Uni e sidade do Minho
Escola de Economia e Ges ão
Menna-Allah Mohamed Mos a a
Online Beha io al Ad e ising – The In luence
o Pe sonaliza ion, Consume
Cha ac e is ics, B and Familia i y, B and
T us , and T anspa ency on Consume s’
A i udes, In en ions and Beha io s
Online Beha io al Ad e ising – The In luence o Pe sonaliza ion, Consume Cha ac e is ics, B and
Familia i y, B and T us , and T anspa ency on Consume s’ A i udes, In en ions and Beha io s
Menna-Allah Mohamed Mos a a
U
Minho
|
2024
June 2024
Uni e sidade do Minho
Escola de Economia e Ges ão
Menna-Allah Mohamed Mos a a
Online Beha io al Ad e ising – The
In luence o Pe sonaliza ion, Consume
Cha ac e is ics, B and Familia i y,
B and T us , and T anspa ency on
Consume s’ A i udes, In en ions and
Beha io s
Doc o al Thesis In
Business Adminis a ion
Conduc ed unde he supe ision o :
P o esso a Ana Ma ia Soa es
June 2024
ii
DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS
Es e é um abalho académico que pode se u ilizado po e cei os desde que espei adas as eg as e boas
p á icas in e nacionalmen e acei es, no que conce ne aos di ei os de au o e di ei os conexos.
Assim, o p esen e abalho pode se u ilizado nos e mos p e is os na licença abaixo indicada.
Caso o u ilizado necessi e de pe missão pa a pode aze um uso do abalho em condições não p e is as
no licenciamen o indicado, de e á con ac a o au o , a a és do Reposi ó iUM da Uni e sidade do Minho.
A ibuição-NãoCome cial-SemDe i ações
CC BY-NC-
ND
h
p
s
:/
/
c ea
i e
c
o
mm
on
s
.o
g
/
lic
en
s
es
/
by
-
nc
-
n
d/
4.0/
iii
ACKNOWLEDGMENT
I would like o hank God, The G ea Lo d, wi hou you I would ha e achie ed no hing in my li e.
I would like o hank my dea supe iso , P o esso a Ana Ma ia Soa es, you ha e been he bes mo i a o ,
suppo e , and iend. You pa ience and conside a ion we e endless. Thank you o being he pe son you
a e. I will o e e che ish ou ela ionship.
I would also like o hank my Mom Amani Mos a a, o e e y hing you ha e done o me o e he pas
yea s, you a e he s onges pe son I know. I would like o hank my amily, Uncle Khalid Khalil, Mo he -
in-law Ma iam Hussein and My Husband Ahmed Moheib. You lo e and suppo we e wha li ed me up
du ing my da kes imes.
My F iends, Bassan Saeed, and Dina Yahia, how can I e e hank you enough! You made my PhD jou ney
a e y ema kable one. I am e y hank ul o you being in my li e.
I would like o ex end my hanks o all my p o esso s in he Ph.D. cou se, and all he eam membe s o
he School o Economics and Managemen , who we e always suppo i e and eadily a ailable whene e
I needed hem.
Finally, I would like o hank my G andpa Mos a a Khalil o e e y hing you e e did o me. You aised
me up like he dad I ne e had and always wan ed o see me a PhD holde , his is o you. May you es
in peace. Amen.
i
STATEMENT OF INTEGRITY
I he eby decla e ha ing conduc ed his academic wo k wi h in eg i y. I con i m ha I ha e no used plagia ism o any
o m o undue use o in o ma ion o alsi ica ion o esul s along he p ocess leading o i s elabo a ion.
I u he decla e ha I ha e ully acknowledged he Code o E hical Conduc o he Uni e si y o Minho.
RESUMO
A ecnologia pe mi e às emp esas comp eende melho os seus clien es-al o. A ualmen e, os anuncian es
podem isa os seus consumido es com base no conhecimen o eal de quem, quando, como e po que
é mais p o á el que comp em um p odu o especí ico. As emp esas u ilizam a publicidade
compo amen al online (Online Beha iou al Ad e ising - OBA) - ou seja, moni o iza o compo amen o
online dos consumido es pa a ap esen a anúncios pe sonalizados aos consumido es isados - pa a
aumen a o e o no a ual e u u o das emp esas. No en an o, a OBA ambém le a a um aumen o das
p eocupações com a p i acidade, descon iança e eação à comp a do p odu o ou à u ilização do se iço
que a emp esa o e ece.
A pe sonalização é conside ada a p incipal abo dagem à OBA, uma ez que ajuda a o na os anúncios
mais apela i os do pon o de is a pessoal. No en an o, a in es igação empí ica mos a esul ados
con adi ó ios no que espei a aos e ei os da pe sonalização e à o ma como es a pode a é causa
eacções e a i udes nega i as po pa e dos consumido es de ido ao aumen o das p eocupações com a
p i acidade.
É p opos a e es ada empi icamen e uma modelo conce ual que engloba o papel da pe sonalização, das
ca ac e ís icas do consumido , da amilia idade com a ma ca, da con iança na ma ca e da anspa ência
nas a i udes e in enções dos consumido es e no compo amen o p o e o em elação às ma cas. Além
disso, o es udo explo a a in luência mediado a da pe ceção de in usi idade, da pe ceção de p eocupação
com a p i acidade e da eac ância nas a i udes e in enções dos consumido es e no compo amen o de
p o eção, udo is o com base nos undamen os eó icos da
Social Exchange Theo y
.
Os esul ados mos am que alguns dos cons u os p opos os, nomeadamen e a pe sonalização, as
ca ac e ís icas do consumido , a amilia idade com a ma ca e a con iança na ma ca, in luenciam as
a i udes dos consumido es em elação à ma ca, as a i udes em elação ao OBA, as in enções de clica
no anúncio, as in enções de comp a o p odu o anunciado e o compo amen o de e i a o OBA. Além
disso, os esul ados con i mam que es a in luência é a e ada po uma sé ie de elemen os mediado es,
nomeadamen e a pe ceção da in usi idade pelos consumido es, a pe ceção das p eocupações com a
p i acidade e a pe ceção da ameaça à escolha.
Es es esul ados êm implicações eó icas e p á icas. No que espei a à eo ia, o es udo ap o unda a
comp eensão do concei o de OBA e da sua e icácia. A in es igação isou con ibui pa a os es udos
académicos e p opo ciona uma melho comp eensão da ques ão em es udo e da sua elação com a
eo ia das ocas sociais e a eo ia da eação psicológica. Além disso, o es udo o neceu alguns
conhecimen os impo an es sob e as p á icas de OBA pa a ajuda os p o issionais de ma ke ing a c ia
campanhas de OBA bem-sucedidas e a ul apassa as p eocupações associadas à OBA.
PALAVRAS-CHAVE: Publicidade compo amen al online, Pe sonalização, In usi idade pe cebida,
e i amen o de OBA, Social Exchange Theo y.
i
ABSTRACT
Technology enables businesses o be e unde s and hei a ge ed cus ome s. Nowadays, ad e ise s
can a ge hei consume s based on ac ual knowledge o who, when, how, and why hey a e mo e likely
o buy a speci ic p oduc . Businesses use Online Beha io al Ad e ising (OBA) – i.e. moni o ing
consume s’ online beha io in o de o p esen indi idually ailo ed ads o a ge ed consume s – o
inc ease business cu en and u u e e u ns, howe e , OBA also leads o inc eased p i acy conce ns,
dis us , and eac ance o buying he p oduc o using he se ice he business o e s.
Pe sonaliza ion is conside ed he main app oach o OBA since i helps make ads mo e pe sonally
appealing. Howe e , empi ical esea ch shows mixed esul s ega ding he e ec s o pe sonaliza ion and
how i can e en cause nega i e consume esponses and a i udes due o inc eased p i acy conce ns.
A concep ual amewo k encompassing he ole o pe sonaliza ion, consume cha ac e is ics, b and
amilia i y, b and us , and anspa ency on consume s’ a i udes and in en ions and p o ec i e beha io
on b ands is p oposed and empi ically es ed. Mo eo e , he s udy explo es he media ing in luence o
pe cei ed in usi eness, pe cei ed p i acy conce n, and eac ance on consume s’ a i udes and in en ions
and p o ec i e beha io , all based on he heo e ical ounda ions o he social exchange heo y.
The indings show ha some o he p oposed cons uc s, namely pe sonaliza ion, consume
cha ac e is ics, b and amilia i y, b and us in luence consume s’ a i udes owa ds he b and, a i udes
owa ds OBA, in en ions o click he ad, in en ions o pu chase he ad e ised p oduc , and OBA a oidance
beha io . Also, he indings suppo ed ha his in luence is a ec ed by se e al media o elemen s namely
consume s’ pe cei ed in usi eness, pe cei ed p i acy conce ns, and pe cei ed h ea o choice.
These indings ha e implica ions o heo y and p ac ice. Fo heo y, he s udy deepens he unde s anding
o he OBA concep and i s e ec i eness. The esea ch aimed o con ibu e o he academic schola ship
and gi e a be e unde s anding o he issue unde in es iga ion and i s ela ion o he Social Exchange
Theo y and Psychological Reac ance Theo y. Also, he s udy p o ides some impo an insigh s o OBA
p ac ices o help ma ke ing p o essionals c ea e success ul OBA campaigns and o e come conce ns
associa ed wi h OBA.
KEYWORDS: Online Beha io al Ad e ising, Pe sonaliza ion, Pe cei ed In usi eness, OBA A oidance,
Social Exchange Theo y.
ii
ACRONYMS
ATB – A i ude Towa ds he B and
ATO – A i ude owa ds OBA
AVE – A e age Va iance Ex ac ed
BF – B and Familia i y
BT – B and T us
CB-SEM – Co a iance-based S uc u al Equa ion Modeling
CC – Consume Cha ac e is ics
CMB – Common Me hod Bias
F2 – E ec Size
ICA – In en ion o Click he Ad
IPP – In en ion o Pu chase he P oduc
OBA – Online Beha io al Ad e ising
PB – P o ec i e Beha io
PER – Pe sonaliza ion
PIN – Pe cei ed In usi eness
PLS-SEM – Pa ial Leas Squa es S uc u al Equa ion Modeling
PRC – P i acy Conce ns
PRT - Psychological Reac ance Theo y
Q2
– P edic i e Rele ance
R2 – Coe icien De e minan s
REA – Reac ance
SEM – S uc u al Equa ion Modeling
SET – Social Exchange Theo y
SPSS – S a is ical Package o Social Science
TA – T anspa ency App oach
VIF – Va iance In la ion Fac o
2
1.1. In oduc ion
This chap e p esen s he heme unde s udy, he concep o online beha io al ad e ising (OBA) and he
a iables iden i ied in he li e a u e as a ec ing OBA, such as pe sonaliza ion, consume cha ac e is ics,
b and amilia i y, b and us , anspa ency, pe cei ed in usi eness, p i acy conce ns, eac ance,
consume a i udes and in en ions and p o ec i e beha io . The chap e includes he a ionale and scope
o he s udy de i ing om he gap iden i ied in ex an li e a u e. The esea ch backg ound suppo s he
eme gence o he gap and he de elopmen o he main esea ch ques ions and objec i es. Finally, his
chap e b ie ly add esses he expec ed con ibu ions o he hesis.
1.2. Resea ch Backg ound
In oday’s digi al wo ld, de elopmen s in echnology ha e enabled businesses o be e unde s and hei
a ge cus ome s. Businesses a e now able o iden i y cus ome s wi h pu chase in en ions, wha hey a e
likely o pu chase, epu chase o e en ecommend o o he s (Kuma and Gup a, 2016). Ad e ise s
moni o online consume s’ ac i i ies, collec hei pe sonal da a, and use sophis ica ed analysis
mechanisms o a ge cus ome s wi h indi idually ailo ed ads in o de o maximize hei cus ome s’ base
and acco dingly maximize business e u ns. Collec ed da a includes websi es isi ed, con en iewed (e.g.
a icles ead and ideos wa ched), in addi ion o sea ch engine que ies and online pu chases. This is
e e ed o as online beha io al ad e ising – OBA (Boe man e . al, 2017).
Mo eo e , unexpec ed changes in he wo ld, like he Co id-19 pandemic, esul ed in people s aying a
home and/o wo king emo ely om home, which esul ed in massi e g ow h in online a ic. The
consequence o his massi e g ow h was mo e oppo uni ies o online businesses o engage in online
digi al ma ke ing ac i i ies. The In e ac i e Ad e ising Bu eau epo ed a o al e enue om digi al
ad e ising o $209.7 billion be ween he yea 2021-2022, while i was epo ed a a o al o only $40.1
billion in he i s hal o he yea 2017 (IAB, 2022).
Ad e ise s as well as hi d pa y businesses (e.g. Google and Facebook) use consume s’ online ac i i y
his o y o in e consume s’ in e es s and demog aphics o ailo indi idually a ge ed ads based on hese
in e ences ha appea in he o m o sea ch esul s, social eeds, and ads (Maye and Mi chell, 2012).
Al aweel e al. (2012) s a ed ha he 100 mos popula si es collec mo e han 6,000 cookies, 83% o
which a e hi d-pa y cookies, and mo e han 350 collec ed by indi idual websi es. Roesne e al. (2012)
a i med ha se e al online acke s cap u e mo e han 20% o each use ’s b owsing beha io .
Fu he mo e, Engleha d e al. (2015) claimed ha ad e ise s can ec ea e 62-73% o each online use
3
b owsing his o y. In 2022, he In e na ional Associa ion o P i acy P o essionals eleased a epo s a ing
ha nowadays a ound 80% o use s’ b owsing his o ies a e cap u ed by an a e age o 177 di e en
online acke s only wi hin he i s wo hou s o b owsing he In e ne (IAPP, 2022). The p ac ice o OBA
in ol es he collec ion, use and sha ing o pe sonal da a. Howe e , since consume s ha e li le o no
knowledge o he ex en o which hei online beha io is acked o hei pe sonal in o ma ion is being
collec ed (U e al. 2012), i is said o aise p i acy conce ns and eeling o eac ance (Chen and S allae ,
2014). Reac ance s ems om Psychological Reac ance Theo y and can be de ined as a s a e whe e
consume s ejec some hing o ced on hem by beha ing opposi e o in ended (Tucke , 2014).
Pe sonaliza ion is conside ed a main base o OBA. Inc eased pe sonaliza ion inc eases ailo ed ad
ele ance which in e u n inc eases p oduc pu chases o se ice adop ion (Tucke , 2014), bu
pa adoxically i also inc eases p i acy conce ns abou being wa ched and how in o ma ion is being
collec ed, which may lead o a eeling o ulne abili y and eac ance. Tucke (2014) a gued ha al hough
pe sonalized ad e isemen s can be wice as e ec i e as impe sonalized ads, consume s end o
expe ience eac ance due o inc eased pe cei ed p i acy conce ns and pe cep ions o in usi eness. This
calls o he need o unde s anding o he impac o pe sonaliza ion, consume cha ac e is ics and
p i acy conce ns on consume s’ a i udes, in en ions and beha io (Agui e e al., 2015).
Also, how ad e ise s collec consume s’ da a – ei he o e o co e – has an e ec on he way
consume s pe cei e pe sonalized ads and e en ually on he way hey beha e owa ds hese ads. The
way ma ke e s collec consume ela ed da a online is e e ed o as anspa ency app oach (Boe man
e al., 2017). T anspa ency is belie ed o inc ease p i acy conce ns and eeling o in usi eness, i
consume s a e p esen ed by a pe sonalized ad ha con ains iden i iable pe sonal in o ma ion ha has
been collec ed wi hou hei consen (Milne e al., 2008). This calls o a be e unde s anding o he
e ec o anspa ency on OBA (Boe man e al., 2017).
B and amilia i y can be de ined as consume s’ knowledge abou a b and, de eloped h ough p e ious
in e ac ions wi h ha b and, i is also conside ed as a men al sho cu used by consume s o make a
pu chasing decision whe e b and amilia i y in luences consume s’ us (Phelan e al., 2016). Kuma
and Gup a (2016) claimed ha us is a men al s a e ha mo i a es consume s o accep a isky si ua ion
in ade o a posi i e ou come om ha si ua ion. This can be explained in he ligh o he Social
Exchange Theo y (SET). SET posi s ha conce ned pa ies in a social ela ionship conside he cos o
engaging in o he ela ionship and y o maximize hei gains and minimize hei cos s om he
exchanges in ha ela ionship (Eme son, 1976), he e o e, online consume s p e e o engage in
4
ansac ions wi h people hey us due o he lack o ace- o- ace communica ion in he online
en i onmen (Chou and Hsu, 2016)
Thus, al hough ad pe sonaliza ion causes a o able ou comes, i can also cause un a o able consume
esponses. A pe sonalized ad wi h ma ching consume p e e ences, can cause inc eased eelings o loss
o choice and manipula ion (Tucke , 2012). Also, since pe sonaliza ion is clea e idence ha a use ’s
beha io was acked and analyzed, i aises use s’ p i acy conce ns and eelings o in usi eness
(Gold a b and Tucke , 2011). I is he e o e no su p ising ha a la ge majo i y o online consume s end
o a oid OBA and/o de elop nega i e pu chase in en ions owa ds b ands using OBA (Guild, 2013).
Re e ing o wha was p e iously men ioned, p e ious li e a u e is no unanimous o whe he a
pe sonalized message is always mo e e ec i e han a s anda dized one (Noa e al., 2009). The e also
appea s a dea h o s udies add essing how di e en consume cha ac e is ics a ec consume s’
pe sonal p i acy conce ns, eelings o in usi eness and h ea s o choices (Boe man e al., 2017). All
he p e ious in e wind wi h how much he ad e ised b and is us ed, amilia o consume s’ memo y
and ecall, he anspa ency app oach used by he ad e ise o collec consume s’ in o ma ion and how
hese ac o s e en ually a ec consume s’ a i udes and beha io s owa ds he ad message. Hence
exis ing li e a u e lacks comp ehensi e s udies on he e ec o pe sonaliza ion, di e en consume
cha ac e is ics, b and- ela ed ac o s and anspa ency. This gap highligh s he need o u he s udies
illumina ing he usage o OBA and i s e ec s.
1.3. Resea ch P oblem, Ques ions, Objec i es and Con ibu ion
The esea ch in hand aims o explo e he ole o some speci ic a iables on consume s’ beha io al
ou comes o OBA. The esea ch p oblem can be summa ized in he ques ion o “Wha a e he ac o s
ha in luence consume s' a i udes, in en ions, and beha io s in he online en i onmen owa ds an
online beha io ally a ge ed ad e isemen ? And wha ac o s can media e such e ec ?”.
Speci ically, his esea ch aims o explo e he ole o pe sonaliza ion, consume cha ac e is ics, b and
amilia i y, b and us , and anspa ency on a i udes owa d OBA, a i udes owa d he b and, in en ion
o click on he ad, in en ion o pu chase he ad e ised p oduc and inally on OBA a oidance. Mo eo e ,
he s udy also explo es he media ing in luence o pe cei ed in usi eness, pe cei ed p i acy conce ns,
and eac ance on consume s’ a i udes, in en ions and beha io al ou comes.
Thus, his esea ch aims o answe he ollowing ques ions:
5
1. How do pe sonaliza ion, consume cha ac e is ics, b and amilia i y, b and us , and
anspa ency in luence consume s’ a i udes, in en ions and OBA a oidance in he online
en i onmen ?
2. How do pe cei ed in usi eness, p i acy conce ns, and eac ance media e such in luence?
The inding o his s udy is expec ed o ad ance he knowledge o OBA on wo sides. Fi s , i aims a
con ibu ing o he academic schola ship, o gi e a be e unde s anding o he issue unde in es iga ion
– OBA and o he a iables – and i s ela ion o he Social Exchange heo y and Psychological Reac ance
Theo y. Ou esea ch adds o he body o knowledge on how hese heo ies could be used o desc ibe
consume s’ a i udes and beha io s owa d OBA. Also, his s udy deepens he unde s anding o he OBA
concep and i s e ec i eness. Thus, his esea ch p o ides use ul in o ma ion o u u e schola s o build
on in hei coming esea ch.
Second, his esea ch has impo an p ac ical implica ions o b ands, ma ke ing, and communica ion
manage s. This esea ch p o ides some impo an insigh s in o he online beha io al ad e ising
p ac ices. These insigh s a e expec ed o help ma ke ing p o essionals in c ea ing success ul OBA
campaigns and o e come conce ns and isks associa ed wi h online beha io al ad e ising p ac ices.
1.4. Thesis S uc u e
The o e all s uc u e o his disse a ion akes he o m o i e chap e s. A e he p esen Gene al
In oduc ion chap e , Chap e wo p esen s he Li e a u e Re iew ha suppo s he heo e ical a gumen s
o he hesis. This chap e add esses he p ima y cons uc s o he s udy, based on he p e ious
heo e ical backg ound, he end o he chap e p esen ed he esea ch model and he esea ch
hypo heses. Chap e h ee in oduces he esea ch me hodology pa . A e p esen ing he esea ch
design, he da a collec ion me hods o he empi ical s udy (quan i a i e me hod) a e discussed. Chap e
ou p esen s he esul s and a discussion abou he main indings. Finally, chap e i e discusses he
esea ch's main conclusions and poin s ou he cu en s udy's limi a ions and hough s ha may assis
in u u e esea ch. The hesis ou line is p esen ed in he ollowing igu e (Figu e 1).
6
Figu e 1: Thesis S uc u e
7
LITERATURE REVIEW
8
2.1. In oduc ion
The Li e a u e Re iew chap e consis s o h ee main sec ions. The i s sec ion discusses one o he
p ima y concep s o he cu en s udy, Online Beha io al Ad e ising (OBA). I s a s wi h explo ing he
gene al concep o online beha io al ad e ing, b oadening he concep by add essing di e en OBA
de ini ions, echniques and componen s as s a ed in pas esea ch. The second sec ion in oduces he
heo e ical amewo k o he s udy add essing he Social Exchange Theo y and he Psychological
Reac ance Theo y.
The hi d sec ion concen a es on he o he p ima y concep s o he s udy, which we e ound o ha e a
di ec /indi ec e ec on OBA in p e ious li e a u e. De ini ions o he concep s o Pe sonaliza ion,
Consume Cha ac e is ics, B and Familia i y, B and T us , T anspa ency, Pe cei ed In usi eness,
P i acy Conce ns, Reac ance, Consume A i udes and In ensions, and P o ec i e Beha io s a e
p esen ed as s a ed in p e ious li e a u e and hen he links be ween hose concep s and he concep o
OBA a e discussed.
Finally, he ou h sec ion is de o ed o de eloping and p esen ing he heo e ical amewo k o he
esea ch, he li e a u e e iew chap e hen ends wi h he de eloped hypo heses o he cu en s udy.
2.2. The Concep o OBA
Online beha io al ad e ising (OBA) can be simply de ined as “ he p ac ice o acking an indi idual’s
online ac i i ies in o de o deli e ad e ising ailo ed o he consume ’s in e es s” (Fede al T ade
Commission, 2009). Since OBA has become manda o y o many businesses o su i e in he online
ma ke nowadays, mos o he la ge po als (ex. Amazon), sea ch si es (ex. Google AdSense), and social
ne wo king si es (ex. Facebook) ha e c ea ed hei own OBA se ices (Kim and Huh, 2017). Mo eo e ,
almos all ee web se ices ely on mone izing pe sonal in o ma ion h ough OBA (Mikians e al., 2012).
Online beha io al ad e ising is o en e e ed o as “Re a ge ed Ad e ising” because i has some unique
cha ac e is ics ha di e i om any o he ype o a ge ed ads (Bou ial, 2015). Gold a b (2014) s a ed
ha OBA e e s only o ad e ising ha is based on consume s pas online beha io s and in e ac ions
wi h websi es o sea ch engines, while a egula a ge ed ad e isemen can be based on gene al
consume cha ac e is ics such as demog aphics (age, gende , educa ion o employmen s a us) o
loca ion.
The a ailabili y o Web Analy ics and Big Da a mining echniques has d ama ically enhanced he le el o
indi idualized consume a ge ing, hese echniques enabled ad e ising i ms o p edic wha hei
9
cu en and po en ial u u e cus ome s a e in e es ed in buying based on hei pas online sea ch
beha io s (Kim and Huh, 2017). Ad e ise s, agg ega o s and ad-ne wo ks can collec consume
in o ma ion o e ime while consume s’ b owse he web and build consume s’ p o iles based on hei
o e all ac i i y. These p o iles can hen be used o show each indi idual consume ad e ising messages
ma ching hei in e es s mo e closely; ins ead o wha hey migh be sea ching o , which e en ually
would lead o highe click- h ough a es and in u n, highe e enues o he ad e ise (Ca ascosa e al.,
2015).
While OBA has p o ed i s impo ance as a echnique o inc ease he e ec i eness o online ad e ising,
i was also claimed ha OBA aised some consume s’ p oblems including bu no limi ed o p i acy
conce ns, pe cei ed h ea o choice and pe cei ed in usi eness (Ca ascosa e al., 2015). Kim and Huh
(2017) claimed ha OBA can be seen as double-edged swo d, wi h a possibili y o ha ing ei he a o able
o un a o able impac s on consume esponses (Bleie and Eisenbeiss 2015).
In he nex h ee sec ions, di e en OBA de ini ions om p e ious li e a u e a e discussed, he echniques
o how OBA wo ks, and he ypes o da a collec ed a e men ioned.
2.2.1. OBA De ini ions
The e a e many de ini ions o OBA, which can be also e e ed o as online p o iling, beha io al a ge ing
(Benne , 2010) and e a ge ed ad e ising (Bou ial, 2015).
The wo k o Boe man e al. (2017) ci ed many de ini ions o OBA including he de ini ion McDonald and
C ano (2010:2) ha de ined OBA as “
he p ac ice o collec ing da a abou an indi idual’s online ac i i ies
o use in selec ing which ad e isemen o display
”, Smi e al. (2014:15) de ini ion o OBA as “
adjus ing
ad e isemen s o p e ious online su ing beha io
”, and Ham and Nelson (2016:690) de ini ion as “
a
echnology-d i en ad e ising pe sonaliza ion me hod ha enables ad e ise s o deli e highly ele an
ad messages o indi iduals
”. Table (1) below sums up de ini ions om p e ious li e a u e. Al hough OBA
de ini ions a e no consis en ac oss p e ious li e a u e, Boe man e al. (2017) no ed ha hese
de ini ions sha e wo signi ican aspec s: (1) Moni o ing o consume s’ online beha io and (2) he use
o he collec ed da a o gene a e pe sonally a ge ed ads. Thus, Boe man e al. (2017:364) in oduced a
new de ini ion o OBA as:
“The p ac ice o moni o ing people’s online beha io and using he collec ed
in o ma ion o show people indi idually a ge ed ad e isemen s”.
10
Table 1: Online Beha io al Ad e ising De ini ions.
OBA De ini ion
Au ho /Yea
The p ac ice o acking an indi idual’s online ac i i ies in o de o
deli e ad e ising ailo ed o he consume ’s in e es s.
(Fede al T ade Commission,
2009)
The p ac ice o collec ing da a abou an indi idual’s online
ac i i ies o use in selec ing which ad e isemen o display.
McDonald and C ano
(2010:2)
Adjus ing ad e isemen s o p e ious online su ing beha io .
Smi e al. (2014:15)
A echnology-d i en ad e ising pe sonaliza ion me hod ha
enables ad e ise s o deli e highly ele an ad messages o
indi iduals.
Ham and Nelson (2016:690)
The p ac ice o moni o ing people’s online beha io and using he
collec ed in o ma ion o show people indi idually a ge ed
ad e isemen s.
Boe man e al. (2017:364)
Sou ce: Own Elabo a ion
2.2.2. OBA Techniques
OBA in ol es acking cus ome s’ online ac i i ies and associa ing hese ac i i ies wi h a speci ic
compu e o de ice (Be ge , 2010). Fi ms who employ his ype o ad e ising can use mul iple
app oaches o collec consume s’ in o ma ion, including cookie-based app oach, spywa e-based
app oach, deep packe inspec ion-based app oach (Be ge , 2010), i s pa y model app oach and
b owse inge p in ing app oach (Le ne e al., 2016). A desc ip ion o hese app oaches is p o ided in
( able 2).
Using he p e iously men ioned echniques, Engleha d e al. (2015) claimed ha companies ha e he
abili y o econs uc 62-73% o use s’ b owsing his o y. Al hough beha io al ad e ise s do no di ec ly
collec pe sonally iden i ying da a like a consume ’s name o physical add ess; Zhonghao e al. (2016)
epo ed ha 78% o si es con ained acke s ha a emp ed o ans e unsa e pe sonal da a, hese da a
make i possible o iden i y he eal consume ’s iden i y in he o line wo ld (Ba ba o e al., 2006).
11
Table 2:Online Beha io al Ad e ising Techniques and App oaches
Technique Name
How i wo ks
Cookie-based app oach
A cookie is a named piece o da a ha a isi ed websi e eques s
he use ’s web b owse o accep and keep sa ed. Then, each
ime he consume uses he web b owse o isi he cookie
owne websi e, he consume ’s web b owse sends he con en
o he cookie back o he owne websi e wi h he sa ed
consume online beha io in o ma ion (Be ge , 2010).
Spywa e-based app oach
Re e s o willingly o unwillingly, ins alling a acking so wa e
di ec ly o he consume ’s compu e o ack and eco d he
consume da a and beha io on he In e ne . The ins alla ion
can happen h ough he In e ne Se ice P o ide (ISP) o e en
a websi e he consume isi s (Be ge , 2010).
Deep Packe Inspec ion-based
app oach (DPI)
Using his app oach, he consume ’s ISP ins alls powe ul
ha dwa e de ices o examine all o he a ic going in and ou
o he consume s’ compu e s, which will be la e used o
compile in o ma ion and sell his in o ma ion o ma ke e s o
c ea e addi ional e enues. DPI is p ac ically impossible o s op
as i does no equi e any so wa e o be ins alled on he
consume ’s compu e , no would an an i-spywa e p og am be
able o easily de ec i (Be ge , 2010).
Fi s Pa y Model app oach
Re e s o a ge ing based on da a collec ed a and by a single
websi e. Fi s Pa y a ge ing does no pose a h ea o
consume p i acy as i does no in ol e he sha ing o da a
wi h hi d pa ies o ac oss mul iple websi es (Be ge , 2010).
B owse Finge p in ing app oach
Re e s o a s a eless acking echnique ha uses consume ’s
de ice con igu a ion in o ma ion unp o ec ed by he b owse
h ough Ja aSc ip APIS and HTTP heade s. B owse
18
2.8. T anspa ency
T anspa ency e e s o he way ma ke e s collec consume ela ed da a online, o he pu pose o
p esen ing consume s wi h pe sonalized ads associa ed wi h hei collec ed da a. T anspa ency app oach
includes bo h o e and con e da a collec ion echniques (Boe man e al., 2017).
An o e da a collec ion app oach means in o ming consume s ha hei da a is being collec ed. The
undamen al assump ion he e is ha a e i ms ha e in o med consume s abou hei da a collec ion,
he con inued use o he se ices g an s a consen o u he da a collec ion by he i m in he u u e
(Sunda and Ma a he, 2010). While a co e da a collec ion app oach occu s when i ms does no in o m
use s abou hei da a being collec ed, ins ead ma ke e s disc ee ly collec and s o e consume da a
while consume s su he Web (Mon gome y and Smi h, 2009). Ad oca es o co e da a collec ion
echniques claim hey a e mo e bene icial o he i ms, as hey enable hem o collec unbiased da a,
and gain a deepe unde s anding o hei cus ome s (Ve hoe e al. 2010). On he o he hand, i also
bene i s consume s, by no dis u bing he low o hei online su ing expe ience (Milne e al., 2008).
Al hough In e ne use s gene ally know ha i ms a e collec ing hei in o ma ion, being exposed o an
ad message ha con ains ecognizable pe sonal in o ma ion signals cus ome s ha hei in o ma ion has
been collec ed, wi hou hei consen . This esul s in nega i e consume s' eac ions and inc eased
consume s' discom o , leading o lowe esponse a es o he ad message (Milne e al., 2008). This calls
o a be e unde s anding o he e ec o anspa ency on OBA (Boe man e al., 2017).
Fi ms seek o gain consume us h ough cues ha show hei anspa ency and eliabili y (Kim and
Kim 2011). Al e na i ely, p i acy laws also equi e companies o be anspa en abou hei da a
collec ion p ac ices (EU Da a P o ec ion Di ec i e, 1995; EU Gene al Da a P o ec ion Regula ion, 2016).
These e o s can be ealized in he inclusion o p i acy s a emen s on a ma ke e 's websi e. A p i acy
s a emen is a documen ha e eals which pe sonal da a was collec ed h ough he websi e, as well as
how and why (Milne e al., 2008). Howe e , p i acy s a emen s a e no o en ead due o he complexi y
o hei language and hei leng h, cus ome s o en ei he ag ee o igno e hem (McDonald and C ano ,
2010; Milne e al., 2006).
Addi ionally, in o de o imp o e anspa ency p ac ices and inc ease consume s’ accep ance o OBA,
he Digi al Ad e ising Alliance (DAA, 2011) de eloped a se o icons o be shown on ad e isemen s
ailo ed by i s membe s. These icons (e.g. he s anda d icon and an “as e isk man” icon) we e de eloped
o explain OBA mechanisms and o p o ide op ou op ions o consume s (Sama e al., 2017). Howe e ,
he wo k o Leon e al. (2012) showed how pa icipan s misunde s ood o a ely ecognized disclosu e
19
icons, 53% o hei pa icipan s emembe ed OBA disclosu e icons and only 12% emembe ed seeing a
agline “AdChoices”. Mo eo e , he agline was inco ec ly unde s ood o in end selling ad e ising space
ins ead o being a link o pages whe e hey can make choices abou OBA. Van Noo e al. (2013)
sugges ed o complemen he s anda d OBA icon wi h a label s a ing “
This ad is based on you su ing
beha io
” in o de o inc ease OBA awa eness.
2.9. Pe cei ed In usi eness
Li e al. (2002:39) de ined in usi eness as
“a psychological eac ion o ads ha in e e e wi h a
consume ’s ongoing cogni i e p ocessing”
.
Pe cei ed in usi eness is ela ed o he pe cei ed alue o
he ad message (Ying e al., 2009). This can be explained in he ligh o he Social Exchange heo y,
when consume s weigh he bene i s o an ad message mo e han he h ea s i possesses, hey pe cei e
i as less in usi e (Rony, 2018). Howe e , consume s’ pe cep ions o OBA appea o be mixed; while
some see e a ge ed ads as bene icial (McDonald and C ano 2010), o he s ind hem in usi e and
c eepy (Smi e al., 2014). McDonald and C ano (2010) epo ed ha some consume s ac ually change
hei online beha io when hey know hei da a a e being collec ed.
In usi eness was add essed in p e ious li e a u e in ela ion o OBA. Li e al. (2002) a gued ha ad
messages wi h high le el o pe sonaliza ion a ec s consume s’ cogni i e abili ies and can induce in usi e
eelings owa ds he ad, Van Doo n and Hoeks a (2013) discussed he nega i e consume a i udes
associa ed wi h eeling o in usi eness, gene a ed by ads based on consume s pas online beha io s.
Zhao e al. (2017) claimed ha ad con en , b and amilia i y, and b and lo e could lead consume s o
expe ience he ad as less in usi e.
2.10. P i acy Conce ns
P i acy has a mul idimensional meaning, ha depends on he con ex in which i is de ined and used.
Wes in (1967:7) de ined p i acy as “
he abili y o he indi iduals o con ol he e ms unde which
pe sonal in o ma ion is acqui ed and used
”. In ma ke ing li e a u e, a p i acy conce n is de ined as “
he
deg ee o which a consume is wo ied abou he po en ial in asion o he igh o p e en he disclosu e
o pe sonal in o ma ion o o he s
” (Baek and Mo imo o, 2012:63). Rognehaugh (1999:125) de ined
in o ma ion p i acy as “
use s’ igh s o keep in o ma ion abou hemsel es om being disclosed o o he s
[ma ke e s and o he unknown people]
”.
In he online en i onmen , in o ma ion p i acy became a c i ical opic because o he collec ion, s o ing,
and use o massi e amoun s o pe sonal da a (Okazaki e al., 2009). OBA ends o gene a e high le els
20
o p i acy conce ns, compa ed o o he o ms o online ad e ising. When a e a ge ed ad message
con ains ex ensi e consume ela ed unique in o ma ion, i ends o ac i a e conce ns o e p i acy,
esul ing in a nega i e consume eac ion owa ds OBA (Jai e al., 2013). When consume s become
conce ned abou hei pe sonal da a, hey end o emo e hei in o ma ion (Sheehan and Hoy, 1999),
o hey e en migh p o ide incomple e and inaccu a e da a abou hemsel es, leading o ine icien
a ge ing and was ed ad e ising e o s (Awad and K ishnan, 2006).
Empi ical esea ch add essed p i acy conce ns and p o ided e idence o i s nega i e in luence on
consume a i udes and in en ions owa ds OBA. Baek and Mo imo o (2012) s a ed how p i acy conce ns
inc eased ad e ising skep icism and e en ually ad a oidance. Van Doo n and Hoeks a (2013) showed
a dec ease in consume s pu chase in en ions as a esul o high p i acy conce ns. Bleie and Eisebeiss
(2015) discussed how consume s wi h high p i acy conce ns showed lowe ole ance owa ds
pe sonalized ad messages. Agui e e al. (2015) epo ed huge declines in click h ough a es a e
consume s ealized hei da a was co e ly collec ed o deli e pe sonalized banne s, while Smi e al.
(2014) epo ed ha consume s wi h less p i acy conce ns showed a mo e posi i e a i ude owa d a
pe sonalized ad message. Also, Phelan e al. (2016) claimed ha some ac o s such as us and b and
lo e can in luence consume s’ p i acy conce ns le els.
In a ecen s udy by Sama e al. (2017:303), he esul showed p i acy conce ns as a main eason o
a oiding OBA, he au ho s concluded a wide ange o nega i e opinions owa ds a ge ed ads, and
epo ed he second mos common eason obse ed om a hypo he ical scena io was “
because i eels
like an in asion o my p i acy.
”
Howe e , despi e such empi ical e idences abou p i acy conce ns, consume s’ ac ual beha io is, in
many cases, no consis en wi h o e en opposi e o hei a i ude. Con adic o y o wha migh seem
logical, Acquis i e al. (2013) ound ha people who claim high p i acy conce ns ag ee o disclose hei
pe sonal in o ma ion o small in ensi es such as ee se ices o coupons in wha is known as ‘p i acy
pa adox’ (Boe man e al., 2017).
2.10.1. P i acy Pa adox
P io esea ch, such as he wo k o Dha and Va shney (2011) and Su an o e al. (2013), used he e m
‘pe sonaliza ion-p i acy pa adox’ o desc ibe he con adic o y eelings consume s expe ience be ween
hei desi e o pe sonaliza ion and hei inc eased p i acy conce ns and need o p o ec hei pe sonal
in o ma ion. Smi h e al. (2011:993) desc ibed he phenomenon as “
despi e epo ed high p i acy
conce ns, consume s s ill eadily submi hei pe sonal in o ma ion in a numbe o ci cums ances
”. The
21
concep o p i acy pa adox s ems om he idea ha indi iduals do no hing o p o ec hei p i a e
in o ma ion, hough hey a e awa e ha he online en i onmen possess a h ea o hei p i acy (No be g
e al., 2007).
The P i acy Calculus Model, is buil upon a heo y called “Calculus o Beha io ”, de eloped by Lau e
and Wol e (1977). The heo y sugges s ha consume s’ pe sonal in o ma ion disclosu e beha io is
in luenced by hei pe cep ions abou he bene i s hey expec om ha disclosu e beha io . The e o e,
consume s a e only expec ed o sha e hei pe sonal in o ma ion online, i hey pe cei e he bene i o
sha ing such in o ma ion - equal o o g ea e han - he isk o he disclosu e beha io (Culnan and Bies,
2003).
The P i acy Calculus Model is also oo ed o he Social Exchange Theo y (Schumann e al., 2014), whe e
people e alua e social exchanges in e ms o cos s and bene i s, and a e only expec ed o join he
exchange when bene i s exceed he cos s (Schumann e al., 2014).
2.11. Reac ance
The concep o eac ance s ems om he Psychological Reac ance Theo y (PRT) (B ehm, 1966; B ehm
and B ehm, 1981). F om a consume esea ch pe spec i e, Reac ance “
desc ibes a s a egy ha
consume s use o a oid complying wi h a pe suasion a emp bu does no di ec ly cha ac e ize he cause
o discom o associa ed wi h a p i acy in asion
” (Tucke 2012:2).
Boe man e al. (2017:367) a gued ha “
Highly pe sonalized ads lead people o pe cei e a loss o choice,
con ol, o owne ship, and hus cause nega i e eelings and esponses
”. Thus, al hough consume s migh
ind use ulness in a pe sonalized ad message, he highly pe suasi e message migh also igge eelings
o eac ance, which leads o un a o able a i udes and nega i e beha io al ou comes owa ds he ad
message o he ad e ised b and (Bleie , & Eisenbeiss, 2015; Van Noo e al., 2014).
P e ious consume esea ch ound ha highly pe sonalized ads lead consume s o eel loss o hei
pe cei ed choice and inc eased eelings o being manipula ed (Tucke , 2012). Whi e e al. (2008),
claimed ha highe le els o pe sonaliza ion wi h no jus i ica ion o why all hese consume in o ma ion
has been used, is conside ed way less e ec i e and gene a es mo e eac ance eelings, han when a
jus i ica ion is p o ided. Leon e al. (2012) s a ed ha i use s el loss o con ol o e hei da a h ough
a a ge ed ad message, hey will e alia e by a oiding he exac p oduc s shown o hem in he a ge ed
ad e isemen . Tucke (2014) ound a link be ween in oducing p i acy con ols and how hey help o
educe eac ance owa ds pe sonalized ad e isemen s.
22
2.12. Consume A i udes and In en ions
A i udes can be de ined as gene al eac ions – usually bo h cogni i e and a ec i e – owa ds a s imulus
(Haug ed and Kasme , 2008). Rony (2018) claimed ha a i ude is a pe son’s logical and emo ional
e alua ion o some hing ha is o med a e exposu e o di e en si ua ions. In en ions, on he o he
hand; can be de ined as “
he mo i a ional ac o s ha in luence a beha io ; hey a e indica ions o how
ha d people a e willing o y, and how much o an e o hey a e planning o exe o pe o m he
beha io
” (Ajzen, 1991:181).
A i udes can be ei he posi i e o nega i e, hey a e o med and changed egula ly unde ce ain
ci cums ances (Haug ed and Kasme , 2008). Pas esea ch ecognized a i ude-beha io consis ency
in many s udies; hus, i can be said ha a i udes can p edic beha io s (C ano and P islin, 2006). Ajzen
(2008) claimed ha in en ions we e also a di ec cause o human beha io al ou come.
P e ious esea ch shows a ied consume s’ a i udes and in en ions owa d OBA, Lee e al. (2015),
discussed how OBA igge s a ec i e esponses such as p i acy conce ns and eac ance, which
e en ually a ec s he beha io al ou comes o a ge ed consume s (e.g., lowe in en ions o click he ad).
The schola s u he mo e, discussed how o he ac o s such as pe sonaliza ion, us , le el o p i acy
conce ns, and anspa ency mode a es he e ec o OBA on consume s’ beha io and pu chase
in en ions.
O he se e al s udies also es ablished how he le el o pe sonaliza ion in OBA in luences click- h ough
in en ions and click- h ough a es. Van Doo n and Hoeks a (2013), discussed how highe le els o
pe sonaliza ion in an ad message, had led o an inc eased eelings o in usi eness and hus nega i ely
a ec ed he pu chase in en ions. Tucke (2014) s udied Facebook ads, and showed how a Facebook ad
based on a pe son’s in e es gene a ed highe click- h ough a es, han hose o an ad based on a
pe son’s pe sonal cha ac e is ics. Bleie and Eisenbeiss (2015) ound ha banne ads wi h high le el o
pe sonaliza ion (e.g. showing i ems om consume s’ shopping ca ) inc eased click- h ough a es
compa ed o ads wi h a lowe le el o pe sonaliza ion bu only wi hin he bounda ies o a us ed e aile
o b and.
Fu he mo e, se e al s udies es ablished how anspa ency and consume awa eness o OBA guide
consume s’ a i udes and in en ions owa ds online beha io al ads. Agui e e al. (2015) discussed how
o e da a collec ion mechanisms inc eased click- h ough a es compa ed o co e da a collec ion
mechanisms. The schola s also discussed how he inclusion o an OBA icon posi i ely inc eased us in
he ad e ise , and he ad e ised b and and e en ually inc eased he ad e ising ou comes.
23
Consume s’ a i udes and unde s anding o OBA has as well been add essed in a numbe o s udies, o
example, mul iple s udies discussed how people wi h lowe p i acy conce ns le els end o be mo e
accep ing owa ds OBA (Miyazaki, 2008; Baek and Mo imo o, 2012; Smi e al., 2014). Lamb ech and
Tucke (2013) showed how OBA gene a ed be e beha io al ou comes and pu chase in en ions when
consume had na owly cons ued p e e ences. Leon e al. (2013) a gued ha consume s only wan o
sha e ce ain amoun o pe sonal in o ma ion bu a e uncom o able sha ing he es . Simila ly, Yao e
al. (2017) showed ha consume s ca ed abou he ypes o pe sonal in o ma ion collec ed mo e han
hey we e conce ned wi h who was collec ing i . Plane e al. (2017) ound use s we e mo e conce ned i
an ad was a ge ed based on demog aphic in o ma ion, such as age, gende , o ace, han based on
in e es s.
2.13. P o ec i e Beha io
Smi e al. (2014) s a ed ha p o ec i e beha io depends mainly on a ying consume cha ac e is ics
whe e consume s wi h mo e p i acy conce ns p ac ice inc eased p o ec i e beha io o p o ec hei
online p i acy. Howe e , he wo k o Leon e al. (2012) shows ha people ha e p oblems p o ec ing hei
online p i acy as hey do no unde s and o know how o use he a ailable p o ec i e ools (e.g. ools ha
block access o ad e ising websi es) o op ou OBA. Only ew consume s dele e cookies and sea ch
his o y in an a emp o p o ec hei online in o ma ion (McDonald and C ano 2010).
Empi ical esea ch also men ioned o he ac o s ha can explain OBA accep ance and a oidance. Baek
and Mo imo o (2012) discussed how p i acy conce ns and ad i i a ion leads o a oidance o OBA due
o inc eased ad skep icism. The schola s also discussed how pe cei ed pe sonaliza ion can lead o less
ad a oidance. Jai e al. (2013) explained how anspa ency abou consume s’ da a collec ion echniques
and easons in luences OBA accep ance and a oidance, whe e in o ming consume s hei da a was being
sha ed wi h hi d-pa y websi es led o lowe epu chase in en ions compa ed o sha ing hei da a
in e nally wi hin he same websi e. In addi ion, Schumann e al. (2014) indings showed ha consume
belie ed ha ecei ing a ee web se ice is an accep able ade-o o hei collec ed in o ma ion.
2.13. Hypo hesis De elopmen
In he pas wo decades, he e ec o pe sonaliza ion has been widely examined in many p io s udies
ac oss di e en domains such as ma ke ing, mass communica ion, in o ma ion echnology, and public
heal h (Li and Kalyana aman, 2013). Howe e , he e ec i eness o pe sonalized messages compa ed o
s anda dized ones emains ambiguous and unclea in he exis ing li e a u e (Noa e al., 2009). The
24
e ec o pe sonaliza ion was di e en and some imes e en opposi e in p io esea ch, whe e some
esea ch con i med he e ec i eness o pe sonaliza ion (e.g., Beam and Kosicki, 2014; Ha and Janda,
2014), while o he s udies showed he e was no signi ican di e ence be ween pe sonaliza ion and
s anda diza ion (e.g., Li, 2016; Po e and Whi comb, 2003). Pe sonaliza ion has been ea ed as a
double-edged swo d.
Resea che s, such as, Li (2016) ad ised esea che s o u he in es iga e why he e ec o
pe sonaliza ion looked so unp edic able, andom and uns able. This calls o he i s hypo hesis o he
s udy:
➢
H1: Pe sonaliza ion has a posi i e impac on a) Consume s’ a i udes owa ds he b and, b)
Consume s’ a i udes owa ds OBA, c) Consume s’ in en ion o click on he Ad, d) Consume s’
in en ions o pu chase he p oduc and e) Consume s’ p o ec i e beha io .
Some p e ious s udies showed ha consume cha ac e is ics had an e ec on pe cep ions, a i udes and
beha io owa ds OBA, esea che s poin ed wha is called “chilling e ec ”, whe e online use s change
hei online beha io when hey know hei da a is being collec ed (McDonald and C ano , 2010).
Howe e , he a ailable s udies a e sca ce, and mos p e ious s udies did no s udy he condi ion o
implemen ing app op ia e online pe sonaliza ion s a egies, al e ed based on consume cha ac e is ics
(Smi h e al., 2011). This calls o he second hypo hesis o he s udy:
➢
H2: Conside ing consume cha ac e is ics has a posi i e impac on a) Consume s’ a i udes
owa ds he b and, b) Consume s’ a i udes owa ds OBA, c) Consume s’ in en ion o click on
he Ad, d) Consume s’ in en ions o pu chase he p oduc and e) Consume s’ p o ec i e
beha io .
P e ious li e a u e sugges s ha b and amilia i y a ec s consume a i udes and beha io s, i also ac s
as a sho cu o online pu chase decision making o online consume s. Familia i y also has an e ec
online consume s’ us in he ad e ise (Shoenbe ge and Tho son, 2014). This suppo s he hi d
hypo hesis o he s udy:
➢
H3: B and Familia i y has a posi i e impac on a) Consume s’ a i udes owa ds he b and, b)
Consume s’ a i udes owa ds OBA, c) Consume s’ in en ion o click on he Ad, d) Consume s’
in en ions o pu chase he p oduc and e) Consume s’ p o ec i e beha io .
P e ious esea ch poin ed ou how unde s anding he dynamics o consume us was c ucial in he
online en i onmen , due o he lack o ace- o- ace ansac ions, hus sugges ing ha online consume s
25
a e mo e likely o engage wi h i ms o b ands hey us (Rony, 2018). In ac , p e ious esea ch on us
mos ly ocused on ins i u ional us , his us ends o be c ea ed based on he epu a ion o he company
o he b and (Keh e al., 2015). Kuma and Pansa i (2016) claimed ha when consume s us he
ad e ising i m o he b and, hey will be mo e willing o sha e hei p i a e in o ma ion, which hen can
be used o c ea ing an e ec i e online ad e ising campaign, ha would e en ually gene a e be e
e enues o he i m. This suppo s he ou h hypo hesis o he s udy:
➢
H4: B and us has a posi i e impac on a) Consume s’ a i udes owa ds he b and, b)
Consume s’ a i udes owa ds OBA, c) Consume s’ in en ion o click on he Ad, d) Consume s’
in en ions o pu chase he p oduc and e) Consume s’ p o ec i e beha io .
P e ious li e a u e showed ha online consume s app ecia ed hones y, and wan ed o be in o med abou
he collec ion, usage, and sha ing o hei pe sonal da a (Tu ow e al. 2009). Da a collec ion app oaches,
in addi ion o, implemen ing anspa ency mechanisms we e ound o a ec eelings o ulne abili y,
which in u n a ec s consume s’ a i udes and beha io s owa ds OBA (Agui e e al. 2015). Acco dingly,
when ad e ise s include anspa ency mechanisms in o an ad message, consume s may expe ience
mo e us o become mo e willing o accep ulne abili y, so ha hey would accep o click a highly
pe sonalized ad e isemen despi e hei p i acy conce ns and eelings o pe cei ed in usi eness
(Boe man e al., 2017). This suppo s he i h hypo hesis o he s udy:
➢
H5: Being anspa en abou da a collec ion mechanisms has a posi i e impac on a)
Consume s’ a i udes owa ds he b and, b) Consume s’ a i udes owa ds OBA, c) Consume s’
in en ion o click on he Ad, d) Consume s’ in en ions o pu chase he p oduc and e) Consume s’
p o ec i e beha io .
P e ious esea ch highligh s ha he le el o pe sonaliza ion o he online ad causes a o able as well as
un a o able consume a i udes and beha io s owa ds he ad (Tucke 2012b). Tucke (2014) claims
pe sonaliza ion causes eeling o in usi eness o inc ease, leading o nega i e a i udes and beha io s
om consume s in esponse o he pe sonalized ad e isemen . Ano he s udy by Hühn e al. (2017),
suppo ed he no ion ha hose eelings o in usi eness gene a e ad annoyance and a oidance beha io
among use s in he online en i onmen and mobile media. This suppo s he six h hypo hesis o he
s udy:
➢
H6: Pe cei ed in usi eness media es he in luence o i) pe sonaliza ion, ii) consume
cha ac e is ics, iii) b and amilia i y i ) b and us and ) anspa ency app oach on a)
Consume s’ a i udes owa ds he b and, b) Consume s’ a i udes owa ds OBA, c) Consume s’
26
in en ion o on click he Ad, d) Consume s’ in en ions o pu chase he p oduc and e) Consume s’
p o ec i e beha io .
O e he pas decade, li e a u e has shown how consume s ha e become mo e conce ned abou hei
online p i acy, especially owa ds OBA p ac ices. Resea ch highligh ed how p i acy conce ns play an
impo an ole in consume accep ance o OBA, whe e he le el o pe sonaliza ion a ec s consume -
ela ed ac o s such as p i acy conce ns which acco dingly a ec a i udes and beha io s (Bleie and
Eisenbeiss, 2015).
Also, p e ious esea ch indings sugges ha since OBA is based on acking consume s’ online beha io
and collec ing consume s’ in o ma ion, he e o e, he nega i e impac o p i acy conce ns would be e en
highe in he con ex o OBA leading o nega i e a i udes and beha io al esponses o he e a ge ed
ad e isemen . This calls o he se en h hypo hesis o he s udy:
➢
H7: P i acy conce ns media e he in luence o i) pe sonaliza ion, ii) consume cha ac e is ics,
iii) b and amilia i y i ) b and us and ) anspa ency app oach on a) Consume s’ a i udes
owa ds he b and, b) Consume s’ a i udes owa ds OBA, c) Consume s’ in en ion o click on
he Ad, d) Consume s’ in en ions o pu chase he p oduc and e) Consume s’ p o ec i e
beha io .
P e ious li e a u e discussed how highly pe sonalized ads can gene a e eelings o loss o choice and
con ol among consume s, and e en ually cause nega i e a i udes and esponses o he online
e a ge ed ad e isemen . Thus, i can be expec ed ha he le el o online ad pe sonaliza ion in luences
consume - ela ed ac o s, such as eelings o eac ance (Bleie and Eisenbeiss, 2015), which in u n
in luences OBA ou comes, such as pu chase in en ions and click- h ough a es (Agui e e al., 2015).
The e o e, mo e esea ch is needed o disclose he ole eac ance plays in in luencing consume s’
a i udes, in en ions and beha io s owa ds OBA con ex . This calls o he eigh h hypo hesis o he s udy:
➢
H8: Reac ance media es he in luence o i) pe sonaliza ion, ii) consume cha ac e is ics, iii)
b and amilia i y i ) b and us and ) anspa ency app oach on a) Consume s’ a i udes
owa ds he b and, b) Consume s’ a i udes owa ds OBA, c) Consume s’ in en ion o click on
he Ad, d) Consume s’ in en ions o pu chase he p oduc and e) Consume s’ p o ec i e
beha io .
27
2.14. Resea ch Model
Boe man e al. (2017) claimed ha OBA esea ch lacks a clea unde s anding o he concep due o he
in e disciplina y na u e o he ield and he numbe o in e es ed pa ies including ad e ise s, consume s,
compu e scien is s, and policymake s. Howe e , OBA esea ch has s udied a ious independen ,
media ing, mode a ing, and ou come a iables.
Pe sonaliza ion p o ed o ha e a g ea impac on consume s’ esponses o OBA. Howe e , i is no well
unde s ood up o his da e wha le els o pe sonaliza ion consume s ind accep able and wha hey
conside c eepy and oo in asi e (Boe man e al. 2017). Mo eo e , p io esea ch ecommends i is
impo an o know who is being a ge ed, as he e seem o be indi idual di e ences in esponses o OBA.
The e o e, he pu pose o his s udy is o explo e he ole o pe sonaliza ion, consume cha ac e is ics,
b and amilia i y, b and us , and anspa ency on consume s’ a i udes owa ds he b and, consume s’
a i udes owa ds OBA, consume s’ in en ion o click on he Ad, consume s’ in en ions o pu chase he
p oduc and consume s’ p o ec i e beha io . Mo eo e , he s udy explo es he media ing in luence o
pe cei ed in usi eness, pe cei ed p i acy conce ns, and eac ance on consume s’ a i udes and
in en ions and p o ec i e beha io .
34
3.6. Resea ch App oach
Resea ch app oaches a e classi ied in o wo main app oaches; he
deduc i e app oach
whe e he
esea che de elops a heo y, o mula es hypo heses and hen designs a concep ual amewo k o es
he hypo heses using da a, and he
induc i e app oach
, in which da a is collec ed and analyzed o de elop
heo y. In he deduc i e app oach p e ious li e a u e is used o iden i y heo ies and o build he
heo e ical amewo k ha will be es ed using da a (Saunde s e al., 2009).
Seka an and Bougie (2016) claimed ha deduc i e easoning wo ks om he mo e gene al o he mo e
speci ic, whe e esea che s s a wi h a gene al heo y ha is hen na owed down in o speci ic
hypo heses ha can be es ed, hese hypo heses a e hen u he na owed down when speci ic da a is
collec ed and analyzed.
Saunde s e al. (2009) s a ed ha deduc ion is he dominan esea ch app oach in na u al sciences, and
a e mo e o en used in causal and quan i a i e s udies whe e deduc ion owes mo e o posi i ism esea ch
philosophy.
An impo an cha ac e is ic o deduc i e esea ch is gene aliza ion. I is essen ial o selec samples o
adequa e nume ical size in o de o be able o o m a s a is ical gene aliza ion abou egula i ies in human
social beha io (Saunde s e al., 2009).
The esea ch in hand ollows a deduc i e app oach, whe e p e ious li e a u e is used o gene a e
hypo heses based on he heo e ical ounda ions o he Social Exchange Theo y and he Psychological
Reac ance Theo y. Hypo heses a e hen es ed using quan i a i e esea ch me hod. The nex sec ion,
sec ion (3.7) en ails he esea ch me hodology chosen in his s udy, he da a collec ion echnique and
he ins umen s used o collec he da a.
3.7. Resea ch Me hodology
The e m me hods e e o echniques and p ocedu es used o ob ain and analyze da a. This includes
esea ch ins umen s such as ques ionnai es, obse a ion and in e iews as well as bo h s a is ical
quan i a i e and nons a is ical quali a i e analysis echniques. Whe eas he e m me hodology e e s o
he heo y o how esea ch should be unde aken (Saunde s e al., 2009).
The nex sec ions (3.7.1, 3.7.2, and 3.7.3) discusses he esea ch me hod choice, da a collec ion
echnique, he esea ch ins umen s and he sampling design.
35
3.7.1. Resea ch Me hod Choice (Quan i a i e Me hod)
The s udy adop s a mono-me hod choice, which is he quan i a i e me hod. Mono me hods can be
de ined as he use o a single da a collec ion echnique and co esponding analysis p ocedu es as a way
o answe he esea ch ques ions. Quan i a i e me hods a e used o de ine any da a collec ion echnique
(such as a ques ionnai e) o da a analysis p ocedu e (such as s a is ics) ha uses o p oduces nume ical
da a (Saunde s e al., 2009). This choice is widely used wi hin business and managemen esea ch
(Cu an and Blackbu n 2001).
C onbach and Webb (1975) a gued ha quan i a i e me hods use co ela ional designs o educe e o ,
bias, and o he noise ha keeps one om clea ly unde s anding social ac s. Saunde s e al. (2009)
claimed ha using a quan i a i e me hod such as a ques ionnai e o collec da a and es hypo heses
p o ide a mo e e icien da a collec ion echnique om a la ge sample because each esponden will
espond o he same ques ion.
The main da a collec ion ins umen used in he s udy is a su ey using a ques ionnai e. The da a is
conside ed p ima y da a collec ed by he ques ionnai e. P e ious li e a u e was ho oughly in es iga ed
o iden i y measu emen scales o he ques ions included in he ques ionnai e.
The ollowing sec ion (3.7.2) p esen s he da a collec ion me hod o his s udy, ollowed by a desc ip ion
o he esea ch ins umen used o collec he da a.
3.7.2. Da a Collec ion Technique (Expe imen + Su ey)
An expe imen al s udy ollowed by a su ey was conduc ed o es he hypo heses and answe he
esea ch ques ions s a ed in he p e ious sec ion (3.2).
The key app oach o posi i is esea che s is he expe imen , and i is usually associa ed wi h a
hypo he ico‐deduc i e app oach o esea ch, whe e he esea che manipula es he independen a iable
o s udy he e ec o his manipula ion on he dependen a iable (Seka an and Bougie, 2016).
Expe imen s he e o e a e used o answe ‘how’ and ‘why’ ques ions (Saunde s e al., 2009).
Expe imen al esea ch is app op ia e o his s udy because “
expe imen s a e well-sui ed o s udying
causal ela ionships
” (Shadish e al., 2002:7).
36
Shadish e al. (2002:6) u he a gued ha by using expe imen s:
(1)
We manipula e he p esumed cause and obse e an ou come a e wa d; (2) we see whe he
a ia ion in he cause is ela ed o a ia ion in he e ec ; and (3) we use a ious me hods
du ing he expe imen o educe he plausibili y o o he explana ions o he e ec along
wi h ancilla y me hods o explo e he plausibili y o hose we canno ule ou .
The s udy u ilized scena io-based online expe imen s. Scena ios we e chosen because simila p e ious
s udies showed ha scena io-based expe imen s a e common and app op ia e in his a ea o esea ch
(e.g., Agui e e al., 2015; Bleie & Eisenbeiss, 2015; Shoenbe ge & Tho son, 2014; Whi e e al., 2008).
Appendix (I) includes eigh di e en e sions o he expe imen scena ios designed o he s udy.
To design he expe imen , eigh scena ios ea u ing di e en combina ions o amilia iza ion,
pe sonaliza ion, and anspa ency we e c ea ed whe e esponden s we e asked o imagine ha hey a e
in e es ed in weigh -loss p og ams and hey sea ched online abou he opic. Following ha , he nex
mo ning while eading he news on an online news websi e, hey saw an ad abou weigh -loss medica ion.
Appendix (I) includes he eigh e sions o he ads ha we e designed o i each condi ion (Familia –
Non-Pe sonalized – Non-T anspa en , Familia – Non-Pe sonalized – T anspa en , Familia –
Pe sonalized – Non-T anspa en , Familia – Pe sonalized – T anspa en , Un amilia – Non-Pe sonalized
– Non-T anspa en , Un amilia – Non-Pe sonalized – T anspa en , Un amilia – Pe sonalized – Non-
T anspa en , Un amilia – Pe sonalized – T anspa en ).
The esea ch hen adop s he su ey s a egy as a me hod o collec p ima y da a. The su ey s a egy is
usually associa ed wi h he deduc i e app oach, i allows he collec ion o a la ge amoun o da a om a
la ge popula ion in a highly cos -e ec i e way (Saunde s e al., 2009). Su eys a e o en conduc ed by
using a ques ionnai e ha a esponden comple es on his o he own, ei he on pape o ia he compu e
(Seka an and Bougie, 2016). The su ey s a egy enables he collec ion o quan i a i e da a ha would
be analyzed quan i a i ely using desc ip i e and in e en ial s a is ics (Saunde s e al., 2009).
Appendix (II) includes he English language e sion o he su ey ins umen . The ollowing subsec ion
(3.7.3) in oduces he ins umen o da a collec ion (online expe imen and ques ionnai e), ollowed by
he desc ip ion o he expe imen scena ios, ques ionnai e s uc u e, and con en .
3.7.3. Expe imen al design
As men ioned be o e in sec ion (3.7.2), he s udy used a scena io-based online expe imen ollowed by
a ques ionnai e o collec p ima y da a. The expe imen was eplica ed 8 imes using a eal b and (Lipo
37
6) and a ic i ious b and (Mi acle Cu e), unde di e en pe sonaliza ion and anspa ency se ings. Weigh
loss was selec ed as he expe imen opic as i is conside ed a sensi i e ye impo an opic o consume s.
A eal amilia weigh loss medica ion (LIPO 6) was selec ed o he eal b and while a ic i ious un amilia
b and (Mi acle Cu e) was in oduced o esponden s. Expe imen manipula ions a e included in Appendix
(I).
Replica ion helps o gene alize esea ch indings and expe imen al esea ch should be eplica ed o
ex e nal alidi y (Lynch, 1982). Also, since b and amilia i y is one o he a iables ha need o be es ed,
using a eal b and and a ic i ious b and was a mus o conside he di e ences in pa icipan s’ b and
amilia i y and b and us esponses.
The esea ch ook place ONLINE using he GORILLA SC expe imen builde . The expe imen was hen
ollowed by a ques ionnai e on he same pla o m o ga he expe imen da a. A simple link ha di ec s
esponden s o he online expe imen websi e was popula ed h ough email, social media si es (Facebook
and Ins ag am), and iends.
A e pa icipan s clicked he expe imen link, hey we e e-di ec ed o he Go illa SC websi e. Pa icipan s
we e asked o ag ee on a consen o pa icipa e in an online expe imen , a e wa ds, hey we e andomly
assigned one o he eigh expe imen scena ios. Fo hose who we e andomly assigned o one o he
pe sonalized scena ios; pa icipan s we e asked o inpu hei i s names, hese i s names we e used
in c ea ing he ad e ising message.
A e being exposed o he online scena io, pa icipan s a e hen e-di ec ed o he nex page o answe
he online ques ionnai e which measu ed hei iew o ad pe sonaliza ion, amilia i y wi h he ad e ised
b and, us in he ad e ised b and, p i acy conce ns, eac ance, pe cei ed in usi eness, a i udes
owa ds b and, a i udes owa ds OBA, in en ion o click he Ad, in en ions o pu chase he p oduc and
p o ec i e beha io . Pa icipan s’ demog aphic in o ma ion was also collec ed in he ques ionnai e,
including age, gende , income le el, educa ional backg ound, and employmen s a us.
3.8. Ques ionnai e De elopmen
A ques ionnai e is a w i en se o ques ions o which esponden s eco d hei answe s in a
p ede e mined o de , usually wi hin a he closely de ined al e na i es (De Vaus, 2002).
Using online ques ionnai es yields many ad an ages ha include, gaining a deep unde s anding o he
subjec unde in es iga ion, wide geog aphical co e age and each, and pa icipan s inc eased
con enience. Also, he dis ibu ion o online ques ionnai es is as and easy (Seka an and Bougie, 2016).
38
The ques ionnai e was applied in English since he esponden s a e p o icien in his language. B yman
(2003) claimed ha he ques ionnai e design is mo e c ucial han he ques ionnai e leng h in ecei ing
a high esponse a e. Gi en ha , he ques ions had o be unde s andable, wi h no use o ambiguous
language also o limi he possibili y o misunde s anding he ques ions, se e al ques ions we e asked
o each scale. Responden s’ anonymi y and con iden iali y was assu ed by he esea che h ough he
ph ase ‘
The in o ma ion collec ed ia he ques ionnai e is ANONYMOUS and used o academic
pu poses ONLY.
’, so ha hey won’ eel hei p i acy in aded.
The esea che designed he ques ionnai e o be illed-in in abou i een minu es maximum. The
ques ionnai e encompasses se e al ques ions, he upcoming sec ions include u he de ail abou hese
ques ions.
The ollowing sec ions discuss he ques ionnai e di e en sec ion, ques ions and scales used o measu e
he a iables.
3.8.1. Demog aphic Ques ions
Demog aphic ques ions we e used o ga he esponden s’ pe sonal da a. Demog aphic da a ga he ed by
he ques ionnai e include age, gende , income le el, educa ional le el and employmen s a us. The only
case when esponden s we e asked o en e hei i s name, i was only o ad pe sonaliza ion easons
bu was no sa ed o eco ded.
Demog aphic ques ions appea a he end o his s udy ques ionnai e, ye esea che s can add
demog aphic ques ions a he beginning o he ques ionnai e.
3.8.2. Iden i ica ion o OBA Ads
Since he no ion – OBA – is no e y common among In e ne use s, he ollowing de ini ion o OBA was
p esen ed o ensu e esponden s unde s ood he opic in hand be o e answe ing OBA ela ed ques ions:
‘Online beha io al ad e ising (OBA) can be de ined as:
The p ac ice o acking indi iduals' online
ac i i ies in o de o deli e ad e ising ailo ed o he consume s' in e es s.’
(Fede al T ade Commission,
2009)
3.8.3. The Ques ionnai e Main Sec ions
The ques ionnai e included se e al sec ions, answe s o each o hese sec ions we e manda o y.
Responden s could no mo e o he nex sec ion wi hou inishing he ques ions o he p e ious one, o
gua an ee he comple ion o he ques ionnai e and assu e app op ia e da a o he analysis. As men ioned
39
be o e, Appendix (II) con ains an English language e sion o he ques ionnai e. The nex sec ions a e
classi ied in o ables wi h a iables name, he scale used o measu e each a iable, o igin and desc ip ion
o he scale and he i ems in each scale.
Pe sonaliza ion
The ques ionnai e s a s igh a e he esponden was subjec ed o he scena io o he expe imen . The
i s ques ion he esponden was asked abou is he le el o pe sonaliza ion o he ad. Responden s
indica e on a single i em se en-poin scale whe he hey ag ee o disag ee wi h he ques ion ‘
This ad is
di ec ed o me pe sonally
’ adap ed om De Keyze , Dens, and De Pelsmacke (2015).
Ano he aspec o pe sonaliza ion - which is how consume s iew he impo ance o he amoun o
pe sonalized in o ma ion an ad con ains – is measu ed using a i e-i em se en-poin Like ype scale (1-
S ongly Disag ee, 7-S ongly Ag ee) as shown he able below.
Table 3: Pe sonaliza ion (In o ma i eness) Scale I ems
Va iable
Sou ce:
Ozcelik, and Va nali (2019)
Pe sonaliza ion (In o ma i eness)
1. Cus omized online ad gi es me quick and
easy access o la ge olumes o
in o ma ion.
2. In o ma ion ob ained om he cus omized
online ad is use ul.
3. I lea ned a lo om cus omized online ads.
4. I hink in o ma ion ob ained om
cus omized online ads is help ul.
5. Cus omized online ad makes acqui ing
in o ma ion inexpensi e.
40
B and Familia i y
Nex , esponden s use a single bipola scale o indica e whe he hey we e amilia o un amilia , wi h
he b and. This answe s he ques ion ‘
Rega ding he b and, I am:
’
Table 4: B and Familia i y Scale I ems
Va iable
Sou ce:
Bake , Hu chinson, Moo e and
Nedungai (1986)
B and amilia i y
A unidimensional cons uc ha is di ec ly ela ed
o he amoun o ime ha has been spen
p ocessing in o ma ion abou he b and,
ega dless o he ype o con en o he p ocessing
ha was in ol ed.
The esponden s indica ed on a single bipola
scale whe he hey we e amilia o un amilia .
Consume A i udes
Consume a i udes can be classi ied in o
Consume A i udes owa ds OBA
and
Consume A i udes
owa ds he b and
. To measu e hese cons uc s, scales sou ced om he li e a u e we e used as shown
in he ables below. As a measu emen scale, se en-poin scale we e used.
Table 5: Consume A i udes owa ds OBA Scale I ems
Va iable
Sou ce:
MacKenzie and Lu z (1983)
A i ude Towa d OBA
(Gene al and Speci ic)
1. 1= bad s. 7 = good,
2. 1 = unpleasan s. 7 = pleasan ,
3. 1 = un a o able s. 7 = a o able.
41
Table 6: Consume A i udes owa ds he B and Scale I ems
Va iable
Sou ce:
Mi chell and Olson (1981)
A i ude Towa d he B and
1. 1= bad s. 7 = good,
2. 1 = unpleasan s. 7 = pleasan ,
3. 1 = dislike e y much s. 7 = like e y much,
4. 1 = poo quali y s. 7 = high quali y.
Consume In en ions
Consume in en ions include
In en ions o Click on he Ad
and
In en ions o Pu chase he Ad e ised
P oduc
. Th ee-i em se en-poin scale used o measu e in en ion owa d he OBA.
Table 7: Consume In en ions Scale I ems
Va iable
Sou ce:
MacKenzie, Lu z, and Belch (1983).
In en ion o Click on he Ad
In en ion o Pu chase he P oduc
.
1. 1= unlikely s. 7 = likely,
2. 1 = imp obable s. 7 = p obable,
3. 1 = impossible s. 7 = possible.
B and T us
B and us was measu ed wi h a ou -i em se en-poin Like ype scale (1-S ongly Disag ee, 7-S ongly
Ag ee). The scales sou ced om he li e a u e we e used as shown in he able below.
Table 8: B and T us Scale I ems
Va iable
Sou ce:
Chaudhu i and Holb ook (2001).
B and T us
1. I us XYZ
2. I ely on XYZ.
42
3. XYZ is an hones b and.
4. The XYZ b and is sa e.
Pe cei ed In usi eness
Pe cei ed in usi eness was measu ed using Li, Edwa ds and Lee (2002) Se en-i em, se en-poin Like
ype scale (1-S ongly Disag ee, 7-S ongly Ag ee) as shown in he able below.
Table 9: Pe cei ed In usi eness Scale I ems
Va iable
Sou ce:
Li, Edwa ds, and Lee (2002).
Pe cei ed In usi eness
1. Dis ac ing.
2. Dis u bing.
3. Fo ced.
4. In e e ing.
5. In usi e.
6. In asi e.
7. Ob usi e.
P i acy Conce ns
P i acy conce ns we e measu ed using Baek and Mo imo o (2012) scale o six i ems, se en-poin Like
ype scale (1-S ongly Disag ee, 7-S ongly Ag ee) as shown he able below.
Table 10: P i acy Conce ns Scale I ems
Va iable
Sou ce:
Baek and Mo imo o (2012)
P i acy Conce ns
1. I eel uncom o able when in o ma ion is
sha ed wi hou pe mission.
43
2. I am conce ned abou misuse o pe sonal
in o ma ion.
3. I bo he s me o ecei e oo much
ad e ising ma e ial o no in e es .
4. I eel ea ha in o ma ion may no be sa e
while s o ed.
5. I belie e ha pe sonal in o ma ion is o en
misused.
6. I hink companies sha e in o ma ion wi hou
pe mission.
Reac ance
Reac ance was cap u ed using Ga dne and Leshne (2016) scale o pe cei ed h ea o choice. The
scale was adap ed because i con ained e y easy o unde s and s aigh o wa d ques ions. The scale
ques ions a e shown in he able below.
Table 11: Reac ance Scale I ems
Va iable
Sou ce:
Ga dne and Leshne (2016)
Reac ance (Pe cei ed Th ea o Choice)
1. The message h ea ened my eedom o
choose.
2. The message ied o make a decision o
me.
3. The message ied o manipula e me.
4. The message ied o pe suade me.
50
model is e lec i e. Indica o s a e conside ed e lec i e i hey es ablish a ep esen a i e se o all possible
i ems wi hin he cons uc (Diaman opoulos and Winklho e , 2001). Reliabili y and alidi y a e he
common es s applied o assess he p op ie y o he measu emen model (ou e model). (Hai e al.,
2019; Hai J e al., 2017).
I he measu emen model mee s all he equi ed c i e ia, hen he esea che needs o e alua e he
s uc u al model which will be elabo a ed in sec ion (3.13.7) (Hai e al., 2019).
3.12.1. In e nal Consis ency Reliabili y
Reliabili y es s a e used o iden i y he ex en o which an ins umen is consis en and s able in
measu ing a concep . The i s pa o a eliabili y assessmen is he in e nal consis ency o he measu ing
ins umen , which is a measu e o he homogenei y o he i ems o a ce ain cons uc . While es ing a
e lec i e model, all he measu emen i ems should be highly co ela ed and in e changeable o make
su e hey measu e he cons uc om all aspec s (Seka an & Bougie, 2016). C onbach’s alpha is he
mos popula es o in e nal consis ency. A a alue o a leas 0.7 he cons uc is conside ed eliable
(Nunnally J, 1978).
The second pa o a eliabili y assessmen is a composi e eliabili y es , which akes in o conside a ion
he dis inc ou e loadings o he indica o s and hus i p o ides ela i ely highe eliabili y es ima es. A
composi e eliabili y alue anging om 0.6 o 0.9 is conside ed accep able eliabili y (Diaman opoulos
e al., 2012). The ue alue o eliabili y lies be ween he lowe bound ep esen ed by C onbach’s alpha
and he uppe bound ep esen ed by he composi e eliabili y (Hai J e al., 2017; Sa s ed e al., 2021).
The cu en s udy will apply bo h es s o assess he eliabili y o he measu ing ins umen .
3.12.2. Indica o Reliabili y
Indica o eliabili y is he second s ep in ensu ing he eliabili y o he e lec i e model. I examines how
much a iance o each indica o is explained by i s cons uc . Indica o eliabili y is measu ed by
compu ing he squa e o he indica o loading, which is he bi a ia e co ela ion be ween he indica o
and i s cons uc , since e lec i e indica o s ela e o he cons uc h ough loadings. Indica o loadings
o 0.7 and abo e a e he ecommended h eshold, as his alue indica es ha he cons uc explains
mo e han 50% o he indica o ’s a iance, hence, achie ing accep able indica o eliabili y (Sa s ed e
al., 2021).
Howe e , in social science esea ch, weake indica o loadings (<0.7) a e o en ob ained, especially in
he case o using newly de eloped scales (Hulland, 1999). Indica o s wi h e y low loadings (<0.4) should
51
be emo ed om he measu emen model. While, loadings lying be ween 0.4 and 0.7, a e no emo ed
unless hei dele ion will imp o e he alue o in e nal consis ency eliabili y o con e gen alidi y and
will no a ec he con en alidi y. O he wise, hese indica o s a e e ained (Hai e al., 2022).
3.12.3. Validi y
Validi y assesses he ce ain y o he measu ing ins umen ; i ensu es how accu a ely an i em measu es
a pa icula a iable. Cons uc alidi y examines how well he measu ing ins umen i s he heo e ical
ounda ion o he cons uc (Seka an & Bougie, 2016). Con e gen and disc iminan alidi y a e applied
o es cons uc alidi y.
Con e gen Validi y indica es he ex en o which a cons uc comes oge he o explain he a iance o
i s indica o s, and is e alua ed by assessing he co ela ion o he cons uc wi h an al e na i e measu e
o he same concep . The a e age a iance (AVE) was used o e alua e con e gen alidi y. The minimum
accep ed alue o AVE o achie e con e gen alidi y is 0.50, and abo e (Hai e al., 2022).
Disc iminan alidi y is a es applied o ensu e ha a cons uc is uly dis inc om o he cons uc s in
he model empi ically. Disc iminan alidi y was assessed using he Fo nell-La cke c i e ion which
compa es he squa e oo o each cons uc ’s AVE o i s co ela ion wi h o he cons uc s. Fo
disc iminan alidi y o be achie ed, he squa e oo alue o he cons uc ’s AVE should be g ea e han
i s co ela ions wi h all o he cons uc s (Hai J e al., 2017; Fo nell-La cke , 1981)
3.13. S uc u al Model Assessmen
S uc u al model assessmen is he nex s ep in e alua ing PLS-SEM esul s a e he measu emen model
has been assessed and sa is ac o y esul s we e achie ed.
The s anda d assessmen p ocedu es o be conside ed when analyzing he s uc u al model a e (a)
assessing he s uc u al model o collinea i y issues, (b) assessing he signi icance and ele ance o he
model ela ionships, (c) assessing he le el o coe icien de e mina ion (
R2
), (d) assessing he e ec size
(
2
) and (e) assessing he p edic i e ele ance (
Q2
) (Beni ez e al., 2020; Hai e al., 2011, 2019; Hai J
e al., 2017).
3.13.1. Collinea i y Assessmen
Collinea i y e e s o he exis ence o high co ela ion be ween wo independen a iables in eg ession
analysis. I is assessed by calcula ing he Va iance In la ion Fac o (VIF) alues o each se o p edic o
cons uc s in e e y sub-pa o he model. De e mining he mul icollinea i y p oblems is conside ed he
52
i s s ep in e alua ing he s uc u al model in PLS-SEM (Hai e al., 2020). VIF alues g ea e han 5
indica e c i ical le els o collinea i y. In such cases, collinea i y mus be ea ed ei he by emo ing ce ain
cons uc s o me ging se e al p edic o s in o one cons uc . Low le els o collinea i y mus be achie ed
o ensu e ha e e y independen a iable possesses a unique pe cen age o he explained a iance o
he dependen a iable (Hai J e al., 2010, 2017).
3.13.2. S uc u al Model Pa h Coe icien s
Examining he size and signi icance o he pa h coe icien s is he nex s ep a e sol ing he c i ical
collinea i y p oblems. Pa h coe icien s a e he es ima es ob ained o e i y he hypo hesized ela ionships
be ween he a iables in he s uc u al model.
The pa h coe icien s alues ange om -1 o +1 and a ely be -1 o +1; alues close o +1 indica e a
s ong posi i e ela ionship, while hose close o -1 indica e a s ong nega i e ela ionship. The mo e he
alue ge s close o 0, he weake he ela ionship is. Howe e , he ue signi icance o he pa h coe icien
is de e mined acco ding o i s s anda d e o . A pa h coe icien s anda d e o is ob ained by applying
he boo s apping echnique ha compu es he
alues and
p
alues ha uly indica e he signi icance
o he ela ionships (Hai J e al., 2017).
Boo s apping is a esampling echnique ha es ima es he signi icance o he hypo hesized ela ionships
by d awing a sample om he o iginal sample wi h eplacemen s. The numbe o obse a ions o each
boo s ap sample is equal o he numbe o obse a ions o he o iginal sample. P io s udies
ecommended he use o 5,000 boo s ap samples (Hensele e al., 2009; Hai J e al., 2017).
The
alues compu ed om he boo s apping ep esen he signi icance o he pa h coe icien s, while
he
p
is he p obabili y o e oneously ejec ing a ue null hypo hesis (p obabili y o e o ). The p e-
de e mined c i ical
alues o a one- ailed es a e 1.28, 1.65, 2.33 o signi icance le els 10%, 5%, and
1% espec i ely. Fo wo- ailed es s, c i ical
alues a e 1.65, 1.96, 2.57 o signi icance le els 10%, 5%,
and 1% espec i ely. The e o e, when he compu ed
alue exceeds hese c i ical alues, acco ding o
he chosen signi icance le el ( he mos common is 5%), he hypo hesized ela ionship is alida ed o be
s a is ically signi ican . The choice o he signi icance le el and ype o es (one o wo ails) depends on
he ield o s udy and he s udy’s objec i e (Hai J e al., 2017).
3.13.3. Coe icien o De e mina ion (
R2
Value)
The mos used me ic o assess s uc u al model p edic ion is he coe icien o de e mina ion (
R2
alue),
which is calcula ed as he squa e co ela ion be ween he dependen a iable’s ac ual and p edic ed
53
alues. This coe icien measu es he accumula ed e ec s o independen a iables on he dependen
a iable.
R2
alue e alua es he p edic i e accu acy o he s uc u al model as i de e mines he amoun
o he dependen a iable’s a iance explained by all independen a iables (Hai e al., 2019).
R2
is
measu ed in a ange om 0 o 1, a highe alue o
R2
indica es highe p edic i e accu acy deno ing ha
he es ed independen a iables a e indeed causing mos o he a iance o he dependen a iable (Hai
J e al., 2017).
R
2 alues o 0.75, 0.50, and 0.25 a e conside ed subs an ial, mode a e, and weak, espec i ely.
Howe e , hese alues a e ecommended o ma ke ing esea ch (Hai J e al., 2010). In o he
disciplines, an
R
2 alue s a ing om 0.10 is conside ed accep able (Falk and Mille , 1992; Rai hel e al.,
2012). Acco ding o Chin (1998),
R
2 alues 0.67, 0.33 and 0.19 a e conside ed subs an ial, mode a e,
and weak espec i ely (Pu wan o and Suda gini, 2021). Mo eo e , Cohen (1988) sugges ed
R
2 alues o
endogenous la en a iables as ollows: 0.26 (subs an ial), 0.13 (mode a e), 0.02 (weak). Fu he mo e,
oo high alues o
R
2 migh indica e a model o e i (Sa s ed e al., 2021).
3.13.4. E ec Size (
2
)
The e ec size (
2
) is conside ed a second measu emen o he signi icance o he hypo hesized
ela ionship in addi ion o he s a is ical signi icance p o ided by he
R2
.
2
es ima es he p edic i e abili y
o each cons uc in he model, i measu es he change in he
R2
alue when a speci ic independen
a iable is emo ed om he s uc u al model, he eby assess whe he he emo ed a iable had a
subs an ia i e e ec on he dependen a iable (Hai J e al., 2017).
An e ec is conside ed small, medium, o la ge a
2
alues o 0.02, 0.15, and 0.35, espec i ely (Cohen,
1998). In mos cases he
2
is simila o he size o he pa h coe icien s (Sa s ed e al., 2021).
3.13.5. P edic i e Rele ance (
Q2
)
Ano he way o assess he model p edic i e accu acy is by calcula ing i s p edic i e ele ance (
Q2
alue).
The p edic i e ele ance (
Q2
) measu es i he pa h model has an adequa e p edic i e powe on he “ou
o sample” popula ion. Meaning, he model can p edic da a no used in he es ima ion (Hai J e al.,
2017). This app oach is based on he blind olding p ocedu e ha elimina es single poin s in he da a
ma ix, assigns he emo ed poin s wi h he mean and es ima es he model pa ame e s (Rigdon, 2014).
Q2
alue is he di e ence be ween he o iginal and he p edic ed da a poin s. The p ocedu e is epea ed
o eplace each da a poin and e-es ima e he model epea s o omi each da a poin and e-es ima e
he model.
54
Q2
alues g ea e han ze o o a speci ic endogenous a iable indica es ha he pa h model has a
p edic i e ele ance o ha dependen a iable, while less han ze o indica es a lack o p edic i e
ele ance, mo eo e , alues wi hin 0.25 and 0.50 indica es medium and la ge p edic i e ele ance o
PLS-SEM model (Hai J e al., 2017).
3.14. Media ion Analysis
Media ion happens when a a iable - e e ed o as a media o a iable - in e enes be ween wo o he
ela ed a iables. In pa icula , he change in he exogenous cons uc (independen a iable) causes a
change in he media o a iable, in u n, esul s in a change in he endogenous cons uc (dependen
a iable) in he PLS pa h model (Hai J e al., 2017).
MacKinnon e al. (2007), claimed ha media ion analysis assumes ha a sequence o ela ionships in
which an independen a iable a ec s he media ing a iable, which hen a ec s he dependen a iable.
Media ing e ec s can be ei he di ec o indi ec . Di ec e ec s a e he ela ionships ha links wo
cons uc s wi h a single a ow, while indi ec e ec s a e a sequence o wo o mo e ela ionships
ep esen ed by mul iple a ows, hey include a se ies o ela ionships wi h a leas one in e ening
cons uc in ol ed in he ela ionship (Sa s ed e al., 2020). Media ion analysis e eals whe he he
media ion has a di ec e ec o indi ec e ec and es s he signi icance le el o he media o (s) (Hai J
e al., 2017).
Hai J e al. (2017) desc ibe wo ypes o non-media ion: 1. Di ec -only non-media ion, whe e he di ec
e ec is signi ican bu no he indi ec e ec . 2. No-e ec non-media ion, whe e nei he he di ec no
indi ec e ec a e signi ican . In addi ion, hey iden i y h ee ypes o media ion: 1. Complemen a y
media ion, whe e he indi ec e ec and he di ec e ec bo h a e signi ican and poin in he same
di ec ion. 2. Compe i i e media ion, whe e he indi ec e ec and he di ec e ec bo h a e signi ican
and poin in opposi e di ec ions. And, 3. Indi ec -only media ion, whe e he indi ec e ec is signi ican
bu no he di ec e ec .
The cu en s udy includes h ee media o a iables, be ween exogenous and endogenous cons uc s;
his ype o analysis is e e ed o as mul iple media ion analysis (Sa s ed e al., 2020).
3.14.1. Tes ing Media ing E ec s
Many p io s udies elied on he 1982 Sobel es , o es he signi icance o media ing e ec s. Sobel es
compa es he di ec ela ionship be ween he independen a iable and he dependen a iable wi h he
indi ec ela ionship be ween he independen a iable and dependen a iable ha includes he
55
media ion cons uc (Helm e al., 2010). Howe e , Sobel es assumes a no mal da a dis ibu ion which
is no consis en wi h he nonpa ame ic PLS-SEM me hod. Also, he pa ame ic assump ions o he Sobel
es usually do no ega d he indi ec e ec be ween cons uc s. Besides, Sobel es equi es
uns anda dized pa h coe icien s as inpu o he es s a is ic and lacks s a is ical powe , especially when
applied o small sample sizes (Hai J e al., 2017; Sa le e al., 2010). Fo hese easons, he cu en
esea ch has dismissed he Sobel es o e alua ing media ion analysis.
Since he cu en s udy applies PLS-SEM, he e o e, he media ion analysis is conduc ed using
boo s apping es , o es he di ec and indi ec e ec s o he media ion. Indi ec E ec s p o ide an
o e iew o esul s, including s anda d e o s, boo s ap mean alues, alues, and p alues. In addi ion,
i shows he con idence in e al as de i ed om he BCa me hod (Ni zl e al., 2016).
56
DATA ANALYSIS AND FINDINGS
57
4.1. In oduc ion
The main indings o he empi ical in es iga ion a e p esen ed in his chap e . This chap e i s in oduces
he expe imen manipula ion check, ollowed by he main indings and esul s a e unning he da a
analysis, including alidi y and eliabili y es , desc ip i e analysis, and s uc u al equa ional modelling
(SEM). The indings a e epo ed using SPSS 26 and pa ial leas squa ed-s uc u al equa ion modelling
(PLS-SEM) using Sma PLS4. SPSS is used in he desc ip i e analysis while PLS is used o es he
esea ch hypo heses, as indica ed in he esea ch model.
4.2. Expe imen Manipula ion Check
A manipula ion check as de ined by Hoewe (2017:1) is "
a es used o de e mine he e ec i eness o a
manipula ion in he expe imen al design
." I he manipula ion check is success ul (p omp an expec ed
di e ence be ween o among expe imen al condi ions), he esea che can sa is ac o ily conclude
pa icipan s co ec ly unde s and, in e p e , o espond o he s imulus and d aw mo e accu a e
conclusions abou he ela ionship be ween he independen and dependen a iables (Hoewe, 2017)
The cu en s udy has h ee manipula ion scena ios: Pe sonaliza ion (Pe sonalized s. Non-Pe sonalized),
B and Familia i y (Familia s. Un amilia ), and T anspa ency App oach (T anspa en s. Non-
T anspa en ). Du ing he expe imen pa icipan s a e subjec ed o a scena io con aining he s imulus
ma e ials, ollowed by a su ey o check he manipula ions. An independen samples - es using SPSS26
was used in o de o de e mine whe he he manipula ion check was success ul o no .
In he cu en s udy, pa icipan s’ i s name was used o manipula e pe sonaliza ion whe e hey we e
asked o en e hei names be o e hey we e subjec o he s imulus sc een. In he pe sonalized condi ion
pa icipan s we e g ee ed using hei i s names in he ad, while in he non-pe sonalized condi ion, hey
we e g ee ed using he ph ase “Hello The e,”. Pe sonaliza ion manipula ion was checked using a one-
i em scale: “This ad is di ec ed o me pe sonally,” adap ed om De Keyze , Dens, and De Pelsmacke
(2015), and a 7-poin Like - ype scale (1-S ongly Disag ee, 7- S ongly Ag ee) was used o measu e hei
esponses. An independen sample - es was conduc ed o check he manipula ion. Pa icipan s in he
pe sonalized condi ion epo ed highe ag eemen (M = 5.619, SD = 1.411) han hose in he non-
pe sonalized condi ion (M = 2.511, SD = 1.422), (470) = 24.113, p≤0.001. The esul s show
s a is ically signi ican di e ence be ween he wo scena ios; he e o e, he manipula ion was success ul.
58
Table 13: Pe sonalized/Non-Pe sonalized G oup S a is ics
N
Mean
S d. De ia ion
S d. E o Mean
Non-Pe sonalized
266
2.5113
1.42281
0.08724
Pe sonalized
221
5.6199
1.41111
0.09492
Table 14: Pe sonalized/Non-Pe sonalized Independen Samples Tes
Le ene's Tes
o Equali y o
Va iances
95% Con idence
In e al o he
Di e ence
F
Sig.
d
Sig. (2-
ailed)
Mean
Di e e
nce
S d.
E o
Di e e
nce
Lowe
Uppe
Equal
a iances
assumed
0.074
0.786
-24.094
485
0.000
-3.108
0.129
-3.362
-2.855
Equal
a iances
no
assumed
-24.113
470
0.000
-3.108
0.128
-3.361
-2.855
B and Familia i y was manipula ed using an image as a s imulus in he ad, he image ei he showed a
amilia b and o an un amilia one. B and Familia i y manipula ion was checked using a one-i em scale:
“Rega ding he b and, I am:” adap ed om Bake , Hu chinson, Moo e and Nedungai (1986), and a 7-
poin Like - ype scale (1-Un amilia , 7- Familia ) was used o measu e hei esponses. An independen
sample - es was conduc ed o check he manipula ion. Pa icipan s in he amilia condi ion epo ed
highe ag eemen (M = 5.876, SD = 0.914) han hose in he un amilia condi ion (M = 2.250, SD =
0.989), (485) = 42.010, p≤0.001. The esul s show s a is ically signi ican di e ence be ween he wo
scena ios; he e o e, he manipula ion was success ul.
Table 15:Familia /Un amilia G oup S a is ics
N
Mean
S d. De ia ion
S d. E o Mean
Un amilia
244
2.2500
0.98914
0.06332
Familia
243
5.8765
0.91429
0.05865
59
Table 16: Familia /Un amilia Independen Samples Tes
Le ene's Tes
o Equali y o
Va iances
95% Con idence
In e al o he
Di e ence
F
Sig.
d
Sig.
(2-
ailed)
Mean
Di e en
ce
S d. E o
Di e enc
e
Lowe
Uppe
Equal
a iances
assumed
4.425
0.036
-42.010
485
0.000
-3.626
0.086
-3.796
-3.456
Equal
a iances
no
assumed
-42.016
482
0.000
-3.626
0.086
-3.796
-3.456
Rega ding T anspa ency, some pa icipan s we e andomly subjec ed o a anspa en expe imen
displaying a link wi h
“Why am I seeing his ad?”
ques ion, while o he s we e subjec ed o a non-
anspa en one lacking he same ques ion. The su ey did no ha e an explici ques ion o es he
manipula ion o anspa ency, howe e , mul iple ela ed ques ions we e ied as p oxies o explici
anspa ency ques ion (such as “I hink companies sha e in o ma ion wi hou pe mission”, “The
ad e ised b and is an hones b and”). Ye , all es ed p oxy ques ions yielded s a is ically insigni ican
esul s (manipula ion unsuccess ul), his sugges s ha anspa ency manipula ion is ei he
o e looked/misunde s ood by he esponden s o ac ually has no bea ing on hei esponses. This esul
is ac ually no new as men ioned ea lie in he Li e a u e Re iew chap e .
Fo he i s wo scena ios since he manipula ion checks we e success ul, i could be assumed ha
pa icipan s could iden i y he di e en (non)pe sonalized/(un) amilia scena ios, and hus hei answe s
o he su ey ques ions we e based on ha iden i ica ion. As o anspa ency, he expe imen numbe
e e ing o whe he he expe imen was anspa en o no was eco ded o es he e ec o anspa ency
on use esponses. Su ey da a examina ion was ca ied ou using SPSS26 and Sma PLS4, da a
examina ions a e co e ed in he nex sec ions.
4.3. Da a Examina ion
The da a collec ed om ques ionnai es we e i s en e ed in o he SPSS 26 so wa e o ob ain desc ip i e
s a is ics o he esponses. Then, s a is ical me hods, using SPSS 26, we e employed o e i y ha he
da a is ee o e o s be o e analysis o ensu e eliable and alid da a, o ensu e he inal esul s a e
66
4.5.3. Disc iminan Validi y
Fo nell-La cke (1981) is used o assess he disc iminan alidi y. Fo disc iminan alidi y o be achie ed,
he squa e oo alue o he cons uc ’s AVE should be g ea e han i s co ela ions wi h all o he
cons uc s (Hai J e al., 2017). Resul s suppo ( able 20), he disc iminan alidi y o all a iables.
The co ela ion coe icien s in b acke s deno e he s eng h o ela ionships be ween he cons uc s,
e lec ing hei disc iminan alidi y. No ably, cons uc s like "A i udes owa ds he B and" and "A i ude
owa ds OBA" exhibi high co ela ion coe icien s o 0.914 and 0.954, espec i ely, implying po en ial
o e lap o sha ed a iance, which aises conce ns ega ding hei disc iminan alidi y.
Simila ly, "In en ion o Click on he Ad" and "In en ions o Pu chase he P oduc " e eal high coe icien s
o 0.961 wi h hemsel es, sugges ing obus in e nal consis ency. Addi ionally, cons uc s such as
"P o ec i e Beha io ," "Pe sonaliza ion," "Pe cei ed In usi eness," "P i acy Conce ns," and
"Reac ance" also exhibi high coe icien s wi h hemsel es, indica ing in e nal eliabili y. While mos
cons uc s demons a e s ong in e nal consis ency, he subs an ial co ela ions be ween ce ain
cons uc s, like "A i udes owa ds he B and" and "A i ude owa ds OBA," hin a po en ial o e laps in
hei concep ual domains, wa an ing u he examina ion o hei disc iminan alidi y.
67
Table 20: Co ela ion Coe icien be ween he Resea ch Va iables - Fo nell-La cke
ATB
ATO
BF
BT
CC
ICA
IPP
PB
PER
PIN
PRC
REA
TA
A i ude Towa ds he B and
0.914
A i ude Towa ds OBA
0.854
0.954
B and Familia i y
0.514
0.435
1.000
B and T us
0.733
0.627
0.468
0.965
Consume Cha ac e is ics
0.706
0.721
0.320
0.541
0.933
In en ion o Click on he Ad
0.873
0.858
0.428
0.628
0.740
0.961
In en ion o Pu chase he P oduc
0.857
0.851
0.433
0.627
0.724
0.850
0.961
P o ec i e Beha io
0.865
0.849
0.434
0.631
0.706
0.840
0.849
0.957
Pe sonaliza ion
0.670
0.577
0.291
0.503
0.524
0.604
0.583
0.600
0.962
Pe cei ed In usi eness
0.821
0.656
0.457
0.629
0.509
0.667
0.629
0.661
0.550
0.946
P i acy Conce ns
0.787
0.616
0.397
0.587
0.506
0.670
0.623
0.629
0.526
0.731
0.948
Reac ance
0.823
0.632
0.384
0.583
0.532
0.673
0.664
0.681
0.588
0.767
0.732
0.949
T anspa ency App oach
0.048
0.026
0.108
0.066
-0.032
0.044
0.041
0.054
0.018
0.064
0.051
0.010
1.000
68
4.5.4. Con e gen Validi y and Composi e Reliabili y
This sec ion ep esen s he con e gen alidi y and he composi e eliabili y o he esea ch a iables. Fo
he con e gen alidi y, i is measu ed h ough AVE. AVE should be ex ac ed when he cons uc s o 0.5
o g ea e (Hai e al., 2022). F om he analysis, i is no iced ha he AVE o all he a iables is g ea e
han 0.5.
Fo he eliabili y, i is measu ed h ough C onbach's Alpha and Composi e Reliabili y (Hai e al., 2017;
Sa s ed e al., 2021). A sa is ac o y deg ee o eliabili y is indica ed by alpha o a leas 0.7 (Nunnally,
1978), while composi e eliabili y coe icien s anging om 0.6 o 0.9 a e conside ed accep able eliabili y
(Diaman opoulos e al., 2012). F om hese esul s, i is no iced ha all he alues o C onbach’s Alpha
and Composi e Reliabili y alues a e g ea e han 0.7. The esul s o AVE, C onbach alpha, and
Composi e Reliabili y o esea ch cons uc s a e shown in ( able 21).
Table 21: Con e gen Validi y and Composi e Reliabili y
C onbach's
alpha
Composi e
eliabili y
A e age a iance
ex ac ed (AVE)
A i ude Towa ds he B and
0.934
0.953
0.835
A i ude Towa ds OBA
0.980
0.984
0.910
B and T us
0.975
0.982
0.931
Consume Cha ac e is ics
0.853
0.931
0.870
In en ion o Click on he Ad
0.959
0.973
0.924
In en ion o Pu chase he P oduc
0.959
0.973
0.924
P o ec i e Beha io
0.988
0.990
0.915
Pe sonaliza ion
0.984
0.987
0.925
Pe cei ed In usi eness
0.980
0.983
0.894
P i acy Conce ns
0.977
0.982
0.898
Reac ance
0.963
0.973
0.900
4.6. S uc u al Model Assessmen
S uc u al model assessmen is he nex s ep in e alua ing PLS-SEM esul s a e he measu emen model
has been assessed and sa is ac o y esul s we e achie ed.
The s anda d assessmen p ocedu es o be conside ed when analyzing he s uc u al model a e (a)
assessing he s uc u al model o collinea i y issues, (b) assessing he signi icance and ele ance o he
model ela ionships, (c) assessing he le el o coe icien de e mina ion (
R2
), (d) assessing he e ec size
69
(
2
) and (e) assessing he p edic i e ele ance (
Q2
) (Beni ez e al., 2020; Hai e al., 2011, 2019; Hai J
e al., 2017).
4.6.1. Mul icollinea i y Assessmen
Ve i ica ion o mul icollinea i y be ween independen a iables in he model is conduc ed; he co ela ion
be ween wo o mo e a iables causes di icul y in de e mining a iance explained by independen
a iables and c ea ing a eg ession model (Hai e al., 2017). Redundan dependen a iable in o
p esen ed. The inne VIF shows ha all VIFs a e less han 5, which means ha he e is no p oblem in
mul icollinea i y. The esul s o inne VIF a e shown in ( able 22).
Table 22: Inne VIF
A i udes
owa ds he
B and
A i ude
owa ds OBA
In en ion o
Click on he
Ad.
In en ions o
Pu chase
he P oduc
P o ec i e
Beha io
Pe cei ed
In usi enes
s
p i acy
conce ns
Reac ance
B and Familia i y
1.372
1.372
1.372
1.372
1.372
1.306
1.306
1.306
B and T us
2.063
2.063
2.063
2.063
2.063
1.774
1.774
1.774
Consume Cha ac e is ics
1.703
1.703
1.703
1.703
1.703
1.626
1.626
1.626
Pe sonaliza ion
1.755
1.755
1.755
1.755
1.755
1.524
1.524
1.524
Pe cei ed In usi eness
3.192
3.192
3.192
3.192
3.192
P i acy Conce ns
2.663
2.663
2.663
2.663
2.663
Reac ance
3.122
3.122
3.122
3.122
3.122
T anspa ency App oach
1.025
1.025
1.025
1.025
1.025
1.020
1.020
1.020
4.6.2. S uc u al Model Pa h Coe icien (Tes ing Resea ch Hypo heses)
PLS analysis is done o es he hypo heses h ough making s uc u al equa ion modeling (SEM). The
esul s o analysis a e shown in ( able 26). F om he able, i could be obse ed ha :
➢ H1: Pe sonaliza ion has a posi i e impac on he dependen a iables.
This hypo hesis is di ided in o i e sub-hypo heses:
Fo he i s sub-hypo hesis, H1a, “Pe sonaliza ion has a posi i e impac on Consume s’
A i udes Towa ds he B and”, he esul s shows ha he e is a posi i e signi ican impac o
Pe sonaliza ion on Consume s’ a i udes owa ds he b and (p≤0.001) and he es ima e equals o 0.111
(mo e han 0). Based on he p e ious esul s, he i s sub-hypo hesis is suppo ed.
70
Fo he second sub-hypo hesis, H1b, “Pe sonaliza ion has a posi i e impac on Consume s’
A i udes Towa ds OBA”, SEM shows ha he e is a posi i e signi ican e ec o Pe sonaliza ion on
Consume s’ a i udes owa ds OBA (p≤0.001) and he es ima e equals o 0.104. Based on he p e ious
esul s, he second sub-hypo hesis is suppo ed.
Fo he hi d sub-hypo hesis, H1c, “Pe sonaliza ion has a posi i e impac on Consume s’
In en ion o Click on he Ad”, SEM indica ed a posi i e signi ican in luence o Pe sonaliza ion on
Consume s’ in en ion o click on he Ad (p≤0.001) and he es ima e equals o 0.114. Based on he
p e ious esul s, he hi d sub-hypo hesis is suppo ed.
Fo he ou h sub-hypo hesis, H1d, “Pe sonaliza ion has a posi i e impac on Consume s’
In en ions o Pu chase he P oduc ”, SEM shows ha Pe sonaliza ion has a posi i e signi ican
impac on Consume s’ in en ions o pu chase he p oduc (p≤0.001) and he es ima e equals o 0.103.
Based on he p e ious esul s, he ou h sub-hypo hesis is suppo ed.
Fo he i h sub-hypo hesis, H1e, “Pe sonaliza ion has a posi i e impac on Consume s’
P o ec i e Beha io ”, SEM shows ha he e is a posi i e signi ican e ec o Pe sonaliza ion on
Consume s’ p o ec i e beha io (p≤0.001) and he es ima e equals o 0.123. Based on he p e ious
esul s, he i h sub-hypo hesis is suppo ed.
F om he abo e esul s, i is concluded ha he i s hypo hesis is suppo ed.
➢ H2: Conside ing consume cha ac e is ics has a posi i e impac on he dependen
a iables.
This hypo hesis is di ided in o i e sub-hypo heses:
Fo he i s sub-hypo hesis, H2a, “Conside ing consume cha ac e is ics has a posi i e impac
on Consume s’ A i udes Towa ds he B and”, he analysis shows ha Unde s anding consume
cha ac e is ics has a posi i e signi ican e ec on Consume s’ a i udes owa ds he b and (p≤0.001) and
he es ima e equals o 0.224. Based on he p e ious esul s, he i s sub-hypo hesis is suppo ed.
Fo he second sub-hypo hesis, H2b, “Conside ing consume cha ac e is ics has a posi i e
impac on Consume s’ A i udes Towa ds OBA”, SEM shows ha he e is a posi i e signi ican
in luence o Unde s anding consume cha ac e is ics on Consume s’ a i udes owa ds OBA (p≤0.001)
and he es ima e equals o 0.416. Based on he p e ious esul s, he second sub-hypo hesis is suppo ed.
Fo he hi d sub-hypo hesis, H2c, “Conside ing consume cha ac e is ics has a posi i e impac
on Consume s’ In en ion o Click on he Ad”, SEM indica ed a posi i e signi ican in luence o
71
Unde s anding consume cha ac e is ics on Consume s’ in en ion o click he Ad (p≤0.001), Based on
he p e ious esul s, he hi d sub-hypo hesis is suppo ed.
Fo he ou h sub-hypo hesis, H2d, “Conside ing consume cha ac e is ics has a posi i e
impac on Consume s’ In en ions o Pu chase he P oduc ”, SEM shows ha he e is a posi i e
signi ican e ec o Unde s anding consume cha ac e is ics on Consume s’ in en ion o pu chase he
p oduc (p≤0.001). Acco dingly, he ou h sub-hypo hesis is suppo ed.
Fo he i h sub-hypo hesis, H2e, “Conside ing consume cha ac e is ics has a posi i e impac
on Consume s’ P o ec i e Beha io ”, he analysis p o ed a posi i e signi ican impac o
Unde s anding consume cha ac e is ics on Consume s’ p o ec i e beha io (p≤0.001) and he es ima e
equals o 0.369. Acco dingly, he i h sub-hypo hesis is suppo ed.
F om he abo e esul s, i is concluded ha he second hypo hesis is suppo ed.
➢ H3: B and Familia i y has a posi i e impac on he dependen a iables.
Fo he i s sub-hypo hesis, H3a, “B and Familia i y has a posi i e impac on Consume s’
A i udes Towa ds he B and”, he esul s show ha he e is a posi i e signi ican impac o B and
Familia i y on Consume s’ a i udes owa ds he b and (p≤0.001) and he es ima e equal o 0.087. Based
on he p e ious esul s, he i s sub-hypo hesis is suppo ed.
Fo he second sub-hypo hesis, H3b, “B and Familia i y has a posi i e impac on Consume s’
A i udes Towa ds OBA”, he esul s indica ed a posi i e signi ican in luence o B and Familia i y on
Consume s’ a i udes owa ds OBA as P- alue is 0.011 and he es ima e equals o 0.081. Based on he
p e ious esul s, he second sub-hypo hesis is suppo ed.
Fo he hi d sub-hypo hesis, H3c, “B and Familia i y has a posi i e impac on Consume s’
In en ion o Click on he Ad”, SEM p o ed a posi i e signi ican impac o B and Familia i y on
Consume s’ in en ion o click on he Ad as P- alue is 0.027 and he es ima e equals o 0.063. Based on
he p e ious esul s, he hi d sub-hypo hesis is suppo ed.
The analysis o he ou h sub-hypo hesis, H3d, “B and Familia i y has a posi i e impac on
Consume s’ In en ions o Pu chase he P oduc ”, p o ed a posi i e signi ican impac o B and
Familia i y on Consume s’ in en ions o pu chase he p oduc as P- alue is 0.003 and he es ima e equals
o 0.087. Based on he p e ious esul s, he ou h sub-hypo hesis is suppo ed.
Fo he i h sub-hypo hesis, H3e, “B and Familia i y has a posi i e impac on Consume s’
P o ec i e Beha io ”, he esul s show ha he e is a posi i e signi ican impac o B and Familia i y
72
on Consume s’ p o ec i e beha io as P- alue is 0.011 wi h an es ima e equal o 0.075. Acco dingly, he
i h sub-hypo hesis is suppo ed.
F om he abo e esul s, he hi d hypo hesis is suppo ed.
➢ H4: B and us has a posi i e impac on he dependen a iables.
This hypo hesis is di ided in o i e sub-hypo heses:
Fo he i s sub-hypo hesis, H4a, “B and us has a posi i e impac on Consume s’ A i udes
Towa ds he B and”, he esul s show ha he e is a posi i e signi ican impac o B and us on
Consume s’ a i udes owa ds he b and (p≤0.001) and he es ima e equal o 0.147. Based on he
p e ious esul s, he i s sub-hypo hesis is suppo ed.
Fo he second sub-hypo hesis, H4b, “B and us has a posi i e impac on Consume s’
A i udes Towa ds OBA”, SEM p o ed a posi i e signi ican in luence o B and us on Consume s’
a i udes owa ds OBA, as he P- alue equals o 0.001 and he es ima e equals o 0.124. Based on he
p e ious esul s, he second sub-hypo hesis is suppo ed.
Fo he hi d sub-hypo hesis, H4c, “B and us has a posi i e impac on Consume s’ In en ion
o Click on he Ad.”, SEM shows a posi i e signi ican impac o B and us on Consume s’ in en ion
o click on he Ad., as he P- alue equals o 0.009 and he es ima e equals o 0.089. Acco dingly, he
hi d sub-hypo hesis is suppo ed.
Fo he ou h sub-hypo hesis, H4d, “B and us has a posi i e impac on Consume s’
In en ions o Pu chase he P oduc ”, SEM shows ha B and us has a posi i e signi ican in luence
on in en ions o pu chase he p oduc (p≤0.001) and he es ima e is 0.128. Based on he p e ious
esul s, he ou h sub-hypo hesis is suppo ed.
Fo he i h sub-hypo hesis, H4e, “B and us has a posi i e impac on Consume s’ P o ec i e
Beha io ”, SEM indica ed a posi i e signi ican impac o B and us on Consume s’ p o ec i e
beha io , as he P- alue equals o 0.002 and he es ima e is 0.121. Based on his esul , he i h sub-
hypo hesis is suppo ed.
F om all he abo e esul s, he ou h hypo hesis is p o ed o be suppo ed.
➢ H5: Being anspa en abou da a collec ion mechanisms has a posi i e impac on
he dependen a iables.
This hypo hesis is di ided in o i e sub-hypo heses:
73
Fo he i s sub-hypo hesis, H5a, “Being anspa en abou da a collec ion mechanisms has a
posi i e impac on Consume s’ A i udes Towa ds he B and”, he esul s show an insigni ican
impac o Being anspa en abou da a collec ion mechanisms on Consume s’ a i udes owa ds he
b and, as he P- alue equals o 0.276 (g ea e han 0.05). Based on he p e ious esul s, he i s sub-
hypo hesis is no suppo ed.
Fo he second sub-hypo hesis, H5b, “Being anspa en abou da a collec ion mechanisms has
a posi i e impac on Consume s’ A i udes Towa ds OBA”, SEM p o es ha Being anspa en
abou da a collec ion mechanisms has an insigni ican in luence on Consume s’ a i udes owa ds OBA,
as P- alue equals o 0.433. Acco dingly, he second sub-hypo hesis is no suppo ed.
Fo he hi d sub-hypo hesis, H5c, “Being anspa en abou da a collec ion mechanisms has a
posi i e impac on Consume s’ In en ion o Click on he Ad.”, SEM indica ed an insigni ican
e ec o Being anspa en abou da a collec ion mechanisms on in en ion o click on he Ad., as he P-
alue equals o 0.138. Based on he p e ious esul s, he hi d sub-hypo hesis is no suppo ed.
Fo he ou h sub-hypo hesis, H5d, “Being anspa en abou da a collec ion mechanisms has
a posi i e impac on Consume s’ In en ions o Pu chase he P oduc ”, he esul s show ha
Being anspa en abou da a collec ion mechanisms has an insigni ican e ec on Consume s’ in en ions
o pu chase he p oduc , as P- alue is 0.156. Acco dingly, he ou h sub-hypo hesis is no suppo ed.
Fo he i h sub-hypo hesis, “Being anspa en abou da a collec ion mechanisms has a
posi i e impac on Consume s’ P o ec i e Beha io ”, SEM p o es an insigni ican in luence o
Being anspa en on p o ec i e beha io , as P- alue is 0.086. Based on he p e ious esul s, he i h
sub-hypo hesis is no suppo ed.
F om all he abo e esul s, he i h hypo hesis is no suppo ed.
74
Figu e 3: Measu emen Model
75
Figu e 4: S uc u al Model
82
Fo he media ion ole o Pe cei ed In usi eness be ween independen a iables and
In en ions o Pu chase he P oduc , esul s can be in e p e ed as ollows:
Rega ding he media ing ole o PIN in he ela ionship be ween PER and IPP, he i s media ion
assessmen e eals an insigni ican indi ec e ec o PER on IPP h ough PIN (H6id: β= 0.009, = 0.668,
p=0.025). The e o e, H6id has no media ion e ec , sugges ing ha Pe cei ed In usi eness does no
media e he ela ionship be ween Pe sonaliza ion and consume s’ In en ions o Pu chase he P oduc .
Rega ding he media ing ole o PIN in he ela ionship be ween CC and IPP, he i s media ion
assessmen e eals an insigni ican indi ec e ec o CC on IPP h ough PIN (H6iid: β= 0.005, = 0.645,
p=0.260). The e o e, H6iid has no media ion e ec , sugges ing ha Consume s’ pe cep ions o
In usi eness do no play a ole in in luencing hei In en ions o Pu chase he P oduc .
Rega ding he media ing ole o PIN in he ela ionship be ween BF and IPP, he i s media ion
assessmen e eals an insigni ican indi ec e ec o BF on IPP h ough PIN (H6iiid: β= 0.006, = 0.650,
p=0.258). The e o e, H6iiid has no media ion e ec , sugges ing ha Pe cei ed In usi eness has no ole
in explaining how B and Familia i y shapes consume s’ In en ions o Pu chase he P oduc .
Rega ding he media ing ole o PIN in he ela ionship be ween BT and IPP, he i s media ion
assessmen e eals an insigni ican indi ec e ec o BT on IPP h ough PIN (H6i d: β= 0.012, = 0.670,
p=0.252). The e o e, H6i d has no media ion e ec , sugges ing ha Pe cei ed In usi eness has no ole
in explaining how B and T us shapes consume s’ In en ions o Pu chase he P oduc .
Finally, Rega ding he media ing ole o PIN in he ela ionship be ween TA and IPP, he i s media ion
assessmen e eals an insigni ican indi ec e ec o TA on IPP h ough PIN (H6 d: β= 0.002, = 0.327,
p=0.372). The e o e, H6 d has no media ion e ec , sugges ing ha Pe cei ed In usi eness does no
in luence he ela ionship be ween T anspa ency App oach and consume s’ In en ions o Pu chase he
P oduc .
Fo he media ion ole o Pe cei ed In usi eness be ween independen a iables and
P o ec i e Beha io , esul s can be in e p e ed as ollows:
Rega ding he media ing ole o PIN in he ela ionship be ween PER and PB, he i s media ion
assessmen e eals a signi ican indi ec e ec o PER on PB h ough PIN (H6ie: β= 0.026, = 0.013,
p=0.020). The di ec e ec o PER in he p esence o he media o on PB is signi ican (β=0.123, =
3.829, p≤0.001). The o al e ec o PER on PB is also signi ican (β=0.226, = 7.251, p≤0.001). This
shows a complemen a y pa ial media ing ole o PIN in he ela ionship be ween PER and PB (Hai e
83
al., 2021; Zhao e al., 2010). The e o e, H6ie is pa ially suppo ed. This sugges s ha Pe sonaliza ion's
impac on consume s’ P o ec i e Beha io can explained h ough he e ec o Pe cei ed In usi eness.
Rega ding he media ing ole o PIN in he ela ionship be ween CC and PB, he i s media ion
assessmen e eals a signi ican indi ec e ec o CC on PB h ough PIN (H6iie: β= 0.014, = 1.823,
p=0.034). The di ec e ec o CC in he p esence o he media o on PB is signi ican (β=0.369, =
12.823, p≤0.001). The o al e ec o CC on PB is also signi ican (β=0.426, = 14.295, p≤0.001). This
shows a complemen a y pa ial media ing ole o PIN in he ela ionship be ween CC and PB (Hai e al.,
2021; Zhao e al., 2010). The e o e, H6iie is pa ially suppo ed. The inding sugges ha Consume s’
pe cep ions o In usi eness play a ole in shaping hei P o ec i e Beha io .
Rega ding he media ing ole o PIN in he ela ionship be ween BF and PB, he i s media ion
assessmen e eals a signi ican indi ec e ec o BF on PB h ough PIN (H6iiie: β= 0.018, = 1.910,
p=0.028). The di ec e ec o BF in he p esence o he media o on PB is signi ican (β=0.075, =
2.306, p=0.011). The o al e ec o BF on PB is also signi ican (β=0.121, = 3.300, p≤0.001). This
shows a complemen a y pa ial media ing ole o PIN in he ela ionship be ween BF and PB (Hai e al.,
2021; Zhao e al., 2010). The e o e, H6iiie is pa ially suppo ed. Thus, Pe cei ed In usi eness plays a
signi ican ole in explaining how B and Familia i y shapes consume s’ P o ec i e Beha io .
Rega ding he media ing ole o PIN in he ela ionship be ween BT and PB, he i s media ion
assessmen e eals a signi ican indi ec e ec o BT on PB h ough PIN (H6i e: β= 0.036, = 2.098,
p=0.018). The di ec e ec o BT in he p esence o he media o on PB is signi ican (β=0.121, =
2.891, p=0.002). The o al e ec o BT on PB is also signi ican (β=0.228, = 5.380, p≤0.001). This
shows a complemen a y pa ial media ing ole o PIN in he ela ionship be ween BT and PB (Hai e al.,
2021; Zhao e al., 2010). The e o e, H6i e is pa ially suppo ed. Thus, while B and T us di ec ly impac s
consume s’ P o ec i e Beha io , pa o his ela ionship is media ed by Pe cei ed In usi eness.
In con as , Rega ding he media ing ole o PIN in he ela ionship be ween TA and PB, he i s media ion
assessmen e eals an insigni ican indi ec e ec o TA on PB h ough PIN (H6 e: β= 0.004, = 0.595,
p=0.276). The e o e, H6 e has no media ion e ec , sugges ing ha Pe cei ed In usi eness does no
in luence he ela ionship be ween T anspa ency App oach and consume s’ P o ec i e Beha io .
Fo he media ion ole o P i acy Conce ns be ween independen a iables and A i udes
owa ds he B and, i could be obse ed ha :
84
Rega ding he media ing ole o PRC in he ela ionship be ween PER and ATB, he i s media ion
assessmen e eals a signi ican indi ec e ec o PER on ATB h ough PRC (H7ia: β= 0.041, = 4.595,
p≤0.001). The di ec e ec o PER in he p esence o he media o on ATB is signi ican (β=0.111, =
6.348, p≤0.001). The o al e ec o PER on ATB is also signi ican (β=0.282, = 11.691, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween PER and ATB (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7ia is pa ially suppo ed. This sugges s ha Pe sonaliza ion's
impac on A i udes owa ds he B and can explained h ough he e ec o P i acy Conce ns in his model.
Rega ding he media ing ole o PRC in he ela ionship be ween CC and ATB, he i s media ion
assessmen e eals a signi ican indi ec e ec o CC on ATB h ough PRC (H7iia: β= 0.029, = 3.738,
p≤0.001). The di ec e ec o CC in he p esence o he media o on ATB is signi ican (β=0.224, =
11.688, p≤0.001). The o al e ec o CC on ATB is also signi ican (β=0.322, = 12.727, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween CC and ATB (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7iia is pa ially suppo ed. The inding sugges ha consume
pe cep ions o P i acy play a ole in shaping consume s’ A i udes owa ds he B and.
Rega ding he media ing ole o PRC in he ela ionship be ween BF and ATB, he i s media ion
assessmen e eals a signi ican indi ec e ec o BF on ATB h ough PRC (H7iiia: β= 0.021, = 2.838,
p=0.002). The di ec e ec o BF in he p esence o he media o on ATB is signi ican (β=0.087, =
4.134, p≤0.001). The o al e ec o BF on ATB is also signi ican (β=0.169, = 5.303, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween BF and ATB (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7iiia is pa ially suppo ed. Thus, P i acy Conce ns plays a
signi ican ole in explaining how B and Familia i y shapes consume s’ A i udes owa ds he B and.
Rega ding he media ing ole o PRC in he ela ionship be ween BT and ATB, he i s media ion
assessmen e eals a signi ican indi ec e ec o BT on ATB h ough PRC (H7i a: β= 0.053, = 4.811,
p≤0.001). The di ec e ec o BT in he p esence o he media o on ATB is signi ican (β=0.147, =
5.775, p≤0.001). The o al e ec o BT on ATB is also signi ican (β=0.337, = 9.419, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween BT and ATB (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7i a is pa ially suppo ed. The indings highligh he pi o al
ole o P i acy Conce ns in shaping consume s’ A i udes owa ds he b and based on B and T us .
Finally, Rega ding he media ing ole o PRC in he ela ionship be ween TA and ATB, he i s media ion
assessmen e eals an insigni ican indi ec e ec o TA on ATB h ough PRC (H7 a: β= 0.006, = 0.513,
85
p=0.304). The e o e, H6 a has no media ion e ec , sugges ing ha P i acy conce ns does no media e
he ela ionship be ween T anspa ency App oach and consume s' A i udes owa ds he B and.
Fo he media ion ole o P i acy Conce ns be ween independen a iables and A i udes
owa ds OBA, esul s can be in e p e ed as ollows:
Rega ding he media ing ole o PRC in he ela ionship be ween PER and ATO, he i s media ion
assessmen e eals a signi ican indi ec e ec o PER on ATO h ough PRC (H7ib: β= 0.019, = 1.682,
p=0.046). The di ec e ec o PER in he p esence o he media o on ATO is signi ican (β=0.104, =
3.295, p≤0.001). The o al e ec o PER on ATO is also signi ican (β=0.186, = 5.974, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween PER and ATO (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7ib is pa ially suppo ed. This sugges s ha Pe sonaliza ion's
impac on A i udes owa ds OBA can explained h ough he e ec o P i acy Conce ns.
Rega ding he media ing ole o PRC in he ela ionship be ween CC and ATO, he i s media ion
assessmen e eals an insigni ican indi ec e ec o CC on ATO h ough PRC (H7iib: β= 0.014, = 0.009,
p=0.056). The e o e, H7iib has no media ion e ec , sugges ing ha P i acy Conce ns does no in luence
he ela ionship be ween Consume Cha ac e is ics and A i udes owa ds OBA.
Rega ding he media ing ole o PRC in he ela ionship be ween BF and ATO, he i s media ion
assessmen e eals an insigni ican indi ec e ec o BF on ATO h ough PRC (H7iiib: β= 0.010, =
1.457, p=0.073). The e o e, H7iiib has no media ion e ec , sugges ing ha P i acy Conce ns has no
signi ican ole in explaining how B and Familia i y shapes consume s’ a i udes owa ds OBA.
Rega ding he media ing ole o PRC in he ela ionship be ween BT and ATO, he i s media ion
assessmen e eals an insigni ican indi ec e ec o BT on ATO h ough PRC (H7i b: β= 0.025, =
1.650, p=0.050). The e o e, H7i b has no media ion e ec , sugges ing ha P i acy Conce ns has no
signi ican ole in explaining how B and T us shapes consume s’ A i udes owa ds OBA.
Finally, Rega ding he media ing ole o PRC in he ela ionship be ween TA and ATO, he i s media ion
assessmen e eals an insigni ican indi ec e ec o TA on ATO h ough PRC (H7 b: β= 0.003, = 0.439,
p=0.330). The e o e, H7 b has no media ion e ec , sugges ing ha P i acy Conce ns does no in luence
he ela ionship be ween T anspa ency App oach and consume s' A i udes owa ds OBA.
Fo he media ion ole o P i acy Conce ns be ween independen a iables and In en ions
o Click on he Ad, esul s can be in e p e ed as ollows:
86
Rega ding he media ing ole o PRC in he ela ionship be ween PER and ICA, he i s media ion
assessmen e eals a signi ican indi ec e ec o PER on ICA h ough PRC (H7ic: β= 0.041, = 3.312,
p≤0.001). The di ec e ec o PER in he p esence o he media o on ICA is signi ican (β=0.114, =
3.994, p≤0.001). The o al e ec o PER on ICA is also signi ican (β=0.215, = 7.060, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween PER and ICA (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7ic is pa ially suppo ed. This sugges s ha Pe sonaliza ion's
impac on In en ions o Click on he Ad can be explained h ough he e ec o P i acy Conce ns.
Rega ding he media ing ole o PRC in he ela ionship be ween CC and ICA, he i s media ion
assessmen e eals a signi ican indi ec e ec o CC on ICA h ough PRC (H7iic: β= 0.028, = 2.901,
p=0.002). The di ec e ec o CC in he p esence o he media o on ICA is signi ican (β=0.420, =
15.161, p≤0.001). The o al e ec o CC on ICA is also signi ican (β=0.481, = 17.078, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween CC and ICA (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7iic is pa ially suppo ed. The inding sugges ha consume s'
pe cep ions o P i acy Conce ns play a ole in shaping hei In en ions o Click on he Ad.
Rega ding he media ing ole o PRC in he ela ionship be ween BF and ICA, he i s media ion
assessmen e eals a signi ican indi ec e ec o BF on ICA h ough PRC (H7iiic: β= 0.021, = 2.342,
p=0.010). The di ec e ec o BF in he p esence o he media o on ICA is signi ican (β=0.063, =
1.928, p=0.027). The o al e ec o BF on ICA is also signi ican (β=0.112, = 3.209, p=0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween BF and ICA (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7iiic is pa ially suppo ed. Thus, P i acy Conce ns plays a
signi ican ole in explaining how B and Familia i y shapes consume s' In en ions o Click on he Ad.
Rega ding he media ing ole o PRC in he ela ionship be ween BT and ICA, he i s media ion
assessmen e eals a signi ican indi ec e ec o BT on ICA h ough PRC (H7i c: β= 0.052, = 3.251,
p=0.001). The di ec e ec o BT in he p esence o he media o on ICA is signi ican (β=0.089, =
2.348, p=0.009). The o al e ec o BT on ICA is also signi ican (β=0.205, = 5.394, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween BT and ICA (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7i c is pa ially suppo ed. Thus, while B and T us di ec ly
impac s consume s’ In en ions o Click on he Ad, pa o his ela ionship is media ed by P i acy
Conce ns.
In con as , Rega ding he media ing ole o PRC in he ela ionship be ween TA and ICA, he i s
media ion assessmen e eals an insigni ican indi ec e ec o TA on ICA h ough PRC (H7 c: β= 0.006,
87
= 0.493, p=0.311). The e o e, H7 c has no media ion e ec , sugges ing ha P i acy Conce ns does
no in luence he ela ionship be ween T anspa ency App oach and consume s’ In en ions o Click on
he Ad.
Fo he media ion ole o P i acy Conce ns be ween independen a iables and In en ions
o Pu chase he P oduc , esul s can be in e p e ed as ollows:
Rega ding he media ing ole o PRC in he ela ionship be ween PER and IPP, he i s media ion
assessmen e eals a signi ican indi ec e ec o PER on IPP h ough PRC (H7id: β= 0.021, = 1.777,
p=0.038). The di ec e ec o PER in he p esence o he media o on IPP is signi ican (β=0.103, =
3.368, p≤0.001). The o al e ec o PER on IPP is also signi ican (β=0.193, = 6.478, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween PER and IPP (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7id is pa ially suppo ed. This sugges s ha Pe sonaliza ion's
impac on In en ions o Pu chase he P oduc can be explained h ough he e ec o P i acy Conce ns.
Rega ding he media ing ole o PRC in he ela ionship be ween CC and IPP, he i s media ion
assessmen e eals a signi ican indi ec e ec o CC on IPP h ough PRC (H7iid: β= 0.015, = 1.681,
p=0.046). The di ec e ec o CC in he p esence o he media o on IPP is signi ican (β=0.414, =
14.272, p≤0.001). The o al e ec o CC on IPP is also signi ican (β=0.467, = 15.730, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween CC and IPP (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7iid is pa ially suppo ed. This sugges s ha Consume s’
pe cep ions o P i acy Conce ns impac hei In en ions o Pu chase he P oduc .
Rega ding he media ing ole o PRC in he ela ionship be ween BF and IPP, he i s media ion
assessmen e eals an insigni ican indi ec e ec o BF on IPP h ough PIN (H7iiid: β= 0.011, = 1.547,
p=0.061). The e o e, H7iiid has no media ion e ec , sugges ing ha P i acy Conce ns has no ole in
explaining how B and Familia i y shapes consume s’ In en ions o Pu chase he P oduc .
Rega ding he media ing ole o PRC in he ela ionship be ween BT and IPP, he i s media ion
assessmen e eals a signi ican indi ec e ec o BT on IPP h ough PRC (H7i d: β= 0.028, = 1.791,
p=0.037). The di ec e ec o BT in he p esence o he media o on IPP is signi ican (β=0.128, =
3.309, p≤0.001). The o al e ec o BT on IPP is also signi ican (β=0.218, = 5.665, p≤0.001). This
shows a complemen a y pa ial media ing ole o PRC in he ela ionship be ween BT and IPP (Hai e
al., 2021; Zhao e al., 2010). The e o e, H7i d is pa ially suppo ed. This sugges s ha B and T us
impac s consume s’ In en ions o Pu chase he P oduc in luenced by hei P i acy Conce ns.
88
Finally, Rega ding he media ing ole o PRC in he ela ionship be ween TA and IPP, he i s media ion
assessmen e eals an insigni ican indi ec e ec o TA on IPP h ough PRC (H7 d: β= 0.003, = 0.457,
p=0.324). The e o e, H7 d has no media ion e ec , sugges ing ha P i acy Conce ns does no in luence
he ela ionship be ween T anspa ency App oach and consume s’ In en ions o Pu chase he P oduc .
Fo he media ion ole o P i acy Conce ns be ween independen a iables and P o ec i e
Beha io , esul s can be in e p e ed as ollows:
Rega ding he media ing ole o PRC in he ela ionship be ween PER and PB, he i s media ion
assessmen e eals an insigni ican indi ec e ec o PER on PB h ough PRC (H7ie: β= 0.015, = 1.210,
p=0.113). The e o e, H7ie has no media ion e ec , sugges ing ha P i acy Conce ns does no in luence
he ela ionship be ween Pe sonaliza ion and consume s’ P o ec i e Beha io .
Rega ding he media ing ole o PRC in he ela ionship be ween CC and PB, he i s media ion
assessmen e eals an insigni ican indi ec e ec o CC on PB h ough PRC (H7iie: β= 0.010, = 1.196,
p=0.116). The e o e, H7iie has no media ion e ec , sugges ing ha P i acy Conce ns does no in luence
he ela ionship be ween Consume Cha ac e is ics and consume s’ P o ec i e Beha io .
Rega ding he media ing ole o PRC in he ela ionship be ween BF and PB, he i s media ion
assessmen e eals an insigni ican indi ec e ec o BF on PB h ough PRC (H7iiie: β= 0.007, = 1.112,
p=0.133). The e o e, H7iiie has no media ion e ec , sugges ing ha P i acy Conce ns does no in luence
he ela ionship be ween B and Familia i y and consume s’ P o ec i e Beha io .
Rega ding he media ing ole o PRC in he ela ionship be ween BT and PB, he i s media ion
assessmen e eals an insigni ican indi ec e ec o BT on PB h ough PRC (H7i e: β= 0.019, = 1.229,
p=0.110). The e o e, H7i e has no media ion e ec , sugges ing ha P i acy Conce ns does no in luence
he ela ionship be ween B and T us and consume s’ P o ec i e Beha io .
Finally, Rega ding he media ing ole o PRC in he ela ionship be ween TA and PB, he i s media ion
assessmen e eals an insigni ican indi ec e ec o TA on PB h ough PRC (H7 e: β= 0.002, = 0.393,
p=0.347). The e o e, H7 e has no media ion e ec , sugges ing ha P i acy Conce ns does no in luence
he ela ionship be ween T anspa ency App oach and consume s’ P o ec i e Beha io .
Fo he media ion ole o Reac ance be ween independen a iables and A i udes owa ds
he B and, i could be obse ed ha :
Rega ding he media ing ole o REA in he ela ionship be ween PER and ATB, he i s media ion
assessmen e eals a signi ican indi ec e ec o PER on ATB h ough REA (H8ia: β= 0.076, = 5.759,
89
p≤0.001). The di ec e ec o PER in he p esence o he media o on ATB is signi ican (β=0.111, =
6.348, p≤0.001). The o al e ec o PER on ATB is also signi ican (β=0.282, = 11.691, p≤0.001). This
shows a complemen a y pa ial media ing ole o REA in he ela ionship be ween PER and ATB (Hai e
al., 2021; Zhao e al., 2010). The e o e, H8ia is pa ially suppo ed. This sugges s ha Pe sonaliza ion's
impac on A i udes owa ds he B and can explained h ough he e ec o consume s’ eelings o
Reac ance.
Rega ding he media ing ole o REA in he ela ionship be ween CC and ATB, he i s media ion
assessmen e eals a signi ican indi ec e ec o CC on ATB h ough REA (H8iia: β= 0.042, = 4.213,
p≤0.001). The di ec e ec o CC in he p esence o he media o on ATB is signi ican (β=0.224, =
11.688, p≤0.001). The o al e ec o CC on ATB is also signi ican (β=0.322, = 12.727, p≤0.001). This
shows a complemen a y pa ial media ing ole o REA in he ela ionship be ween CC and ATB (Hai e
al., 2021; Zhao e al., 2010). The e o e, H8iia is pa ially suppo ed. The inding sugges ha consume
pe cep ions o Reac ance play a ole in shaping consume s’ A i udes owa ds he B and.
Rega ding he media ing ole o REA in he ela ionship be ween BF and ATB, he i s media ion
assessmen e eals a signi ican indi ec e ec o BF on ATB h ough PRC (H8iiia: β= 0.025, = 2.353,
p=0.009). The di ec e ec o BF in he p esence o he media o on ATB is signi ican (β=0.087, =
4.134, p≤0.001). The o al e ec o BF on ATB is also signi ican (β=0.169, = 5.303, p≤0.001). This
shows a complemen a y pa ial media ing ole o REA in he ela ionship be ween BF and ATB (Hai e
al., 2021; Zhao e al., 2010). The e o e, H8iiia is pa ially suppo ed. Thus, Reac ance plays a signi ican
ole in explaining how B and Familia i y shapes consume s’ A i udes owa ds he B and.
Rega ding he media ing ole o REA in he ela ionship be ween BT and ATB, he i s media ion
assessmen e eals a signi ican indi ec e ec o BT on ATB h ough REA (H8i a: β= 0.064, = 5.014,
p≤0.001). The di ec e ec o BT in he p esence o he media o on ATB is signi ican (β=0.147, =
5.775, p≤0.001). The o al e ec o BT on ATB is also signi ican (β=0.337, = 9.419, p≤0.001). This
shows a complemen a y pa ial media ing ole o REA in he ela ionship be ween BT and ATB (Hai e
al., 2021; Zhao e al., 2010). The e o e, H8i a is pa ially suppo ed. The indings highligh he pi o al
ole o Reac ance in shaping consume s’ pe cep ions o B and T us and hei A i udes owa ds he
B and.
In con as , Rega ding he media ing ole o REA in he ela ionship be ween TA and ATB, he i s
media ion assessmen e eals an insigni ican indi ec e ec o TA on ATB h ough REA (H8 a: β= -
90
0.009, = 0.598, p=0.275). The e o e, H8 a has no media ion e ec , sugges ing ha Reac ance does
no media e he ela ionship be ween T anspa ency App oach and A i udes owa ds he B and.
Fo he media ion ole o Reac ance be ween independen a iables and A i udes owa ds
OBA, esul s can be in e p e ed as ollows:
Rega ding he media ing ole o REA in he ela ionship be ween PER and ATO, he i s media ion
assessmen e eals an insigni ican indi ec e ec o PER on ATO h ough REA (H8ib: β= 0.018, =
1.061, p=0.144). The e o e, H8ib has no media ion e ec , sugges ing ha Reac ance does no in luence
he ela ionship be ween Pe sonaliza ion and A i udes owa ds OBA.
Rega ding he media ing ole o REA in he ela ionship be ween CC and ATO, he i s media ion
assessmen e eals an insigni ican indi ec e ec o CC on ATO h ough REA (H8iib: β= 0.010, =
01.045, p=0.148). The e o e, H8iib has no media ion e ec , sugges ing ha Reac ance does no
in luence he ela ionship be ween Consume Cha ac e is ics and A i udes owa ds OBA.
Rega ding he media ing ole o REA in he ela ionship be ween BF and ATO, he i s media ion
assessmen e eals an insigni ican indi ec e ec o BF on ATO h ough REA (H8iiib: β= 0.006, =
0.911, p=0.181). The e o e, H8iiib has no media ion e ec , sugges ing ha Reac ance has no signi ican
ole in explaining how B and Familia i y shapes consume s’ A i udes owa ds OBA.
Rega ding he media ing ole o REA in he ela ionship be ween BT and ATO, he i s media ion
assessmen e eals an insigni ican indi ec e ec o BT on ATO h ough REA (H8i b: β= 0.015, = 1.048,
p=0.147). The e o e, H8i b has no media ion e ec , sugges ing ha Reac ance has no signi ican ole in
explaining how B and T us shapes consume s’ A i udes owa ds OBA.
Finally, Rega ding he media ing ole o REA in he ela ionship be ween TA and ATO, he i s media ion
assessmen e eals an insigni ican indi ec e ec o TA on ATO h ough PRC (H8 b: β= -0.002, =
0.411, p=0.340). The e o e, H8 b has no media ion e ec , sugges ing ha Reac ance does no in luence
he ela ionship be ween T anspa ency App oach and A i udes owa ds OBA.
Fo he media ion ole o Reac ance be ween independen a iables and In en ions o Click
on he Ad, esul s can be in e p e ed as ollows:
Rega ding he media ing ole o REA in he ela ionship be ween PER and ICA, he i s media ion
assessmen e eals a signi ican indi ec e ec o PER on ICA h ough REA (H8ic: β= 0.034, = 2.002,
p=0.023). The di ec e ec o PER in he p esence o he media o on ICA is signi ican (β=0.114, =
3.994, p≤0.001). The o al e ec o PER on ICA is also signi ican (β=0.215, = 7.060, p≤0.001). This
91
shows a complemen a y pa ial media ing ole o REA in he ela ionship be ween PER and ICA (Hai e
al., 2021; Zhao e al., 2010). The e o e, H8ic is pa ially suppo ed. This sugges s ha Pe sonaliza ion's
impac on In en ions o Click on he Ad can be explained h ough he e ec o Reac ance.
Rega ding he media ing ole o REA in he ela ionship be ween CC and ICA, he i s media ion
assessmen e eals a signi ican indi ec e ec o CC on ICA h ough REA (H8iic: β= 0.019, = 1.912,
p=0.028). The di ec e ec o CC in he p esence o he media o on ICA is signi ican (β=0.420, =
15.161, p≤0.001). The o al e ec o CC on ICA is also signi ican (β=0.481, = 17.078, p≤0.001). This
shows a complemen a y pa ial media ing ole o REA in he ela ionship be ween CC and ICA (Hai e
al., 2021; Zhao e al., 2010). The e o e, H8iic is pa ially suppo ed. The inding sugges ha consume
pe cep ions o Reac ance play a ole in shaping hei In en ions o Click on he Ad.
Rega ding he media ing ole o REA in he ela ionship be ween BF and ICA, he i s media ion
assessmen e eals an insigni ican indi ec e ec o BF on ICA h ough REA (H8iiic: β= 0.011, = 1.497,
p=0.067). The e o e, H8iiic has no media ion e ec , sugges ing ha Reac ance does no in luence he
ela ionship be ween B and Familia i y and In en ions o Click on he Ad.
Rega ding he media ing ole o REA in he ela ionship be ween BT and ICA, he i s media ion
assessmen e eals a signi ican indi ec e ec o BT on ICA h ough REA (H8i c: β= 0.029, = 1.981,
p=0.024). The di ec e ec o BT in he p esence o he media o on ICA is signi ican (β=0.089, =
2.348, p=0.009). The o al e ec o BT on ICA is also signi ican (β=0.205, = 5.394, p≤0.001). This
shows a complemen a y pa ial media ing ole o REA in he ela ionship be ween BT and ICA (Hai e
al., 2021; Zhao e al., 2010). The e o e, H8i c is pa ially suppo ed. Thus, while B and T us di ec ly
impac s consume s’ In en ions o Click on he Ad, pa o his ela ionship is media ed by Reac ance.
In con as , Rega ding he media ing ole o REA in he ela ionship be ween TA and ICA, he i s
media ion assessmen e eals an insigni ican indi ec e ec o TA on ICA h ough REA (H8 c: β= -0.004,
= 0.548, p=0.292). The e o e, H8 c has no media ion e ec , sugges ing ha Reac ance does no
in luence he ela ionship be ween T anspa ency App oach and consume s’ In en ions o Click on he
Ad.
Fo he media ion ole o Reac ance be ween independen a iables and In en ions o
Pu chase he P oduc , esul s can be in e p e ed as ollows:
Rega ding he media ing ole o REA in he ela ionship be ween PER and IPP, he i s media ion
assessmen e eals a signi ican indi ec e ec o PER on IPP h ough REA (H8id: β= 0.061, = 3.128,
98
Hypo hesis
β
SD
p
Hypo heses
Decision
H7ia: Pe sonaliza ion -> P i acy Conce ns -
> A i ude owa ds he B and
0.041
0.009
4.595
0.000
Pa ially
Suppo ed
H7ib Pe sonaliza ion -> P i acy Conce ns -
> A i ude owa ds OBA
0.019
0.012
1.682
0.046
Pa ially
Suppo ed
H7ic Pe sonaliza ion -> P i acy Conce ns -
> In en ion o Click on he Ad
0.041
0.012
3.312
0.000
Pa ially
Suppo ed
H7id Pe sonaliza ion -> P i acy Conce ns -
> In en ion o Pu chase he P oduc
0.021
0.012
1.777
0.038
Pa ially
Suppo ed
H7ie Pe sonaliza ion -> P i acy Conce ns -
> P o ec i e Beha io
0.015
0.012
1.210
0.113
No Media ion
H7iia: Consume Cha ac e is ics -> P i acy
Conce ns -> A i ude owa ds he B and
0.029
0.008
3.738
0.000
Pa ially
Suppo ed
H7iib Consume Cha ac e is ics -> P i acy
Conce ns -> A i ude owa ds OBA
0.014
0.009
1.587
0.056
No Media ion
H7iic Consume Cha ac e is ics -> P i acy
Conce ns -> In en ion o Click on he Ad
0.028
0.010
2.901
0.002
Pa ially
Suppo ed
H7iid Consume Cha ac e is ics -> P i acy
Conce ns -> In en ion o Pu chase he
P oduc
0.015
0.009
1.681
0.046
Pa ially
Suppo ed
H7iie Consume Cha ac e is ics -> P i acy
Conce ns -> P o ec i e Beha io
0.010
0.009
1.196
0.116
No Media ion
H7iiia: B and Familia i y -> P i acy
Conce ns -> A i ude owa ds he B and
0.021
0.007
2.838
0.002
Pa ially
Suppo ed
H7iiib B and Familia i y -> P i acy
Conce ns -> A i ude owa ds OBA
0.010
0.007
1.457
0.073
No Media ion
H7iiic B and Familia i y -> P i acy
Conce ns -> In en ion o Click on he Ad
0.021
0.009
2.342
0.010
Pa ially
Suppo ed
H7iiid B and Familia i y -> P i acy
Conce ns -> In en ion o Pu chase he
P oduc
0.011
0.007
1.547
0.061
No Media ion
H7iiie B and Familia i y -> P i acy
Conce ns -> P o ec i e Beha io
0.007
0.007
1.112
0.133
No Media ion
H7i a: B and T us -> P i acy Conce ns ->
A i ude owa ds he B and
0.053
0.011
4.811
0.000
Pa ially
Suppo ed
H7i b B and T us -> P i acy Conce ns ->
A i ude owa ds OBA
0.025
0.015
1.650
0.050
No Media ion
H7i c B and T us -> P i acy Conce ns ->
In en ion o Click on he Ad
0.052
0.016
3.251
0.001
Pa ially
Suppo ed
99
Hypo hesis
β
SD
p
Hypo heses
Decision
H7i d B and T us -> P i acy Conce ns ->
In en ion o Pu chase he P oduc
0.028
0.015
1.791
0.037
Pa ially
Suppo ed
H7i e B and T us -> P i acy Conce ns ->
P o ec i e Beha io
0.019
0.015
1.229
0.110
No Media ion
H7 a: T anspa ency App oach -> P i acy
Conce ns -> A i ude owa ds he B and
0.006
0.012
0.513
0.304
No Media ion
H7 b T anspa ency App oach -> P i acy
Conce ns -> A i ude owa ds OBA
0.003
0.006
0.439
0.330
No Media ion
H7 c T anspa ency App oach -> P i acy
Conce ns -> In en ion o Click on he Ad
0.006
0.012
0.493
0.311
No Media ion
H7 d T anspa ency App oach -> P i acy
Conce ns -> In en ion o Pu chase he
P oduc
0.003
0.007
0.457
0.324
No Media ion
H7 e T anspa ency App oach -> P i acy
Conce ns -> P o ec i e Beha io
0.002
0.005
0.393
0.347
No Media ion
H8ia: Pe sonaliza ion -> Reac ance ->
A i ude owa ds he B and
0.076
0.013
5.759
0.000
Pa ially
Suppo ed
H8ib Pe sonaliza ion -> Reac ance ->
A i ude owa ds OBA
0.018
0.017
1.061
0.144
No Media ion
H8ic Pe sonaliza ion -> Reac ance ->
In en ion o Click on he Ad
0.034
0.017
2.002
0.023
Pa ially
Suppo ed
H8id Pe sonaliza ion -> Reac ance ->
In en ion o Pu chase he P oduc
0.060
0.019
3.128
0.001
Pa ially
Suppo ed
H8ie Pe sonaliza ion -> Reac ance ->
P o ec i e Beha io
0.062
0.018
3.406
0.000
Pa ially
Suppo ed
H8iia: Consume Cha ac e is ics ->
Reac ance -> A i ude owa ds he B and
0.042
0.010
4.213
0.000
Pa ially
Suppo ed
H8iib Consume Cha ac e is ics ->
Reac ance -> A i ude owa ds OBA
0.010
0.010
1.045
0.148
No Media ion
H8iic Consume Cha ac e is ics ->
Reac ance -> In en ion o Click on he Ad
0.019
0.010
1.912
0.028
Pa ially
Suppo ed
H8iid Consume Cha ac e is ics ->
Reac ance -> In en ion o Pu chase he
P oduc
0.033
0.011
2.908
0.002
Pa ially
Suppo ed
H8iie Consume Cha ac e is ics ->
Reac ance -> P o ec i e Beha io
0.034
0.011
3.052
0.001
Pa ially
Suppo ed
H8iiia: B and Familia i y -> Reac ance ->
A i ude owa ds he B and
0.025
0.010
2.353
0.009
Pa ially
Suppo ed
100
Hypo hesis
β
SD
p
Hypo heses
Decision
H8iiib B and Familia i y -> Reac ance ->
A i ude owa ds OBA
0.006
0.006
0.911
0.181
No Media ion
H8iiic B and Familia i y -> Reac ance ->
In en ion o Click on he Ad
0.011
0.007
1.497
0.067
No Media ion
H8iiid B and Familia i y -> Reac ance ->
In en ion o Pu chase he P oduc
0.019
0.010
1.922
0.027
Pa ially
Suppo ed
H8iiie B and Familia i y -> Reac ance ->
P o ec i e Beha io
0.020
0.010
1.962
0.025
Pa ially
Suppo ed
H8i a: B and T us -> Reac ance ->
A i ude owa ds he B and
0.064
0.013
5.014
0.000
Pa ially
Suppo ed
H8i b B and T us -> Reac ance -> A i ude
owa ds OBA
0.015
0.015
1.048
0.147
No Media ion
H8i c B and T us -> Reac ance ->
In en ion o Click on he Ad
0.029
0.015
1.981
0.024
Pa ially
Suppo ed
H8i d B and T us -> Reac ance ->
In en ion o Pu chase he P oduc
0.050
0.016
3.056
0.001
Pa ially
Suppo ed
H8i e B and T us -> Reac ance ->
P o ec i e Beha io
0.052
0.016
3.268
0.001
Pa ially
Suppo ed
H8 a: T anspa ency App oach ->
Reac ance -> A i ude owa ds he B and
-0.009
0.015
0.598
0.275
No Media ion
H8 b T anspa ency App oach ->
Reac ance -> A i ude owa ds OBA
-0.002
0.005
0.411
0.340
No Media ion
H8 c T anspa ency App oach ->
Reac ance -> In en ion o Click on he Ad
-0.004
0.008
0.548
0.292
No Media ion
H8 d T anspa ency App oach ->
Reac ance -> In en ion o Pu chase he
P oduc
-0.007
0.013
0.574
0.283
No Media ion
H8 e T anspa ency App oach ->
Reac ance -> P o ec i e Beha io
-0.008
0.013
0.589
0.278
No Media ion
101
DISCUSSION AND CONCLUSION
102
5.1. In oduc ion
This chap e p esen s an o e all analysis and discussion o he p ima y inding o he cu en esea ch, in
compa ison wi h p e iously e iewed li e a u e. Followed by he heo e ical and p ac ical con ibu ions o he
s udy. Then, esea ch limi a ions and sugges ions o u u e wo k a e discussed. Finally, a sho conclusion
o he whole s udy is p esen ed.
5.2. Discussion
The pu pose o he s udy was o explo e he ole o some speci ic a iables on consume s’ beha io al
ou comes o OBA. Speci ically, he aim o his esea ch was o explo e he ole o pe sonaliza ion, consume
cha ac e is ics, b and amilia i y, b and us , and anspa ency on consume s’ a i udes owa d OBA,
a i udes owa d he b and, in en ion o click on he ad, in en ion o pu chase he ad e ised p oduc and
inally on OBA a oidance. Mo eo e , he s udy also explo ed he media ing in luence o pe cei ed
in usi eness, pe cei ed p i acy conce ns, and eac ance on consume s’ a i udes, in en ions and beha io al
ou comes.
P e ious s udies ha e add essed hese p e iously men ioned a iables, bu no o all o hem simul aneously.
Also, some speci ic a iables we e no e en s udied be o e in ela ion o OBA bu only o online ad e ising
in gene al. In addi ion, we build upon he Social Exchange Theo y and he Phycological Reac ance Theo y o
explain he concep o OBA and i s an eceden s, The esea ch p oblem can be summa ized in he ques ion
o “Wha a e he ac o s ha in luence consume s a i udes, in en ions, and beha io s in he online
en i onmen owa ds an online beha io ally a ge ed ad e isemen ? And wha ac o s can media e such
e ec ?”. F om his p oblem and a e e iewing p e ious li e a u e and selec ing he app op ia e a iables,
he esea ch ques ions we e de eloped as:
1. How do pe sonaliza ion, consume cha ac e is ics, b and amilia i y, b and us , and anspa ency
in luence consume s’ a i udes, in en ions and OBA a oidance in he online en i onmen ?
2. How do pe cei ed in usi eness, p i acy conce ns, and eac ance media e such in luence?
In o de o add ess he esea ch p oblem and answe he esea ch ques ions, a quan i a i e me hodology
was adop ed; whe e an online expe imen and a s uc u ed ques ionnai e su ey using Go illa SC pla o ms
we e used as esea ch ins umen s. The ques ionnai e was dis ibu ed h ough a non-p obabili y sampling
me hod (Con enience sampling), ha esul ed in 487 alid esponses o online In e ne use s. The conduc ed
103
su ey ques ions we e based on he adop ion o scales widely es ed in p e ious li e a u e, and hus he aim
was o con i m o wha ex en he model p esen ed in his s udy i ed he in o ma ion p esen ed in he
li e a u e. In his ligh , s uc u al equa ion analysis was used using Sma PLS. The esul s ob ained
h oughou his esea ch will now be discussed in ela ion o he esul s o he exis ing li e a u e, hen
heo e ical and p ac ical implica ions will be p esen ed.
H1:
Pe sonaliza ion has a posi i e impac on a) Consume s’ a i udes owa ds he b and, b) Consume s’
a i udes owa ds OBA, c) Consume s’ in en ion o click on he Ad, d) Consume s’ in en ions o pu chase he
p oduc and e) Consume s’ p o ec i e beha io .
This hypo hesis is di ided in o i e sub-hypo heses, whe e all he sub-hypo heses we e suppo ed. This inding
goes in line wi h some p e ious li e a u e, ha sugges ha di e en le els o pe sonaliza ion can in luence
consume ’s a i udes owa ds he ad and he b and, and in en ions o click he pe sonalized ad (Boe man e
al., 2017). The esul s o wo k o Agui e e al. (2015) poin s ou ha highe pe sonaliza ion leads o highe
click- h ough in en ion. Bleie (2015) s a ed highe pe sonaliza ion in luence OBA ou comes exp essed in
click h ough a es and pu chase in en ions, as i enhances online use s’ pe cei ed in o ma i eness by
ecei ing in o ma ion ele an o hei in e es s wi hou going h ough loads o i ele an in o ma ion. In
con as , Van Doo n & Hoeks a (2013) ound ha a highe deg ee o pe sonaliza ion in luence eeling o
inc eased in usi eness by online cus ome s which in u n leads o OBA a oidance.
H2:
Conside ing consume cha ac e is ics has a posi i e impac on a) Consume s’ a i udes owa ds he
b and, b) Consume s’ a i udes owa ds OBA, c) Consume s’ in en ion o click on he Ad, d) Consume s’
in en ions o pu chase he p oduc and e) Consume s’ p o ec i e beha io .
This hypo hesis is di ided in o i e sub-hypo heses, whe e all he sub-hypo heses we e suppo ed. This inding
goes in line wi h some p e ious li e a u e, Tu ow e al. (2009) poin s ou ha pe cep ions o OBA and a i udes
owa ds pe sonalized ads depend on consume cha ac e is ics such as age, whe e younge consume s a e
mo e accep ing o OBA compa ed o olde consume s. Also, he wo k o Smi e al. (2014) discussed how
consume s’ cha ac e is ics such as age, educa ion le el and online expe ience in luenced hei a i udes and
esponses owa ds OBA, whe e he olde emale consume s wi h a low le el o educa ion and a low amily
income ended o a oid OBA.
104
H3:
B and Familia i y has a posi i e impac on a) Consume s’ a i udes owa ds he b and, b) Consume s’
a i udes owa ds OBA, c) Consume s’ in en ion o click on he Ad, d) Consume s’ in en ions o pu chase he
p oduc and e) Consume s’ p o ec i e beha io .
This hypo hesis is di ided in o i e sub-hypo heses, whe e all he sub-hypo heses we e suppo ed. This inding
goes in line wi h some p e ious li e a u e, whe e he wo k o Dodds and Mon oe (1985) epo ed ha b and
amilia i y in luenced consume s’ a i udes and pe cep ions abou he p oduc quali y and e en ually
consume s’ in en ions o pu chase he p oduc . La e , Dodds e al. (1991) in ano he s udy, epo ed ha
b and name had a posi i e e ec on consume s’ pu chase in en ion. Also, Shoenbe ge (2014) claimed ha
b and amilia i y is a key heu is ic used by consume s o make a pu chasing decision. Finally, Zhao e al.
(2017) claimed ha b and amilia i y led consume s o pe cei e a pe sonalized ad as less in usi e and hus
dec eased OBA a oidance.
H4:
B and us has a posi i e impac on a) Consume s’ a i udes owa ds he b and, b) Consume s’ a i udes
owa ds OBA, c) Consume s’ in en ion o click on he Ad, d) Consume s’ in en ions o pu chase he p oduc
and e) Consume s’ p o ec i e beha io .
This hypo hesis is di ided in o i e sub-hypo heses, whe e all he sub-hypo heses we e suppo ed. This inding
goes in line wi h some p e ious li e a u e, whe e he wo k o McS ay (2011) con i med he us is key
de e minan o consume s’ pe cep ions, a i udes and beha io s in he online en i onmen . Kuma and Gup a
(2016) s a ed ha us can be c ea ed h ough b and name. Also, a s udy by Ea p and Baume (2003)
ound ha b and us leads consume s o disclose hei pe sonal in o ma ion online be e han an unknown
websi e wi h good p i acy policy. Rony (2018) assumed ha online consume s will only engage wi h i ms o
b ands ha hey us mo e. Bleie (2015) also ound ha inc eased click- h ough in en ions a es o highly
pe sonalized ads when online cus ome s us he ad e ise . Finally, Kuma and Pansa i (2016) claimed
ha i online consume s us ed he ad e ising b and, hen hey will be mo e willing o sha e hei p i a e
in o ma ion.
H5:
Being anspa en abou da a collec ion mechanisms has a posi i e impac on a) Consume s’ a i udes
owa ds he b and, b) Consume s’ a i udes owa ds OBA, c) Consume s’ in en ion o click on he Ad, d)
Consume s’ in en ions o pu chase he p oduc and e) Consume s’ p o ec i e beha io .
105
This hypo hesis is di ided in o i e sub-hypo heses, whe e all he sub-hypo heses we e no suppo ed.
Su p isingly, his inding also goes in line wi h some p e ious li e a u e, whe e he wo k o Leon e al. (2012)
showed how pa icipan s misunde s ood o a ely ecognized disclosu e icons, and inco ec ly unde s ood
he agline o in end selling ad e ising space ins ead o being a link o pages whe e hey can make choices
abou OBA. The same happened in he cu en s udy whe e pa icipan s ei he did no ecognize o did no
unde s and he ques ion “Why am I seeing his ad?” associa ed wi h he anspa en expe imen hey we e
subjec ed o.
H6: Pe cei ed in usi eness media es he in luence o i) pe sonaliza ion, ii) consume cha ac e is ics, iii)
b and amilia i y i ) b and us and ) anspa ency app oach on a) Consume s’ a i udes owa ds he b and,
b) Consume s’ a i udes owa ds OBA, c) Consume s’ in en ion o click on he Ad, d) Consume s’ in en ions
o pu chase he p oduc and e) Consume s’ p o ec i e beha io .
H7: P i acy conce ns media e he in luence o i) pe sonaliza ion, ii) consume cha ac e is ics, iii) b and
amilia i y i ) b and us and ) anspa ency app oach on a) Consume s’ a i udes owa ds he b and, b)
Consume s’ a i udes owa ds OBA, c) Consume s’ in en ion o click on he Ad, d) Consume s’ in en ions o
pu chase he p oduc and e) Consume s’ p o ec i e beha io .
H8: Reac ance media es he in luence o i) pe sonaliza ion, ii) consume cha ac e is ics, iii) b and amilia i y
i ) b and us and ) anspa ency app oach on a) Consume s’ a i udes owa ds he b and, b) Consume s’
a i udes owa ds OBA, c) Consume s’ in en ion o click on he Ad, d) Consume s’ in en ions o pu chase he
p oduc and e) Consume s’ p o ec i e beha io .
Each one o he p e iously men ioned media ing hypo heses is di ided in o wen y- i e sub-hypo heses, whe e
all he sub-hypo heses had ei he a pa ial media ion ole o a no media ion ole a all.
Fi s , all o he h ee media o s namely, Pe cei ed In usi eness, P i acy Conce ns, and Reac ance, had no
media ion e ec a all be ween T anspa ency App oach and all he dependen a iables. This can be aced
back o he ac ha consume s did no ecognize/ unde s and he anspa ency ques ion associa ed o he
andom expe imen hey we e subjec o, which may in u n ha e a ec ed hei answe s o he su ey
ques ions. This is con adic ing wi h p e ious li e a u e ha claimed online consume s app ecia ed hones y,
whe e implemen ing anspa ency mechanisms had an e ec on consume s’ eelings o ulne abili y, which
in u n a ec ed consume s’ us , a i udes and beha io s owa ds OBA (Agui e e al. 2015). Boe man e al.
106
(2017) also claimed ha including anspa ency mechanisms in o an ad message, lead consume s o accep
o click a highly pe sonalized ad e isemen despi e hei p i acy conce ns and eelings o pe cei ed
in usi eness.
Second, Pe cei ed In usi eness pa ially media ed he e ec o Pe sonaliza ion, Consume Cha ac e is ics,
B and Familia i y and B and T us on Consume s’ A i udes owa ds he B and, Consume s’ A i udes
owa ds OBA, Consume s’ In en ion o Click on he Ad, and Consume s’ p o ec i e beha io . Howe e , he e
was no media ion e ec o Pe cei ed In usi eness be ween Pe sonaliza ion, Consume Cha ac e is ics,
B and Familia i y, B and T us and Consume s’ In en ions o Pu chase he P oduc . This goes in line wi h
p e ious li e a u e, whe e he wo k o Doo n & Hoeks a (2013) implied ha pe sonaliza ion leads o be e
consume s’ a i udes and highe click- h ough in en ions bu also o highe le els o pe cei ed in usi eness,
hese eelings o in usi eness in e up he cogni i e p ocessing o online cus ome s which may lead o lowe
click- h ough a es and in en ions. Hühn e al. (2017), suppo ed ha hose eelings o in usi eness gene a e
ad annoyance and a oidance beha io among use s in he online en i onmen . While, Zhao e al. (2017)
claimed ha b and amilia i y, and b and us could lead consume s o expe ience he ads as less in usi e.
Thi d, P i acy Conce ns had no media ion e ec be ween Pe sonaliza ion, Consume Cha ac e is ics, B and
Familia i y, B and T us and consume s’ P o ec i e Beha io . Also, no media ion e ec was ound be ween
Consume Cha ac e is ics, B and T us , B and Familia i y and A i udes owa ds OBA. Finally, no media ion
e ec was ound be ween B and Familia i y and consume s’ In en ions o Pu chase he P oduc . While,
P i acy Conce ns pa ially media ed he e ec o he es o he independen a iables on he dependen
a iables. This goes in line wi h p e ious li e a u e, Xu e al. (2011) claimed ha pe sonalized ads had
signi ican e ec s on consume s’ p i acy conce ns when consume s we e p esen ed by an ad esembling
hei pe sonal in o ma ion o sea ch his o y. Baek and Mo imo o (2012) s a ed ha when consume s’ eelings
o p i acy conce ns inc ease, his leads o inc eased ad skep icism, which consequen ly leads o mo e
a oidance o OBA. Addi ionally, Fo es e al. (2017) claimed ha use s’ pe cep ion o p i acy issues a ec
hei a i udes and in en ions o click online ads.
Finally, Reac ance had no media ion e ec be ween Pe sonaliza ion, Consume Cha ac e is ics, B and
Familia i y, B and T us and consume s’ A i udes owa ds OBA. Also, no media ion e ec was ound be ween
B and Familia i y and consume s’ In en ions o Click on he Ad. Howe e , Reac ance pa ially media ed he
e ec o he es o he independen a iables on he dependen a iables. This goes in line wi h p e ious
107
li e a u e, Bleie and Eisenbeiss (2015) claimed ha al hough pe sonalized ads a e use ul, hey igge
eelings o eac ance, which may lead o nega i e beha io al ou comes owa ds he ad message o he
ad e ised b and, and lowe pu chase in en ions and click- h ough a es (Agui e e al., 2015).
5.3. Theo e ical Implica ions
This s udy deepens he unde s anding o OBA concep and i s e ec i eness. The esea ch aimed o con ibu e
o he academic schola ship and gi e be e unde s anding o he issue unde in es iga ion – OBA and
selec ed a iables – and i s ela ion o he Social Exchange Theo y and Phycological Reac ance Theo y.
The esea ch used he Social Exchange Theo y (SET) and he Phycological Reac ance Theo y o desc ibe
consume s’ a i udes and beha io owa ds OBA. SET sugges s ha people assess social exchanges in e ms
o cos s and bene i s. Acco dingly, consume s would accep OBA i hey weigh he bene i s a ising om
online ads g ea e han i s pe cei ed cos (Schumann e al., 2014). Also, he esea ch s udied OBA as an
online ma ke ing mechanism ha is based on he collec ion o online pe sonal da a which could esul in
gene a ing eelings o eac ance among online consume s, and how hese eelings can a ec consume s'
a i udes, in en ions and beha io s acco dingly.
In addi ion o building on heo ies, he s udied a iables in his esea ch namely pe sonaliza ion, consume
cha ac e is ics, b and amilia i y, b and us , and anspa ency app oach, we e no s udied be o e combined
oge he and also hei collec i e in luence on OBA a oidance beha io is no el o li e a u e. P e ious s udies
also used he media ing a iables o es hei e ec on consume s’ a i udes, in en ions o click on he ad,
and pu chase in en ions bu no on p o ec i e beha io agains OBA. Thus, his esea ch p o ides use ul
in o ma ion o u u e schola s o build on in hei coming esea ch.
5.4. P ac ical Implica ions
This esea ch p o ides some impo an insigh s o he online beha io al ad e ising p ac ices. These insigh s
a e discussed below o help ma ke ing p o essional in c ea ing success ul OBA campaigns and o e come
conce ns associa ed wi h OBA.
Fi s , online ad message pe sonaliza ion p o ed o be an impo an aspec o OBA, whe e pe sonaliza ion
and pe sonaliza ion in o ma i eness showed posi i e e ec s on consume s’ a i udes owa ds he ad and
in en ions o pu chase he ad e ised p oduc . In addi ion o lowe OBA a oidance beha io in he online
en i onmen . Howe e , he cu en esea ch also ound ha OBA wo ks be e o a amilia b and a he
114
Delgado-Balles e , E., Na a o, A., & Sicilia, M. (2012). Re i alising b ands h ough communica ion
messages: The ole o b and amilia i y.
Eu opean Jou nal o Ma ke ing, 46
(1/2), 31-51.
Dha , S., & Va shney, U. (2011). Challenges and business models o mobile loca ion-based se ices and
ad e ising.
Communica ions o he ACM, 54
(5), 121-128.
Diaman opoulos, A., & Winklho e , H. M. (2001). Index cons uc ion wi h o ma i e indica o s: An al e na i e
o scale de elopmen .
Jou nal o ma ke ing esea ch
,
38
(2), 269-277.
Diaman opoulos, A., Sa s ed , M., Fuchs, C., Wilczynski, P., & Kaise , S. (2012). Guidelines o choosing
be ween mul i-i em and single-i em scales o cons uc measu emen : a p edic i e alidi y
pe spec i e.
Jou nal o he Academy o Ma ke ing Science
,
40
, 434-449.
Digi al Ad e ising Alliance (2011). “You Ad Choices”, h p://www.you adchoices.com
Dilla d, J. P., & P au, M. (2002).
The pe suasion handbook: De elopmen s in heo y and p ac ice
. Sage
Publica ions.
Dine , T., & Ha , P. (2006). An ex ended p i acy calculus model o e-comme ce ansac ions.
In o ma ion
Sys ems Resea ch, 17
(1), 61-80.
Dodds, W. B., & Mon oe, K. B. (1985). The e ec o b and and p ice in o ma ion on subjec i e p oduc
e alua ions.
Ad ances in Consume Resea ch., 12
(1), 85- 90.
Dodds, W., Mon oe, K., & G ewal, D. (1991). E ec s o p ice, b and, and s o e in o ma ion on buye s' p oduc
e alua ions.
Jou nal o Ma ke ing Resea ch, 28
(3), 307-319.
Dolin, C., Weinshel, B., Shan, S., Hahn, C. M., Choi, E., Mazu ek, M. L., & U , B. (2018, Ap il). Unpacking
pe cep ions o da a-d i en in e ences unde lying online a ge ing and pe sonaliza ion. In
P oceedings o he
2018 CHI con e ence on human ac o s in compu ing sys ems
(pp. 1-12).
Ea p, J.B., & Baume , D. (2003). Inno a i e web use o lea n abou consume beha io and online
p i acy.
Communica ions o he ACM
, 46(4), 81-83.
Edwa ds, S. M., Li, H., & Lee, J. H. (2002). Fo ced exposu e and psychological eac ance: An eceden s and
consequences o he pe cei ed in usi eness o pop-up ads.
Jou nal o Ad e ising, 31
(3), 83-95.
Eme son, R. M. (1976). Social exchange heo y.
Annual Re iew o Sociology
,
2
(1), 335-362.
Engleha d , S., Reisman, D., Eubank, C., Zimme man, P., Maye , J., Na ayanan, A., & Fel en, E. W. (2015,
May). Cookies ha gi e you away: The su eillance implica ions o web acking. In
P oceedings o he 24 h
In e na ional Con e ence on Wo ld Wide Web
(pp. 289-299). In e na ional Wo ld Wide Web Con e ences
S ee ing Commi ee.
EU Da a P o ec ion Di ec i e (1995), “Di ec i e 95/46/EC o he Eu opean Pa liamen and o he Council o
24 Oc obe 1995 on he P o ec ion o Indi iduals wi h Rega d o he P ocessing o Pe sonal Da a and on he
F ee Mo emen o Such Da a”,
O icial Jou nal o he Eu opean Communi ies, 38
(281), 31–50.
EU Gene al Da a P o ec ion Regula ion (2016), “Regula ion (EU) 2016/679 o he Eu opean Pa liamen and
o he Council o 27 Ap il 2016 on he P o ec ion o Na u al Pe sons wi h Rega d o he P ocessing o Pe sonal
Da a and on he F ee Mo emen o such Da a, and Repealing Di ec i e 95/46/EC (Gene al Da a P o ec ion
Regula ion)”,
O icial Jou nal o he Eu opean Union, 119
(1).
115
Falk, M., & Mille , A. G. (1992). In a ed spec um o ca bon dioxide in aqueous solu ion.
Vib a ional
spec oscopy
,
4
(1), 105-108.
Faul, F., E d elde , E., Lang, A. G., & Buchne , A. (2007). G* Powe 3: A lexible s a is ical powe analysis
p og am o he social, beha io al, and biomedical sciences.
Beha io esea ch me hods
,
39
(2), 175-191.
Fede al T ade Commission. (2009). Sel - egula o y p inciples o online beha io al ad e ising.
Fede al T ade
Commission S a Repo , Washing on, DC.
Fede al T ade Commission. 2009. “FTC S a Re ises Online Beha io al Ad e ising P inciples.” Accessed
Decembe 29, 2023. h p://www. c.go /opa/2009/02/beha ad.sh m/.
Fi zsimons, G. J., & Lehmann, D. R. (2004). Reac ance o ecommenda ions: When unsolici ed ad ice yields
con a y esponses. Ma ke ing Science, 23(1), 82-94.
Ga dne , L., & Leshne , G. (2016). The ole o na a i e and o he - e e encing in a enua ing psychological
eac ance o diabe es sel -ca e messages.
Heal h communica ion, 31
(6), 738-751.
Ge en, D. (2000). E-comme ce: The ole o amilia i y and us .
Omega, 28
(6), 725- 737.
Gill, J. and Johnson, P. (2002).
Resea ch Me hods o Manage s
(3 d ed.). London: Sage Publica ions.
Gold a b, A. (2014). Wha is di e en abou online ad e ising?
Re iew o Indus ial O ganiza ion
,
44
(2), 115-
129.
Gold a b, A., & Tucke , C. (2011). Online display ad e ising: Ta ge ing and ob usi eness.
Ma ke ing
Science, 30
(3), 389-404.
Guild, B. (2013). “2013 Choice s eam Su ey: Consume Opinions on Online Ad e ising & Audience
Ta ge ing”. Accessed Janua y 18, 2024), h p://choices eam.com/2013 S aging/wp-
con en /uploads/2013/10/2013-Su ey.pd .
Ha, H. Y., & Janda, S. (2014). The e ec o cus omized in o ma ion on online pu chase in en ions.
In e ne
Resea ch, 24
(4), 496-519.
Hai J , J.F., Black, W.C., Babin, B.J., & Ande son, R.E. (2010). Ad anced diagnos ics o mul iple eg ession:
A supplemen o mul i a ia e da a analysis.
Ad anced Diagnos ics o Mul iple Reg ession: A Supplemen o
Mul i a ia e Da a Analysis
.
Hai J , J.F., Hul , G.T.M., Ringle, C.M., & Sa s ed , M. (2017).
A p ime on pa ial leas squa es s uc u al
equa ion modeling (PLS-SEM), 2nd ed
. Sage publica ions.
Hai , J. F., Hul , G. T. M., Ringle, C. M., & Sa s ed , M. (2022).
A P ime on Pa ial Leas Squa es S uc u al
Equa ion Modeling (PLS-SEM)
. Sp inge In e na ional Publishing.
Hai , J. F., Ringle, C. M., & Sa s ed , M. (2011). PLS-SEM: Indeed a sil e bulle .
Jou nal o Ma ke ing heo y
and P ac ice
,
19
(2), 139-152.
Hai , J. F., Rishe , J. J., Sa s ed , M., & Ringle, C. M. (2019). When o use and how o epo he esul s o
PLS-SEM.
Eu opean business e iew
,
31
(1), 2-24.
Ham, C. D., & Nelson, M. R. (2016). The ole o pe suasion knowledge, assessmen o bene i and ha m,
and hi d-pe son pe cep ion in coping wi h online beha io al ad e ising.
Compu e s in Human Beha io
,
62
,
689-702.
116
Ham, C. D. (2017). Explo ing how consume s cope wi h online beha io al ad e ising.
In e na ional Jou nal
o Ad e ising, 36
(4), 632–658.
Haug ed , C. P., & Kasme , J. A. (2008). A i ude change and pe suasion. In C. P. Haug ed , P. M. He ,
& F. P. Ka des (Eds.),
Handbook o consume psychology
(pp. 419-435. New Yo k, NY: Law ence E lbaum
Associa es.
Hee wegh, D., Vanho e, T., Ma hijs, K., & Loos eld , G. (2005). The e ec o pe sonaliza ion on esponse
a es and da a quali y in web su eys.
In e na ional Jou nal o Social Resea ch Me hodology
, 8(2), 85-99.
Helm, R., Mau one , O., & Dowling, M. (2010). Inno a ion as media o be ween en ep eneu ial o ien a ion
and spin-o en u e pe o mance.
In e na ional Jou nal o En ep eneu ship and Small Business, 11
(4), 472-
491.
Hensele , J., Ringle, C. M., & Sa s ed , M. (2015). A new c i e ion o assessing disc iminan alidi y in
a iance-based s uc u al equa ion modeling.
Jou nal o he academy o ma ke ing science, 43
, 115-135.
Hensele , J., Ringle, C. M., & Sinko ics, R. R. (2009). The use o pa ial leas squa es pa h modeling in
in e na ional ma ke ing. In
New challenges o in e na ional ma ke ing
(Vol. 20, pp. 277-319). Eme ald G oup
Publishing Limi ed.
Hoewe, J. (2017). Manipula ion check.
The in e na ional encyclopedia o communica ion esea ch me hods
,
1-5.
Hoye , W., & B own, S. (1990). E ec s o b and awa eness on choice o a common, epea - pu chase
p oduc .
Jou nal o Consume Resea ch, 17
(2), 141-148.
Huang, E. (2012). Online expe iences and i ual goods pu chase in en ion.
In e ne Resea ch
,
22(3)
, 252-
274.
Huang, E. (2012). Online expe iences and i ual goods pu chase in en ion.
In e ne Resea ch, 22(3)
, 252-
274.
Hühn, A. E., Khan, V. J., Ke elaa , P., an' Rie , J., Konig, R., Rozendaal, E., ... & Ma kopoulos, P. (2017).
Does loca ion cong uence ma e ? A ield s udy on he e ec s o loca ion-based ad e ising on pe cei ed AD
in usi eness, ele ance & alue.
Compu e s in Human Beha io
.
Hulland, J. (1999). Use o pa ial leas squa es (PLS) in s a egic managemen esea ch: A e iew o ou
ecen s udies.
S a egic managemen jou nal
,
20
(2), 195-204.
In e ac i e Ad e ising Bu eau (2022). “In e ne Ad e ising Re enue Repo : Full Yea 2022”. Accessed Ap il
12, 2024. h ps://www.iab.com/insigh s/in e ne -ad e ising- e enue- epo - ull-yea -2022/
In e na ional Associa ion o P i acy P o essionals (2022). “80% o use s' b owse his o y is cap u ed by online
acke s, epo inds” Accessed Ap il 12, 2024. h ps://iapp.o g/news/a/80-pe cen -o -use s-b owse -
his o y-is-cap u ed-by-online- acke s- epo /
Iqbal, U., Engleha d , S., & Sha iq, Z. (2021, May). Finge p in ing he inge p in e s: Lea ning o de ec
b owse inge p in ing beha io s. In
2021 IEEE Symposium on Secu i y and P i acy (SP)
(pp. 1143-1161).
IEEE.
117
Jai, T. M. C., Bu ns, L. D., & King, N. J. (2013). The e ec o beha io al acking p ac ices on consume s’
shopping e alua ions and epu chase in en ion owa d us ed online e aile s.
Compu e s in Human
Beha io , 29
(3), 901-909.
Kalyana aman, S., & Sunda , S. S. (2006). The psychological appeal o pe sonalized con en in web po als:
Does cus omiza ion a ec a i udes and beha io ?.
Jou nal o Communica ion, 56
(1), 110-132.
Keh , F., Kowa sch, T., Wen zel, D., & Fleisch, E. (2015). Bliss ully igno an : he e ec s o gene al p i acy
conce ns, gene al ins i u ional us , and a ec in he p i acy calculus.
In o ma ion Sys ems Jou nal, 25
(6),
607-635.
Kim, H., & Huh, J. (2017). Pe cei ed ele ance and p i acy conce n ega ding online beha io al ad e ising
(OBA) and hei ole in consume esponses.
Jou nal o Cu en Issues & Resea ch in Ad e ising, 38
(1), 92-
105.
Kim, K., & Kim, J. (2011). Thi d-pa y p i acy ce i ica ion as an online ad e ising s a egy: An in es iga ion
o he ac o s a ec ing he ela ionship be ween hi d-pa y ce i ica ion and ini ial us .
Jou nal o In e ac i e
Ma ke ing, 25
(3), 145-158.
Kissme ics Blog (n.d.).
Behind he scenes o beha io al ad e ising
. Re ie ed om
h ps://blog.kissme ics.com/beha io al-ad e ising/ , Accessed Decembe 29, 2023
Komiak, S. X., & Benbasa , I. (2006). The e ec s o pe sonaliza ion and amilia i y on us and adop ion o
ecommenda ion agen s.
MIS Qua e ly
,
30
(4), 941- 960.
Ko ha i, C. R. (2004).
Resea ch me hodology: Me hods and echniques
. New Age In e na ional.
Kuma , V., & Gup a, S. (2016). Concep ualizing he e olu ion and u u e o ad e ising
. Jou nal o
ad e ising, 45
(3), 302-317.
Kuma , V., & Pansa i, A. (2016). Compe i i e ad an age h ough engagemen .
Jou nal o ma ke ing
esea ch, 53
(4), 497-514.
Lamb ech , Anja and Ca he ine E. Tucke (2013), “When Does Re a ge ingWo k? In o ma ion Speci ici y in
Online Ad e ising”,
Jou nal o Ma ke ing Resea ch, 50
(5), 561–76.
Lau e , R. S., & Wol e, M. (1977). P i acy as a concep and a social issue: A mul idimensional de elopmen al
heo y.
Jou nal o Social Issues
,
33
(3), 22- 42.
Lee, S., Lee, Y., Lee, J. I., & Pa k, J. (2015). Pe sonalized e-se ices: consume p i acy conce n and
in o ma ion sha ing.
Social Beha io and Pe sonali y: an in e na ional jou nal, 43
(5), 729-740.
Leon, P. G., U , B., Shay, R., Wang, Y., Balebako, R., & C ano , L. F. (2012). “Why Johnny Can' Op Ou : A
Usabili y E alua ion o Tools o Limi Online Beha io al Ad e ising,”
In P oceedings o he SIGCHI Con e ence
on Human Fac o s in Compu ing Sys ems
, 589-98.
Leon, P. G., U , B., Wang, Y., Sleepe , M., Balebako, R., Shay, R., ... & C ano , L. F. (2013, July). Wha
ma e s o use s? Fac o s ha a ec use s' willingness o sha e in o ma ion wi h online ad e ise s.
In
P oceedings o he nin h symposium on usable p i acy and secu i y
(pp. 1-12).
Le ne , A., Simpson, A.K., Kohno, T., & Roesne , F. (2016, Augus ). In e ne Jones and he Raide s o he
Los T acke s: An A chaeological S udy o Web T acking om 1996 o 2016. In
USENIX Secu i y Symposium
.
118
Li, C. (2016). When does web-based pe sonaliza ion eally wo k? The dis inc ion be ween ac ual
pe sonaliza ion and pe cei ed pe sonaliza ion.
Compu e s in human beha io , 54
, 25-33.
Li, C., & Kalyana aman, S. (2013). “I, me, mine” o “Us, we, ou s?”: The in luence o cul u al psychology
on web-based cus omiza ion.
Media Psychology, 16
(3), 272-294.
Li, C., & Liu, J. (2017). A name alone is no enough: A eexamina ion o web-based pe sonaliza ion
e ec .
Compu e s in Human Beha io
,
72
, 132-139.
Li, H., Edwa ds, S. M., & Lee, J. H. (2002). Measu ing he in usi eness o ad e isemen s: Scale
de elopmen and alida ion.
Jou nal o Ad e ising
,
31
(2), 37-47.
Li, W., & Huang, Z. (2016). The esea ch o in luence ac o s o online beha io al ad e ising
a oidance.
Ame ican Jou nal o Indus ial and Business Managemen , 6
(09), 947.
Liu, Z., Min, Q., Zhai, Q., & Smy h, R. (2016). Sel -disclosu e in Chinese mic o-blogging: A social exchange
heo y pe spec i e.
In o ma ion & Managemen , 53
(1), 53-63.
Luo, X. (2002). T us p oduc ion and p i acy conce ns on he In e ne : A amewo k based on ela ionship
ma ke ing and social exchange heo y.
Indus ial ma ke ing managemen , 31
(2), 111-118.
Lynch J ., J. G. (1982). On he ex e nal alidi y o expe imen s in consume esea ch.
Jou nal o Consume
Resea ch
,
9
(3), 225-239.
MacKenzie, S. B., & Lu z, R. J. (1983). Tes ing compe ing heo ies o ad e ising e ec i eness ia s uc u al
equa ion models. In
P oceedings o Ame ican Ma ke ing Associa ion Resea ch Me hods and Causal Modeling
Con e ence, o hcoming.
MacKinnon, D. P., Fai child, A. J., & F i z, M. S. (2007).
Media ion analysis. Annu. Re . Psychol., 58
, 593-
614.
Ma í-Pa eño, J., Be mejo-Be os, J., & Aldás-Manzano, J. (2017). P oduc placemen in ideo games: he
e ec o b and amilia i y and epe i ion on consume s' memo y.
Jou nal o In e ac i e Ma ke ing
,
38
, 55-
63.
Ma o e, E. M., & Khai ani, A. Z. (2020). The Pa e n o Skewness and Ku osis Using Mean Sco e and Logi
in Measu ing Ad e si y Quo ien (AQ) Fo No mali y Tes ing.
In e na ional Jou nal o Fu u e Gene a ion
Communica ion and Ne wo king, 13
(1), 688–702.
Maye , J. R., & Mi chell, J. C. (2012, May). Thi d-pa y web acking: Policy and echnology. In
2012 IEEE
symposium on secu i y and p i acy
(pp. 413-427). IEEE.
McDonald, A., & C ano , L. F. (2010). Belie s and Beha io s: In e ne Use s’ Unde s anding o Beha io al
Ad e ising. SSRN Schola ly Pape ID 1989092.
Social Science Resea ch Ne wo k, Roches e , NY.
h p://pape s. ss n. com/abs ac , 1989092.
McS ay, A. (2011),
The Mood o In o ma ion: A C i ique o Online Beha io al Ad e ising,
New Yo k:
Con inuum.
Mikians J., Gya ma i L., E amilli V., and Laou a is N. (2012). De ec ing p ice and sea ch disc imina ion on
he In e ne . In
P oc.
ACM Ho Ne s
.
Milne, G. R., Bahl, S., & Rohm, A. (2008). Towa d a amewo k o assessing co e ma ke ing
p ac ices.
Jou nal o Public Policy & Ma ke ing, 27
(1), 57-62.
119
Milne, G. R., Culnan, M. J., & G eene, H. (2006). A longi udinal assessmen o online p i acy no ice
eadabili y.
Jou nal o Public Policy & Ma ke ing, 25
(2), 238-249.
Mi chell, A. A., & Olson, J. C. (1981). A e p oduc a ibu e belie s he only media o o ad e ising e ec s on
b and a i ude?.
Jou nal o ma ke ing esea ch, 18
(3), 318-332.
Miyazaki, A. D. (2008). Online p i acy and he disclosu e o cookie use: E ec s on consume us and
an icipa ed pa onage.
Jou nal o Public Policy & Ma ke ing, 27
(1), 19-33.
Mon gome y, A. L., & Smi h, M. D. (2009). P ospec s o pe sonaliza ion on he In e ne .
Jou nal o In e ac i e
Ma ke ing
,
23
(2), 130-137.
Moon, Y. (2002). Pe sonaliza ion and pe sonali y: Some e ec s o cus omizing message s yle based on
consume pe sonali y.
Jou nal o Consume Psychology
, 12(4), 313-326.
Moo man, C., Ge ald, Z., & Rohi , D. (1992). “Rela ionships Be ween P o ide s and Use s o Ma ke
Resea ch: The Dynamics o T us ”.
Jou nal o Ma ke ing, 29
(3), 314–28.
Mu hi, B. P. S., & Sa ka , S. (2003). The ole o he managemen sciences in esea ch on
pe sonaliza ion.
Managemen Science, 49
(10), 1344-1362.
Ni zl, C., Roldan, J. L., & Cepeda, G. (2016). Media ion analysis in pa ial leas squa es pa h modeling:
Helping esea che s discuss mo e sophis ica ed models.
Indus ial managemen & da a sys ems, 116
(9),
1849-1864.
Noa , S. M., Ha ing on, N. G., & Ald ich, R. S. (2009). The ole o message ailo ing in he de elopmen o
pe suasi e heal h communica ion messages. In C. S. Beck (Ed.),
Communica ion yea book 33
(pp. 73-133).
New Yo k, NY: Rou ledge.
No be g, P. A., Ho ne, D. R., & Ho ne, D. A. (2007). The p i acy pa adox: Pe sonal in o ma ion disclosu e
in en ions e sus beha io s.
Jou nal o Consume A ai s
,
41
(1), 100-126.
Nunnally, Jum C. (1978). Psychome ic Theo y, 2nd ed. New Yo k: McG aw-Hill
Nu in, J. M. (1985). Na cissism beyond Ges al and awa eness: The name le e e ec .
Eu opean Jou nal
o Social Psychology
,
15(3)
, 353-361.
Okazaki, S., Li, H., & Hi ose, M. (2009). Consume p i acy conce ns and p e e ence o deg ee o egula o y
con ol: A s udy o mobile ad e ising in Japan.
Jou nal o Ad e ising, 38
(4), 63-77.
Ozcelik, A. B., & Va nali, K. (2019). E ec i eness o online beha io al a ge ing: A psychological
pe spec i e.
Elec onic Comme ce Resea ch and Applica ions, 33
, 100819.
Phelan, C., Lampe, C., & Resnick, P. (2016). I 's c eepy, bu i doesn' bo he me. In
P oceedings o he
2016 CHI Con e ence on Human Fac o s in Compu ing Sys ems
(pp. 5240-5251). ACM.
Phelps, J. E., & Hoy, M. G. (1996). The Aad‐Ab‐PI ela ionship in child en: The impac o b and amilia i y
and measu emen iming.
Psychology & Ma ke ing
,
13
(1), 77-105.
Plane, A. C., Redmiles, E. M., Mazu ek, M. L., & Tschan z, M. C. (2017). Explo ing use pe cep ions o
disc imina ion in online a ge ed ad e ising. In
26 h USENIX Secu i y Symposium
(USENIX Secu i y 17)
(pp.
935-951).
120
Podsako , P. M., MacKenzie, S. B., Lee, J. Y., & Podsako , N. P. (2003). Common me hod biases in
beha io al esea ch: a c i ical e iew o he li e a u e and ecommended emedies.
Jou nal o applied
psychology
,
88
(5), 879-903.
Po e , S. R., & Whi comb, M. E. (2003). The impac o con ac ype on web su ey esponse a es.
Public
Opinion Qua e ly
, 67(4), 579-588.
Pu wan o, A., & Suda gini, Y. (2021). Explo ing Fac o s A ec ing he Pu chase In en ion o Halal Food
P oduc s: An Empi ical S udy on S uden Consume s.
In e na ional Jou nal o Social and Managemen
S udies
,
2
(4), 13-21.
Rai hel, S., Sa s ed , M., Scha , S., & Schwaige , M. (2012). On he alue ele ance o cus ome sa is ac ion.
Mul iple d i e s and mul iple ma ke s.
Jou nal o he academy o ma ke ing science
,
40
, 509-525.
Reina z, W., Haenlein, M., & Hensele , J. (2009). An empi ical compa ison o he e icacy o co a iance-
based and a iance-based SEM.
In e na ional Jou nal o esea ch in Ma ke ing
,
26
(4), 332-344.
Richa dson, H. A., Simme ing, M. J., & S u man, M. C. (2009). A ale o h ee pe spec i es: Examining pos
hoc s a is ical echniques o de ec ion and co ec ion o common me hod a iance.
O ganiza ional Resea ch
Me hods
,
12
(4), 762-800.
Richins, M. L., & Bloch, P. H. (1986). A e he new wea s o : The empo al con ex o p oduc in ol emen .
Jou nal o Consume Resea ch
,
13(2)
, 280-285.
Rigdon, E. E. (2014). Re hinking pa ial leas squa es pa h modeling: b eaking chains and o ging
ahead.
Long ange planning
,
47
(3), 161-167.
Ringle, C. M., Sa s ed , M., & S aub, D. W. (2012). A c i ical look a he use o PLS-SEM in" MIS
Qua e ly".
MIS qua e ly
,
36
(1), iii-xi .
Ringle, C., Da Sil a, D., & Bido, D. (2015). S uc u al equa ion modeling wi h he Sma PLS.
Bido, D., da
Sil a, D., & Ringle, C. (2014). S uc u al Equa ion Modeling wi h he Sma pls. B azilian Jou nal O
Ma ke ing
,
13
(2), 56-73.
Robson, C. (2002).
Real wo ld esea ch: A esou ce o social scien is s and p ac i ione - esea che s
.
(No
Ti le)
(2nd ed.). Ox o d: Blackwell.
Roesne , F., Kohno, T., & We he all, D. (2012). De ec ing and De ending Agains {Thi d-Pa y} T acking on
he Web. In
9 h USENIX Symposium on Ne wo ked Sys ems Design and Implemen a ion (NSDI 12)
(pp. 155-
168).
Rognehaugh, R. 1999.
The Heal h In o ma ion Technology Dic iona y
, Gai he sbu g, MD: Aspen.
Rony, M. N. (2018). Online Beha io al Ad e ising (OBA): The In luence o Recip oci y, Pe sonaliza ion, And
Ad Con en Type on Consume s’ a i ude And In en ion.
Rousseau, D.M., Si kin, S.B., Bu , R.S., & Came e , C. (1998). No so di e en a e all: A c oss-discipline
iew o us .
Academy o managemen e iew
, 23(3), 393-404.
Sama , S., Acquis i, A., & Babcock, L. (2017). Raise he cu ains: The e ec o awa eness abou a ge ing on
consume a i udes and pu chase in en ions.
In Thi een h Symposium on Usable P i acy and Secu i y
(SOUPS 2017)
(pp. 299-319).
121
Sa s ed , M., & Mooi, E. (2014).
A concise guide o ma ke esea ch:The P ocess, Da a, and Me hods using
IBM SPSS S a is ics (2nd ed)
. Sp inge
.
Sa s ed , M., Hai J , J. F., Ni zl, C., Ringle, C. M., & Howa d, M. C. (2020). Beyond a andem analysis o
SEM and PROCESS: Use o PLS-SEM o media ion analyses!.
In e na ional Jou nal o Ma ke
Resea ch, 62
(3), 288-299.
Sa s ed , M., Ringle, C. M., & Hai , J. F. (2021). Pa ial leas squa es s uc u al equa ion modeling.
In
Handbook o ma ke esea ch
(pp. 587-632). Cham: Sp inge In e na ional Publishing.
Sa le , H., Völckne , F., Riedige , C., & Ringle, C. M. (2010). The impac o b and ex ension success d i e s
on b and ex ension p ice p emiums.
In e na ional Jou nal o Resea ch in ma ke ing, 27
(4), 319-328.
Saunde s, M., Lewis, P., & Tho nhill, A. (2009).
Resea ch me hods o business s uden s
. Pea son educa ion.
Schumann, J. H., Von Wangenheim, F., & G oene, N. (2014). Ta ge ed online ad e ising: Using ecip oci y
appeals o inc ease accep ance among use s o ee web se ices.
Jou nal o Ma ke ing, 78
(1), 59-75.
Seka an, U., & Bougie, R. (2016).
Resea ch me hods o business: A skill building app oach
. John wiley &
sons.
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002).
Expe imen al and quasi- expe imen al designs o
gene alized causal in e ence
(2nd ed.). Bos on, MA: Hough on Mi lin Company.
Shang, R. A., Chen, Y. C., & Liao, H. J. (2006). The alue o pa icipa ion in i ual consume communi ies
on b and loyal y.
In e ne Resea ch, 16(4)
, 398-418.
Sheehan, K. B., & Hoy, M. G. (1999). Flaming, complaining, abs aining: How online use s espond o p i acy
conce ns.
Jou nal o ad e ising, 28
(3), 37-51.
Shoenbe ge , H. (2014).
Risky business: Using heu is ic-sys ema ic p ocessing heo y o unde s and
consume da a p i acy conce ns in an online beha io al ad e ising con ex
(Unpublished doc o al
disse a ion). Uni e si y o Missou i, Columbia, MO.
Shoenbe ge , H., & Tho son, E. (2014). p edic ion o pe cei ed online shopping bene i s and isks om us
and knowledge o a ge ing.
Ame ican Academy o Ad e ising. Con e ence. P oceedings (Online),
155-162.
Smi , E. G., Van Noo , G., & Voo eld, H. A. (2014). Unde s anding online beha iou al ad e ising: Use
knowledge, p i acy conce ns and online coping beha iou in Eu ope.
Compu e s in Human Beha io , 32
,
15-22.
Smi h, A. D. (2006). Explo ing se ice ma ke ing aspec s o e-pe sonaliza ion and i s impac on online
consume beha io .
Se ices Ma ke ing Qua e ly, 27
(2), 89-102.
Smi h, H., Dine , T., & Xu, H. (2011). In o ma ion p i acy esea ch: An in e disciplina y e iew.
MIS Qua e ly,
35
(4), 989-1015.
S a is a [ONLINE, 2023]. (h ps://www.s a is a.com/s a is ics/1410056/egyp -digi al-
popula ion/#:~: ex =As%20o %20Janua y%202023%2C%20Egyp , hose%20we e%20using%20social%20me
dia.). Accessed Janua y, 20 2024.
Sunda , S. S., & Ma a he, S. S. (2010). Pe sonaliza ion e sus cus omiza ion: The impo ance o agency,
p i acy, and powe usage.
Human communica ion esea ch, 36
(3), 298-322.
122
Su an o, J., Palme, E., Tan, C., & Phang, C. (2013). Add essing he pe sonaliza ion- p i acy pa adox: An
empi ical assessmen om a ield expe imen on sma phone use s.
MIS Qua e ly, 37
(4), 1141-1164.
Tam, K. Y., & Ho, S. Y. (2006). Unde s anding he impac o web pe sonaliza ion on use in o ma ion
p ocessing and decision ou comes.
MIS qua e ly
, 865-890.
Te’eni-Ha a i, T., Lehman-Wilzig, S. N., & Lampe , S. I. (2009). The impo ance o p oduc in ol emen o
p edic ing ad e ising e ec i eness among young people.
In e na ional Jou nal o Ad e ising, 28(2)
, 203-
229.
Tucke , C. E. (2014). Social ne wo ks, pe sonalized ad e ising, and p i acy con ols.
Jou nal o ma ke ing
esea ch, 51
(5), 546-562.
Tucke , Ca he ine E. (2012), “The Economics o Ad e ising and P i acy,”
In e na ional Jou nal o Indus ial
O ganiza ion, 30
(3),326–9.
Tu ow, J., King, J., Hoo nagle, C. J., Bleakley, A., & Hennessy, M. (2009). Ame icans ejec ailo ed
ad e ising and h ee ac i i ies ha enable i .
A ailable a SSRN 1478214
.
U , B., Leon, P. G., C ano , L. F., Shay, R., & Wang, Y. (2012, July). Sma , use ul, sca y, c eepy: pe cep ions
o online beha io al ad e ising. In
p oceedings o he eigh h symposium on usable p i acy and secu i y
(pp.
1-15).
U ban, G.L., Amyx, C., & Lo enzon, A. (2009). Online us : s a e o he a , new on ie s, and esea ch
po en ial.
Jou nal o In e ac i e Ma ke ing
, 23(2), 179-190.
Van Doo n, J., & Hoeks a, J. C. (2013). Cus omiza ion o online ad e ising: The ole o
in usi eness.
Ma ke ing Le e s, 24
, 339-351.
Van Noo , G., Smi , E. G., & Voo eld, H. A. (2013). The online beha iou al ad e ising icon: wo use
s udies.
Ad ances in Ad e ising Resea ch (Vol. IV) The Changing Roles o Ad e ising
, 365-378.
Ve hoe , P. C., Reina z, W. J., & K a , M. (2010). Cus ome engagemen as a new pe spec i e in cus ome
managemen .
Jou nal o se ice esea ch, 13
(3), 247-252.
Wes in, A. F. (1967).
P i acy and eedom
. New Yo k, NY: A heneum.
Whi e, T., Zahay, D., Tho bjø nsen, H., & Sha i , S. (2008). Ge ing oo pe sonal: Reac ance o highly
pe sonalized email solici a ions. Ma ke ing Le e s, 19(1), 39-50.
Xu, H., Luo, X. R., Ca oll, J. M., & Rosson, M. B. (2011). The pe sonaliza ion p i acy pa adox: An explo a o y
s udy o decision-making p ocess o loca ion-awa e ma ke ing.
Decision suppo sys ems, 51
(1), 42-52.
Yan, Z., Wang, T., Chen, Y., & Zhang, H. (2016). Knowledge sha ing in online heal h communi ies: A social
exchange heo y pe spec i e.
In o ma ion & managemen , 53
(5), 643-653.
Yao, Y., Lo Re, D., & Wang, Y. (2017, Feb ua y). Folk models o online beha io al ad e ising.
In P oceedings
o he 2017 ACM Con e ence on Compu e Suppo ed Coope a i e Wo k and Social Compu ing
(pp. 1957-
1969).
Yap, J. E., Be e land, M., & Bo e, L. (2009). A concep ual amewo k o he causes and consequences o
he p i acy pa adox. In
The Aus alian and New Zealand Ma ke ing Academy (ANZMAC) Con e ence 2009
P oceedings
(pp. 1-9). Melbou ne, VIC: Monash Uni e si y.
123
Ying, L., Ko neliussen, T., & G ønhaug, K. (2009). The e ec o ad alue, ad placemen and ad execu ion on
he pe cei ed in usi eness o web ad e isemen s.
In e na ional Jou nal o Ad e ising, 28
(4), 623-638.
Yoon, C., & Rolland, E. (2012). Knowledge-sha ing in i ual communi ies: amilia i y, anonymi y and sel -
de e mina ion heo y.
Beha iou & In o ma ion Technology
,
31
(11), 1133-1143.
Yu, J., & Cude, B. (2009). “Hello, M s. Sa ah Jones! We ecommend his p oduc !” Consume s’ pe cep ions
abou pe sonalized ad e ising: Compa isons ac oss ad e isemen s deli e ed ia h ee di e en ypes o
media.
In e na ional Jou nal o Consume S udies
,
33(4)
, 503-514.
Yu, S., Hudde s, L., & Caube ghe, V. (2017). Ta ge ing he luxu y consume : A ice o i ue? A c oss-cul u al
compa ison o he e ec i eness o beha io ally a ge ed ads.
Jou nal o Fashion Ma ke ing and
Managemen , 21
(2), 187-205.
Zaichkowsky, J. L. (1985). Measu ing he in ol emen cons uc .
Jou nal o Consume Resea ch, 12(3)
, 341-
352.
Zhao, X., Yang, J., Xie, T., & Wang, Z. (2017). Examining ad e ising in usi eness on Ins ag am: Hedonic
and u ili a ian a ibu es o b and and sponso ed con en .
Ame ican Academy o Ad e ising. Con e ence.
P oceedings (Online),
243-255.
Zhonghao Yu, Sam Macbe h, Kona k Modi, and Josep M. Pujol. (2016). T acking he T acke s.
In P oceedings o he 25 h In e na ional Con e ence on Wo ld Wide Web (pp. 121-132).
Zuide een Bo gesius, F. (2015). Imp o ing p i acy p o ec ion in he a ea o beha iou al a ge ing.
A ailable
a SSRN 2654213.
130
APPENDIX II – QUESTIONNAIRE | ENGLISH
131
Consen o Pa icipa e in an Online Expe imen
This esea ch is being conduc ed a The School o Economics and Managemen , Uni e si y o Minho. The
pu pose o his esea ch is o unde s and consume s’ iews owa ds online beha io al ad e ising. I you
ag ee o ake pa in his esea ch, you will be edi ec ed o a page ha con ains a scena io explaining he
online ad e isemen and hen asked o answe some ques ions abou you opinion owa ds he ad. You will
be also asked o answe some gene al usage and demog aphic ques ions. Please no e ha he e a e no igh
o w ong answe s. The su ey will ake abou 15 minu es o comple e. The in o ma ion collec ed ia he
ques ionnai e is ANONYMOUS and used o academic pu poses ONLY.
Thank you e y much o you collabo a ion!
I unde s and and ag ee o pa icipa e.
132
[Scena io]:
Imagine you a e in e es ed in weigh -loss p og ams and so, Las nigh you sea ched online
o weigh -loss p og ams and exe cises. Nex mo ning you opened a news websi e o ead he mo ning
news and you saw he ollowing banne ad.
Based on you unde s anding o he p e ious scena io, please indica e you opinion by
selec ing he app op ia e op ion o each ques ion:
1. This ad is di ec ed o me pe sonally [Pe sonaliza ion]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
2. Rega ding he b and, I am: [B and Familia i y]
1- Un amilia
2
3
4
5
6
7- Familia
3. In gene al, my eeling owa d ad e ising is: [A i udes owa ds OBA]
1- Bad
2
3
4
5
6
7- Good
1- Unpleasan
2
3
4
5
6
7- Pleasan
1- Un a o able
2
3
4
5
6
7- Fa o able
133
The nex wo ques ions a e based on you unde s anding o he ollowing in o ma ion:
Online beha io al ad e ising (OBA) can be de ined as:
The p ac ice o acking indi iduals' online ac i i ies
in o de o deli e ad e ising ailo ed o he consume s' in e es s.
4. Do you emembe seeing his ype o ad in he las six mon hs?
a. Yes.
b. No.
5. App oxima ely, how many Online Beha io al Ad e ising (OBA) do you see e e y day?
a. None.
b. One.
c. Two.
d. Th ee.
e. Fou .
. Fi e.
g. Mo e han i e.
134
Based on you unde s anding o he p e ious scena io and OBA de ini ion, please indica e
you opinion by selec ing he app op ia e op ion o each ques ion.
6. Thinking abou he ad, I eel: [A i ude owa ds OBA]
1- Bad
2
3
4
5
6
7- Good
1- Unpleasan
2
3
4
5
6
7- Pleasan
1- Un a o able
2
3
4
5
6
7- Fa o able
7. Thinking abou he b and, I eel: [A i ude owa ds he B and]
1- Bad
2
3
4
5
6
7- Good
1- Unpleasan
2
3
4
5
6
7- Pleasan
1- Dislike e y
much
2
3
4
5
6
7- Like e y much
1- Poo Quali y
2
3
4
5
6
7- High Quali y
8. My in en ion o click he ad is: [In en ion o Click he Ad]
1- Unlikely
2
3
4
5
6
7- Likely
1- Imp obable
2
3
4
5
6
7- P obable
1- Impossible
2
3
4
5
6
7- Possible
9. My in en ion o pu chase he p oduc is: [In en ion o Pu chase he P oduc ]
1- Unlikely
2
3
4
5
6
7- Likely
1- Imp obable
2
3
4
5
6
7- P obable
1- Impossible
2
3
4
5
6
7- Possible
135
10. I us he ad e ised b and: [B and T us ]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
11. I can ely on he ad e ised b and: [B and T us ]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
12. The ad e ised b and is an hones b and: [B and T us ]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
13. The ad e ised b and is sa e: [B and T us ]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
14. I hink he ad was dis ac ing: [Pe cei ed In usi eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
15. I hink he ad was dis u bing: [Pe cei ed In usi eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
16. I hink he ad was o ced: [Pe cei ed In usi eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
17. I hink he ad was in e e ing: [Pe cei ed In usi eness]
136
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
18. I hink he ad was in usi e: [Pe cei ed In usi eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
19. I hink he ad was in asi e: [Pe cei ed in usi eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
20. I hink he ad was ob usi e: [Pe cei ed In usi eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
Fo he ollowing ques ions, please indica e you opinion by selec ing he app op ia e op ion
o each ques ion:
21. Cus omized online ad gi es me quick and easy access o la ge olumes o in o ma ion.
[Pe sonaliza ion - In o ma i eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
22. In o ma ion ob ained om he cus omized online ad is use ul o me. [Pe sonaliza ion - In o ma i eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
137
23. I lea ned a lo om cus omized online ads. [Pe sonaliza ion - In o ma i eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
24. I hink in o ma ion ob ained om cus omized online ads is help ul. [Pe sonaliza ion - In o ma i eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
25. Cus omized online ad makes acqui ing in o ma ion inexpensi e o me. [Pe sonaliza ion -
In o ma i eness]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
26. I eel uncom o able when in o ma ion is sha ed wi hou my pe mission [P i acy Conce n]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
27. I am conce ned abou misuse o pe sonal in o ma ion [P i acy Conce n]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
28. I bo he s me o ecei e oo much ad e ising ma e ial o no in e es [P i acy conce n]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
29. I eel ea ha in o ma ion may no be sa e while s o ed [P i acy Conce n]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
138
30. I belie e ha pe sonal in o ma ion is o en misused [P i acy Conce n]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
31. I hink companies sha e in o ma ion wi hou pe mission [P i acy Conce n]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
32. The message h ea ened my eedom o choose [Pe cei ed h ea o Choice, Reac ance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
33. The message ied o make a decision o me [Pe cei ed h ea o Choice, Reac ance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
34. The message ied o manipula e me [Pe cei ed h ea o Choice, Reac ance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
35. The message ied o pe suade me [Pe cei ed h ea o Choice, Reac ance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
36. I delibe a ely igno e online beha io al ads when I’m su ing on he In e ne . [P o ec i e Beha io , OBA
A oidance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
139
37. I delibe a ely igno e online beha io al ads when I’m b owsing on he shopping si es. [P o ec i e
Beha io , OBA A oidance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
38. I delibe a ely igno e online beha io al ads when I open my mailbox. [P o ec i e Beha io , OBA
A oidance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
39. Online beha io al ads a e annoying. [P o ec i e Beha io , OBA A oidance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
40. Online beha io al ads make me eel dis u bed. [P o ec i e Beha io , OBA A oidance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
41. Online beha io al ads a e unappealing. [P o ec i e Beha io , OBA A oidance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
42. I online beha io al ads pop ou , I will close hem. [P o ec i e Beha io , OBA A oidance]
1- S ongly
Disag ee
2
3
4
5
6
7- S ongly
Ag ee
