Measuring behavioural intention through the use of greenwashing: a study of the mediating effects and variables involved

sus ainabili y A icle Measu ing Beha iou al In en ion h ough he Use o G eenwashing: A S udy o he Media ing E ec s and Va iables In ol ed Eloy Gil-Co de o * , Juan Ped o Cab e a-Sánchez , Ignacio Cepeda-Ca ión and Jaime O ega-Gu ie ez   Ci a ion: Gil-Co de o, E.; Cab e a-Sánchez, J.P.; Cepeda-Ca ión, I.; O ega-Gu ie ez, J. Measu ing Beha iou al In en ion h ough he Use o G eenwashing: A S udy o he Media ing E ec s and Va iables In ol ed. Sus ainabili y 2021,13, 6720. h ps://doi.o g/ 10.3390/su13126720 Academic Edi o s: Gianluigi Guido, C is ian Rizzo and Luigi Pipe Recei ed: 30 Ap il 2021 Accep ed: 10 June 2021 Published: 14 June 2021 Publishe ’s No e: MDPI s ays neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a il- ia ions. Copy igh : © 2021 by he au ho s. Licensee MDPI, Basel, Swi ze land. This a icle is an open access a icle dis ibu ed unde he e ms and condi ions o he C ea i e Commons A ibu ion (CC BY) license (h ps:// c ea i ecommons.o g/licenses/by/ 4.0/). Depa men o Business Adminis a ion and Ma ke ing, Uni e si y o Se ille, 41018 Se ille, Spain; [email p o ec ed] (J.P.C.-S.); [email p o ec ed] (I.C.-C.); [email p o ec ed] (J.O.-G.) *Co espondence: egco [email p o ec ed] Abs ac : The wo k aims o achie e a be e unde s anding o i ms’ g een s a egy, and speci ically, in he alse g een s a egy called g eenwashing, and he ela ionships be ween g eenwashing (GW) and beha iou in en ion (BI), and how his ela ionship is a ec ed by wo d o mou h (WOM) and co po a i e social esponsabili y (CRS). A su ey was conduc ed and 198 alid and comple e online ques ionnai es we e collec ed om use s o u ban mobili y apps (Blablaca and Amo ens) in Spain. The s uc u al equa ion modeling echnique pa ial leas squa es (PLS-SEM) was used o es he p oposed esea ch model and hypo hesized ela ionships. The esul s o ou s udy indica e ha he di ec ela ionship be ween GW and BI is no suppo ed, al hough he indi ec ela ionship h ough WOM and CRS is signi ican , so ha bo h become media ing a iables o he GW and BI ela ionship. The pape also analyzes he di ec ela ionships be ween GW, CRS, WOM and BI, so ha he di ec e ec s GW and CRS; CRS and WOM; and WOM and BI a e signi ican . This empi ical s udy analyzes he e ec o GW, which has no been s udied much, especially in empi ical esea ch. The s udy analyzes se e al a iable consequences o GW and analyzes media ing e ec s o CRS and WOM on he GW and BI ela ionship. The s udy also includes wo beha io al indica o s, WOM and BI, in a esea ch model, and, addi ionally, he s udy demons a es he ela ionship be ween GW and pe cei ed CRS. Keywo ds: g eenwashing; wo d o mou h; co po a i e social esponsibili y; beha iou in en ion; pa ial leas squa e 1. In oduc ion In a e iew o he li e a u e since he 1990s, we ound he many businesses [ 1 – 12 ] and esea che s ha bene i ed om g een ma ke ing con inue o ely on hese bene i s nea ly 30 yea s la e [ 13 – 16 ]. We de ine g een ma ke ing as he se o ac i i ies designed o gene a e and acili a e any exchange aimed a sa is ying human needs o desi es, in such a way ha hese needs and desi es a e me , wi h a minimum ha m ul impac on he na u al en i onmen . As Ma ikyan e al. (2019) [ 17 ] ell us, he economic p o i abili y o sma homes is o en coupled wi h en i onmen al gains. Howe e , he e is a end in he li e a u e on he opposi e side o g een ma ke ing since, as a consequence o he concep , consume s a e disco e ing he oppo unis ic ad- an ages ha companies ha e owa ds hese en i onmen al ad an ages [ 18 ]. In his sense, companies ha p omo e hei p oduc s using he suppo o he en i onmen ha e appa - en ly become “g eene ” in ecen yea s [ 19 ]. As a esul o all o he abo e, he concep o “g eenwashing” appea s, which was i s used by Jay Wes e eld in he 1980s, wi h mos esea che s in he ield ag eeing wi h wha was said ea lie , ha companies a e becoming “g eene ” [ 20 – 24 ]. One o he concep s abou g eenwashing (GW) mos accep ed by he esea ch communi y was aken up by esea che s Pa guel e al. [ 25 ], de ining i as “ he p ac ices ca ied ou by companies o appeal he excessi e unc ionali y o hei p oduc on Sus ainabili y 2021,13, 6720. h ps://doi.o g/10.3390/su13126720 h ps://www.mdpi.com/jou nal/sus ainabili y Sus ainabili y 2021,13, 6720 2 o 12 he en i onmen and which canno be sus ained”. Ano he o he mos widely accep ed concep s abou g eenwashing is he one p o ided by he au ho s Delmas and Bu bano [ 26 ], s a ing ha g eenwashing is an in e sec ion o wo i m beha io s: Poo en i onmen al pe o mance and posi i e communica ion abou en i onmen al pe o mance. Wi h he ise o he concep and i s implica ion in socie y, nume ous in es iga ions ha e eme ged on he accumula ed incon eniences o he use o g eenwashing, some o hem e e ing o he nega i e implica ion ha a ec s beha iou in en ion [ 27 ] and co po a e social esponsibili y [28]. The main objec i e o his esea ch is o de elop a model o measu e he impac o g eenwashing on CSR and WOM and hese in u n on beha iou in en ion. I is also ou aim o b ing o he scien i ic communi y he oppo unis ic aspec ha many companies ha e adop ed conce ning he use o a non- eal g een s a egy, in o de o bene i hemsel es and impac hei bo om line. Speci ically, we wan o highligh hose companies ha use “g een” as a business s a egy, a he han as a company philosophy, which is eally wha he ma ke demands. The s uc u e o his a icle begins i s wi h an in oduc ion o he opic o g een ma ke ing and g eenwashing, de ining bo h concep s and ou lining when hey i s ap- pea ed in he li e a u e. In he second pa , we de elop he heo e ical pa ocusing on he di e en a iables ha we include in he model, as well as he de elopmen and app oach o he se o hypo heses. Finally, we p esen he esul ing model o ou esea ch g aphically, he esul s and he conclusions. 2. Theo e ical F amewo k 2.1. Pe cei ed CSR Co po a e Social Responsibili y (CSR) is a se o olun a y ac ions ha bene i he company and socie y as a whole [ 29 ]. Inc eased co po a e social esponsibili y is based on a conce n wi h esponding app op ia ely o socie y, ha being pa o he c i e ia ha s akeholde s ha e o judge business beha io . Acco ding o K aus e al. [ 30 ] he aspec o social and en i onmen al sus ainabili y is a po en ial sou ce o ideas o imp o e he en i onmen . Mo eo e , he lack o compliance wi h CSR can become a h ea o he epu a ion and p ospe i y o companies [31]. Consequen ly, he heo ies ha suppo pe cei ed CSR in he li e a u e ecognize ha companies mus ake in o accoun he in e es s o di e en s akeholde s, includ- ing cus ome s, pa ne s, supplie s, employees and communi ies, in business decision- making [ 32 , 33 ]. In his sense, he e is a g ea ea among di ec o s ha hey will be accused o g eenwashing [ 34 ] and ha , he e o e, he company’s pe cei ed CSR may be unde s ood as such [ 28 ]. This is de e mined when people pe cei e ha a company ha is p oud o i s social wo k can hide an oppo unis ic in en ion [ 35 ]. The e o e, i is necessa y o know wha in luence g eenwashing can ha e on co po a e social esponsibili y. In addi ion o he abo e, and once he ela ionship be ween g eenwashing and pe - cei ed co po a e social esponsibili y is known, i is necessa y o ake in o accoun ha pe cei ed CSR is a de e mining alue wi hin he company ha in u n no only mo i a es cu en cus ome s o adop a a o able a i ude, bu also po en ial cus ome s. Po en ial cus ome s lea n o co po a ions om hei ad e isemen s and o he sou ces, o exam- ple, wo d o mou h om exis ing cus ome s [ 36 ]. O he au ho s such as Chom ilailuk and Bu che es ablish ha pe cei ed CSR is an in luen ial ac o in he wo d o mou h o cus ome s [ 37 ], bu i is necessa y o know whe he he sample o his las s udy is co obo a ed in o he coun ies; he e o e, and in addi ion o knowing he ela ionship be ween g eenwashing and pe cei ed CSR, i is also necessa y o know he in luence o he la e on WOM. Pe i e al. (2020) [ 38 ] show in hei esul s ha emedies aimed a inc easing he awa eness o pu chasing le el s a can be bene icial, while ini ia i es ha in ol e expe s in he ield may be signi ican ly less use ul. Finally, in he li e a u e we ind ha esea che s such as Bigne-Alcañiz and Cu ás- Pé ez in 2008 sugges ha pe cei ed CSR is an in luen ial ac o in beha iou in en ion Sus ainabili y 2021,13, 6720 3 o 12 (BI) [ 39 ], bu gi en ha pe cei ed CSR is a cons uc ha is cons an ly e ol ing and depends on he cu en si ua ion o he business ecosys em [ 40 ] i is necessa y o e i y whe he his ela ionship is cu en ly con i med. Fo all he abo e easons, we p opose he ollowing hypo heses: Hypo hesis 1. G eenwashing has a nega i e in luence on pe cei ed CSR. Hypo hesis 2. Pe cei ed CSR has a posi i e in luence on BI. Hypo hesis 3. Pe cei ed CSR has a posi i e in luence on WOM. 2.2. WOM WOM has adi ionally been concep ualized as he exchange o in e pe sonal in o ma- ion be ween indi iduals who know each o he [ 41 ]. In his ega d, i has been es ablished om he ou se as he exchange o ma ke ing in o ma ion be ween consume s in such a way ha i plays a undamen al ole in shaping hei beha io and changing hei a i udes owa ds p oduc s and se ices [42]. This concep has e ol ed along wi h he g owing conce n abou global wa ming, so consume s a e mo e awa e o en i onmen al conside a ions [ 43 ]. Some companies a e using his scena io o selec i ely e eal posi i e in o ma ion abou he en i onmen al cha ac e is ics o hei p oduc s wi hou e ealing nega i e in o ma ion in o de o gene a e a g een image [ 44 ], he eby gene a ing nega i e communica ions om consume s esul ing in wha is conside ed “g een wom”, de ined by Chen e al. as he ex en o which a cus ome would in luence iends, ela i es and colleagues on he posi i e o nega i e en i onmen al messages o a p oduc o b and [ 45 ]. Thus, i should be no ed ha esea che s claim ha consume - o-consume communica ion, such as wo d o mou h, igge s a s ong and c edible in luence on consume beha io [ 46 , 47 ], and ha i we use en i onmen al bene i s in a nega i e way, we may incu wha is known as g eenwashing. The e o e, hey can cause damage o he pe cei ed quali y o a b and. This s a emen has been e i ied by Chen, exposing he nega i e ela ionship be ween g eenwashing and WOM [ 45 ], bu has no conside ed whe he his pa e n is e i ied wi h BI. On he o he hand, o he ecen s udies such as ha o Zhang e al. ha e e i ied whe he consume pe cep ions o g eenwashing ha e a nega i e e ec on BI o g een p oduc s in addi ion o s udying in he same esea ch he nega i e indi ec e ec s h ough WOM [ 48 ]. Howe e , his las s udy is based on a sample loca ed in China, so i is necessa y o know i hese esul s a e ul illed in o he coun ies. Fo all he abo e easons, we p opose he ollowing hypo heses: Hypo hesis 4. G eenwashing has a nega i e in luence on WOM. Hypo hesis 5. WOM has a posi i e in luence on he BI. 2.3. Beha iou In en ion Beha iou in en ion (BI) is de ined in he li e a u e as hose signals ha show whe he a cus ome con inues o use a company’s se ice o changes i s supplie [ 49 ]. This concep has been widely s udied by se e al au ho s o de e mine he backg ound a ec ing consume s when choosing a p oduc o se ice [50–52]. I is now widely accep ed ha consume s’ beha iou in en ion is in luenced by emo ions du ing he s ages be o e, du ing and a e consump ion o he p oduc o se - ice [ 53 , 54 ], which, oge he wi h he in o ma ion he consume has abou he company’s en i onmen ally sus ainable beha io , de e mine an impo an ole in he u u e BI [ 55 ]. Bal a-Ozkan e al. (2014) [ 56 ] ake his wo k as an example in a pape whe e hey compa ed u al and u ban a eas in se e al coun ies whe e ci izens show di e en a i udes owa ds en i onmen al bene i . The esea ch showed ha en i onmen al sus ainabili y has become a mo e signi ican elemen o u ban dwelle s. This is explained by he g ea e ole o Sus ainabili y 2021,13, 6720 4 o 12 economic bene i o u ban dwelle s, which a ou weighs en i onmen al conce ns. In his line, p e ious s udies ha e shown he posi i e in luence ha he BI has wi h espec o p oduc s ha a e esponsible wi h he en i onmen [ 57 ] and, a he same ime, hey wa n ha no all companies make good use o his in o ma ion, leading o g eenwashing [ 26 ]. In his sense, some p e ious s udies es ablish ha he ela ionship be ween g eenwashing and BI is media ed by us and WOM [ 58 ], bu hey do no inco po a e in his media ion he pe cei ed CSR despi e he ac ha some esea che s sugges ha he pe cei ed CSR is an in luen ial ac o in he BI [ 39 ]. In he p io li e a u e, i has been in es iga ed which a e possible an eceden s o beha io al in en ion, and, speci ically, he s udies o Song e al. (2008) [ 59 ] show how communica ion be ween consume s (WOM) posi i ely impac s on beha io al in en ion, and on he o he hand whe he , i cus ome s ha e a posi i e pe cep- ion o he company as socially esponsible, hey end o adop a o able beha io owa ds ha o ganiza ion [ 60 ]. In addi ion, i would be impo an o ex end he sample o his esea ch o check i i is e i ied in o he coun ies. Fo all he abo e easons, we p opose he ollowing hypo heses: Hypo hesis 6. G eenwashing has a nega i e in luence on BI. These 6 hypo heses a e di ec hypo heses and because o ha a e d awn in con inuous lines in he p oposed model shown in Figu e 1. Sus ainabili y 2021, 13, x FOR PEER REVIEW 4 o 13 ole o economic bene i o u ban dwelle s, which a ou weighs en i onmen al conce ns. In his line, p e ious s udies ha e shown he posi i e in luence ha he BI has wi h espec o p oduc s ha a e esponsible wi h he en i onmen [57] and, a he same ime, hey wa n ha no all companies make good use o his in o ma ion, leading o g eenwashing [26]. In his sense, some p e ious s udies es ablish ha he ela ionship be ween g een- washing and BI is media ed by us and WOM [58], bu hey do no inco po a e in his media ion he pe cei ed CSR despi e he ac ha some esea che s sugges ha he pe - cei ed CSR is an in luen ial ac o in he BI [39]. In he p io li e a u e, i has been in es i- ga ed which a e possible an eceden s o beha io al in en ion, and, speci ically, he s udies o Song e al. (2008) [59] show how communica ion be ween consume s (WOM) posi i ely impac s on beha io al in en ion, and on he o he hand whe he , i cus ome s ha e a pos- i i e pe cep ion o he company as socially esponsible, hey end o adop a o able be- ha io owa ds ha o ganiza ion [60]. In addi ion, i would be impo an o ex end he sample o his esea ch o check i i is e i ied in o he coun ies. Fo all he abo e easons, we p opose he ollowing hypo heses: Hypo hesis 6. G eenwashing has a nega i e in luence on BI. These 6 hypo heses a e di ec hypo heses and because o ha a e d awn in con inu- ous lines in he p oposed model shown in Figu e 1. Figu e 1. Resea ch p oposed model. Addi ionally we p opose wo mo e hypo heses bu in his case; hey a e indi ec e - ec s. Tha is why hey a e d awn in do ed and dashed lines. These hypo heses a e: Hypo hesis 7. The pa h om G eenwashing o BI is media ed by he nex sequence o media o s: CRS and WOM. Hypo hesis 8. The pa h om G eenwashing o BI is media ed by WOM. Figu e 1. Resea ch p oposed model. Addi ionally we p opose wo mo e hypo heses bu in his case; hey a e indi ec e ec s. Tha is why hey a e d awn in do ed and dashed lines. These hypo heses a e: Hypo hesis 7. The pa h om G eenwashing o BI is media ed by he nex sequence o media o s: CRS and WOM. Hypo hesis 8. The pa h om G eenwashing o BI is media ed by WOM. 3. Me hodology 3.1. Da a Collec ion and Sampling The popula ion o his s udy was he cus ome s o use s o he Blablaca and Amo ens u ban mobili y mobile apps. They a e all 18 yea s old o olde cus ome s. The ques ion- nai es we e sen by email wi h a link o access he ques ions. Ques ionnai es we e sen o 345 use s o hese apps, and a o al o 198 alid ques ionnai es we e collec ed be ween Feb ua y and Ma ch 2020. Sus ainabili y 2021,13, 6720 5 o 12 Responses om a o al o 105 women (53%) and 93 men (47%) we e collec ed. Fo each i em, pa icipan s indica ed how hey ag eed o disag eed, and all i ems we e a ed on a se en-poin Like scale anging om 1 (s ongly disag ee) o 7 (s ongly ag ee). 3.2. Measu e In his s udy we used exis ing scales aken om p e ious li e a u e s udies. The ques ionnai e includes: (a) G een Washing (GW). The elemen s we e measu ed using a se en-poin Like scale om he s udy o Lau e [61]. This scale is a one-dimensional scale wi h i e i ems. (b) Pe cei ed Co po a e Social Responsibili y (CSR). This cons uc used he one-dimensional scale o i e i ems om he s udy by Ramasamy e al. [33]. (c) Wo d o mou h (WOM). This s udy used he one-dimensional adap ed scale o six i ems o he s udy o Goye e e al. [62]. (d) Beha io al in en ion (BI). This s udy used he one-dimensional scale o h ee i ems om he s udy by Xu e al. [63]. 3.3. Da a Analysis Fo he da a analysis, his s udy used he echnique o s uc u al equa ion modelling (SEM) based on a iance, Pa ial Leas Squa e (PLS-SEM). In ecen imes, SEM echniques ha e been g owing and e ol ing wi h echnological de elopmen . The SEM echnique belongs o he amily o mul i a ia e analysis, whe e mul iple a iables a e analyzed ha in ou esea ch ield a e ela ed o companies, man- age s, clien s, e c. This SEM echnique is also conside ed a second gene a ion o da a analysis echniques and has been used by social esea che s o he las 30 yea s. The e a e wo ypologies o SEM echniques, hose based on co a iance (CB-SEM) and hose based on a iance (pa ial leas squa e—PLS-SEM). The i s uses common ac o models o ep esen la en a iables, and he second uses compound ac o models [64]. In his ega d, Rigdon (2016) iden i ies se e al easons ha should no be used as a p e ex o using PLS-SEM: The small sample size; an unusual da a dis ibu ion; he exis ence o o ma i e indica o s; and he s a emen o explo a o y esea ch. Mo e ecen ly, Ali e al. [ 65 ], Rigdon [ 64 ], and Rigdon e al. [ 66 ] emphasized ha esea che s should use ha echnique o analyze he da a ha is consis en wi h he ype o model hey wan o y o es ima e, and he e o e hey wan o say ha i is e y impo an ha esea che s indica e he ype o model o es ima e, and hus de e mine he ype o echnique o ea he da a mos app op ia ely. Mo e gene ally, he PLS-SEM esea che s ag ee ha in hose models ha only use common ac o s, echniques based on co a iance (CB-SEM) would be mainly used. While hose esea che s using composi es would use me hods based on PLS-SEM ype compos- i es [ 67 ], and in he case o mixed models, whe e bo h ypes a e used, he op ion would be o use he consis en PLS e sion (PLSc) [ 68 ]. Based on hese conside a ions, i is easonable o assume ha he i s eason o using PLS-SEM is acco ding o he ype o la en a iable in he model ha he esea che s a e ying o es ima e. In his s udy, all measu emen s we e ope a ionalized as composi es [ 66 , 69 ], he e o e, we chose o use a s uc u al equa ion modeling echnique, pa ial leas squa es (PLS-SEM), o es he model p oposed by he esea ch and he hypo hesized ela ionships. Two easons suppo his echnique: Fi s , ollowing Hai e al. [ 70 ] and Rigdon e al. [ 66 ], we used Mode A composi es; second, acco ding o Hensele [ 71 ], we chose an explana o y app oach. F om he e, we de eloped a wo-s ep p ocess wi h PLS o he e alua ion o he p oposed model. Fi s , he measu emen model is analyzed, and second, he s uc u al model will be analyzed [72]. In o de o ind he impo ance o he indices, we ca ied ou a boo s apping p ocess [ 73 ]. Fo he analysis o he da a we used he so wa e Sma PLS 3.2.6 acco ding o Ringle e al. (2015) and o he analysis o he media ing e ec s we ollowed he p ocess de eloped by Cepeda-Ca ion e al. [74]. Sus ainabili y 2021,13, 6720 6 o 12 Acco ding o Hensele e al. [ 75 ], he mos sui able measu es o in e nal eliabili y would be ρ A, Jö eskog’s ho and C onbach’s alpha. Values abo e 0.7 indica e an adequa e le el o eliabili y [ 76 ]. Th ough he AVE (mean a iance ex ac ed) we analyzed he unidi- mensionali y acco ding o Fo nell-La cke [ 77 ]. Finally, ollowing Hai e al. [ 78 ], o e alua e disc iminan alidi y, we ollowed he HTMT c i e ion (he e o ai –mono ai a io). 4. Resul s F om he indica ed PLS-SEM echnique, he ollowing esul s we e ob ained by ana- lyzing on one hand he measu emen model and on he o he hand he s uc u al model. 4.1. Measu emen Model As o he measu emen model, i should be no ed ha all i ems mee he p e equisi e o eliabili y since all loads a e g ea e han 0.7 (Table 1) and consis en measu emen s a e g ea e han 0.8 [79]. Table 1. Indica o co ela ion weigh s. GW BI CSR WOM GW1 0.867 GW2 0.914 GW3 0.834 GW4 0.866 GW5 0.802 BI1 0.846 BI2 0.904 BI3 0.865 CSR1 0.789 CSR2 0.848 CSR3 0.795 CSR4 0.903 CSR5 0.718 WOM1 0.896 WOM2 0.901 WOM3 0.844 WOM4 0.871 WOM5 0.727 WOM6 0.911 No es: GW: G eenwashing; BI: Beha iou in en ion; CSR: Co po a e social esponsibili y; WOM: Wo d o mou h. The eliabili y o he scales was e alua ed acco ding o he esul s ob ained ( his is e lec ed in Table 2) by checking i he C onbach’s Alpha o each scale is highe han he h eshold o 0.70 [ 80 ]. Composi e eliabili y (CR) [ 81 ] is conside ed a be e measu e o eliabili y since i does no assume he equaliza ion o he i ems, and he alue o he a e age a iance (AVE) is g ea e han 0.5 in all cases [72]. Table 2. Cons uc eliabili y and con e gen alidi y coe icien s. CA ho_A CR AVE GW 0.912 0.946 0.933 0.735 BI 0.842 0.846 0.905 0.760 CSR 0.870 0.879 0.907 0.661 WOM 0.929 0.935 0.945 0.741 No es: CA = C onbach’s alpha; CR = composi e eliabili y; AVE = a e age a iance ex ac ed. In Table 3, we can see ha all a iables p esen disc iminan alidi y based on he e o ai –mono ai a io (HTMT) c i e ia. Sus ainabili y 2021,13, 6720 7 o 12 Table 3. Disc iminan alidi y. He e o ai –mono ai a io (HTMT). GW BI CSR WOM GW BI 0.202 CSR 0.246 0.198 WOM 0.169 0.763 0.312 No es: GW: G eenwashing; BI: Beha iou in en ion; CSR: Co po a e social esponsibili y; WOM: Wo d o mou h. 4.2. S uc u al Model The alida ion o he s uc u al model will be done acco ding o he signi icance o he model’s coe icien s, using he boo s ap me hod, wi h 5000 samples wi h eplacemen o he same size as he o iginal sample (198 obse a ions) [82]. Da a om he analysis o he s uc u al model a e indica ed in Table 4and in he Figu e 2. Table 4. Cons uc e ec s on endogenous a iables (including lowe and uppe limi s o 95% con idence in e al). E ec s on Endogenous Va iables Pa h Coe . Con idence In e als (95%) Signi icance o E ec (p-Value) R [2] Dependen Cons uc s 5%CIlo 95%CIhi GW →CSR (H1) −0.216 −0.369 −0.041 Yes (0.009) 0.046 CSR →BI (H2) −0.039 −0.138 0.063 No (0.447) 0.450 CSR →WOM (H3) 0.263 0.109 0.415 Yes (0.001) 0.105 GW→WOM (H4) −0.115 −0.281 0.064 No (0.186) 0.150 WOM →BI (H5) 0.673 0.573 0.756 Yes (0.000) 0.450 GW →BI (H6) −0.086 −0.196 0.022 No (0.125) 0.450 GW →CSR →WOM →BI (H7) −0.038 −0.027 −0.075 Full Media ion 0.450 GW →WOM →BI (H8) −0.077 −0.224 0.010 No (0.185) 0.450 No es: GW: G eenwashing; BI: Beha iou in en ion; CSR: Co po a e social esponsibili y; WOM: Wo d o mou h. Sus ainabili y 2021, 13, x FOR PEER REVIEW 8 o 13 Figu e 2. Resea ch p oposed model. Table 4 shows he analysis o he p oposed hypo heses in ou esea ch model. In ou esea ch model, he analysis o di ec ela ionships (hypo heses 1 o 6) and indi ec ela ionships (media ing e ec s o hypo heses 7 and 8) we e p oposed. The indi- ec ela ionships a e shown in Figu e 2 wi h do ed and dashed lines. 5. Discussion A e pe o ming he co esponding analysis wi h he p og am Sma PLS we ha e o emphasize ha ou o he eigh p oposed hypo heses a e ul illed. Hypo heses H1, H3, H5 and H7 a e ul illed, his las one being he hypo hesis o mula ed abou o al media- ion o GW o e BI h ough CSR and WOM. F om he esul s ob ained, we can indica e in he i s place, in ela ion o di ec e - ec s, H1, H3 and H5 a e signi ican and he e o e H1, H3 and H5 a e suppo ed. Howe e , hypo heses H2, H4 and H6 u ned ou no o be signi ican and he e o e ha e no been suppo ed. Secondly, and in ela ion o media ing e ec s, H7 is signi ican , while H8 is no signi ican , and he e o e H7 is suppo ed and H8 is no . In ela ion o H7, and acco ding o Hai e al. (2017) and Ni zl e al. (2016), he ype o media ion is ull media ion because he di ec e ec be ween GW and BI (H6) is no signi ican , and he indi ec e ec , h ough he media ion o CSR and WOM, is signi ican . I is no ewo hy ha he H2 hypo hesis is no ul illed since a good CSR policy im- p o es he co po a e image and epu a ion agains wha au ho s like Bigné-Alcañiz and Cu ás-Pé ez say [39] bu i does no a ec BI. Hypo heses H4 and H6 a e no ul illed ei he , so a p io i he GW only signi ican ly a ec s CSR. This e ec is in line wi h Mahoney e al. (2013). The p oposed media ion o GW wi h BI h ough WOM is no signi ican ei- he , bu he o al media ion (H7) is, his ime acco dingly o Bigné-Alcañiz and Cu ás- Pé ez [39]. Among he signi ican ela ionships, i should be no ed ha all o hem exceed he minimum ecommended alue o 0.2. H1 has a pa h o -0.216 while H3 has 0.263 and H5 has he g ea es con ibu ion o he in en ion o use wi h 0.673. We can also e i y ha he g ea es con ibu ion o he in en ion o use is made by wo d o mou h [83]. We mus also highligh he s eng h o he ela ionship o hypo hesis 5 since i s 2 is 0.766, which is much highe han he ecommended alue [73]. This, oge he wi h he ac ha he pa h be ween WOM and BI is 0.673, makes his ela ionship he clea es o all hose shown in he model. Figu e 2. Resea ch p oposed model. Table 4shows he analysis o he p oposed hypo heses in ou esea ch model. In ou esea ch model, he analysis o di ec ela ionships (Hypo heses 1 o 6) and indi ec ela ionships (media ing e ec s o Hypo heses 7 and 8) we e p oposed. The indi ec ela ionships a e shown in Figu e 2wi h do ed and dashed lines. Sus ainabili y 2021,13, 6720 8 o 12 5. Discussion A e pe o ming he co esponding analysis wi h he p og am Sma PLS we ha e o emphasize ha ou o he eigh p oposed hypo heses a e ul illed. Hypo heses H1, H3, H5 and H7 a e ul illed, his las one being he hypo hesis o mula ed abou o al media ion o GW o e BI h ough CSR and WOM. F om he esul s ob ained, we can indica e in he i s place, in ela ion o di ec e ec s, H1, H3 and H5 a e signi ican and he e o e H1, H3 and H5 a e suppo ed. Howe e , hypo heses H2, H4 and H6 u ned ou no o be signi ican and he e o e ha e no been suppo ed. Secondly, and in ela ion o media ing e ec s, H7 is signi ican , while H8 is no signi ican , and he e o e H7 is suppo ed and H8 is no . In ela ion o H7, and acco ding o Hai e al. (2017) and Ni zl e al. (2016), he ype o media ion is ull media ion because he di ec e ec be ween GW and BI (H6) is no signi ican , and he indi ec e ec , h ough he media ion o CSR and WOM, is signi ican . I is no ewo hy ha he H2 hypo hesis is no ul illed since a good CSR policy imp o es he co po a e image and epu a ion agains wha au ho s like Bigné-Alcañiz and Cu ás-Pé ez say [ 39 ] bu i does no a ec BI. Hypo heses H4 and H6 a e no ul illed ei he , so a p io i he GW only signi ican ly a ec s CSR. This e ec is in line wi h Mahoney e al. (2013). The p oposed media ion o GW wi h BI h ough WOM is no signi ican ei he , bu he o al media ion (H7) is, his ime acco dingly o Bigné-Alcañiz and Cu ás-Pé ez [39]. Among he signi ican ela ionships, i should be no ed ha all o hem exceed he minimum ecommended alue o 0.2. H1 has a pa h o -0.216 while H3 has 0.263 and H5 has he g ea es con ibu ion o he in en ion o use wi h 0.673. We can also e i y ha he g ea es con ibu ion o he in en ion o use is made by wo d o mou h [83]. We mus also highligh he s eng h o he ela ionship o hypo hesis 5 since i s 2 is 0.766, which is much highe han he ecommended alue [ 73 ]. This, oge he wi h he ac ha he pa h be ween WOM and BI is 0.673, makes his ela ionship he clea es o all hose shown in he model. 6. Conclusions We can emphasize ha , once he model is p oposed and he da a collec ed in ou ield wo k is ob ained, he esul s show us ha he ma ke knowledge o he une hical ac i i ies o he companies has a signi ican in luence on di e en a iables, which can in luence he BI in a di ec o indi ec way. GW has a signi ican ly nega i e impac on he CSR. This in luence can be due o he ac ha he company may p ojec a nega i e image o he ou side, which in luences consume beha io . I he company ansmi s an image ha is no e y bene icial o i , he consume will ac acco dingly, esul ing in a dec ease in he company’s own income accoun . CSR does no show an in luence on BI. This shows us a somewha su p ising esul no in line wi h he li e a u e and we hink ha his may be due o he ac ha he clien o his ype o business does no alue much he esponsibili y o he company wi h he su ounding en i onmen . Mo eo e , s eng hening he p e ious a gumen , we add ha o he ac o s also ha e a g ea e in luence on BI, so i ends up nulli ying o co e ing up he e ec o CSR. While we ha e o emphasize ha he e is an indi ec e ec on BI media ed by CSR, so BI has indi ec e ec s e en in his ype o business. CSR has a signi ican and posi i e e ec on WOM. A good image o a company is commen ed on and alued by consume s. GW shows us ha i does no ha e an in luence on WOM. He e we disco e ano he o he esul s o ou wo k ha is qui e s iking since i does no ollow he line we expec ed. We hink ha his can be explained because ei he people do no unde s and well he philosophy o he GW concep o be ansmi ed o hey conside ha he a gumen is no impo an enough o ansmi o o he people o consume s. We could ind ha pe haps he clien does no ha e enough in o ma ion o judge and con ey a nega i e o posi i e idea o he o ganiza ion ca ying ou a alse g een s a egy and ha i only cons i u es a claim o sell mo e by con using indi iduals. We could also hink ha in his ype o business Sus ainabili y 2021,13, 6720 9 o 12 he cus ome looks he o he way in o de o sa e money o ob ain he se ice a a e y compe i i e p ice. WOM has an impo an and posi i e in luence on BI. This seems logical, we unde - s and ha i is caused by he powe ul e ec gene a ed by he opinions o o he indi idu- als/cus ome s on o he people’s beha iou in en ion. The ideas ha people ansmi and ha a e ecei ed by o he indi iduals a e sp ead by indi iduals ha a e us ed by he ecei e s and who can in luence hei BI since hese a e people ha he ecei e us s and belie es in. Con inuing wi h he abo e, cus ome s us he assessmen and opinions o people who a e in hei closes ci cle, as hey espec and alue hei opinion ega ding hei BI, a abo e he ad e ising claims o an ad o he s a emen s o he companies’ selle s o o he easons p o ided by o he means o he i ms. GW does no ha e a signi ican ela ionship wi h BI, which may come as a su p ise o us. To explain his, we belie e ha GW can be alued in a lesse deg ee in compa ison wi h o he a iables ha would camou lage i s e ec . Tha is, i is conside ed by he indi idual as mo e impo an han GW when gene a ing his BI. Many imes, he clien is in a si ua ion whe e he mus choose be ween di e en easons o gene a e his pu chase in en ion ha can in luence his decision, and his leads him o disca d op ions ha a e less e hical conce ning he en i onmen , because he eels ha o he incen i es such as p ice, dis ibu ion o o he a iables a e mo e in luen ial in decision making and he e o e he e ec o GW is no conside ed. Howe e , we mus poin ou ha he e is an indi ec e ec be ween GW and BI ha is signi ican ly in line wi h he p oposed model and ha i does emind cus ome s ha , al hough o he easons such as hose se ou in he p e ious pa ag aph may p e ail, i is impo an no o ha e a GW policy. 7. Limi a ions and Fu u e Resea ch Fu u e lines o esea ch would be o apply his esea ch o o he ypes o businesses, and also o o he ypes o consume s wi h he aim o e-measu ing he same ela ionships. We also conside ha s udying his now wi h he pandemic ha is hi ing us would be e y app op ia e o see i he esul s would change o be modi ied, which we see as e y signi i- can o ca ying ou a longi udinal s udy. I would also mean u u e esea ch o see he ela ionship o GW wi h o he a iables such as sales o income accoun s, o o he a iables, and o use samples om o he coun ies o compa e esul s a a c oss-cul u al le el. The ela ionship be ween CSR and BI could be measu ed in u u e s udies using di e en samples o consume s, di e en ypes o companies and e en using he opinion o company manage s, and es ablishing compa isons wi h hese esul s, which is a g ea u u e o esea ch. As limi a ions we would ind wha we ha e indica ed p e iously, ha i is a ans e sal s udy, ha we only use a sample o consume s and ha i is only o Spanish consume s, which limi s he gene aliza ion o he esul s. The e is also limi a ion conce ning he companies selec ed o pe o m he wo k. Wi h hese collabo a i e economy companies whe e low p ice p e ails o e almos any o he conside a ion, i is di icul o gene alize he esul s. I would be in e es ing o check companies om o he sec o s ha a e also suscep ible o pe o m GW. Au ho Con ibu ions: Concep ualiza ion, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; me hodology, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; so wa e, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; alida ion, E.G.-C., J.P.C.- S., I.C.-C. and J.O.-G.; o mal analysis, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; in es iga ion, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; esou ces, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; da a cu a ion, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; w i ing—o iginal d a p epa a ion, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; w i ing— e iew and edi ing, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; supe ision, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; p ojec adminis a ion, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; unding acquisi ion, E.G.-C., J.P.C.-S., I.C.-C. and J.O.-G.; All au ho s ha e ead and ag eed o he published e sion o he manusc ip . Funding: This esea ch ecei ed no ex e nal unding.