luencing sus ainable con-
sump ion in en ions. In a sepa a e s udy, Alam e al. [38] employed he
ex ended TPB amewo k o asce ain he ac o s in luencing sus ainable
ood consump ion beha io s in Malaysia. Thei indings indica ed ha
social no ms, pe cei ed alue, and pe cei ed consume e ec s and a -
i udes signi ican ly impac ed consump ion in en ions.
In con as , pe cei ed a ailabili y and pe cei ed consume e ec s
and in en ions signi ican ly a ec ed ac ual beha io . In China, Qi and
Ploege [39] e ised and ex ended he TPB by inco po a ing ace con-
sciousness and g oup consis ency in o he model, eplacing subjec i e
no ms o enhance he modelʼs applicabili y in Chinese consume s’in-
en ions o pu chase g een ood. They also included con idence and
pe sonal cha ac e is ics as model componen s o be e e lec he cu -
en consume en i onmen and ea u es. Du ing he COVID-19
pandemic, Qi and Ploege [40] ex ended he TPB o include he e ec s
o e hical a i udes, heal h awa eness, and COVID-19 o explain Chinese
consume s’g een ood pu chase in en ions du ing he cu en and
pos -pandemic pe iods. Acco ding o he indings, Chinese consume s’
in en ions o pu chase g een oods a e be e explained and p edic ed by
he expanded TPB model han by he o iginal TPB model.
In conclusion, he TPB and i s ex ended model p o ide a solid
heo e ical ounda ion o unde s anding consume s’sus ainable ood
consump ion beha io s unde di e en cul u al and en i onmen al
condi ions. Howe e , ew s udies ha e used he model o analyze he
public’s willingness and beha io o pu chase CAP and hei links wi h
each o he . Based on hese esul s, we p opose he ollowing hypo heses:
H1. A i udes owa d he beha io (ATT) posi i ely a ec pu chase
in en ion (PI) in he pu chase o CAP.
H2. Subjec i e no ms (SN) ha e a posi i e impac on PI o CAP.
H3. Pe cei ed beha io al con ol (PBC) has a a o able e ec on PI
in CAP pu chasing.
H4. PBC posi i ely in luences pu chase beha io (PB) in CAP
pu chasing.
H5. PI posi i ely in luences PB in he pu chasing o CAP.
2.3. No m ac i a ion model
NAM is a beha io al heo y ha Schwa z [41] de eloped o explain
and p edic indi idual p o-social beha io . NAM con ends ha people’s
p o-social/al uis ic in en ions and beha io s, such as olun ee ing hei
ime and helping o he s, esul om hei no ms, which a e inspi ed by
hei awa eness o p oblems and obliga ions [41]. Pe sonal No ms (PN),
Awa eness o Consequences (AC), and Asc ip ion o Responsibili y (AR)
a e he h ee p ima y elemen s o NAM. The e m PN deno es an in-
di idual’s sel -expec a ions o beha io s in each si ua ion, and AC is he
p opensi y o be awa e o he consequences o one’s ac ions on o he s
[41]. The mo e likely indi iduals a e o pe cei e si ua ions in e ms o
he consequences o hei ac ions on o he s, he mo e likely hey a e o
a end o he alues and no ms associa ed wi h hose in e pe sonal
consequences, c ea ing a sense o obliga ion o exp ess hose no ms. AR
is a pe son’s sense o whe he hey a e esponsible o he consequences
o hei ac ions [41].
NAM has become one o he mos in luen ial heo ies [42]. I has
been used o s udy a a ie y o en i onmen ally iendly beha io s,
including d one ood deli e y se ices [43], li e educ ion and li e
picking beha io s [44], and en i onmen ally iendly pes con ol
adop ion beha io s [45]. S eg and G oo [46] applied NAM o explain
a ious p o-social and p o-en i onmen al beha io al in en ions. They
disco e ed ha he in e ac ion be ween AC, AR, PN, and beha io al
in en ions is a chain-media ed model, meaning ha AC ac i a es PN
h ough AR, leading o PI. Acco ding o some esea che s, ATT, SN, and
PN all indi ec ly impac consume s’in en ions o ecycle [47]. Song
e al. [48] inco po a ed SN and en i onmen al conce n in o an ex ended
NAM, a guing ha PN, es ablished by consume s’AC and AR, signi i-
can ly impac s hei beha io al in en ion. In he con ex o he opic
s udied in his pape , consume s’AC and AR may inc ease hei PN, hus
leading o suppo o pu chasing CAP.
The s udy pu s ou he ollowing hypo hesis conside ing he analyses
men ioned abo e.
H6. AC has a a o able impac on AR in pu chases o CAP.
H7. AC has a posi i e in luence on PN when pu chasing CAP.
H8. AR has a signi ican posi i e impac on PN when pu chasing
CAP.
H9. When consume s buy CAP, PN has a posi i e e ec on PI.
H10. SN has a posi i e in luence on PN when pu chasing CAP.
2.4. Ex ension o TPB and NAM
Applying and expanding heo e ical amewo ks a e essen ial o a
mo e p o ound comp ehension o consume beha io wi hin sus ainable
ood consump ion. The TPB and NAM amewo ks ha e been ex ensi ely
u ilized o p edic and explain he en i onmen al beha io o in-
di iduals [49]. Ne e heless, he ex ension and in eg a ion o hese
heo ies become especially impo an as social and en i onmen al
challenges become complex. Ajzen [36] s a es ha TPB is an open heo y
o which addi ional a iables can be added i hey cap u e a signi ican
po ion o beha io al di e ences. Recen s udies such as he TPB-NAM
in eg a ion model ha e p o en supe io o he o iginal TPB model
when s udying ac o s a ec ing Vie namese a me s’in en ion owa d
o ganic ag icul u al p oduc ion [50]. He and Sui [51] in es iga ed he
willingness o Chinese college s uden s o consume g een ood by in e-
g a ing he TPB wi h he NAM. They ound ha SN, ATT, and PN we e
he key ac o s in luencing s uden s’willingness o buy, wi h ATT ha -
ing he s onges di ec e ec on willingness o buy. This wo k illus a es
he unc ion o NAM in p omo ing he o ma ion o PN while also
expanding he applica ion o TPB. The empi ical s udy by Vie nam Le
and Nguyen [52] con i ms he impo ance o ATT, social no ms, and
pe sonal no ms in o ganic ood pu chase in en ions. The s udy o e s
new empi ical e idence in suppo o NAM, emphasizing he signi ican
impac o en i onmen al awa eness and knowledge abou o ganic ood
on consume pu chase in en ions h ough a i udes.
Fu he mo e, Salmi aa a e al. [53] p o ide no el insigh s in o un-
de s anding sus ainable ood choices by delinea ing he dis inc ion be-
ween desc ip i e and no ma i e social no ms. Thei esea ch e eals
he impo ance o desc ip i e no ms in ac ual and expec ed ood choices,
while SN ails o show he expec ed ele ance. These indings challenge
adi ional heo ies and o e new s a egies o in luencing consume
beha io h ough desc ip i e no ms. Acco ding o hese esea ches,
combining TPB and NAM o e s a mo e ho ough analy ical amewo k
and highligh s he ela i e signi icance o a ious mo i a ing ac o s in
ce ain cul u al and ma ke con ex s [54]. We can be e unde s and
consume psychological and beha io al mechanisms when aced wi h
Y. Xu e al. Sus ainable Fu u es 8 (2024) 100363
3
sus ainable ood choices due o his heo e ical ex ension and in-dep h
empi ical s udy. This will enable us o de elop mo e use ul ecom-
menda ions and ac ics o encou aging sus ainable consump ion.
Awa eness is ealizing o comp ehending a si ua ion o ac [55].
Residen accep ance and beha io al implemen a ion a e posi i ely
impac ed by hei gene al pe cep ion, sa is ac ion, and posi i e a i udes
[56]. Acco ding o Xia e al. [57], aising consume knowledge o low
ca bon emissions is ad an ageous o ca bon educ ion. Pe sonal
knowledge can be de ined as he ex en o which a consume is awa e o
a speci ic p oduc , se ice, o si ua ion [58]. Consume s’le el o pe -
sonal knowledge ega ding a p oduc can in luence hei a i udes. Fo
example, Li e al. [10] ound ha consume s wi h in-dep h knowledge o
low-ca bon p oduc s a e mo e likely o hold posi i e a i udes, which is
consis en wi h he “ATT”componen o he heo e ical planned
beha io (TPB) p oposed by Ajzen [36]. In he con ex o sus ainable
consump ion, Ding e al. [49] conduc ed a comp ehensi e e iew o he
ole o pe sonal knowledge in in luencing consume choices owa ds
en i onmen ally iendly p oduc s. S udies by Xu and Lin [59] and Qi
and Ploege [39] ha e demons a ed ha pe sonal knowledge is a
pi o al ac o in luencing consume pu chasing in en ions, pa icula ly
in g een and heal hy p oduc s. In his s udy, “pe sonal knowledge”de-
sc ibes he esponden s’knowledge and unde s anding o CAP. This
knowledge can po en ially in luence hei a i udes, decision-making
p ocesses, and, ul ima ely, hei pu chasing beha io . Acco ding o
esea ch on consume beha io and willingness o pay, isk pe cep ion
and isk p e e ence a e signi ican ac o s o ood accep abili y, which
has p o ound implica ions o consume beha io and hei willingness
o pay [58,60,61]. Bha i and U Rehman [62] examined he ela ion-
ship be ween di e en ac o s, including pe cei ed bene i s, pe cei ed
isks, and online shopping beha io , wi h he media ing ole o con-
sume pu chase in en ion. The indings show ha isk pe cep ion ha ms
online shopping beha io and ha hese isks mus be minimized o
inc ease hei sense o secu i y.
Based on he abo e heo ies, his pape in eg a es he o iginal TPB
and NAM models. I adds h ee ex e nal a iables (i.e., pe sonal
knowledge, low-ca bon awa eness, and isk pe cep ion) o analyze he
en i onmen ally iendly beha io o he public’s beha io al willingness
o pu chase CAP. The ollowing hypo heses ha e eme ged, and he
esea ch amewo k is depic ed in Fig. 1.
H11. Consume s’ATT is posi i ely impac ed by pe sonal knowledge
(PK) o CAP.
H12. Low-ca bon awa eness (LA) among consume s in luences hei
AC o buy CAP.
H13. The associa ion be ween WTP and PB o CAP is mode a ed by
isk pe cep ion (RP).
3. Ma e ial and me hods
3.1. Con ingen alua ion me hod
The e a e se e al app oaches o elici indi iduals’willingness o pay:
con ingen alua ion, choice expe imen , and expe imen al auc ion
[63–65]. The con ingen alua ion me hod (CVM) was ini ially
employed in 1958 o analyze non-ma ke p ices o ec ea ional se ices
in he Delawa e Ri e Basin egion o he Uni ed S a es [66]. CVM has
been u ilized in a ious applica ions [59,67]. Typically, CVM employs
ques ionnai es ha le esponden s explici ly exp ess hei p e e ences
o speci ic goods in mone a y e ms [68]. The mone a y alue epo ed
by esponden s is gene ally exp essed in e ms o willingness o pay, i.e.,
he maximum mone a y alue ha an indi idual would be p epa ed o
pay o a hypo he ical imp o emen p og am [68]. Al e na i ely, will-
ingness o pay can be he minimum amoun an indi idual will accep as
compensa ion o he change [68]. Also, CVM has mos ly been used o
measu e “p e e ences o goods o se ices o which a con en ional
ma ke does no exis ”[69]. The use o CVM is deemed app op ia e, and
hence, CVM has been chosen o his s udy.
3.2. Econome ic model
S uc u al Equa ion Modeling (SEM) is a s a is ical me hod o
analyzing complex ela ionships among a iables, combining ac o and
pa h analysis. O igina ing in he 1970s, SEM has become popula in
economics [70]. I consis s o wo pa s: he measu ed model, which
links obse ed and la en a iables, and he s uc u al model, which
connec s la en a iables. Obse ed a iables a e di ec ly measu able,
Fig. 1. Resea ch amewo k.
Y. Xu e al. Sus ainable Fu u es 8 (2024) 100363
4
while la en a iables canno be measu ed di ec ly and mus be
measu ed wi h he help o obse ed a iables. The equa ion o he
measu emen model is as ollows:
X=Λxξ+δ(1)
Y=Λy
η
+
ε
(2)
The equa ion o he s uc u al model is as ollows:
η
=B
η
+Γξ+ζ(3)
Table A1 displays he symbols and explana ions o he h ee equa-
ions abo e. Eigh la en a iables in luence consume s’pu chase
in en ion and pu chase beha io , and a s uc u al equa ion model is
es ablished o s udy his in luence.
3.3. Ques ionnai e design
Ques ionnai es a e commonly u ilized da a collec ion su ey ools
wi hin con ingen alua ion me hod s udies [71]. The me hodology
measu es he alue consume s place on non-ma ke p oduc s and en-
ables esea che s o de e mine a i udes and iews. Consequen ly, his
esea ch uses a ques ionnai e as a da a collec ion ins umen .
The ques ionnai e is spli in o ou blocks. The i s pa is an
in oduc ion o he su ey, which in oduces he esponden s o he
opic o he ques ionnai e, he esea ch o ganiza ion, and he goal and
ele ance o he s udy. S ep 1 asks esponden s o comple e indi idually
a se ies o ques ions ha cap u ed hei ini ial p e e ences and willing-
ness o pay wi hou en e ing any in o ma ion. In S ep 2, e ms in ol ed
in he ques ionnai e a e explained o imp o e he comp ehensibili y o
he ques ionnai e. Based on he heo e ical amewo k o TPB-NAM and
combined he cu en ma ke si ua ion o ag icul u al p oduc s wi h
pas esea ch indings ([72]; O eng-Pep ah e al. [73,74]), we ha e
designed a o al o 35 measu emen ques ions. All ques ions we e
measu ed on a i e-poin Like scale anging om “comple ely
disag ee” o “comple ely ag ee” ega ding low ca bon awa eness,
asc ip ion o esponsibili y, pe cei ed beha io al con ol, pe cep ion o
isk, and so on (Table 1).
CVM is he main opic o he hi d sec ion, which concen a es on
how o e alua e consume s’willingness o pay and beha io . A e
examining and con as ing se e al WTP boo s apping echniques, we
choose o employ he paymen ca d as a boo s apping mechanism o
p e en non- e lec i e bias in he ques ionnai e. We also c ea e a su ey
wi h wo ques ions o accoun o p o es answe s. I consis s o sample
selec ion ques ions and heu is ic ques ions as ollows. Rega ding “ca -
bon neu al”milk—cu en ly o e ed o sale in China’s ag icul u al
ma ke — esponden s a e in o med ha labeling he goods would
necessi a e measu ing g eenhouse gas emissions du ing p oduc ion and
ob aining ce i ica ion om a hi d-pa y o ganiza ion, which would
come a an ex a expense o he manu ac u e . Nex , we asked e-
sponden s i hey would pay mo e o CAP. I he answe is “yes”, he
nex leading ques ion is o ask how much ex a hey/he would like o pay
o each CAP. The poll indica es ha he e a e cu en ly ca bon-labeled
ag icul u al p oduc s on he ma ke , such as “ze o-ca bon” ege ables
and ca bon-neu al milk. We choose he ep esen a i e ype o ag icul-
u al p oduc s among many a ie ies and alk wi h he pa icipan s in
he in e iews, conside ing ac o s like he ypes o ag icul u al p od-
uc s, he a ailabili y o each ype in di e en egions, and he equency
o daily consump ion by consume s. Using he p ice da a al eady
collec ed, we in o m each esponden abou he coun yʼs a e age
ma ke p ice o se e al ag icul u al p oduc s. A he same ime, based on
he p ice, we gi e speci ic amoun s in pe cen age inc emen s ha he
esponden s chose hei la ges willingness o pay. I a esponden an-
swe s “no” o he sample selec ion ques ion, esponden s who indica e
hey a e unwilling o pay will be gi en a se ies o easons o explo e hei
in en ions. The a e age willingness o pay in he paymen ca d ques-
ionnai e can be de e mined o disc e e a iables using he
Table 1
De ini ion and desc ip ion o a iables.
Va iable Code I em Re e ences
Awa eness o
Consequences
AC01 Pu chasing high ca bon dioxide
emissions om ag icul u al
p oduc s will cause se ious
pollu ion and en i onmen al
damage.
[75]
AC02 Pu chasing ca bon-labeled
ag icul u al p oduc s can educe
en i onmen al pollu ion.
AC03 Pu chasing ca bon-labeled
ag icul u al p oduc s can ensu e
he quali y and sa e y o
ag icul u al p oduc s, which
bene i s us all.
Asc ip ion o
Responsibili y
AR01 As a consume , I should bea some
esponsibili y o educing ca bon
dioxide emissions.
[76]
AR02 I eel esponsible o
en i onmen al issues caused by
no pu chasing ca bon-labeled
ag icul u al p oduc s.
AR03 I would eel guil y i I didn’ buy
ag icul u al i ems wi h ca bon
labels, con ibu ing o inc eased
ca bon dioxide emissions.
AR04 I belie e ha e e y consume
bea s some esponsibili y o he
en i onmen al and social
p oblems caused by he
p oduc ion and consump ion o
ag icul u al p oduc s.
A i udes owa d
he Beha io
ATT01 I belie e ha pu chasing ca bon-
labeled ag icul u al p oduc s is
bene icial o he en i onmen .
[77]
ATT02 I belie e pu chasing ca bon-
labeled ag icul u al p oduc s is a
wise choice.
ATT03 Pu chasing ca bon-labeled
ag icul u al p oduc s will make
me eel physically and men ally
pleasan .
Low-ca bon
awa eness
LA01 When shopping, I keep plas ic
shopping bags and euse hem.
[78]
LA02 When p in ing, I ac i ely use bo h
sides o each shee o pape .
LA03 I o en pay a en ion o a icles o
epo s ela ed o en i onmen al
issues.
Pe cei ed
Beha io al
Con ol
PBC01 The p ice o ca bon-labeled
p oduce signi ican ly in luences
my decision o pu chase i .
[73]
PBC02 I hink ha consume s ind i
di icul o pu chase i ems wi h a
ca bon label because o he high
p ices.
PBC03 I am willing o buy ca bon-labeled
p oduc s when I ha e con idence
in hei en i onmen al bene i s.
Pu chase
In en ion
PI01 I am glad o pu chase ca bon-
labeled ag icul u al p oduc s.
[73,76]
PI02 I am likely o pu chase ca bon-
labeled ag icul u al p oduc s in
he u u e.
PI03 I plan o buy mo e ca bon-labeled
ag icul u al p oduc s in he
u u e.
PI04 I would ecommend ca bon-
labeled ag icul u al p oduc s o
my ela i es and iends.
Pe sonal
knowledge o
CAP
PK01 I unde s and he concep o
ca bon-labeled ag icul u al
p oduc s (such as “ze o-ca bon”
ege ables, ca bon-neu al milk,
e c.).
[79]
(con inued on nex page)
Y. Xu e al. Sus ainable Fu u es 8 (2024) 100363
5
ma hema ical expec a ion o mula.
E(WTP) = ∑
n
i=1
Pibi(4)
whe e E(WTP) is he maximum a e age alue o willingness o pay o
each CAP; P
i
is he p obabili y o esponden s selec ing each bid alue; b
i
is he bid amoun . The maximum a e age willingness o esponden s o
pay can be calcula ed by Eq. (4). In his pape , we combine he ele an
li e a u e [81] o measu e consume s’ac ual pu chase o CAP using wo
measu es, i.e., whe he hey pu chased CAP in he pas yea and he
exac equency o consump ion. The las pa is he socioeconomic
cha ac e is ics o he esponden s, including gende , age, cu en esi-
dence, educa ion le el, mon hly income, e c.
3.4. Da a collec ion
To co ec inaccu a e o eadily miscons ued con en , a pilo poll
was ca ied ou wi h 60 andomly chosen pa icipan s in Oc . 2022. The
o mal su ey was conduc ed om Ap il o June 2023 in Shanghai,
Nanjing, Wuhan, and Xi’an, China. The esea che s a e highly ained
g adua e and doc o al s uden s. Ta ge esponden s a e 18 yea s o age
o olde . A o al o 641 esponden s ag eed o pa icipa e in he esea ch.
A e excluding ques ionnai es wi h missing alues o ou lie s based on
he a iables equi ed o his esea ch, 580 alid samples we e eco -
e ed, wi h an e ec i e eco e y a e o 90.48 %. The demog aphic
cha ac e is ics o he sample da a a e as ollows. In e ms o gende ,
he e is an equal dis ibu ion o males and women (51.72 % and 48.28
%, espec i ely); in e ms o age, he p opo ion o hose aged 28 and
below eaches 70.86 %; and in e ms o educa ion, 90.69 % o he
samples ha e a high school diploma abo e. Table 2 p esen s a b ie
o e iew o demog aphic da a.
4. Resul s
4.1. Willingness o pay o ca bon-labeled ag icul u al p oduc s
This sec ion examines esponden s’WTP o i e dis inc CAP.
Desc ip i e s a is ics a e hen used o u he explo e dispa i ies in e-
sponden s’WTP and possible easons o consume s’ eluc ance o
pu chase CAP. Only 29.83 % o he 580 esponden s a e unwilling o pay
he CAP p emium, lea ing 70.17 % eage o do so. This sugges s ha
he e is al eady some basis o mo ing o wa d wi h he ca bon labeling
sys em in e ms o ini ial willingness o do so.
As shown in Table 3, he esul s demons a e ha emale esponden s
ha e sligh ly highe WTP on ca bon-labeled milk and co n han male
esponden s. In con as , male esponden s ha e a highe p emium WTP
on ca bon-labeled ui s, ege ables, and eggs han emale esponden s,
indica ing ha WTP does no appea o be in luenced by gende . Re-
sponden s unde 40 ha e he highes WTP o mos o CAP, and
Table 1 (con inued)
Va iable Code I em Re e ences
PK02 I unde s and he quali y
cha ac e is ics o ca bon-labeled
ag icul u al p oduc s (such as
“ze o-ca bon” ege ables, ca bon-
neu al milk, e c.).
PK03 I am amilia wi h he cos s
associa ed wi h ca bon-labeled
ag icul u al p oduc s (such as
“ze o-ca bon” ege ables, ca bon-
neu al milk, e c.).
Pe sonal No ms PN01 To sa egua d he en i onmen , I
should limi he numbe o
ag icul u al i ems I buy ha
elease much ca bon dioxide.
[75]
PN02 I eel mo ally obliga ed o
con ibu e o educing ca bon
dioxide emissions by pu chasing
ca bon-labeled ag icul u al
p oduc s.
PN03 I conside i a mo al du y o
socie y o pu chase ca bon-
labeled ag icul u al p oduc s o
educe ca bon dioxide emissions.
PN04 E e yone is esponsible o
conside ing he en i onmen al
impac when pu chasing
ag icul u al p oduc s.
Risk Pe cep ion PR01 I wo y ha I migh be was ing
money when buying ca bon-
labeled ag icul u al p oduc s.
Ma inho
e al., 2022;
[80]
PR02 I am conce ned ha ca bon-
labeled ag icul u al p oduc s may
no p o ide he necessa y
bene i s.
PR03 I wo y ha ca bon-labeled
ag icul u al p oduc s may no
pe o m as well as ad e ised.
PR04 Conside ing a ious ac o s, I
hink he e a e isks associa ed
wi h pu chasing ca bon-labeled
ag icul u al p oduc s.
Subjec i e No ms SN01 I p e e ca bon-labeled
ag icul u al p oduc s because my
iends and amily app o e
pu chasing hem.
[77]
SN02 My iends and amily hope I will
buy ag icul u al i ems wi h a
ca bon label, so I would like o do
so.
SN03 I p e e o buy ca bon-labeled
ag icul u al p oduc s because he
go e nmen encou ages me.
SN04 I p e e o buy ca bon-labeled
ag icul u al p oduc s because o
he en i onmen ally iendly
examples.
Table 2
Desc ip i e s a is ics o socio-economic cha ac e is ics o esponden s.
Cha ac e is ic F equency Pe cen (%)
Gende
Female 280 48.28
Male 300 51.72
Age
[18,28] 411 70.86
[29,39] 112 19.31
40 and abo e 57 9.83
Occupa ion
Voca ional o blue-colla wo ke s 30 5.17
Ci il se an 50 8.62
Company s a 172 29.65
S uden 264 45.52
F eelance 35 6.04
No wo king/Re i ed 29 5
Region
Eas 301 51.90
Middle 243 41.90
Wes 36 6.20
Mon hly Income (CNY)
5000 o below 323 55.69
5001–10,000 107 18.45
10,001–20,000 101 17.41
20,000 abo e 49 8.45
Educa ion
High School and below 54 9.31
Unde g adua e 372 64.14
Mas e o abo e 154 26.55
Household Size
Two people o below 55 9.48
3–4 people 404 69.66
O e i e people 121 20.86
Y. Xu e al. Sus ainable Fu u es 8 (2024) 100363
6
esponden s o e 40 ha e he highes WTP ega ding he p emium o
exclusi ely ca bon-labeled eggs. This migh be because di e en age
g oups ha e di e en opinions on he same issue. Eas e n Chinese e-
sponden s ha e he g ea es p emium WTP, ollowed by cen al and
wes e n Chinese esponden s. This may be because he eas e n pa o
he coun y is economically de eloped, and i s esiden s ha e ela i ely
s onge inancial esou ces and highe incomes. The household loca ion
is an impo an geog aphical ac o o consume s, and i is ela ed o he
o ma ion o hei pe cep ions, belie s, e c., which in u n a e ela ed o
ood consump ion beha io .
Meanwhile, esponden s wi h mon hly incomes o ¥20,000 o mo e
ha e he highes p emium WTP o 80 % o CAP men ioned in he su -
ey. I is no di icul o unde s and ha p ice is one o he main ac o s
in luencing consume s’decisions o buy p oduc s. Acco ding o he
s udy, he amily dimension appea s o impac consume s’WTP, wi h
esponden s wi h smalle amily sizes epo ing highe WTP. A one-way
analysis o a iance (ANOVA) was pe o med o explo e whe he socio-
s a is ical cha ac e is ics ha e a ole in esiden s’WTP o CAP. Six in-
dependen a iables, such as gende , age, and educa ion, and one
dependen a iable, willingness o pay, we e examined. The analysis’s
indings e eal ha h ee a iables— esponden s’place o esidence,
deg ee o educa ion, and occupa ion—subs an ially impac hei WTP
o CAP. In con as , he emaining a iables ailed o pass he es . This
implies ha an indi idual’s inclina ion o buy a CAP is somewha
in luenced by hei place o esidence’s ai s, le el o educa ion, and
line o wo k.
Among he esponden s who a e no willing o pay a p emium o
CAP, he mos common answe (almos 20 %) is ha hey canno pay o
pe sonal inancial easons bu would be willing o pay i hei income
inc eased. This is ollowed by a eluc ance o spend mo e money (16 %)
and a belie ha he go e nmen should shoulde mo e esponsibili y o
educing emissions han he esponden s should (11 %). Mo e han 9 %
o pa icipan s eel ha ag icul u al p oduc s had no signi ican en i-
onmen al impac a any poin along he supply chain, om p oduc ion
o consump ion, and ano he 8 % s a es ha he eason o hei e usal
is ha hey a e unwilling o y new hings. Some pa icipan s o e
“o he s”jus i ica ions o declining o pay a p emium o CAP, wi h a
signi ican po ion ci ing hei belie ha ag icul u al p oduc s a e
necessa y o ou daily li es, ha hey shouldn’ be excessi ely p iced,
and ha we should look o ways o cu cos s. Cu en ly, he e a en’
enough op ions on he ma ke , and he ca bon labeling ma ke should be
expanded. Some esponden s also no e ha hei decisions would be
impac ed i hey didn’ know he au hen ici y and sa e y o CAP a ail-
able on he ma ke .
4.2. Measu emen model es ing
As seen h ough he esul s in Table 4, all he ac o loadings ha e
alues in he ange o 0.631–0.913, which mee s he c i e ia p oposed by
Hai e al. [82]. A e analyzing he ac o loadings, he eliabili y and
alidi y o he ques ionnai e we e analyzed. C onbach’s alpha was used
o con i m he eliabili y o he ac o s, and 0.7 was chosen as he
s anda dized c i ical alue, as indica ed by Nunnally and Be ns ein [83].
The o e all C onbach’s alpha o all la en a iables was 0.960, and he
es ima es o he ac o s’C onbachʼs alpha anged om 0.722 o 0.928,
all o which we e abo e 0.7. This sugges s ha he esponden s’in e nal
consis ency in assessing he obse ed a iables eached an accep able
le el, as did he eliabili y o he ac o s. Addi ionally, he s udy con-
duc ed KMO and Ba le ’s es s on he ac o s (Table 5). KMO es ima-
ions g ea e han 0.7 implied good adequacy o he sample collec ed
[84]. The a e age ex ac ed analysis o a iance (AVE) anged om
0.479 o 0.766, wi h nea ly all o hem being g ea e han 0.5, and he
composi e eliabili y (CR) o all la en a iables anged om 0.733 o
0.929, all o which we e g ea e han 0.7. The s udy demons a es good
alidi y and eliabili y in examining he p oposed hypo heses as all he
indica o s me he c i e ia [82].
4.3. S uc u al model assessmen
Se e al key me ics o de e mining model i a e he NC alue
(CMIN/DF), oo mean squa e o he app oxima ion e o (RMSMA),
goodness-o - i index (GFI), compa a i e i index (CFI) and no ma i e
i index (NFI). In gene al, a easonable i can be de ined as ha ing NC
alues be ween 1 and 5, RMSEA <0.08, and GFI, CFI, and NFI g ea e
han 0.8 [85–87]. Table 6 shows ha , ollowing a ew model modi i-
ca ions, all i ness indica o s pe o m well, demons a ing ha he
model o he pu chasing beha io o consume s o CAP and he collec ed
da a i well.
The s udy’s hypo heses a e pu o he es based on he s uc u al
modeling. Fig. 2 displays he speci ic isualiza ion indings. The s udy
Table 3
Willingness o pay o ca bon-labeled ag icul u al p oduc s.
Va iables Ca ego y WTP-Milk (%) WTP-Co n (%) WTP-Banana (%) WTP-Toma o (%) WTP-Egg (%)
To al 27.50 29.73 26.86 26.51 24.26
Gende Male 27.06 29.59 27.15 27.33 24.91
Female 27.96 29.88 26.55 25.63 23.56
Age [18,28] 27.83 29.59 26.74 26.22 24.01
[29,39] 27.22 31.36 27.55 27.60 24.55
40 and abo e 25.70 27.65 26.63 26.57 25.62
Occupa ion Voca ional o blue-colla wo ke s 31.45 30.46 30.19 31.07 29.12
Ci il se an 24.78 27.04 39.52 26.80 23.92
Company s a 27.83 25.70 27.03 26.84 24.29
S uden 26.18 27.98 25.65 24.82 22.67
F eelance 32.46 35.59 32.70 32.68 31.43
No wo king/Re i ed 32.09 33.38 29.53 27.21 25.50
Region Eas 28.85 31.53 27.74 27.24 25.01
Middle 26.38 28.17 26.39 26.24 23.76
Wes 23.71 25.20 22.71 22.16 21.34
Mon hly Income (CNY) 5000 o below 26.57 28.02 26.00 25.13 23.21
5001–10,000 29.66 32.44 27.86 27.61 25.25
10,001–20,000 27.66 29.99 27.59 28.49 25.56
20,000 abo e 28.55 34.57 28.86 29.09 26.38
Educa ion High School and below 28.70 31.69 26.73 28.57 28.45
Unde g adua e 27.74 29.80 27.01 26.45 23.94
Mas e o abo e 26.48 28.87 26.57 25.92 23.56
Household Size Two people o below 30.48 33.18 28.61 28.44 26.06
3–4 people 27.28 29.55 27.05 26.50 24.50
O e i e people 26.85 28.76 25.46 21.35 22.64
Y. Xu e al. Sus ainable Fu u es 8 (2024) 100363
7
esul s indica e ha ATT signi ican ly and posi i ely in luences PI
(β=0.219, P<0.001) wi hin he heo e ical amewo k o TPB, hus
suppo ing H1. SN posi i ely in luences PI (β=0.404, P<0.001) and H2
is suppo ed. H3 is con i med, wi h a signi ican posi i e e ec o PBC on
PI (β=0.408, P<0.001). Among hem, he s anda dized pa h coe icien
o PBC is signi ican ly highe han ha o ATT and SN, indica ing ha
he main ac o s in luencing consume s’WTP abou CAP a e hei con-
ol o e ele an esou ces like money, ime, knowledge, and skills, as
well as hei access o in o ma ion abou CAP. PBC has a signi ican
posi i e e ec on PB (β=0.261, P=0.002), and H4 is alid. This sugges s
ha PBC can also di ec ly in luence he occu ence o a consume ’s
beha io when he o she can pe cei e he objec i e cons ain s, such as
esou ces and capabili ies, ha he o she needs o pe o m a ce ain
beha io . The signi icance es e uses H5 wi h he esul ha PI ha ms
PB (β=− 0.307, P<0.001). In he amewo k o NAM, consume s’AC o
pu chasing CAP posi i ely a ec s hei AR (β=0.683, P<0.001) and PN
(β=0.435, P<0.001), suppo ing H6 and H7. Consume s’AR posi i ely
a ec s PN (β=0.389, P<0.001), implying ha a consume ’s sense o
esponsibili y o he ad e se consequences o ailing o pu chase CAP
ac i a es his o he sense o mo al obliga ion o pu chase hem, and H8
is alida ed. The posi i e e ec o consume s’PN on PI passes he sig-
ni icance es (β=0.113, p=0.002); hus, H9 holds, meaning ha people
wi h g ea e e hical obliga ions ha e a highe CAP pu chase in en ion.
In he ex ended TPB-NAM heo e ical amewo k, he hypo hesized e-
la ionships o SN posi i ely a ec ing PN (β=0.164, P<0.001), con-
sume s’pe cep ions o CAP a ec ing hei ATT (β=0.6, P<0.001), and
consume s’low-ca bon awa eness a ec ing hei AC (β=0.903, P<
0.001) all pass he signi icance es , and H10, H11, and H12 a e
con i med, indica ing ha p essu e om socie y posi i ely a ec s hei
sense o mo al obliga ion owa ds he ac o pu chasing CAP and ha he
le el o consume s’knowledge o he backg ound and exis en ial sig-
ni icance o he p oposed CAP signi ican ly and posi i ely a ec s hei
ATT. A he same ime, he mo e en i onmen ally conscious cus ome s
hemsel es a e, he mo e hey a e awa e o he ha m ha no buying
CAP can do o he en i onmen .
4.3.1. Mode a ing e ec s
The SEM esul s indica e ha pu chase in en ion nega i ely in-
luences pu chase beha io . To unde s and he mechanism o RPʼs in-
luence in he p ocess o PI’s e ec on consume s’CAP pu chase,
s uc u al equa ion modeling wi h la en a iable in e ac ion e ms was
cons uc ed using AMOS 26.0, and he mode a ing e ec o he
in en ion-beha io gap was analyzed based on he heo y o likelihood.
Table 4
Validi y and eliabili y o he s udy.
Va iable Coding Fac o loading C onbach’s alpha Composi e eliabili y A e age a iance ex ac ed (AVE)
Awa eness o Consequences AC01 0.631 0.796 0.804 0.581
AC02 0.795
AC03 0.845
Asc ip ion o Responsibili y AR01 0.649 0.851 0.857 0.602
AR02 0.831
AR03 0.826
AR04 0.784
A i udes owa d he Beha io ATT01 0.855 0.890 0.892 0.735
ATT02 0.889
ATT03 0.826
Low-ca bon awa eness LA01 0.660 0.730 0.733 0.479
LA02 0.735
LA03 0.678
Pe cei ed Beha io al Con ol PBC01 0.668 0.722 0.762 0.519
PBC02 0.66
PBC03 0.822
Pu chase In en ion PI01 0.859 0.928 0.929 0.766
PI02 0.87
PI03 0.891
PI04 0.88
Pe sonal Knowledge o CAP PK01 0.795 0.874 0.8772 0.705
PK02 0.913
PK03 0.806
Pe sonal No ms PN01 0.822 0.886 0.889 0.666
PN02 0.86
PN03 0.805
PN04 0.776
Risk Pe cep ion PR01 0.686 0.877 0.879 0.647
PR02 0.85
PR03 0.855
PR04 0.816
Subjec i e No ms SN01 0.907 0.905 0.909 0.715
SN02 0.912
SN03 0.82
SN04 0.73
Table 5
KMO and Ba le ’s es esul s o ac o analysis.
KMO and Ba le ’s Tes
Kaise –Meye –Olkin Measu e o Sampling Adequacy. 0.958
Ba le ’s Tes o Sphe ici y App ox. Chi-Squa e 16,148.911
Deg ee o eedom 630
Signi icance 0.000
Table 6
Resul s o model i ness es .
Measu e Th eshold Es ima e In e p e a ion
CMIN/DF (NC) 1<NC <5 4.957 Accep able
RMSEA <0.1 0.083 Accep able
GFI >0.8 0.810 Accep able
CFI >0.8 0.878 Accep able
NFI >0.8 0.852 Accep able
PCFI >0.5 0.765 Accep able
PNFI >0.5 0.742 Accep able
Y. Xu e al. Sus ainable Fu u es 8 (2024) 100363
8
