A restaurant quality model based on marketing factors

157 1, XX, 2017 Ma ke ing and T ade DOI: 10.15240/ ul/001/2017-1-011 In oduc ion In he highly compe i i e es au an indus y, sa is ying gues s should be he c i ical objec i e o all businesses ha wish o p ospe and encou age epea pu chases. Howe e , ensu ing p ope quali y in es au an s is limi ed by a numbe o indus y-speci i c ac o s, including ola ile demand, small businesses, in ense compe i ion, he wide ange o ood and be e age p oduc s o e ed, he insepa abili y o p oduc ion and consump ion, in angibili y o se ices, labou -in ensi e p oduc ion, he impo ance o employees’ a i ude owa ds gues s and many o he ac o s ha signi i can ly a ec he le el o o e all se ice quali y. A c ucial challenge o all es au a eu s oday is how o p o ide a quali y o e ing ha is no only compelling o gues s bu also supe io o ha o business compe i o s. In o de o gain an ad an ageous edge in his highly compe i i e en i onmen , he ma ke ing li e a u e has consis en ly emphasized he impo ance o ma ke ing o ien a ion as a s a egic ool. The g owing ecogni ion o he cus ome - based ma ke ing app oach (i.e., business o consume ) has sugges ed ha implemen ing quali y as a ma ke ing ool is he essen ial elemen in os e ing cus ome ela ionships and sus ainable ma ke sha e (Wang, Law, Hung, & Guille , 2014). Unde s anding cus ome s’ needs is he i s s ep in deli e ing quali y se ices. The bes way o manage cus ome s’ expec a ions is o i nd ou wha hei needs and wan s a e, s a egize how o mee hem and implemen hese s a egies in p ac ice. In he scien i i c li e a u e, he e a e se e al heo e ical models o explo e cus ome s’ expec a ions and assess se ice quali y. A e he in oduc ion o he gene ic SERVQUAL ins umen in 1988 by Pa asu aman, Zei haml and Be y, he issue o es au an se ice quali y has ecei ed conside able c i ical a en ion. Se e al a emp s ha e been made (Kim, Ng, & Kim, 2009; Sulek & Hensley, 2004; Vannia ajan, 2009) o imp o e and de elop speci i c quali y measu emen echniques sui ed o he needs o he es au an indus y (e.g., DINESERV, SERVPERF, CIERM, TANGSERV). All o hese echniques ocus on speci i c aspec s o se ice deli e y ( echniques a e p esen ed in chap e 1.3). As ood is he undamen al componen o he dining expe ience, i undoub edly has a signi i can impac on gues s’ sa is ac ion (Vannia ajan, 2009). Despi e he impo ance o ood quali y, some schola s ha e ocused on he se ice encoun e aspec o se ice quali y (Ayeh & Chen, 2013; Ed a dsson, 2005; Han, Back, & Ba e , 2010). Fu he examina ion e eals ha mos s udies (Andaleeb & Conway, 2006; Ba be , Goodman, & Goh, 2011; Mosa i & Ghaedi, 2012; Raajpo , 2002; Voon, 2012) ha e empi ically in es iga ed he ole o di e en and speci i c quali y ac o s, such as en i onmen , angibles, cleanliness, p ice pe cep ion, and so o h. One in e es ing i nding ha eme ges om he compa ison o hese s udies is ha esea ch esul s a e mu ually inconsis en , as hey emphasize he impo ance o di e en quali y ac o s. None heless, no s udy has de e mined which ma ke ing ac o (7P) is he mos impo an in assessing he quali y o he dining expe ience. The e o e, i would be o academic signi i cance o summa ize he esul s o di e en s udies and o empi ically in es iga e he impo ance o di e en ma ke ing ac o s in assessing he quali y o es au an o e ings. To i ll his esea ch gap, he cu en s udy ocuses on se en ma ke ing ac o s (ma ke ing mix) ha , acco ding o Ko le (2004), o m he undamen al pa o es au an o e ings. We assume ha all ma ke ing ac o s ha e a signi i can impac on gues s’ sa is ac ion as hey o m an insepa able pa o he dining expe ience. The pu pose o his a icle is o A RESTAURANT QUALITY MODEL BASED ON MARKETING FACTORS Ma ko Kukanja, Do is Gomezelj Ome zel, Bo is Buko ec EM_1_2017.indd 157EM_1_2017.indd 157 13.3.2017 16:59:0613.3.2017 16:59:06 158 2017, XX, 1 Ma ke ing a obchod desc ibe he de elopmen o a ma ke ing- o ien ed model o measu ing es au an quali y and o discuss i s p ope ies and po en ial applica ions. Mo e speci i cally, esea ch cons uc cla i i ca ion is mainly ela ed o ou main esea ch ques ion: How do di e en ma ke ing quali y ac o s in l uence he assessmen o o e all es au an quali y (7P) om he gues s’ pe spec i e? We he e o e hypo hesize: H1: All se en ma ke ing quali y dimensions (7P) ha e a s a is ically signi i can in l uence on gues s’ quali y pe cep ion. Based on Ko le ’s ma ke ing mix s a egic app oach, Edwa ds (2013) and Sedmak (2011) highligh ed he impo ance o he o e all quali y expe ience. Acco ding o hese au ho s, all quali y ac o s ha e o be conside ed oge he when e alua ing he es au an expe ience, as hey all in l uence gues s’ quali y expec a ions and pe cep ions. The es au an o e ing mus he e o e be consis en , cohe en and designed in acco dance wi h gues s’ expec a ions (Sedmak, 2011). We hypo hesize ha : H2: Gues s pe cei e he quali y o deli e ed es au an o e ings as cohe en . Tes ing hese hypo heses calls o a li e a u e e iew inqui y in o ecen esea ch on es au an quali y. Based on a ho ough li e a u e e iew, a quali y model was de eloped based on ma ke ing ac o s o measu ing cus ome s’ pe cep ions o es au an quali y. In he second pa o he s udy, he model was empi ically es ed. The o e all s uc u e o he s udy consis s o ou chap e s, including his in oduc o y chap e . Chap e one begins by laying ou he heo e ical dimensions. Chap e wo is conce ned wi h me hodology – i desc ibes he gene a ion o model i ems and p o ides an e alua ion o he model’s eliabili y and ac o s uc u e. Finally, he conclusion p esen ed in he las chap e gi es a b ie summa y and c i ique o he i ndings. 1. Theo e ical Backg ound 1.1 Se ice Quali y De i ning se ice quali y equi es a speci i c app oach o quali y measu emen , as i is no based on gene al objec i i y and measu abili y. The app oach om he s andpoin o he cus ome is based on a highly subjec i e pe spec i e. While a a ie y o de i ni ions ha e been sugges ed (G ön oos, 1984, 1990; Lange , 1997; Pa asu aman e al., 1988; Ree es & Bedna , 1995), his pape is based on he de i ni ion sugges ed by Pa asu aman, Zei haml, and Be y (1985), who de i ned se ice quali y as he abili y o a se ice o ul i l and exceed gues s’ expec a ions. The common cha ac e is ic o all se ice quali y de i ni ions (Pa asu aman e al., 1988; Ree es & Bedna , 1995; Ryu & Jang, 2007; Van Vae enbe gh, La i iè e, & Ve mei , 2012) is he consume - based concep , which makes se ice quali y a highly subjec i e and ela i e phenomenon ha di e s based on who is judging he se ice. In ou s udy, speci i c ma ke ing ac o s in ol ed in he ma ke ing mix (7P) a e used as key quali y dimensions. 1.2 Theo e ical Models o Se ice Quali y A la ge and g owing body o scien i i c li e a u e has in es iga ed he heo e ical concep o se ice quali y. Se e al a emp s ha e been made o cap u e he essen ial cha ac e is ics o se ice quali y in heo e ical models. These models a e especially impo an because hey p o ide a heo e ical basis o a ious echniques (ins umen s) o measu ing se ice quali y. The Ame ican school (Pa asu aman, Be y, & Zei haml, 1993; Pa asu aman, Zei haml, & Be y, 1994) is mainly ocused on iden i ying he c i e ia ha consume s use in e alua ing he quali y o se ices. Resea che s ha e con ibu ed a i e-s ep model o se ice quali y and an ins umen o measu ing se ice quali y – he SERVQUAL ins umen – in which hey de i ned i e dimensions o se ice quali y: Reliabili y, Assu ance, Tangibles, Empa hy, and Responsi eness. Meanwhile, esea che s om he Scandina ian school (G ön oos, 1990; Leh inen & Leh inen, 1991) ha e iden i i ed wo majo aspec s o se ice quali y: echnical quali y ( he angible aspec o he quali y) and unc ional quali y. D awing on an ex ensi e ange o sou ces, he schola s (Candido & Mo is, 2000; Lin, Chan, & Tsai, 2009) used a ious me hods in an a emp o c ea e alid and o e all-se ice quali y models. Candido and Mo is (2000) de i ned a new model wi h 14 s eps, bu an in-dep h analysis e ealed ha he model is mainly based on he i e-s ep model. O e all, none o hese modi i ed models EM_1_2017.indd 158EM_1_2017.indd 158 13.3.2017 16:59:0613.3.2017 16:59:06 159 1, XX, 2017 Ma ke ing and T ade ecei ed a signi i can scien i i c alida ion. Con e sely, se e al au ho s highligh ed he need o b eak he link be ween he adi ional Ame ican and Scandina ian schools and p oposed al e na i e quali y models. Lin e al. (2009) upg aded he adi ional IPA (Impo ance Pe o mance Analysis) model and de eloped a new model called IPGA. The IPGA model is designed o op imize he use o p oduc ion esou ces wi h he aim o imp o ing he quali y o se ices o e ed. Ne e heless, all hese s udies highligh he need o u u e de elopmen o se ice quali y managemen . 1.3 Tools o Measu ing Se ice Quali y In ou s udy, we ocus on ools ( echniques) ha collec quali y in o ma ion based on p e-de e mined s anda ds, al hough he cus ome s’ eedback can also be ob ained by a numbe o quali a i e echniques. Despi e he unques ionable signi i cance o quali a i e echniques, we decided o use quan i a i e echniques. Some o hese echniques (o en also e e ed o as models) measu e se ice quali y basing on he quali y gaps ha occu as a esul o di e ences be ween gues s’ expec a ions and pe cep ions (SERVQUAL, DINESERV); some a e one-dimensional and ocus solely on se ice pe o mance e alua ion (SERVPERF, Dinese .pe ); some combine quali y and impo ance measu emen o di e en se ice ac o s (SERVIMPERF); some ocus on employees’ esponses o speci i c c i ical si ua ions (C i ical Inciden Technique – he ex ended model); and, i nally, some add ess ex e nal e alua o s (AAA Diamond, Michelin S a s). The p edominan quan i a i e measu emen echnique is he SERVQUAL ins umen (Ma ko ić, Raspo , & Šega ić, 2012), which measu es quali y basing on he gap be ween gues s’ expec a ions and pe cep ions. Acco ding o Aigbedo and Pa ameswa an (2004), all i e dimensions o he SERVQUAL ins umen ha e no ye been ully alida ed. The e o e, he au ho s p opose addi ional me ics ha would be e explain he gap be ween expec a ions and pe cep ions. O he au ho s (Dedeke, 2003; Jensen & Hansen, 2007; Juwahee , 2004; Ryu, 2005) ha e highligh ed he need o a ailo ed app oach o se ice quali y measu emen . Despi e hese c i icisms, howe e , SERVQUAL emains one o he mos commonly used se ice quali y measu emen echniques (Ma ko ić e al., 2012). Inclusion o quali y dimensions in di e en es au an se ice quali y models is p esen ed in Table 1. As can be seen om he able abo e, speci i c echniques, such as Tangse and CIERM, ha e mo ed away om he adi ional RATER dimensions o he SERVQUAL ins umen . Ou i ndings a e consis en wi h Dimension Model Tangibili y Reliabili y Responsi eness Assu ance Empa hy Nu i ional alue Ing edien s Food p epa a ion ech- nique SERVQUAL x xxxx DINESERV x xxxx Dinese .pe x xxxx Tangse x CIERM x x x Sou ce: own Tab. 1: Inclusion o quali y dimensions in di e en se ice quali y models EM_1_2017.indd 159EM_1_2017.indd 159 13.3.2017 16:59:0713.3.2017 16:59:07 160 2017, XX, 1 Ma ke ing a obchod hose o o he s udies (C onin & Taylo , 1994; Llosa, Chandon, & O singhe , 1998; T ibe & Snai h, 1998) in sugges ing he necessi y o mo ing away om hese adi ional dimensions. 2. Me hodology 2.1 Resea ch P ocess and Sample Desc ip ion Following he concep ualiza ion and ope a ionaliza ion o he se ice quali y cons uc (see Fig. 1), a 35-i em ins umen o assessing cus ome pe cep ion o es au an quali y was o mula ed and empi ically es ed. Al hough in many ques ionnai es (Ma ko ić e al., 2012) indi idual quali y ac o s a e subs an i ely combined o exp ess cha ac e is ics o se e al ac o s in a single, uni o m quali y ac o (e.g., a ac i eness o ca pa ks and su ounding a eas), in ou s udy we ha e exclusi ely used one quali y cha ac e is ic o he desc ip ion o each quali y ac o (see Tab. 2). The le el o cus ome s’ pe cep ions was measu ed on a i e-poin Like scale, anging om 1 (s ongly disag ee) o 5 (s ongly ag ee). The model is based on he pe o mance aspec o quali y measu emen . In doing so ( ocusing on pe o mance assessmen ), we suppo he heo e ical i ndings o se e al au ho s (Abdullah & Roza io, 2009; Kei h & Simme s, 2011; Land um, P ybu ok, & Zhang, 2007) whose wo ks emphasize he impo ance o a one- dimensional (pe o mance-only) app oach o measu ing quali y. The ques ionnai e was p e es ed on en gues s and i e es au an manage s who we e in i ed o pa icipa e in he o ma ion o he esea ch ins umen . Based on hei sugges ions, some mino changes we e made. Ou s udy was conduc ed om Janua y o June 2014. The esea ch was pe o med by en in e iewe s in di e en es au an se ings in Slo enia – acco ding o he o i cial (na ional) classi i ca ion he ollowing ypes o es au an s we e included in he esea ch – “ es a acije” ( o mal – luxu y es au an s); “gos ilne” (a kind o na ional Inns), and “ok epče alnice” (in o mal – as ood acili ies, snack ba s and simila ). A o al o 323 independen ly ope a ed es au an s we e included in he s udy, ep esen ing 10% o he popula ion o Slo enia. The esea ch was conduc ed by di ec in e iews wi h domes ic cus ome s a e hey dined in he selec ed es au an s. We ob ained pe mission om he es au an manage s be o e in e iewing gues s. In e iewing o en ook place a he able o , in some cases, in he es au an lobby be o e gues s` depa u e, as some manage s did no allow in e iews in he dining oom. Randomly chosen gues s we e kindly asked o i ll in he ques ionnai e. Ou o a o al o 2003 collec ed ques ionnai es, he i nal analysis is based on 1998 alid ques ionnai es. In he i s s ep, desc ip i e s a is ics analysis was used o analyse esponden s’ demog aphic cha ac e is ics. The majo i y o esponden s we e an a e age o sligh ly less han 40 yea s o age, and he sample was composed o almos equal numbe s o male (49.4%) and emale gues s (50.6%). The highes numbe o gues s had comple ed seconda y ( oca ional) educa ion (47.2%), while 41.5% o gues s had acqui ed a high school educa ion, 4.9% had only i nished elemen a y school and 6.3% o he gues s had ob ained a Mas e ’s deg ee o PhD. A o al o 14% o gues s we e isi ing he es au an o he i s ime, while 86% o gues s had isi ed he es au an h ee o mo e imes. In addi ion o demog aphic da a, he pu pose o he isi (as a psychog aphic a iable) was in oduced in o he s udy. Resul s show ha mo e han hal (53.9%) o gues s isi ed he chosen acili y wi h he aim o sa is ying hunge and hi s (physiological needs). The need o ea was ollowed by he pu pose o socializing (27.1%), celeb a ions (9.9%), business meals (5.2%) and o he pu poses (4.1%). 2.2 Ques ionnai e De elopmen P oduc (Food) Nume ous s udies ha e epo ed ha ood is he mos impo an quali y dimension ha a ec s bo h he es au an selec ion p ocess and gues s’ sa is ac ion (Gup a, McLaughlin, & Gomez, 2007; Sulek & Hensley, 2004; Vannia ajan, 2009). All hese s udies ou line a c i ical ole o ood quali y e alua ion in co ela ion wi h o he quali y ac o s. Ne e heless, in se e al in e na ional s udies (Kim e al., 2009; Sulek & Hensley, 2004; Vannia ajan, 2009), ood was iden i i ed as he mos impo an quali y dimension. Sulek and Hensley (2004) p oposed ha he quali y o ood should be simply de i ned by h ee key cha ac e is ics: ood sa e y, a ac i eness and diges ibili y. Based on li e a u e e iew, we used he ollowing quali y ac o s in ou esea ch model: o e olume (selec ion o dishes), size o po ions, as e, appea ance, and pe cep ion o ood sa e y. EM_1_2017.indd 160EM_1_2017.indd 160 13.3.2017 16:59:0713.3.2017 16:59:07 161 1, XX, 2017 Ma ke ing and T ade People A la ge olume o published s udies (Andaleeb & Conway, 2006; Jaa a , Lumbe s, & E es, 2008; Mosa i & Ghaedi, 2012; Voon, 2012) desc ibe he ole o people as he mos impo an quali y dimension in es au an quali y managemen . Resea che s who ha e looked a psychological aspec s o pe cep ions ha e con i med he co ela ion be ween gues s’ quali y e alua ion p ocess and demog aphic cha ac e is ics o se ice s a . Fo example, Luohe and Tsau (2011) con i med he link be ween gues s’ pe cep ions and age; Ma ínez-Tu , To de a, Pei ó, and Po ocnik (2011) emphasized he impo ance o o ganiza ional clima e; and Wall and Be y (2007) concluded ha gues s’ quali y pe cep ions hea ily depend on he ype o es au an i sel . The p e iously men ioned i nding o Wall and Be y (2007) co obo a es he idea o Kim and Kache sky (2006), and Meng and Ellio (2008) who sugges ed ha gues s o i ne dining es au an s a e mo e sensi i e o he a i udes o se ice s a . In con as , a s udy conduc ed by Waxman (2006) in Ca e shops no only s essed he impo ance o s a ’s a i ude bu also e ealed a ich se o social quali y ac o s associa ed wi h s a , including wa m h, a sense o belonging, amilia i y, espec , us , and suppo . The design o ou ques ionnai e has been based on he esul s o he p esen ed s udies, and some quali y ac o s we e logically in oduced om he SERVQUAL ins umen (employees’ poli eness), DINESERV (well- ained, compe en and expe ienced s a ; numbe o s a ) and he Tangse model (cus ome s’ in e ac ions wi h o he people). Speci i c ac o s ha ha e been in oduced o ou model o assessing he quali y o people a e: hospi ali y o s a , compe ences o se ice s a , su i cien numbe o s a o ensu e quali y se ice, impo ance o he p esence o he es au an manage , and he dis ac ing p esence o o he gues s. P ice Subjec i e assessmen o quali y is pa icula ly p oblema ic in e ms o indi idual pe cep ion o p ice. Kim and Kache sky (2006) s a e ha he pe cei ed p ice le el and i s co ela ion o quali y a e exclusi ely a esul o an indi idual psychological p ocess. This iew is suppo ed by many au ho s (Bha achnaya & F iedman, 2001; Meng & Ellio , 2008) who ha e a gued o he impo ance o indi idualism in p ice pe cep ion (especially ”Fai p ice” pe cep ion). Acco ding o hese au ho s, a ai p ice has a signi i can impac on gues s’ sa is ac ion and hei pe cep ion o quali y. The es au an indus y ends o be highly p ice-elas ic, as a small change in p ice is accompanied by a la ge change in quan i y demanded (Sedmak, 2011). We ha e no ed, ha p ice was no iden i i ed as he mos impo an quali y dimension in any o he p esen ed s udies. Ne e heless, many s udies s essed he impo ance o di e en indi idual p ice ac o s (e.g., an accu a e bill is also a quali y ac o in he DINESERV model). Following he abo e discussion, we may suppose ha p ice quali y can be measu ed based on he ollowing p ice ac o s: unde s andabili y o p ices, accu a e bill, alue o money, p ice compe i i eness, and expec ed p ice le el s. ac ual p ice le el. P ocess The quali y o his dimension is mos o en assessed acco ding o di e en ac i i ies o se ice s a (Cousins, Foske , & Gillespie, 2002). Ha and Jang (2010) ha e p ima ily ea ed he quali y o he se ice encoun e as a mul idimensional cons uc ha is mos o en he esul o gues s’ subjec i e e alua ion o se e al quali y ac o s, such as he p ocess o welcoming gues s, accep ance o o de s, gues a endance, and so o h. Heung, Wong, and Qu (2000) epo ed ha he speed o se ice is he mos impo an ac o in de e mining gues s’ pe cep ion o quali y. Nam, Ekinci, and Whya (2011) s a e ha i s ill emains unknown how many quali y ac o s he e eally a e. Based on he li e a u e e iew, he ollowing quali y ac o s we e included in ou model: s a esponsi eness o ques ions, s a help ulness in se ing gues s’ needs, s a esponsi eness, and es au an wo king hou s. Physical E idence The impo ance o he physical, angible en i onmen has been s essed by se e al au ho s (Kim & Moon, 2009; Ma ila & Wi z, 2001; Namasi ayam & Ma ila, 2007; Ryu & Jang, 2007; Yunkyong, 2007), as i ep esen s an impo an basis o e alua ing he quali y o se ices. A numbe o au ho s (Cheng e al., 2012; Mosa i & Ghaedi, 2012; Ryu & Han, 2011; Wall & Be y, 2007) ha e epo ed ha physical e idences ha e a di e en impac on gues s who a e isi ing he es au an o he EM_1_2017.indd 161EM_1_2017.indd 161 13.3.2017 16:59:0713.3.2017 16:59:07 162 2017, XX, 1 Ma ke ing a obchod i s ime (i.e., i s - ime buye s) han on gues s who ha e al eady isi ed he es au an in he pas . In pa icula , ac o s associa ed wi h cleanliness and noise signi i can ly in l uence gues s’ pe cep ion o quali y (Ba be e al., 2011; Ba be & Sca celli, 2010). Acco ding o i ndings p esen ed in his sec ion and based on se e al quali y models (SERVQUAL, DINESERV, Tangse , SERVPERF), we ha e included he ollowing quali y ac o s in ou ques ionnai e: cleanliness o he p emises, nea and p esen able s a , com o , design in acco dance wi h ood o e ing, and sense o secu i y. P omo ion Acco ding o Sedmak (2011), he mos common o ms o p omo ion in he es au an indus y a e ad e ising, sales p omo ions, public ela ions, discoun s and special o e s, ou doo ligh ed signs and boa ds, menus and wine lis s, di ec sales, in i a ions and announcemen s and special e en s. Di ec sales p esen he only o m o ma ke ing communica ion ha p o ides ins an eedback om gues s. The success o di ec (pe sonal) sales hea ily depends on he compe ence, p o essionalism and cha isma o he se ice s a . Aside om he unc ional aspec o quali y, which hea ily depends on se ice s a sales ac i i ies ( ecommenda ions) and gues s’ p omo ional ac i i ies (i.e., wo d o mou h; Kelle , 2007), he quali y o p omo ional ac i i ies is also pe cei ed h ough he quali y o echnical ac o s, such as menu design (Din, Zaha i, O hman, & Abas, 2012; Sha ma, Wagle, Suche , & Bugwadia, 2011) and discoun s (Taylo & Long-Tolbe , 2002). The e idence p esen ed in his sec ion sugges s ha he e a e no majo di e ences be ween di e en o ms o p omo ional ac i i ies in ensu ing o e quali y and gues s’ ( e)pu chase in en ion. In ligh o he abo e, we ha e decided o include he ollowing p omo ional quali y ac o s in ou ques ionnai e: isible ma ke ing signs, signs o special a en ion and complimen s (small gi s, e c.), se ice s a ecommenda ions, olume o sales campaigns and special o e s, and ad e ising ac i i ies in social media. Placemen In he es au an indus y, unlike he ho el indus y, adi ional channels o dis ibu ion ha e emained unde de eloped. In he es au an indus y, channels o dis ibu ion a e mos o en di ec The mos impo an channels o dis ibu ion a e: loca ion, di ec dis ibu ion and indi ec dis ibu ion h ough a el agencies and o he p o ide s ha include es au an o e ings in hei o e ings (Sedmak, 2011). The impo ance o geog aphical loca ion was emphasized by Bowie and Bu le (2004) and Pa sa, Sel , Sydno -Busso, and Yoon (2011). We ha e decided o include he ollowing quali y ac o s in ou s udy ( he angible ac o s we e logically in oduced om he Tangse model): en ance accessibili y, accessible pa king a eas, nea and clean su oundings, pe cep ion o whe he he es au an is wo h he dis ance a elled, and indi ec dis ibu ion. Di e en s udies emphasize he impo ance o di e en quali y ac o s, as esea che s base hei s udies on di e en (speci i c) quali y ac o s. Ne e heless, no s udy has empi ically in es iga ed and compa ed all se en ma ke ing quali y dimensions (7P). 2.3 Analysis and Findings The nex sec ion o he su ey was conce ned wi h he e alua ion o he pe cei ed quali y o he es au an o e ing. The esul s p esen ed in Table 2 show ha all quali y ac o s we e e alua ed ela i ely highly ( he a e age mean alue is 3.98). Among he se en quali y dimensions, he highes - a ed dimension was p oduc & ood quali y (mean 4.32), wi h ood sa e y i s highes a ed ac o (mean alue 4.47). The esul s indica e ha he lowes pe cep ions a e ela ed o he dimension o ma ke ing communica ion (mean 3.57), wi h he lowes sco es ela ed o he ac o “ he es au an is p ope ly ad e ised in he media (2.95)”. The coe i cien s o a ia ion show how homogeneous gues s a e in he e alua ion o indi idual quali y ac o s. In he nex sec ion o he s udy, an explo a o y ac o analysis was pe o med o assess he ac o s uc u e o pe cei ed es au an quali y. Wi h his ac o analysis, we ha e ied o answe ou main esea ch ques ion and o es ou i s hypo hesis (H1), which sugges s ha all se en ma ke ing quali y dimensions (7P) ha e a s a is ically signi i can in l uence on gues s’ quali y pe cep ion. E idence o he scale’s eliabili y, ac o s uc u e and alidi y on he basis o he analysed da a is p esen ed nex . Fi s , we checked whe he he answe s o he abo e 35 a iables (quali y ac o s) we e no mally dis ibu ed. Because we EM_1_2017.indd 162EM_1_2017.indd 162 13.3.2017 16:59:0713.3.2017 16:59:07 163 1, XX, 2017 Ma ke ing and T ade Quali y dimensions (7P) Mean Coe i cien o a ia ion (%) P oduc ( ood) Selec ion o dishes 4.26 18.64 Ex en o po ions 4.27 19.11 Tas e 4.35 17.70 Appea ance 4.28 18.88 Food sa e y pe cep ion 4.47 16.82 Physical e idences Cleanliness o he p emises 4.34 18.04 Nea (p esen able) s a 4.28 19.30 Com o 4.22 20.14 Sense o secu i y 4.49 16.35 Design in acco dance wi h ood o e ed 4.26 20.16 People Su i cien numbe o s a o ensu ing quali y se ice 4.13 21.40 Impo ance o he p esence o es au an manage o ensu ing quali y o e ing 3.68 32.80 Dis ac ing p esence o o he gues s 2.51 54.26 Hospi able s a 4.26 20.52 Compe ences o se ice s a 4.13 21.84 P omo ion Visible ma ke ing signs 3.97 25.69 Signs o special a en ion and complimen s 3.41 36.86 Recommenda ions om se ice s a 4.00 23.68 Volume o sales campaigns and special o e s 3.53 34.31 Ad e ising ac i i ies in he social media 2.95 43.90 P ice Unde s andabili y o p ice i ems 4.52 16.17 Accu a e bill 4.67 13.88 Value o money 4.26 19.95 P ice compe i i eness 4.11 22.21 Expec ed p ice le el s. ac ual p ice le el 2.12 57.50 Placemen Accessible en ance 4.01 26.38 Accessible pa king a ea 3.78 33.99 Nea (clean) su oundings 3.99 24.51 The es au an is wo h he dis ance a elled 3.99 24.71 The es au an enhances indi ec dis ibu ion 3.37 38.43 Tab. 2: Analysis o he assessmen s o quali y (desc ip i e s a is ics) – Pa 1 EM_1_2017.indd 163EM_1_2017.indd 163 13.3.2017 16:59:0713.3.2017 16:59:07 164 2017, XX, 1 Ma ke ing a obchod Quali y dimensions (7P) Mean Coe i cien o a ia ion (%) P ocesses P omp esponsi eness o s a o ques ions 4.24 21.84 Help ulness o s a in sa is ying clien s’ needs 4.04 24.80 Responsi eness o se ice s a 4.18 21.10 Res au an opening hou s 4.29 19.81 Wai ing ime 4.10 22.00 Sou ce: own Quali y ac o s (QF) Quali y dimensions People Placemen P oduc and Physical e idences Help ulness o s a in sa is ying gues s’ needs 0.879 –0.096 0.047 Immedia e esponsi eness o s a o gues s’ ques ions 0.751 –0.045 –0.049 Signs o special a en ion and complimen s 0.740 0.024 0.144 Responsi eness o se ice s a 0.709 –0.006 –0.096 Recommenda ions om se ice s a 0.649 0.094 –0.034 Compe ences o se ice s a 0.646 0.038 –0.137 Hospi able s a 0.551 0.054 –0.158 Su i cien numbe o s a o ensu ing quali y se ice 0.473 0.019 –0.187 Value o money 0.450 0.084 –0.178 The es au an is wo h he dis ance a elled 0.259 0.448 –0.074 Accessible pa king a ea –0.104 0.759 0.038 Accessible en ance 0.082 0.733 0.020 Nea (clean) su oundings 0.094 0.602 –0.104 Tas e 0.027 –0.041 –0.747 Appea ance 0.002 –0.016 –0.743 Food sa e y pe cep ion –0.072 0.022 –0.729 Sense o secu i y –0.041 0.097 –0.676 Ex en o po ions –0.026 –0.008 –0.667 Selec ion o dishes 0.108 –0.060 –0.649 Com o 0.138 0.088 –0.502 Cleanliness o he p emises 0.211 0.094 –0.495 Nea (p esen able) s a 0.328 0.054 –0.416 Design in acco dance wi h ood o e 0.320 0.044 –0.395 Explained a iance % 40.973 5.058 4.221 Sou ce: own Tab. 2: Analysis o he assessmen s o quali y (desc ip i e s a is ics) – Pa 2 Tab. 3: Ro a ed ac o solu ion EM_1_2017.indd 164EM_1_2017.indd 164 13.3.2017 16:59:0713.3.2017 16:59:07 165 1, XX, 2017 Ma ke ing and T ade could no con i m a no mal dis ibu ion o any o he selec ed quali y ac o s o he i s se (Kolmogo o Smi no Tes (KMO) was used), i was necessa y o us o use he P incipal Axis Fac o ing me hod o he explo a o y ac o analysis. The i s es was pe o med in o de o e alua e he sui abili y o in o ma ion o inclusion in he ac o model. Thus, on he basis o he alue o he KMO measu e o sampling adequacy (0.958), as well as he ou come o he Ba le es o sphe ici y (2 = 31,071.468; deg ees o eedom = 595), we es ima ed ha all included a iables a e sui able o ac o analysis. The majo i y o ac o s had sa is ac o y communali ies (> = 0.50), sugges ing ha he g ea e pa o hei a iabili y can be explained by he in l uence o common ac o s, he h ee a iables wi h oo low communali ies (“p esence o Manage (0.132)”, “dis ac ing p esence o o he gues s (0.190)” and “ad e ising ac i i ies in he social media (0.251)” – we e excluded om he e alua ion p ocess o he ac o model. A e a ew successi e i e a ions o he ac o model e alua ion, we i nally selec ed as mos app op ia e he ac o model wi h 23 ac o s (p esen ed in Tab. 3); while 12 ac o s had o be emo ed om u he analysis. The sui abili y o in o ma ion o inclusion in he i nal ac o model is also suppo ed by he alues o he KMO indica o (0.957) and he ou come o he Ba le es ( = 223,289.398; deg ees o eedom = 253). Based on a o a ed ac o solu ion, we ha e chosen a i nal model wi h h ee ac o g oups (quali y dimensions), as i allows o a mo e meaning ul in e p e a ion o he ac o model. The i nal model wi h h ee quali y dimensions is p esen ed in Table 3. Fac o weigh s wi h ac o loadings abo e 0.3 and ac o s ha con ain mo e han h ee i ems we e e ained. We s a ed ou ac o analysis wi h se en quali y dimensions, as, ollowing Ko le ’s ma ke ing heo y, we ha e hypo hesized ha all se en quali y dimensions ha e a signi i can in l uence on gues s’ o e all quali y pe cep ion. Based on he o a ed ma ix o ac o weigh s shown in he able abo e, i is e iden ha gues s’ pe cep ion o he quali y o es au an o e ings is mainly based on he quali y o he ollowing ma ke ing ac o s: people (40.97%), placemen (5.05%), p oduc ( ood) and physical e idences (4.22%). Based on hese h ee quali y dimensions (see explana ion below) and he alues o hei o al explained a iances, i is clea ly e iden ha he quali y o people (s a ) has he g ea es impo ance (40.97%) in assu ing es au an quali y, ollowed by he quali y o placemen (5.05%) and he quali y o ood and physical e idences (4.22%). O he ma ke ing quali y dimensions a e, in ela ion o gues s’ assessmen o quali y, no s a is ically signi i can . Special a en ion should be paid o he hi d ac o , which is o med by me ging he wo dimensions o ood and physical e idences. As u he dissec ion o he esul s does no con ibu e o he imp o emen o he quali y o he esea ch, we ha e decided o keep he model wi h h ee main ac o s. These esul s clea ly do no suppo H1, as only h ee ma ke ing quali y dimensions ha e a s a is ically signi i can in l uence on gues s’ quali y pe cep ion. The ela ion be ween he heo e ical cons uc and he esea ch esul s is p esen ed in Figu e 1. On he le side, he 7P ma ke ing quali y dimensions a e p esen ed wi h ini ial 35 quali y ac o s, while on he igh side he esea ch esul s, wi h he i nal 3 dimensions and 23 ac o s a e p esen ed. Twel e ac o s had o be emo ed om he ac o analysis, as hey had low communali ies (< 0.5) and low ac o weigh s (< 0.3). Thus, he esul s o his s udy indica e ha gues s pe cei e es au an quali y based on only h ee ma ke ing quali y dimensions and 23 quali y ac o s (see also Tab. 3). Al hough he i ndings p esen ed abo e clea ly indica e non–cohe ency o pe cei ed quali y, we ha e decided o u he empi ically in es iga e he di e ences be ween di e en quali y dimensions. Based on he compa ison o mean alues p esen ed in Table 2, we s a is ically analysed he pe cei ed di e ences be ween all quali y dimensions. Pai ed s a is ics and wo-sided es s we e used o all possible pai s. S a is ically signi i can di e ences we e ound be ween all quali y dimensions ( he only excep ion is he compa ison be ween he mean alue o he dimensions o p oduc and physical e idences, o which he deg ee o signi i cance is 0.510). This indica es ha he o e ed quali y is pe cei ed as highly incohe en ; hus, H2 canno be con i med. Con a y o expec a ions, his s udy did no con i m he cohe ency o pe cei ed quali y in es au an s’ o e ings. Conclusion and Implica ions The pu pose o his s udy was o de e mine how di e en ma ke ing ac o s in l uence es au an gues s’ o e all quali y pe cep ion. EM_1_2017.indd 165EM_1_2017.indd 165 13.3.2017 16:59:0813.3.2017 16:59:08