Content uploaded by Patrick Meyer
Author content
All content in this area was uploaded by Patrick Meyer on Feb 13, 2020
Content may be subject to copyright.
Patrick Meyer is a Research Asso-
ciate at the Chair of Information
Systems – Innovation and Value
Creation at Friedrich-Alexander-
Universität Erlangen-Nürnberg,
Lange Gasse 20, 90403 Nürnberg,
Germany, E-Mail: pat.meyer@
fau.de
* Corresponding Author.
Julia M. Jonas is a Senior Research
Fellow at the Chair of Information
Systems – Innovation and Value
Creation at Friedrich-Alexander-
Universität Erlangen-Nürnberg,
Lange Gasse 20, 90403 Nürnberg,
Germany, E-Mail: julia.jonas@
fau.de
Angela Roth is a Professor at the
Chair of Information Systems –
Innovation and Value Creation at
Friedrich-Alexander-Universität
Erlangen-Nürnberg, Lange Gasse 20,
90403 Nürnberg, Germany,
E-Mail: angela.roth@fau.de
Frontline Employees’ Acceptance of and Resistance to Service
Robots in Stationary Retail – An Exploratory Interview Study
By Patrick Meyer*, Julia M. Jonas, and Angela Roth
Due to rising online competition, increasing cost
pressure and cross-channel customer journeys, sta-
tionary retail has tried to develop innovative value
propositions and co-create value with customers
through new technologies, which are expected to
profoundly change the stationary retail’s service
systems. Among other technologies, service robots
are said to have the potential to revitalise interac-
tive value creation in stationary retail. However, the
integration of such technologies poses new chal-
lenges. Prior research has looked at customers’ ac-
ceptance of service robots in stationary retail set-
tings, but few studies have explored their counter-
parts – the frontline employees’ (FLEs) perspective.
Yet, FLEs’ acceptance of service robots is crucial to
implement service robots for retail innovation. To
explore FLEs’ acceptance of and resistance to ser-
vice robots, a qualitative exploratory interview
study is conducted. It identifies decisive aspects,
amongstotherslossofstatusorroleincongruency.
The findings extend prior studies on technology ac-
ceptance and resistance and reveal i.a. that FLEs
perceive service robots as both a threat and poten-
tial support. Moreover, they feel hardly involved in
the co-creation of use cases for a service robot, al-
though they are willing to contribute.
1. Introduction
Emergingtechnologies foster the transformation ofbusi-
ness models and enable innovation across all sectors. One
promisingtechnology is robotics,which,alongside other
technological advances such as bigdata,cloud computing
and artificial intelligence,is expected toresultin various
profound innovations in service environments(Ivanov
and Webster 2019; Matzner etal. 2018; Wirtz etal. 2018).In
particular,physical service robotsdeveloped for interac-
tion withhumansaresaidtohavegreatpotential for inno-
vationinstationaryretail (Doeringetal. 2015; Grewal et
al. 2017; Iwamura etal. 2011).Stationaryretail has to inno-
vateinterms ofvalue propositions due torisingonline
competition,increasingcostpressure and cross-channel
customer journeys. Todoso,retailers trytoco-createval-
ue withcustomers throughtechnology, such as service ro-
bots. Thus,agrowingnumber ofretailers (e.g.,Nestl´e,
Lowe’s and Marriott) havepiloted service robotsattheir
stores todetermine how customers reacttothem.
Inservice research,the use ofservice robots has been con-
troversiallydiscussed. Service literature is growingin
studies linkingservice robotstoservice environments:Al-
mosttwo decades ago,Schraft and Schmierer (2000) em-
phasised thatservice robotscancreatenewfields ofappli-
cation toaddvalue toservice industries. Afew years later,
Severinson-Eklundh etal. (2003) argued that‘addressing
onlythe primaryuser in service roboticsisunsatisfactory,
and thatthe focus should beonthe setting, activities and
SMR · Journal of Service Management Research · Volume 4 · 1/2020 · p. 21– 34 21
https://doi.org/10.15358/2511-8676-2020-1-21
Generiert durch IP '188.192.241.220', am 13.02.2020, 09:57:10.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.
social interactions ofthe groupofpeople where the robot
is tobeused’(ibid,p.223).This conclusion gained impor-
tance,withvan Doorn etal. (2017) predictingthat‘future
technology infusion [...]enables true relationshipsbe-
tween service robots and humans’(ibid,p.52). Conse-
quently, Wirtz etal. (2018) pointoutspecific research op-
portunities regardingservice robotsandemployee inter-
action,as little attention has been paid tousers’responses
todate.
Sofar service literature discusses waysto incorporateser-
vice robotsintoservice environmentstofoster customer
acceptance. These applications rangefrom functional use
cases,such as navigation (Kanda etal. 2010), carryingof
shoppingbags(Iwamura etal. 2011) or distribution offly-
ers (Shi etal. 2018), to hedonistic use cases,such as enter-
tainmentofcustomers (Aaltonen etal. 2017; Meyer etal.
2018).
Previous studies haveestablished thatservice robotsen-
hance the shoppingexperience and shoppingactivities
(Doeringetal. 2015; Iwamura etal. 2011).However,ahigh
degree ofuser acceptance is required for service robotsto
haveapositiveeffecton their implementation (ˇ
Cai´
cetal.
2018; Kirby etal. 2010) since the interaction partners of
service robotsareconstantlychangingand cannotbe
trained (Goodrich and Schultz 2007).To achieve accep-
tance,human-robotinteractions mustbeuser-friendly
and needs-oriented (Kirby etal. 2010), takingbothdia-
logue (Iwamura etal. 2011) and non-verbal communica-
tion (Doeringetal. 2015) into account.
Emotional and psychological aspectsalsoplaya crucial
role in the acceptance and implementation ofservice ro-
bots(Stockand Merkle 2017; Wirtz etal. 2018).The role
profile ofservice robotsoften shiftsduetochanges in us-
ers’expectations and the technical possibilities ofboth
hardware and software,such as processingofspeech sig-
nals and emotions (Gollnhofer and Schüller 2018).Instead
ofsimplyreceivingcommands,service robotscanbecome
accepted interlocutors,which expand their potential field
ofapplications (Kirchner and Alempijevic 2012).
Yet, scholarlyresearch tends toconfine customers tothe
crucial user group(e.g.Iwamura etal. 2011; Kanda etal.
2010), upon which prior studies haveprimarilyfocused
(Subramonyetal. 2017).There are two opportunities for
research tocontributetothe service field’sknowledge:(1)
First, Subramonyetal. (2017) assume thatthe underex-
ploited potential ofresearch on employees resultsfrom a
lackofawareness thatemployees are keyfor recognising
and resolvingpredominantservice-related issues. It
seems especiallyfruitful toexplore the FLEs’perspective,
as this stakeholder grouphas frequentand directcontact
withcustomers (Jonas etal. 2016) and is responsible for
executingservices and implementingnew processes (Cad-
wallader etal. 2010).Arich understandingofFLEs’accep-
tance ofand resistance toservice robotshasthe potential
toextend the service fields’knowledgeby another per-
spective(Wirtz etal. 2018).
(2) Second,scholarlyresearch can benefitfrom a service
systems perspectiveonthe FLEs’acceptance ofand resis-
tance toservice robots. Amongother technologies,service
robotsrapidlychangethe frontline oforganisations,yet
we lackinsightintohowthese technologies are imple-
mented in service systems (DeKeyser etal. 2019).Asitis
acknowledged thatservice robotshavethe potential to
transform the nature ofservice environments,including
stationaryretail (van Doorn etal. 2017), itseems fruitful to
understand how FLEs perceiveservice robotswithin their
workingenvironment(Kaartemo and Helkkula 2018; Sub-
ramonyetal. 2018).Accordingly, Wirtz etal. (2018) and
DeKeyser etal. (2019) call for research exploringnotonly
customers’butalso FLEs’acceptance ofsuch service sys-
tems.
Based on the above,we aim toexplore FLEs’perceptions
ofservice robotsinservice systems,atopic thatis interest-
ingboththeoreticallyand managerially(Kaartemo and
Helkkula 2018; DeKeyser etal. 2019; Wirtz etal. 2018).
Specifically, we aim toanswerthe followingresearch
question:
From a service systems perspective, what are aspects of FLEs’
acceptance of and resistance to service robots in stationary
retail?
Toanswerthis question,we review relevantliterature and
presentour qualitativeinterview study.Then,we discuss
how the findingsshedlighton FLEs’perceptions ofser-
vice robotsinservice systems.
2. Review of Relevant Literature: Service Robots
in Service Interactions
Interactivevalue creation occurs when a companyand its
customers interact(Reichwald and Piller 2009).Notonly
extrinsic benefits,such as monetaryrewards,butalso in-
trinsic benefits,such as the actual experience ofamean-
ingful interaction,are crucial for interactivevalue creation
to occur (Reichwald and Piller 2009).Inother words,the
experience ofameaningful interaction is necessaryfor
companies toenable their customers totakeupthe com-
pany’svalue proposition and makeitvaluable.
2.1. Value co-creation through service interactions
Consequently, throughdigital innovations and implemen-
tation ofnew technologies,companies,especiallythose in
the retail sector,haveexplored new types ofservice inter-
actions withcustomers toco-createvalue (Grewal etal.
2017).AccordingtoLyons and Tracy (2013), ‘[v]alue is re-
Meyer/Jonas/Roth, Frontline Employees’ Acceptance of and Resistance to Service Robots in Stationary Retail
22 SMR · Journal of Service Management Research · Volume 4 · 1/2020 · p. 21– 34
https://doi.org/10.15358/2511-8676-2020-1-21
Generiert durch IP '188.192.241.220', am 13.02.2020, 09:57:10.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.
alized throughinteractions and co-creation amongservice
systems’(ibid,p.1).Service interactions are also valuable
for customers when theyoccur in the situation ofcustom-
ers’needs since theythen promise appropriatebenefitsfor
customers,as beneficiaries. Aservice interactionassuchis
perceived as positiveandbeneficial when itconveysthe
feelingoffun,competence,exploration and creativity (De-
ci etal. 1999).Inservice systems,service interactions en-
tail a setofformal and informal processes thatdefine the
nature ofthe exchangeandenable value co-creation
(Sandströmetal. 2008).
Companies and their customers co-createvalue within
service systems as their roles are indistinct(Vargo and
Lusch 2017; Var goetal. 2008).Service systems are config-
urations ofinteractingresources,includingpeople and
technologies,thatenable value co-creation duringthe ser-
vice encounter (Larivi`ere etal. 2017), which Solomon etal.
(1985) define as ‘the dyadic interaction between a custom-
er and a service provider’(ibid,p.99) and Larivi`ere etal.
(2017) define as ‘anycustomer-companyinteraction that
resultsfrom a service system thatis comprised ofinterre-
lated technologies (either company-orcustomer-owned),
human actors likeemployees and customers,physical/
digital environments and company/customer processes’
(ibid,p.2).The service encounter has evolved tobecome
dominated by technology (e.g.,mobile sales assistants,
computer terminals,AI-based service agents;Genenniget
al. 2018; Larivi`ere etal. 2017; Wünderlich and Paluch
2017).Thus,itis no longer understood as a strictlyhu-
man-dominated phenomenon in which FLEs serveasthe
face ofthe companytocustomers and specific learned be-
haviours appropriatefor the situation are performed,but
as a balanced composition involvingthe ‘interdependent
roles oftechnology, employees,and customers’(Larivi`ere
etal. 2017, p.1) enabled by aservice system.
2.2. The interplay between technology and FLEs in
service systems
Withreference tothe aforementioned balanced composi-
tion between human and machine actors,the primaryhu-
man actors atthe interface ofservice interactions are cus-
tomers and FLEs. FLEs bringcompetence,knowledgeand
experience tothese interactions,enablingthem toade-
quatelyservecustomers’needs and co-createvalue with
customers. However,the incorporation ofinteractiveser-
vice technologies intoservice systems has influenced in-
teractivevalue creation (Larivi`ere etal. 2017).Interactive
service technologies in a retailer’sservice system have
been classified intovarious schemes (Ahearne and Rapp
2010).Interactiveservice technologies servedistinctivero-
les in the service encounter –(1) augmentation,(2) substitu-
tion and (3) network facilitation –which havedifferentcon-
sequences on the interaction between FLEs and customers
(Larivi`ere etal. 2017).(1) FLEs are augmented by technolo-
gies that‘assistand complement’ them (Larivi`ere etal.
2017, p.3).For example,service robotsmaycollaborate
with human medical staff in the field ofelderlycare (van
Doorn etal. 2017).Also,FLE-centric technologies,such as
mobile computers thatcan takeinventoryofshelves,aug-
ment as theyspeed upthe transactions handled by FLEs,
leavingmore time for them to care for customers. (2) Inter-
activeservice technologies serveasasubstitute for interac-
tions withhumanFLEs (Larivi`ere etal. 2017).For in-
stance,customers mayuse self-checkoutsatretail stores,
reducingthe need for FLEs (Ahearne and Rapp 2010).(3)
FLEs can networkmore comfortablyusinginteractiveser-
vice technologies thatfacilitate networking.These interac-
tiveservice technologies actas ‘an enabler ofconnections
and relationships[...]rather than focusingon replacing
human employees’(Larivi`ere etal. 2017, p.4).
Inconsequence,the growingdependencyon technology
in service encounters has alreadyaltered FLEs’role in the
service system. Yet, as noted by Böhmann etal. (2018),
‘[e]mphatic interaction,creativesolutions,and authentic
experiences all remain mostlythe domain ofhuman actors
in service systems’(ibid,p.1).Nevertheless,itis expected
thatnew technologies will bedeveloped that‘defy limita-
tions toenable organization-customer interactions ofever-
increasingdiversity and consistencyacross multiple
pointsofcustomer contact’ (Singhetal. 2017, p.1).Such
technologies will notonlydigitalise existingservices but
also offer new types ofservice interactions (Grewal etal.
2017).
2.3. The impact of service robots on service
interactions
New types ofservice interactions are onlysome ofthe
manychanges produced by service robots;as Parasura-
man and Colby (2015) note,‘[r]obotswillopen a revolu-
tionaryfrontier thatcould upsettraditional customer-em-
ployee relationships’(ibid,pp.59–60).Inparticular,ser-
vice robotsareexpected toalter FLEs’role in the service
system (Wirtz etal. 2018).
Because these robots can read,understand and respond to
people’semotions withempathetic intelligence (in addi-
tion tointuitive,analytical and mechanical intelligence;
see Huangand Rust 2018), they‘become increasinglyim-
portantduringservice encounters’(Stockand Merkle
2018, p.1056).Theylargelyoperateautonomouslyand in-
teractwithbothFLEs and customers. AccordingtoAhear-
ne and Rapp (2010), the latter characteristic ofthese tech-
nologies offers a wide spectrum ofpossibilities for re-
search on service robotics in retail (e.g.,Goodrich and
Schultz 2007; Schraft and Schmierer 2000; Stockand Merk-
le 2017; van Doorn etal. 2017; Wirtz etal. 2018).
Tosummariseprior literature,service robots are machines
withthe capability tomakeautonomous decisions and
Meyer/Jonas/Roth, Frontline Employees’ Acceptance of and Resistance to Service Robots in Stationary Retail
SMR · Journal of Service Management Research · Volume 4 · 1/2020 · p. 21– 34 23
https://doi.org/10.15358/2511-8676-2020-1-21
Generiert durch IP '188.192.241.220', am 13.02.2020, 09:57:10.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.
sensitivelyadapt tothe given context (Kirchner and
Alempijevic 2012).Inordertodoso,service robotsreceive
datafrom a variety oflocal inputchannels (e.g.,sensors or
cameras)and process these datatoexecuteanintricateset
ofactions (Singer 2009).Westress that, in addition tocus-
tomers,FLEs are a relevantgroupofusers.
Inordertointeractand communicateonanemotional-so-
cial level or todeliver services to human counterparts,ser-
vice robotsrequire a certain degree ofsocial presence (van
Doorn etal. 2017).Wirtz etal. (2018) argue thatthree de-
signattributes are relevantin a service context: (1) pres-
ence,(2) anthropomorphism and (3) task orientation.(1) A ro-
botwithaphysical presence,such as the semi-humanoid ro-
botPepper,makes users feel as thoughtheyare communi-
catingwith another social entity (Jones 2017).Virtual arti-
ficial intelligence (AI) software can bealsocategorised as
aservice robot(Wirtz etal. 2018), butin this article,we fo-
cus on physicallyrepresented and mobile service robots
(Schraft and Schmierer 2000).(2) Service robotsmaybean-
thropomorphic (e.g.,Sophia,a social humanoid robot) or
non-anthropomorphic (e.g.,Walmart’sshelf-scanningrobot)
(Kirby etal. 2010).(3) Service robots can bedesigned for
cognitive-analytical tasks,such as imageanalysis,or emo-
tional-social tasks,such as reception ofcustomers (Wirtz et
al. 2018).The latter tasksaremostrelevanttointeraction
and communication with users in service system environ-
ments.
Based on the previouslymentioned characteristics and ca-
pabilities ofservice robots,we utilise the followingoper-
atingdefinition ofaservice robotin this article:
Service robots are mobile, system-based, autonomous, adapt-
able, physically represented machines that provide service to
an organisation’s customers and FLEs by interacting and
communicating at an emotional-social level.
AsFLEs mayperceiveaservice robot’srolewithin service
interactions differentlythan the organisation adoptingit;
the technological innovation mayfail (Pantano etal. 2018).
Technology adoption depends on organisations’capabili-
ties to accuratelyrespond toFLEs’needs (Lewis and Lo-
ker 2014); makingitcrucial tounderstand their acceptance
ofand resistance tothe distincttechnology, such as service
robots,beforehand.
2.4. Acceptance of and resistance to service robots
Acceptance ofand resistance tonewtechnologies have
been discussed in service literature since the 1970s. The
mostfrequentlyused model tointerpretacceptance of
technologies is Davis’ (1989) technology acceptance model
(TAM), which is based on the theoryofreasoned action
(Fishbein and Ajzen 1975) and social cognitivetheory
(Bandura 1986).Yet, itlackscircumstantial aspectstoclari-
fy how technology is adopted and used (Benbasatand
Barki2007), drawingcriticism.
Technology acceptance theories focus on social and tech-
nological aspectssuchassocialinfluences. Toincorporate
specific influences intothe TAM,two variations (TAM2
and TAM3), were developed from differentperspectives,
such as marketingand sales (e.g.,Lewis and Loker 2014).
Amore integrated model is the unified theoryofaccep-
tance and use oftechnology model (UTAUT), which was
developed by Venkatesh etal. (2003).The UTAUT consid-
ers social influences,such as normativebeliefsaboutpeers
and supervisors,and facilitatingconditions,such as orga-
nisational and technical support, tobesignificantcriteria
affectingthe use ofasystem.
Studies havealsoexamined technology acceptance in the
context ofsales interactions (Ahearne and Rapp 2010).
FLE-specific aspectsareofparticular interestin this re-
gard as FLEs facilitateinteraction between an organisation
and itscustomers,spanningboundaries (Ahearne and
Rapp 2010).Customers appreciatepleasantrelationships
withFLEs who createsocialandemotional value during
service encounters,which is sometimes described as rap-
port, engagementor trust(Wirtz etal. 2018).
Acceptance models havebegun tofocus on service robots
as theyare piloted in an increasingnumber oforganisati-
ons. Based on Solomon etal.’s(1985) role theoryand Da-
vis’ (1989) TAM,Stockand Merkle (2017) developed a the-
oretical social frontline robotacceptance model (SFRAM)
toexamine customers’expectations for an interaction with
afrontline social robotduringaservice encounter. Also,
Wirtz etal. (2018) developed the service robotacceptance
model (sRAM), which builds upon the TAM(Davis 1989).
Theyinclude customers’social-emotional needs,per-
ceived humanness,perceived social interactivity and per-
ceived social presence,relational needs,trustand rapport.
However,bothSFRAM and sRAMfocus onlyon custom-
ers,althoughFLEs’acceptance mustalso beevaluated in
order to orchestratethe use ofservice robotsinaservice
system:‘Addressingonlythe primaryuser in service ro-
boticsisunsatisfactory[...]the focus should beonthe set-
ting, activities and social interactions ofthe groupofpeo-
ple where the robotis tobeused’(Severinson-Eklundh et
al. 2003, p.223).
Adifferentstream ofresearch focuses on resistance to
technologies (Lapointe and Rivard 2005).The sudden in-
fusion ofservice robotsinservice environmentstriggers
various emotional states,from happiness toanxiety; not
all FLEs wanttobeconfronted withnewtechnologies
(Harris and Ogbonna 2002).The uncertainty associated
withbeingreplaced by technology maybe alarming, re-
sultingin fear and resistance (Shah etal. 2017).While
technology acceptance theories tend to assume thatusers
havethe freedom tochoose,technology resistance theo-
ries tend to assume thatusers,includingFLEs,are fre-
quentlyrequired toadapt toanewtechnology the compa-
Meyer/Jonas/Roth, Frontline Employees’ Acceptance of and Resistance to Service Robots in Stationary Retail
24 SMR · Journal of Service Management Research · Volume 4 · 1/2020 · p. 21– 34
https://doi.org/10.15358/2511-8676-2020-1-21
Generiert durch IP '188.192.241.220', am 13.02.2020, 09:57:10.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.
Tab. 1: Qualitative sample description
nyprovides them with(SagaandZmud 1993).Inlinewith
Sheth(1981), we expectthat, due tothe perceived lackof
usefulness ofnew technology, the majority ofindividuals,
includingFLEs,will prefer toavoid change,and a minori-
ty will beinterested in adapting.Toshedlighton this,the
currentstudyexplores aspectsofFLEs’acceptance ofand
resistance toservice robotsinastationaryretail context
from a service systems’perspective.
3. Research Method
Aqualitativeexplorativeapproach was chosen toexamine
aspectsofFLEs’acceptance ofand resistance toservice ro-
botsinaretailer ’sservice system. Aqualitativeapproach
is suitable for obtainingimplicitknowledgeofFLEs (John-
son 2015), and an explorativeapproach ‘can beparticular-
lyuseful in exploringphenomena where little under-
standingexists’(Johnson 2015, p.262).
Aninterview studywas chosen so thatFLEs could share
rich descriptions ofthe meaningascribed toservice robots
while leavingthe datauptothe investigator ’sinterpreta-
tion (Tewksbury 2009).Morespecifically, individual in-
depthinterviews were chosen togain a thorough under-
standingofhow FLEs perceiveandexperience service ro-
botsinaretail workingenvironment, allowingthem ‘to
delvedeeplyinto social and personal matters’(DiCicco-
Bloom and Crabtree 2006, p.315).Overall,the studyaims to
achieveprofound insights and a more complete under-
standingofFLEs’acceptance ofand resistance toservice ro-
botsfrom a retail service system perspective(Johnson 2015).
3.1. Data collection
FLEs were included in the studyiftheymettwo criteria:
(1) the respondent’semployer has tested or implemented
service robotsand(2) the interviewee has had experience
withaservice robotin a retail service system. Bycombin-
ingthe insightsderived from the interviewees withheter-
ogenous service robotexperiences,the reality ofthe retail
context can beappropriatelyrepresented. FLEs are the
mostrelevantsource ofinformation for this studyas they
havepersonal experiences,assessmentsandemotions re-
gardingservice robotsintheir workingenvironment
(Johnson 2015).
The selected interviewees are FLEs ofsixretailers in the
grocery, shoppingcentre,sportinggoods,electronics and
fashion sectors ofthe sales marketsofGermanyand Aus-
tria. At these retailers,service robotsperform service tasks
such as provision ofinformation,entertainmentand navi-
gation.
Between earlyA
ugust 2018 and mid-October 2018, 24 in-
terviews were conducted,recorded and transcribed verba-
tim (Tab . 1 ).For privacyreasons,interviewees’names
Meyer/Jonas/Roth, Frontline Employees’ Acceptance of and Resistance to Service Robots in Stationary Retail
SMR · Journal of Service Management Research · Volume 4 · 1/2020 · p. 21– 34 25
https://doi.org/10.15358/2511-8676-2020-1-21
Generiert durch IP '188.192.241.220', am 13.02.2020, 09:57:10.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.
were anonymised. Since this studyaims tobuild,nottest,
theoretical knowledge,the sample sizeislargelysufficient
(Creswell and Poth2012; Marshall etal. 2013).Datacollec-
tion ceased when theoretical saturation was reached
(Guestetal. 2006).
FollowingYeo etal. (2013), the studyis comprised oftwo
parts. First, abasis for a broad,mature understandingof
potential drivers and barriers toadoption ofservice ro-
botsiscreated by conductingunstructured,open-ended
interviews withsixFLEs from a groceryretailer. Second,
the information gathered in the firstpartis used as a
gatewaytoadditional exploration through semi-struc-
tured interviews (Yeo etal. 2013).Wedeveloped an inter-
view guideline thataligned withextantliterature toen-
sure thatrelevantaspectsofprior technology acceptance
and resistance research and a service system perspective
were considered in the interviews (Maxwell 2008).Three
iterations ofthe guideline were pre-tested withanoppor-
tunity sample (Saunders etal. 2009).These pre-testscon-
firmed thatthe structure ofthe interview guideline was
appropriate.
3.2. Data analysis
FollowingMiles etal. (2014), interpretativedataanalysis
was performed toreducethe complexity ofthe interview
transcriptsandtransform the dataintoalucid,workable
setofconnections. Specifically, f ollowingSalda ˜na (2009), a
transparent, open two-cycle codingprocess was applied
to‘breakupand segmentthe dataintosimpler,general
categories and [...]toexpand and tease outthe data,in or-
der toformulatenewquestions and levels ofinterpreta-
tion’(Coffeyand Atkinson 1996, p.30).
First, descriptivecodingwas performed tocreateanin-
ventoryoftopics to summarise segmentsofdata(Wolcott
1994).Thisisanappropriateapproach for studies witha
wide variety ofdata(Miles etal. 2014; Salda˜na 2009).Sec-
ond,pattern codingwas performed togroupthose sum-
maries intoasmallernumber ofaspects(Miles and Hu-
berman 1994).The codingprocess was conducted using
the software MAXQDA 2018.
Weimplemented several measures toensurethatthe data
analysis was rigorous and trustworthy.First, toensurein-
tersubjectivetraceability, atransparent, open two-cycle
codingprocess was applied. Second,we tested and veri-
fied the resultsby analysingthe datasetfor contradictory
eventsandmodified the findingsover various iterations
ofanalysis and discussion. Third,the objectivity ofthe
analysis was evaluated in terms ofintercoder reliability
(Campbell etal. 2013).Arandomlyselected reliability
sample,representing11.3%ofthe total sample,was cod-
ed by two independentresearchers who reached a broad
consensus on the definition and completeness ofthe cate-
gorisation (Cohen 1960).Particularly, the coefficientkappa
values for each categorywere abovethe 0.61 threshold,
representingsubstantial consensus (Landis and Koch
1977).The overall intercoder reliability was 0.85, indicat-
ingalmostperfectobjectivity in the categorisation (Landis
and Koch 1977).
4. Findings
The interview analysis uncovered 20 aspectsofFLEs’ac-
ceptance ofand resistance toservice robotsfrom a service
system perspective. We assigned these aspectstofive
higher-order categories:(1)loss ofstatus, (2) tension,(3)
required commitment, (4) role incongruencyand (5) advo-
cation (Fig. 1,p.27).Keyquotations were used to describe
the findings(see Tab. 2,p.28) as theyadd transparency
and deepen our understanding(Patton 2007).
Asproposed by Erlingsson and Brysiewicz (2017), the 20
identified aspectsweregrouped intohigher-order catego-
ries accordingtotheir contextual proximity and constitu-
entproperties. Intotal,fivehigher-order categories
emerged. These higher-order categories aim tophrase the
underlyingmeaning, i.e.,latentcontent, found in a group
ofaspects. Moreover,theyaim to communicatethe identi-
fied aspectstothe reader on bothanintellectual and inter-
pretativelevel.
Inparticular, (1) four aspectswereclustered around the
firstnotion of‘loss ofstatus’, relatingtoFLEs’concerns
aboutlosingtheir standingthroughthe use ofservice ro-
bots(Gaudiello etal. 2016; Pellegrini and Scandura 2008).
(2) Fiveaspects were assigned tothe second notion of‘ten-
sion’, related toFLEs’concerns aboutinconvenient
changes in the workingenvironmentwhich are perceived
as unpleasantbyFLEs (Karr-WisniewskiandLu2010;
Spreer and Rauschnabel 2016).(3) Three aspectswereas-
signed tothe third notion of‘required commitment’, relat-
ingtoanincreaseinresponsibilities and psychological
burdens (Boxall and Macky 2014; Lee etal. 2016).(4) Five
aspectswereclustered around the notion of‘role incon-
gruency’, relatingtoFLEs’perception ofservice robotsas
athreatto social relationshipsandanunpleasantinterac-
tion partner (Kamide etal. 2014; Okazakietal. 2010).(5)
The remainder were assigned tothe notion of‘advoca-
tion’, relatingtoFLEs’desire tobesufficientlytrained in
the use ofservice robotstobeable toactivelycontributeto
the organisational changes due tothe use ofservice robots
(Dongetal. 2008; Hanaysha 2016).
(1) Loss of Status.The fear oflosingone’sjobis a major bar-
rier to acceptance ofservice robots. AlthoughFLEs refer to
the operational imperfections ofservice robots and con-
clude thattheir own manpower cannotbereplaced yet,
theystill express several fears. (a)FLEs fear thattheywill
besubstituted withaservice robotand ‘would havetofind
another job’ (see Ta b . 2 ,p.28, John,26years old).FLEs do
Meyer/Jonas/Roth, Frontline Employees’ Acceptance of and Resistance to Service Robots in Stationary Retail
26 SMR · Journal of Service Management Research · Volume 4 · 1/2020 · p. 21– 34
https://doi.org/10.15358/2511-8676-2020-1-21
Generiert durch IP '188.192.241.220', am 13.02.2020, 09:57:10.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.
Fig. 1: Aspects of acceptance of
and resistance to service robots
by frontline employees
notresistchangeper se,butmayresistalossofpay, loss of
comfort, and especiallyalossofstatus (Dentand Gold-
berg 1999).(b) A majority ofFLEs is uncertain about
whether theywill beable tocontinue to workor whether
theywill bedispensable due toservice robots’capabilities.
This situation is differentfrom the certainty ofbeingmade
redundant.Due tothe uncertainty aboutaservice robot’s
future capabilities,FLEs cannotstartpreparingfor redun-
dancyand future (un)employment(Ali etal. 2016) since
theysimply‘don’t know how far itcan gowithtechnolo-
gy’ (Karen, 31).(c)FLEs fear beingdegraded and ‘alittle
less respected’(Nancy, 28) by customers or service robots.
FLEs desire a workingenvironmentin which theyfeel
valued by managementand customers;theydo notwant
tobeperceived as less essential due tothe existence ofser-
vice robots. Fear ofthis typeofdisempowermentcan un-
dermine the perceived raison d’ˆetre ofFLEs. (d)FLEs do
notaccept paternalistic behaviour by aservice robot; they
wanttohavethe final say.Theymayaccept service robots,
as longas the service robotsdonottell them whattodo.
Thus,FLEs would prefer todelegatetaskstoaservice ro-
bot(Patricia, 60).
(2) Tension.The introduction ofservice robots causes FLEs
toface novel challenges within their dailyworkingenvi-
ronmentas theyare unknown actors in the service system.
FLEs are placed in a stress field ofheterogeneous de-
mands by the customer,organisation,t
echnological de-
vices and service robots. (a)This leads todisruption of rou-
tines as FLEs are confronted with new challenges regard-
ingservice robotsasstated,‘no employee expectsarobot
tosuddenlystand behind him’(Barbara, 21).(b) Unpre-
pared,fastincorporation ofservice robotsintoafamiliar
workingenvironmentresultsintechnostress amongama-
jority ofFLEs as ‘itputsyou under additional stress’
(Nancy, 28).(c)Fear of public failure while usingaservice
robotarises simultaneouslytotechnostress. FLEs fear un-
intended malfunctions while interactingwithaservice ro-
botand beingincapable ofprovidinginformation abouta
service robottocustomers. Asstated,FLEs ‘feel a bithu-
miliated’(Sarah, 29) when realisingthattheyare notas
well informed aboutthe service robotas theywish tobe.
(d)Some FLEs cannotcomprehend the retailer’s decision
topilotor implementaservice robot.The perceived lack of
plausibility ofthe decision maylead tointerpersonal hur-
dles. For example,one participantexplained:‘Emotional-
lyIhonestlymustsaythatthere is a certain hatred’(Willi-
am, 26).(e)FLEs noteoperational imperfections in the ser-
vice robots,includingtechnological issues (e.g.,shortbat-
terylife,poor acoustics,slow motor skills or frequent
breakdown)and difficulties related to dailyuse (e.g.,the
service robotsblockpathways or cannotbeswitched off).
Afew FLEs are patientwithservice robots,as theyunder-
stand thatmore innovativetechnologies are more likelyto
bebuggy or tobreakdown (Thomas, 27).However,the
majority ofFLEs are frustrated and annoyed by operation-
al imperfections. For example:‘Inthe course ofthe after-
noon he justwentreallylimpand then shutdown,which
was a bitofapity, because people were there and asked
where Pepper is’(Jessica, 27).
Meyer/Jonas/Roth, Frontline Employees’ Acceptance of and Resistance to Service Robots in Stationary Retail
SMR · Journal of Service Management Research · Volume 4 · 1/2020 · p. 21– 34 27
https://doi.org/10.15358/2511-8676-2020-1-21
Generiert durch IP '188.192.241.220', am 13.02.2020, 09:57:10.
Das Erstellen und Weitergeben von Kopien dieses PDFs ist nicht zulässig.
Tab. 2: Exemplary key quotations of identified aspects of acceptance of and resistance to service robots by FLEs
(3) Required Commitment.The participantsmentioned
three types ofeffort-related barriers toadoption. (a)FLEs
feel an increase in responsibilities;in addition totheir daily
workload,theymustkeepan eyeonthe service robot.As
stated,theyperceiveto‘bear the responsibility for it’ (Bar-
bara, 21).This requires adjustmentofFLEs’duties in order
toavoid overload and the perception thatservice robots
do notsupportFLEs’work.(b) FLEs experience mental
strains related to learninghow to command and interact
withaservice robot.Doingso is seen as necessarybecause
FLEs believeitis their duty to answer customers’ques-
tions,includingthose aboutservice robots(Jessica, 27).(c)
The time efforts required tobecome familiar withaservice
robotmaybeabarrier toadoption. Aptly, one participant
explained that‘you couldn’t stand in frontofthe robotfor
halfan hour and deal withit’ (Jessica,27).
(4) Role Incongruency. Role incongruencyoccurs when an
actor’sperception ofanother actor’sroledoesnotmatch
the latter’sactual behaviour. FLEs mentioned fivepoten-
tial barriers toadoption associated withroleincongruen-
cy.(a)FLEs criticise the social-emotional callousness ofinter-
actions withaservice robot, which is feltbybothFLEs
and customers. ‘This personal,valuable contact, the smile,
the warmth,that’snotwhatarobotcan do’(Patricia, 60).
(b) FLEs tend toanthropomorphise a service robot’sphysi-
cal appearance.Inother words,theyimbue a nonhuman
service robotwithhumanlike characteristics,motivations,
intentions or emotions based on itsrealorimagined be-
haviour,which influences how theyinteractwithit.For
example,one participantexplained:‘You can identify alit-
tle better withsomethingthatlooksmorelikeahuman
than [...]amovingplastic part’ (Nancy, 28).(c)FLEs fear
relational deterioration of interaction within the service sys-
tem caused by service robots. Inother words,theybelieve
thatservice robotsdisturbthe interpersonal relationships
between FLEs and customers,t