Mobile Payment Continuance Intention

Mobile Paymen Con inuance In en ion F ank Bi a F anque A hesis submi ed in pa ial ul illmen o he equi emen s o he deg ee o Doc o in In o ma ion Managemen No embe 2021 NOVA In o ma ion Managemen School Uni e sidade No a de Lisboa In o ma ion Managemen Specializa ion in In o ma ion Technologies II Doc o al P og amme in In o ma ion Managemen P o esso Dou o Tiago And é Gonçal es Félix de Oli ei a, Co-Supe iso ! P o esso Dou o Ca los Tam Chuem Vai, Co-Supe iso III Doc o al P og amme in In o ma ion Managemen Copy igh © by F ank Bi a F anque All igh s ese ed ! IV Doc o al P og amme in In o ma ion Managemen Abs ac The dis up i e de elopmen o in o ma ion and communica ion echnologies o e he las wo decades has e olu ionized he mobile phone indus y, exponen ially inc eased he numbe o mobile phone use s, and encou aged companies o make a ious se ices a ailable h ough a mobile phone. Mobile paymen is one o he as es g owing se ices, enabling use s o pe o m inancial ansac ions o e a mobile phone. The exponen ial g ow h o mobile paymen has a ec ed a numbe o sec o s including inance and echnology, hus ein o cing he need o a deep unde s anding o he impac o he con inued use o mobile paymen se ices. Wi h his disse a ion we con ibu e o a be e unde s anding o he de e minan s o con inuance in en ion o use mobile paymen a he indi idual le el. Fo his eason, we e de eloped ou s udies, one li e a u e e iew, and h ee empi ical s udies. In he i s s udy (Chap e 2) we conduc ed a li e a u e e iew o exis ing s udies on indi idual con inuance in en ion o use an in o ma ion sys em. In Chap e 3 we assessed he con inuance in en ion o use m-paymen employing wo heo e ical models, he DeLone and McLean in o ma ion sys em success model (D&M ISSM) and he expec a ion-con i ma ion model (ECM) in an A ican con ex . The impac o ask echnology i (TTF) and o e all us on ECM o explain he con inuance use o m- paymen is analysed in Chap e 4. In he las s udy, Chap e 5, we assess he impac o cul u e on con inuance in en ion o use m-paymen , combining he ECM and Ho s ede’s cul u al dimensions. This disse a ion p o ides se e al con ibu ions o esea ch and p ac ice, con ibu ing o he ad ancemen o knowledge and implica ions o se ice manage s, se ice p o ide s, use s, and esea che s. The li e a u e e iew applies me a-analysis and weigh analysis om 115 empi ical s udies om con inuance in en ion o use an in o ma ion sys em (IS). The indings e eal ha he ac o s wi h s onges in luence on con inuance in en ion o use an IS a e a ec i e commi men , a i ude, sa is ac ion, hedonic alue, and low. Mo eo e , sample size, indi idualism, unce ain y a oidance, and long- e m o ien a ion mode a e he ela ionship o pe cei ed use ulness on con inuance in en ion. Powe dis ance, masculini y, and indulgence mode a e he V Doc o al P og amme in In o ma ion Managemen ela ionship o sa is ac ion on con inuance in en ion. F om he i s empi ical s udy we examine he in luence indi idual pe o mance d i e s on con inuance in en ion o use m–paymen in an A ican con ex . We ind ha he mos impo an p edic o s o con inuance in en ion o use m-paymen a e indi idual pe o mance, use, and sa is ac ion. The second empi ical s udy in eg a es TTF and o e all us heo ies and e alua es hei ela ionships o con inuance in en ion o use mobile paymen . Findings show ha use, indi idual pe o mance, o e all us , and he mode a ion ole o sa is ac ion a e he mos impo an cons uc s o explain con inuance in en ion. The las empi ical s udy assesses he impac o cul u e on m-paymen con inuance in en ion. The indings e eal ha he ela ionships be ween con i ma ion on sa is ac ion and pe cei ed use ulness, and pe cei ed use ulness on con inuance in en ion a e mode a ed by unce ain y a oidance. Keywo ds: Mobile paymen , con inuance in en ion, ECM, D&M ISSM, TTF, us , cul u e, A ican con ex . VI Doc o al P og amme in In o ma ion Managemen Resumo O desen ol imen o dis up i o das ecnologias de in o mação e comunicação nas úl imas duas décadas e olucionou a indús ia da ele onia mó el, aumen ando exponencialmen e o núme o de u ilizado es de elemó eis, enco ajando des a o ma as emp esas a disponibiliza di e en es se iços a a és de um elemó el. O se iço pagamen o mó el é um dos se iços que se encon a em um ápido c escimen o pe mi indo aos u ilizado es e e ua ansações inancei as a a és de um elemó el. O c escimen o exponencial do se iço de pagamen o mó el em a e ado di e en es sec o es, ais como inanças e ecnologia, e o çando a necessidade de uma comp eensão p o unda do impac o da u ilização con ínua dos se iços de pagamen o mó el. Com o desen ol imen o des a disse ação, espe amos con ibui pa a uma melho comp eensão dos de e minan es da in enção de con inua a usa o se iço de pagamen o mó el a ní el indi idual. De o ma a conc e iza es e obje i o o am desen ol idos um o al de qua o es udos dis in os. No p imei o es udo (Capí ulo 2) ealizámos uma e isão bibliog á ica dos es udos exis en es sob e a in enção de con inua a u iliza um sis ema de in o mação. No capí ulo ês, a aliámos a in enção de con inua a u iliza o se iço de pagamen o mó el, emp egando dois modelos eó icos, o DeLone and McLean in o ma ion sys em success model (D&M ISSM) e o expec a ion-con i ma ion model (ECM) num con ex o a icano. O impac o do ask echnology i (TTF) e o o e all us no modelo ECM pa a explica o uso con ínuo do se iço de pagamen o mó el oi analisado no capí ulo qua o. No úl imo es udo, capí ulo cinco, a aliámos o impac o da cul u a na in enção de con inuação da u ilização do se iço de pagamen o mó el, combinando as dimensões cul u ais de Ho s ede e o modelo ECM. Es a disse ação ap esen a á ias con ibuições pa a a in es igação e pa a a p á ica, con ibuindo pa a o a anço do conhecimen o, p o ocando implicações pa a ges o es de se iços, p es ado es de se iços, u ilizado es e in es igado es. O es udo da e isão bibliog á ica aplicou me a-analysis e weigh analysis a pa i de 115 es udos empí icos de in enção con inua a u iliza um sis ema de in o mação (SI). Os esul ados e elam que os a o es com maio in luência na in enção de con inuação da u ilização de um SI o am o comp omisso a e i o, a i ude, sa is ação, alo hedónico, e low. Além disso, VII Doc o al P og amme in In o ma ion Managemen o amanho da amos a, indi idualismo, p e enção da ince eza, e o ien ação a longo p azo mode am a elação en e pe ceção da u ilidade e in enção de con inua , dis ância do pode , masculinidade e indulgência mode am a elação en e sa is ação e in enção de con inua . Pa a o p imei o es udo empí ico, examinámos a in luência dos a o es de desempenho indi idual na in enção de con inuação da u ilização do m-pagamen o num con ex o a icano. Ve i icámos que os p edi o es mais impo an es da in enção de con inua a u iliza o se iço de pagamen o mó el são o desempenho indi idual, uso e a sa is ação. O segundo es udo empí ico in eg ou as eo ias da TTF e da con iança ge al e a aliou as suas elações pa a a in enção de con inuação da u ilização do pagamen o mó el. Os esul ados mos am que o uso, desempenho indi idual, con iança ge al, o papel de mode ação da sa is ação são os a o es ele an es pa a explica a in enção de con inua a u iliza o se iço de pagamen o mó el. O úl imo es udo empí ico a alia o impac o da cul u a sob e a in enção de con inuação do pagamen o mó el. Os esul ados e elam que as elações en e con i mação, pe ceção de u ilidade com sa is ação, pe ceção de u ilidade com in enção de con inua são mode adas pela p e enção da ince eza. Pala as-cha e: Pagamen o mó el, in enção de con inua , ECM, D&M ISSM, TTF, Con iança, Cul u a, con ex o a icano. VIII Doc o al P og amme in In o ma ion Managemen Publica ions Lis o s udies esul ing om he disse a ion. Pape s (published): F anque, F.B., Oli ei a, T., Tam, C. and San ini, F.d.O. (2021), "A me a-analysis o he quan i a i e s udies in con inuance in en ion o use an in o ma ion sys em", In e ne Resea ch, 31(1), 123-158. h ps://doi.o g/10.1108/INTR-03-2019-0103 F anque, F. B., Oli ei a, T., & Tam, C. (2021). Unde s anding he ac o s o mobile paymen con inuance in en ion: empi ical es in an A ican con ex . Heliyon, 7(8). h ps://doi.o g/10.1016/j.heliyon.2021.e07807 Pape s (unde e iew o submi ed) F anque, F. B., Oli ei a, T., & Tam, C. (2021). Con inuance in en ion o mobile paymen : TTF model wi h T us in an A ican con ex . F anque, F. B., Oli ei a, T., & Tam, C. (2021). Role o he unce ain y a oidance cul u e on he expec a ion con i ma ion model: Mobile paymen case. IX Doc o al P og amme in In o ma ion Managemen Acknowledgemen s To my amily, especially my a he , Gilbe o F anque, and o all my b o he s, who encou aged me om he beginning o pu sue a PhD, and o hei inexhaus ible suppo o his jou ney, mainly emo ional suppo , bea ing in mind ha my amily is in Mozambique, in Te e ci y. To P o . Tiago Oli ei a, supe iso o his disse a ion, o being always a ailable, o all he help, suppo , guidance, and ideas ha signi ican ly con ibu ed o he success o he jou ney. A huge hanks o e e y hing. To P o . Ca los Tam, co-supe iso o his disse a ion, o being always a ailable, o all he help, suppo , guidance, and ideas ha signi ican ly con ibu ed o he success o he jou ney. A huge hanks o e e y hing. To P o . Fe nando de Oli ei a San ini o all he help, sugges ions, and c i iques, which con ibu ed o he en ichmen o Chap e 2. To No a In o ma ion Managemen School, o he oppo uni y gi en o pa icipa e in he PhD p og am. To Enginee ing School, Ca holic Uni e si y o Mozambique, o he oppo uni y gi en o go o Lisbon o pa icipa e in he PhD p og am. To he new iends om Lisbon, o he good momen s o un, ha helped a lo o he success o he jou ney. To my gi l iend, Gisela Manhique, who helped me emendously in he discussions and ansla ion om Po uguese o English o he di e en s udies. A big hanks o e e y hing. To me, o he cou age o go o a comple ely new coun y, wi h new challenges, o he i eless e o du ing he i e yea s o he jou ney. To all my since e hanks. Chap e 1 – In oduc ion 1.1. Mo i a ion Mobile phone ne wo ks a e booming and almos all a e in e connec ed, making people able o communica e and sha e in o ma ion anywhe e in he wo ld. Associa ed wi h his, mobile phone usage is g owing exponen ially, mo i a ing companies o deli e se ices ia mobile phones (Ka jaluo o e al., 2019; Pe saud & Azha , 2012). Taking in o conside a ion ha i can be used anywhe e and any ime, adding mo e alue o se ices, new se ices a e being made a ailable ia mobile phones, such as m-banking, m-paymen , and m-comme ce, he eby b inging cus ome s close o companies and s eng hening hei ela ionship (Oli ei a e al., 2016). In ou wo k we s udy Mobile paymen (m-paymen ). M–paymen is a paymen me hod ha uses mobile phones o make inancial ansac ions such as paying o goods o se ices, ans e ing money, and/o wi hd awing money (Fan, Shao, Li, & Huang, 2018; Zhou, 2013). Whe eas in some egions inancial ins i u ions a e a om he popula ion, o cing people o a el long dis ances o use inancial se ices, m-paymen has become one o he p ominen se ices (Gao & Waech e , 2015; Zhou, 2014), enabling use s o make ansac ions any ime and anywhe e (Lu, Wei, Yu, & Liu, 2017; Zhou, 2014). Socie y has unde gone a dis up i e e olu ion wi h he en ance o his echnology, which a ec s paymen ecosys ems. This se ice is g owing exponen ially all o e he wo ld and is b inging bene i s o use s and p o ide s. Conside ing i s bene i s, companies ealize i s po en ial, and a e p o iding i in di e en ways all o e he wo ld (Fan e al., 2018; Singh, Sinha, & Liébana-cabanillas, 2020). M-paymen echnology o igina ed in he Uni ed S a es and sp ead h oughou he wo ld (Fan e al., 2018). In A ica, his echnology was launched in Kenya and was quickly adop ed by o he coun ies. Mozambique is one o he A ican coun ies ha has adop ed his echnology, helping u al people who do no ha e banking in as uc u e nea hei homes (Ba is a & Vicen e, 2018). M-paymen has a majo impac on de eloping coun ies as i launches basic inancial se ices such as money ans e , paymen o goods and se ices, and/o wi hd awing money, he eby imp o ing people’s li es (Humbani & Wiese, 2018). 2 Doc o al P og amme in In o ma ion Managemen A g ea deal o esea ch has been de eloped o unde s and m-paymen in di e en egions (e.g., Chen & Li, 2017; Oli ei a, Thomas, Bap is a, & Campos, 2016; Sinha, Maj a, Hu chins, & Saxena, 2019). Howe e , he e a e ew s udies o unde s and m- paymen in an A ican con ex (e.g., Chen & Li, 2017; Lin, Fea he man, & Sa ke , 2017). Acco ding o Naba i e al. (2016) and Shaikh and Ka jaluo o (2015) no s udies abou in o ma ion sys em con inuance in en ion we e ound in he A ican con ex . Fo Bha ache jee, (2001) he ea ly s age o in o ma ion sys ems (IS) adop ion is a i al s ep owa d he success o IS, bu pe manen usage o IS and i s success is associa ed wi h con inued use ins ead o i s usage. In his sense, unde s anding wha ac o s in luence an indi idual o con inue using m-paymen has become necessa y and impo an o esea che s and companies (Bha ache jee, 2001; Shaikh & Ka jaluo o, 2015). I is hus ex emely impo an o unde s and he mos impo an d i e s ha in luence con inuous use o mobile paymen in an A ican con ex . 1.2. Con inuance in en ion models S udies on in o ma ion sys em con inuance in en ion (ISCI) ha e used a wide ange o heo ies in combina ion wi h he expec a ion-con i ma ion model (ECM) (Bha ache jee, 2001) such as expec a ion-con i ma ion heo y (ECT) (Oli e , 1986), echnology accep ance model (TAM) (Da is, 1989), uni ied heo y o accep ance and use o echnology (UTAUT) (Venka esh e al., 2003), and Flow heo y (Ge zels & Csikszen mihalyi, 1978), o name a ew. Howe e , con inuance in en ion o IS e e s o ac o s ha con ibu e o IS usage o a long ime. I in ol es unde s anding he long- e m ac o s ha con ibu e o he success o he IS (A. Bha ache jee, 2001; K. Wang, 2015). In he con ex o ISCI, ECM was he i s heo y p oposed by (Bha ache jee, 2001). ECM p oposes ha sa is ac ion o use IS is a c ucial ac o ha impac s con inuance in en ion, ollowed by pe cei ed use ulness o he IS. Also, con i ma ion o he expec a ions and pe cei ed use ulness a e impo an ac o s ha in luence use sa is ac ion. A e ECM appea ed many esea che s es ed and joined i wi h o he models in di e en egions and wi h dis inc echnologies (Ca illo, Sco na acca, & Za, 2017; Hadji & Degoule , 2016; Hong, Tai, Hwang, Kuo, & Chen, 2017b; Hsiao, Chang, 3 Doc o al P og amme in In o ma ion Managemen & Tang, 2016; Zheng, 2019). Mos s udies used ECM as a base heo y. Some used ECM alone (Al aimi e al., 2015; Susan o e al., 2016), and o he s in eg a ed i wi h o he heo ies and sel -cons uc s (Chen e al., 2013; Lee, 2010; Limayem & Cheung, 2008). In ou wo k we in eg a e ECM wi h DeLone & McLean in o ma ion sys em success (D&M ISSM) (DeLone & Mclean, 2003), Task Technology Fi (Goodhue & Thompson, 1995), o e all us (Oli ei a e al., 2017), and cul u e (Ho s ede, 1984). 1.3. Resea ch ocus Mobile paymen se ices a e oday becoming mo e use ul and mo e p esen in people’s daily li es, especially now wi h he COVID-19 pandemic. Conduc ing ansac ions h ough a mobile phone is al eady a eali y in people’s li es. Unde s anding he main ac o s ha a ec he in en ion o con inue using m-paymen is he ocus o his disse a ion. IS ela ed a eas such as e-lea ning, in e ne banking, and e-comme ce a e no wi hin he scope o his wo k. The s udy add esses only he indi idual le el o con inuance in en ion. M-paymen is de ined as a paymen me hod in which a mobile phone is used o pe o m inancial alue exchanges (ini ia e, au ho ize, and con i m) any ime and anywhe e in e u n o goods and se ices (Kujala e al., 2017; Liébana-cabanillas e al., 2018; Shao e al., 2019). The e a e di e en ways o conduc a ansac ion using m-paymen . The simples way is sho -message-based, by which he use can make paymen s using a simple mobile phone (Singh e al., 2020; T. Zhou, 2013). 1.4. Main objec i es To unde s and he mos impo an ac o s o he in en ion o con inue using m-paymen , we di ide ou wo k in o di e en s udies, each p esen ed in sepa a e chap e s. The i s s udy (Chap e 2) add esses weigh and me a-analysis o ISCI. This s udy syn hesizes he esul s o p e ious s udies on ISCI, iden i ying he mos used signi ican ela ionships in he li e a u e, and he mos s udied egions and echnologies, hus con ibu ing o he s a e-o - he-a . 4 Doc o al P og amme in In o ma ion Managemen In Chap e 3 we analyse he impac o indi idual pe o mance o he pu pose o con inuing o use m-paymen . We in eg a e D&M ISSM and ECM in an A ican con ex . In Chap e 4 we analyse he impac o ask and echnology cha ac e is ics, o e all us , and he ole o sa is ac ion as a mode a o on con inuance in en ion o use m-paymen , conside ing ha us is an impo an ac o ha can in luence he usage o m-paymen . As inancial ansac ions a e sensi i e, i is impo an o ha e us in he sys em. In Chap e 5 he ole o cul u e in he ECM model is assessed, conside ing ha cul u e can be an impo an ac o in he adop ion and in en ion o con inue using m-paymen . Ou mo i a ion is o unde s and he impac o cul u e as a mode a o on he ECM model. In Chap e 6 we p esen a summa y o he s udies, hei implica ions, and ecommenda ions o u u e s udies. 1.5. Me hods Acco ding o he li e a u e, he e a e h ee main epis emological pe spec i es (posi i ism, in e p e i ism, and ealism). In his disse a ion we ollow he posi i ism pe spec i e (Smi h, 2006). Se e al me hods we e applied in his esea ch because we de eloped a new and di e en model based on exis ing heo ies o unde s and con inuance in en ion o use mobile paymen . The de elopmen o he heo ies was ollowed by es s in o de o explain he subjec . A se o hypo heses we e de eloped and empi ically es ed. 1.5.1. Theo e ical amewo ks The ECM (Bha ache jee, 2001) is used in all he empi ical s udies, om Chap e s 3 o 5. Chap e 3 is based on ECM combined wi h D&M ISSM. Chap e 4 is based on he in eg a ion o ECM, TTF, and o e all us , and Chap e 5 is based on ECM in eg a ed wi h he cul u al mode a o (Ho s ede, 1984). 5 Doc o al P og amme in In o ma ion Managemen 1.5.2. Quan i a i e esea ch me hods In all o he s udies we use a c oss-sec ion online su ey design o analyse he main ac o s o mobile paymen con inuance in en ion. The da a collec ion was conduc ed in Mozambique. In Chap e 3 we desc ibe he collec ion o 338 alid esponses, in Chap e 4 we examine a sample o 384, and in Chap e 5 apply a mixed-me hods app oach based on 384 alid esponses iangula ed wi h ield in e iews. The da a we e collec ed om July 2018 o Janua y 2019. S uc u e equa ion modelling (SEM) was used o empi ically es he esea ch models. We used PLS-SEM (pa ial leas squa es – s uc u al equa ion modelling) ia Sma PLS 3 so wa e (Ringle e al., 2015). 1.6. Pa h o esea ch This disse a ion epo s he collec ion o di e en in e ela ed s udies on he in en ion o con inue using m-paymen , p esen ed sepa a ely in di e en chap e s. Some o he s udies a e al eady published in in e na ional jou nals wi h a double-blinded e iew p ocess; o he s a e submi ed o publica ion and a e in di e en s ages o e iew and p epa a ion. The s age o each s udy is p esen ed in he Table 1.1. The majo conclusions o he s udies made om Chap e 2 o 5 a e p esen ed a he end o he disse a ion. Table 1.1 - Resea ch s udies s ages Chap e S udy name Jou nal Cu en s age 2 A Me a-analysis o he quan i a i e s udies in con inuance in en ion o use an in o ma ion sys em In e ne Resea ch Published 3 Unde s anding he ac o s o con inuance in en ion o use mobile paymen : Empi ical es o he D&M IS success wi h ECM in he A ican con ex Heliyon Published 4 Con inuance in en ion o mobile paymen : TTF model wi h T us in an A ican con ex Submi ed o a jou nal o qua ile one o Scimago index Unde e iew 6 Doc o al P og amme in In o ma ion Managemen 5 Role o he unce ain y a oidance cul u e on he expec a ion con i ma ion model: mobile paymen case Submi ed o a jou nal o qua ile one o Scimago index Submi ed 7 Doc o al P og amme in In o ma ion Managemen 8 Doc o al P og amme in In o ma ion Managemen Chap e 2 – A me a-analysis o he quan i a i e s udies in con inuance in en ion o use in o ma ion sys ems 2.1. In oduc ion The e olu ion o in o ma ion sys ems (IS) has o e ed he oppo uni y o di e en ypes o ins i u ions o ad ance he capabili y, e iciency, and esponsibili y o hei se ices and p oduc s, he eby s eamlining he day o day ac i i ies o hei cus ome s (Laukkanen 2007). I s use is g owing exponen ially, conside ing he use ulness o IS in socie y. Howe e , he adop ion o IS is no enough o keep hem in he ma ke , con inuous use is necessa y (Bha ache jee, 2001), o b ing a e u n on he in es men s ha companies make, and o help he use s in hei ac i i ies. Con inuous use o IS e e s o he decision o he use o con inue o use he IS. This beha iou is no ed a e he use has he i s expe ience wi h IS. Unde s anding wha ac o s in luence an indi idual o con inue o use IS has become necessa y o esea che s and companies (A Bha ache jee, 2001; Shaikh & Ka jaluo o, 2015). In ecen yea s, he numbe o s udies in con inuance in en ion o use an IS has g own ampan ly and now co e s se e al subjec s such as con inuance in en ion in mobile banking se ices, mobile paymen , e-lea ning, social ne wo king, heal h applica ions, e-go e nmen , mobile comme ce, among o he s. Conside ing ha he numbe o s udies is g owing, di e en echnologies, heo ies and con ex s a e being s udied, he e is plen y o sca e ed in o ma ion and di e en esul s. Wi h ha much in o ma ion in he backg ound, he p ocess o sea ching o s udies became mo e di icul , and he need o comp ehensi e and syn hesised in o ma ion abou IS con inuance in en ion became essen ial. The e o e, i is c ucial and necessa y o highligh , summa ise and cla i y he esul s o exis ing s udies in o de o p o ide a comp ehensi e pic u e o con inuing o use IS (Fe ke, 2006). This p ocess enables heo y de elopmen and e eals new ela ionships and gaps (Hama i & Ke onen, 2017). The e a e some li e a u e e iews on IS con inuous in en ion (e.g., Bha ache jee and Ba a , 2011; Shaikh and Ka jaluo o, 2015; Naba i e al., 2016), ha explo e di e en aspec s o p io s udies such as heo ies, echnologies, and used con ex s. Howe e , mos o hem a e na a i e and desc ip i e; none o hem has used me a-analysis. This s udy will use me a- and weigh 9 Doc o al P og amme in In o ma ion Managemen analysis o de i e o mo e empi ical esul s. The me a-analysis is a p ocess o summa ising, e alua ing, and analysing quan i a i e esea ch indings (L. Zhang e al., 2012), e en i he ou come is non-signi ican o inconsis en , i can con ibu es o a pooled conclusion, ein o cing he gene al alidi y o he in e p e a ions (Hama i & Ke onen, 2017; K. Wu e al., 2011). Acco ding o ea lie esea ch me a-analysis and weigh analysis a e conside ed app op ia e me hods o e iew empi ical da a (Bap is a & Oli ei a, 2016; Rana e al., 2015; Schmid & Hun e , 2016; Y. Zhao e al., 2018). We desc ibe he mos c i ical a iables in he ield, using indings epo ed in exis ing esea ch combined wi h weigh analysis o he cons uc s o iden i y he bes p edic o s (Y. Zhao e al., 2018) o highligh he bes p edic o s o con inuance in en ion o use an IS, imp o emen s o heo ies, and he s eng h o he a iables. Acco ding o ou knowledge, no esea ch add esses: (i) me a-analysis combined wi h weigh analysis in he con ex o con inuance in en ion o use an IS, o (ii) empo al analysis o unde s and he e olu ion o he heo e ical models o e ime. This s udy can ex apola e b oade heo e ical implica ions ela ing o he posi ioning and unde s anding o IS. Con ibu ing o he esea ch, we illus a e he mos used ela ionships, bes p edic o s, mos used echnologies du ing a pe iod, he e olu ion o he numbe o pape s pe yea , and he e olu ion o he heo e ical model. The o e all a iables o be used o p edic con inuance in en ion o use an IS we e illus a ed. Beyond syn hesising he main indings o he s udies, we also c ea ed models o unde s and he empo al e olu ion o he cons uc s be e . Addi ionally, we analysed possible mode a o s in he ela ionship be ween pe cei ed use ulness and sa is ac ion on con inuance in en ion o use IS. The a icle is o ganised as ollows: in Sec ion 2, is p esen ed he li e a u e e iew, Sec ion 3 desc ibes he esea ch me hodology; in he nex sec ion, we p esen he esul s o he esea ch ollowed by a discussion o he indings; he conclusion and u u e ecommenda ions ollow his sec ion. 16 Doc o al P og amme in In o ma ion Managemen 2.3.1. Mode a o analysis Conce ning mode a o s, we in es iga ed some mode a ing e ec s in di e en ela ionships. To selec he ela ionships, we used ela ionships ha ha e enough obse a ions (> 30) (Geyskens e al., 2009; Lipsey & Wilson, 2001). Rega ding he mode a o a iables, we ha e selec ed di e en mode a o s sugges ed by he li e a u e (Ho s ede & Minko , 2010; F. D. O. San ini e al., 2019). Fo ou s udy, we ha e used me hodological, economic, and cul u al mode a o s. This analysis is essen ial o he li e a u e because i p o ides esea che s a be e unde s anding o he e ec s o he ela ionships and p o ides an o e iew o he po en ial mode a o s ha can in luence he con inuance in en ion o use IS in u u e s udies. Appendix D p esen s he mode a o s, he desc ip ions, and he coding s uc u e. Sample size was analysed as a me hodological mode a o , conside ing ha i plays a signi ican ole in a ying he e ec sizes in he s udies (Fe n & Mon oe, 1996). While a small sample size is mo e homogeneous, his aspec ends o o e es ima e he e ec size o he ela ionships (Rosen hal & Rubin, 1982). We ha e also analysed mode a o s in he economic con ex , (1) economic de elopmen , and (2) inno a ion le el. Economic de elopmen can play an impo an ole as i ends o p omo e di e en le els o use o IS. The e o e, we expec ha de eloped economies in luence he beha iou o he in en ion o con inue using IS, compa ed o de eloping economies (Y. Kim & Pe e son, 2017). On ega ds inno a ion le el, i is conside ed a po en ial mode a o because i can in luence he ela ionships, aking in o conside a ion ha coun ies wi h a high le el o inno a ion end o con inue using IS as hey ha e good skills and amilia i y wi h sys ems usage (Y. Kim & Pe e son, 2017). Finally, we ha e analysed six cul u al mode a o s om Ho s ede, powe dis ance, indi idualism, masculini y, unce ain y a oidance, long e m o ien a ion, and indulgence. These cul u al dimensions a e ecognised as he leading indica o s o people's belie s and alues ha impac hei beha iou , so we conside po en ial mode a o s ha can in luence he in en ion o con inue using IS (Ho s ede & Minko , 2010). Ou analysis was suppo ed on a hie a chical linea me a-analysis. This analysis uses he mul i a ia e eg ession o ma o he a iables included in he model and is widely used in me a-analy ic esea ch (Geyskens e al., 2009; F. D. O. San ini e al., 2019). 17 Doc o al P og amme in In o ma ion Managemen 2.4. Findings 2.4.1. Me a-Analysis Table 2.1 ep esen s he me a-analysis and weigh analysis o he 60 ela ionships ha we e mos o en used, and which ha e occu ed h ee o mo e imes ac oss he 115 s udies. Columns 4 o 8 (me a-analysis in o ma ion) o Table 2.1 p esen he numbe o imes ha a ela ionship was analysed ( o al), he sum o samples (sample), an a e age o he co ela ion coe icien (AVG o cc), no mal s anda d de ia ion (Z – alue), and 95% con idence in e al. In addi ion, we hen show he ela ionship be ween he dependen cons uc s and he independen cons uc s. Se e al dependen cons uc s ela e o di e en independen cons uc s, such as con inuance in en ion, which ela es o 16 di e en independen cons uc s, ollowed by sa is ac ion, which applies o 14 di e en cons uc s and pe cei ed use ulness, which applies o 7 di e en cons uc s. The me a-analysis esul s e eal ha he co ela ion coe icien o 60 ela ionships is s a is ically signi ican (p < 0.01). The la ges Z- alues a e sa is ac ion on con inuance in en ion (75.695), con i ma ion on pe cei ed use ulness (55.921), con i ma ion on sa is ac ion (48.176), a i ude on con inuance in en ion (35.602), pe cei ed use ulness on con inuance in en ion (34.287), and pe cei ed use ulness on sa is ac ion (33.995). 2.4.2. Weigh analysis This me hod is used o es ima e he impo ance o a p edic o (i.e. independen cons uc ) and p edic s he s eng h o an independen cons uc (Jeya aj e al., 2006). The weigh s o he 60 mos used ela ionships we e examined and a e p esen ed in columns 9 o 12 in Table 2.1. The alue o weigh was compu ed by di iding he numbe o s a is ically signi ican ela ionships by he o al numbe o s udies used. When he weigh is one (1), i shows ha he ela ionship wi hin he a iables is signi ican in all he esea ch, bu i he weigh is ze o (0), i indica es ha he ela ionship is no signi ican h ough all he s udies examined (Jeya aj e al., 2006). 18 Doc o al P og amme in In o ma ion Managemen Table 2.1 - The mos equen ly used ela ionships o me a-analysis and weigh -analysis (O de ed by dependen cons uc s). Nº (1) Independen Cons uc s (2) Dependen Cons uc s (3) Me a-analysis Weigh analysis To al (4) ∑ Sample (5) AVG o cc (6) Z – alue (7) 95% con idence in e al (low - high) (8) Non- signi i can (9) Signi ican (10) To al (11) Weigh (Signi ican / To al) (12) 1 Hedonic Value A ec i e Commi men 5 1266 0.339 12.536 0.198 0.301 0 5 5 1.000 2 Rela ional Capi al 5 1266 0.258 9.381 0.206 0.309 0 5 5 1.000 3 U ili a ian Value 5 1266 0.112 3.997 0.057 0.166 1 4 5 0.800 4 Pe cei ed Ease o Use A i ude 5 1209 0.162 5.676 0.095 0.210 2 3 5 0.600 5 Pe cei ed Use ulness 13 4535 0.408 29.164 0.362 0.414 1 12 13 0.923 6 Sa is ac ion 6 2580 0.481 26.615 0.451 0.510 0 6 6 1.000 7 Se ice Quali y Con i ma ion 4 1244 0.339 12.434 0.282 0.394 0 4 4 1.000 8 Con inuance In en ion Con inuance Beha iou 5 1526 0.375 15.385 0.331 0.417 0 5 5 1.000 9 A ec i e Commi men Con inuance In en ion 5 1266 0.556 22.284 0.517 0.593 0 5 5 1.000 10 A i ude 14 5657 0.441 35.602 0.409 0.458 0 14 14 1.000 11 E o Expec ancy 3 1075 0.253 8.467 0.196 0.308 1 2 3 0.667 12 Flow 5 1243 0.358 13.207 0.308 0.406 1 4 5 0.800 13 Habi 5 1691 0.255 10.713 0.210 0.299 0 5 5 1.000 14 Hedonic Ou come Expec a ions 7 1951 0.437 20.679 0.435 0.567 0 7 7 1.000 15 In insic Mo i a ion 3 508 0.453 10.977 0.381 0.520 0 3 3 1.000 16 Pe cei ed Beha iou Con ol 7 3021 0.296 16.763 0.263 0.328 0 7 7 1.000 17 Pe cei ed Ease o Use 6 2280 0.074 3.538 0.013 0.104 2 4 6 0.667 18 Pe cei ed Enjoymen 16 5808 0.187 14.417 0.158 0.210 3 13 16 0.813 19 Pe cei ed Use ulness 41 13686 0.285 34.287 0.265 0.297 4 37 41 0.902 20 Pe o mance 5 3707 0.241 14.962 0.210 0.271 1 4 5 0.800 21 Sa is ac ion 74 29220 0.416 75.695 0.399 0.419 2 72 74 0.973 22 Subjec i e No m 11 4379 0.179 11.970 0.150 0.208 1 10 11 0.909 23 T us 7 2353 0.239 11.814 0.175 0.260 1 6 7 0.857 24 U ili a ian Value 9 498 0.242 5.493 0.157 0.323 0 9 9 1.000 25 Pe cei ed Use ulness Discon i ma ion 3 1149 0.133 4.529 0.076 0.189 0 3 3 1.000 26 Sa is ac ion Habi 3 755 0.456 13.499 0.397 0.511 0 3 3 1.000 27 Con i ma ion Pe cei ed Ease o Use 6 1511 0.458 19.214 0.471 0.548 0 6 6 1.000 28 Con ex 3 5121 0.233 16.981 0.207 0.259 0 3 3 1.000 29 Indi idualism 3 5121 0.300 22.143 0.275 0.325 0 3 3 1.000 30 Time pe cep ion 3 5121 0.177 12.797 0.150 0.203 0 3 3 1.000 31 Unce ain y A oidance 3 5121 -0.137 -9.863 -0.164 -0.110 0 3 3 1.000 32 Con i ma ion Pe cei ed Enjoymen 7 2145 0.622 33.705 0.607 0.663 0 7 7 1.000 33 Con ex 3 5121 0.157 11.325 0.130 0.184 0 3 3 1.000 34 Indi idualism 3 5121 0.223 16.226 0.197 0.249 0 3 3 1.000 35 Unce ain y A oidance 3 5121 -0.137 -9.863 -0.164 -0.110 1 2 3 0.667 36 Con ex Pe cei ed Mone a y Value 3 5121 0.200 14.504 0.174 0.226 0 3 3 1.000 37 Indi idualism 3 5121 0.323 23.965 0.298 0.347 0 3 3 1.000 38 Time Pe cep ion 3 5121 0.190 13.760 0.163 0.216 0 3 3 1.000 39 Unce ain y A oidance 3 5121 -0.233 -16.981 -0.259 -0.207 0 3 3 1.000 40 Con i ma ion Pe cei ed Use ulness 33 10168 0.504 55.921 0.492 0.522 0 33 33 1.000 41 Con ex 3 5121 0.223 16.226 0.197 0.249 0 3 3 1.000 42 Discon i ma ion 6 1825 0.555 26.703 0.593 0.652 0 6 6 1.000 43 Indi idualism 3 5121 0.287 21.125 0.262 0.312 0 3 3 1.000 44 Pe cei ed Ease o Use 12 3532 0.327 20.166 0.268 0.329 1 11 12 0.917 45 Time Pe cep ion 3 5121 0.177 12.797 0.150 0.203 0 3 3 1.000 46 Unce ain y A oidance 3 5121 -0.170 -12.281 -0.196 -0.143 0 3 3 1.000 47 Con i ma ion Sa is ac ion 35 10918 0.431 48.176 0.395 0.427 0 35 35 1.000 48 Discon i ma ion 9 2576 0.576 33.299 0.550 0.601 0 9 9 1.000 49 Hedonic Bene i 3 928 0.349 11.080 0.291 0.404 0 3 3 1.000 50 In o ma ion Quali y 5 1735 0.248 10.541 0.203 0.292 1 4 5 0.800 51 Pe cei ed Ease o Use 10 7112 0.188 16.042 0.148 0.194 0 10 10 1.000 52 Pe cei ed Enjoymen 12 8018 0.251 22.962 0.230 0.271 2 10 12 0.833 53 Pe cei ed Use ulness 46 18018 0.248 33.995 0.217 0.245 8 38 46 0.826 54 Pe cei ed Value 3 1158 0.203 6.996 0.147 0.258 0 3 3 1.000 55 Pe o mance 3 2579 0.390 20.901 0.357 0.422 0 3 3 1.000 56 Se ice Quali y 5 2608 0.296 15.574 0.058 0.151 2 3 5 0.600 57 Social Bene i 3 928 0.334 10.563 0.160 0.282 0 3 3 1.000 58 Sys em Quali y 7 2477 0.279 14.255 0.242 0.315 0 7 7 1.000 59 T us 3 880 0.313 9.591 0.252 0.371 0 3 3 1.000 60 U ili a ian Bene i 3 928 0.182 5.598 0.119 0.244 0 3 3 1.000 No e: The highligh ed ela ionships a e he bes p edic o s o he weigh analysis. AVG o cc = a e age o he co ela ion coe icien ; Z- alue = no mal s anda d de ia ion. 19 Doc o al P og amme in In o ma ion Managemen In o de o iden i y he mos e ec i e p edic o s o use IS con inuance in en ion, (Jeya aj e al., 2006) classi ied independen a iables in wo ways: he a iables ha we e e alua ed 5 ( i e) o mo e imes we e classi ied as well-u ilised, and he a iables e alua ed less han 5 ( i e) imes seen as expe imen al. Addi ional de ini ions o Jeya aj e al., (2006) ha e been aken in o conside a ion : bes p edic o s – a e he ela ionships ha we e classi ied as well-u ilised wi h he weigh g ea e han o equal o 0.8; and p omising p edic o s – ela ionships ha we e classi ied as expe imen al wi h he weigh equal o 1. The ou comes o he 60 ela ionships assessed in he weigh analysis show ha 34 we e classi ied as well-u ilised, and 31 as bes p edic o s (highligh ed ela ionships) o he con inuance in en ion o use an IS. Addi ionally, 24 ou o 26 expe imen al ela ionships we e classi ied as p omising p edic o s, equi ing mo e e alua ion o succeed as bes p edic o s. Fo u u e esea ch, we encou age esea che s o e alua e such p omising p edic o s. Howe e , in all he s udies, no ype o ela ionship was ound o be no signi ican . Acco ding o he indings o me a- and weigh analyses, he mos o en used dependen a iables we e con inuance in en ion, sa is ac ion, and pe cei ed use ulness. The e o e, he mos used independen a iables o explain con inuance in en ion (used mo e han en imes) we e sa is ac ion, pe cei ed use ulness, pe cei ed enjoymen , a i ude, and subjec i e no ms. 2.4.3. Mode a o analysis The analysis o po en ial mode a o s was pe o med using ac o s om he me hodological, economic, and cul u al con ex . Thus, hese ac o s we e es ed in he ela ionships ha p edic con inuance in en ion, and p esen a conside able numbe o obse a ions (a leas 30) (Geyskens e al., 2009; Lipsey & Wilson, 2001). Howe e , he ela ionships selec ed we e pe cei ed use ulness and sa is ac ion o con inuance in en ion (Schmid & Hun e , 2016). The esul s o he analysis a e p esen ed in Table 2.2. The mode a o sample size had a signi ican mode a ing e ec on he ela ionship o pe cei ed use ulness o con inuance in en ion ( β = 0.338, M_Low = 0.418, M_High = 20 Doc o al P og amme in In o ma ion Managemen 0.316, p < 0.05), and had no signi ican mode a ing e ec o he o he ela ionships o sa is ac ion o con inuance in en ion ( β = 0.502, M_Low = 0.461, M_High = 0.448). In he economic con ex , he mode a o economic de elopmen had no signi ican mode a ing e ec o pe cei ed use ulness o con inuance in en ion ( β = 0.467, M_Low = 0.364, M_High = 0.380), and sa is ac ion o con inuance in en ion ( β = 0.546, M_Low = 0.451, M_High = 0.466). Simila ly, he mode a o inno a ion le el had no signi ican mode a ing e ec o pe cei ed use ulness o con inuance in en ion ( β = 0.490, M_Low = 0.330, M_High = 0.414), and sa is ac ion o con inuance in en ion ( β = 0.491, M_Low = 0.492, M_High = 0.438). In he cul u al con ex , he powe dis ance mode a o had a signi ican mode a ing e ec o sa is ac ion o con inuance in en ion ( β = 0.455, M_Low = 0.495, M_High = 0.414, p < 0.05), and had no signi ican mode a ing e ec on he o he ela ionship o pe cei ed use ulness o con inuance in en ion ( β = 0.497, M_Low = 0.320, M_High = 0.445). The mode a o indi idualism had a signi ican mode a ing e ec o pe cei ed use ulness o con inuance in en ion ( β = 0.499, M_Low = 0.308, M_High = 0.420, p < 0.1), and had no signi ican mode a ing e ec on he o he ela ionship, sa is ac ion o con inuance in en ion ( β = 0.526, M_Low = 0.393, M_High = 0.449). The mode a o masculini y had a signi ican mode a ing e ec o sa is ac ion o con inuance in en ion ( β = 0.527, M_Low = 0.374, M_High = 0.465, p < 0.1) and had no signi ican mode a ing e ec on he o he ela ionship o pe cei ed use ulness o con inuance in en ion ( β = 0.511, M_Low = 0.336, M_High = 0.435). The mode a o unce ain y a oidance had a signi ican mode a ing e ec on pe cei ed use ulness o con inuance in en ion ( β = 0.361, M_Low = 0.441, M_High = 0.327, p < 0.1), and had no signi ican mode a ing e ec on he ela ionship sa is ac ion o con inuance in en ion ( β = 0.473, M_Low = 0.433, M_High = 0.430). Simila ly, he mode a o long e m o ien a ion had a signi ican mode a ing e ec on pe cei ed use ulness o con inuance in en ion ( β = 0.332, M_Low = 0.420, M_High = 0.308, p < 0.1), and had no signi ican mode a ing e ec on he ela ionship sa is ac ion o con inuance in en ion ( β = 0.540, M_Low = 0.453, M_High = 0.455). Finally, he mode a o indulgence had a signi ican mode a ing e ec on he ela ionship sa is ac ion o con inuance in en ion ( β = 0.584, M_Low = 0.425, M_High = 21 Doc o al P og amme in In o ma ion Managemen 0.484, p < 0.1), and had no signi ican mode a ing e ec on he ela ionship pe cei ed use ulness o con inuance in en ion ( β = 0.394, M_Low = 0.408, M_High = 0.343). Table 2.2 - Mode a ion’s analysis. Mode a o Le el Pe cei ed Use ulness o Con inuance In en ion Sa is ac ion o Con inuance In en ion ! R P_ alue ! R P_ alue Sample size In e cep .338 .001 .502 .001 High 1 .316 1 .448 Low .159 .418 .038** .049 .461 .409 Economic de elopmen In e cep .467 .001 .546 .001 High 1 .380 1 .466 Low -.067 .364 .454 -.026 .451 .712 Inno a ion Le el In e cep .490 .001 .491 .001 High 1 .414 1 438 Low -.125 .330 .163 .094 .492 .113 Powe Dis ance In e cep .497 .001 .455 .001 High 1 .445 1 .414 Low -.125 .320 .158 .143 .495 .015** Indi idualism In e cep .499 .001 .526 .001 High 1 .420 1 .449 Low -.166 .308 .055* .015 .393 .836 Masculini y In e cep .511 .001 .527 .001 High 1 .435 1 .465 Low -.138 .336 .153 -.118 .374 .060* Unce ain y a oidance In e cep .361 .001 .473 .001 High 1 .327 1 .430 Low .159 .441 .096* .018 .433 .772 Long Te m o ien a ion In e cep .332 .001 .540 .001 High 1 .308 1 .455 Low 166 .420 .055* -.020 .453 .769 Indulgence In e cep .394 .001 .584 .001 High 1 .343 1 .484 Low .063 .408 .431 -.115 .425 .053* No e: *** p < 0.01; ** p < 0.05; * p < 0.10 2.5. Discussion Conside ing he numbe o s udies on con inuance in en ion o use an IS using heo ies o models, i becomes signi ican and sui able o analyse and discuss hei collec i e indings. We can e i y ha he a iables and ela ionships used a e qui e dispe sed, as s udies a e analysing di e en IS echnologies, s udies om sepa a e imes, and di e en geog aphical spaces wi h dis inc cul u es. The esul s e eal ha he me a- and weigh analyses o he independen a iables on equi alen dependen a iables a e close . The highe he weigh o an independen a iable, he g ea e is he p obabili y ha i is signi ican in pe o ming he me a- analysis (Rana e al., 2015). In he me a-analysis, all he 31 bes p edic o s, and all he 22 Doc o al P og amme in In o ma ion Managemen 24 p omising p edic o s we e ound o be s a is ically signi ican . The emaining h ee well-u ilised ela ionships, namely se ice quali y on sa is ac ion, pe cei ed ease o use on a i ude, and pe cei ed ease o use on con inuance in en ion, we e also s a is ically signi ican . The esul s e eal ha he mos impo an ela ionships o p edic con inuance in en ion o use an IS a e simul aneously s a is ically signi ican in me a- analysis and bes p edic o s in weigh analysis. Addi ionally, he wid h o he con idence in e al depends on he co ec ness o indi idual s udies along wi h he numbe o he cumula i e s udies (Rana e al., 2015). We can e i y ha all o he bes p edic o and p omising p edic o ela ionships ob ained a na ow in e al, p o iding con idence o he le el o a iance o he co ela ion alues. Acco ding o he esul s, h ee heo e ical models (Figu e 2.3) we e designed o suppo u u e s udies on con inuance in en ion o use an IS. The i s model (A) was c ea ed using all he da a om ou analysis (gene al model), and hen, o unde s and he model's e olu ion, he da a we e di ided in o wo g oups ( om 2001 o 2010, and om 2011 o 2017), and wi h his in o ma ion, wo mo e models we e c ea ed. The second model (B) was c ea ed using da a om 2001 o 2010, and he las model (C) was c ea ed using da a om 2011 o 2017. To gene a e he heo e ical models, i s , signi ican ela ionships om he me a-analysis we e selec ed, second, he bes p edic o s o weigh analysis we e selec ed, and inally, di ec o indi ec a iables ela ed o con inuance in en ion o use an IS we e selec ed. Mo eo e , he ela ionships such as se ice quali y on sa is ac ion, pe cei ed ease o use on a i ude, and pe cei ed ease o use on con inuance in en ion we e e alua ed i e o mo e imes and wi h weigh less han 0.8. Basically, because hey we e s a is ically non-signi ican in indi idual s udies, u u e esea ch is needed o p o e o disapp o e he exis ing end (Jeya aj e al., 2006). Acco ding o he indings, i is possible o unde s and ha he e is an e olu ion o he models, jus by compa ing he o iginal model o IS con inuance in en ion (A Bha ache jee, 2001) wi h ou gene al model. The gene al model is mo e complex and p esen s mo e ela ionships wi h a signi ican se o cons uc s. Going o model B, i is simple han he gene al model (A). This ace means ha , in ha pe iod (2001 o 2010), he key ac o s ha in luenced use s o con inue using IS we e: sa is ac ion, pe cei ed 23 Doc o al P og amme in In o ma ion Managemen use ulness, and a i ude. Con inuance in en ion was used o p edic con inuance beha iou . In model B, Con i ma ion only explains sa is ac ion and pe cei ed use ulness o e ime, and i s a s explaining pe cei ed ease o use and pe cei ed enjoymen . This phenomenon also happened wi h pe cei ed ease o use. I only explains sa is ac ion, o e ime, and explains pe cei ed use ulness. In model C, which is mo e complex han model B and e y simila o he gene al model (A), we can a gue ha in ha pe iod (2011 o 2017), (i) he ype o IS inc eased conside ably, hen, o suppo his phenomenon, (ii) he numbe o cons uc s also inc eased. In model B, we ha e discon i ma ion explaining sa is ac ion and pe cei ed use ulness, bu we do no ha e hese ela ionships in model C. This means ha hese ela ionships ha e been so well-explo ed ha hey became pa o a collec i e body o knowledge. In model B, we ha e pe cei ed ease o use explaining sa is ac ion; o e ime, he same cons uc s a ed explaining pe cei ed use ulness and began o be explained by con i ma ion. In model C, we do no ha e con inuance in en ion explaining con inuance beha iou . F om 2011 o 2017, i is possible o e i y ha he independen cons uc s inc eased signi ican ly. Some cons uc s belong o o he heo e ical models, o example, a i ude, subjec i e no ms, and pe cei ed beha iou con ol (Ajzen, 1985), sys em quali y and in o ma ion quali y (DeLone & McLean, 1992) pe cei ed ease o use (Da is, 1989). Two cons uc s do no appea in ei he model B o C bu appea in he gene al model (A), such as pe o mance and pe cei ed beha iou con ol. This elemen means ha hese ela ionships ha e been explo ed o e he whole ange o ou analysis (2001 - 2017). Mo eo e , when he da ase was di ided, hey did no become s a is ically signi ican in bo h g oups. Howe e , o e ime, he heo e ical models g ow mo e complex, and new cons uc s appea , such as pe cei ed enjoymen , us , low, subjec no ms, and pe o mance, e c. Ne e heless, some cons uc s a e in decadence because hey u ned in o common knowledge, and some cons uc s like sa is ac ion, pe cei ed use ulness, con i ma ion, pe cei ed ease o use, and a i ude emain o e ime. This cha ac e is ic means ha hese cons uc s a e s ill essen ial o p edic IS con inuance in en ion. Now we can a gue ha we ha e a powe ul model (A) o p edic con inuance in en ion o use IS, using 24 Doc o al P og amme in In o ma ion Managemen dis inc ypes o echnologies, in di e en en i onmen s. In addi ion, he heo e ical models had a signi ican e olu ion. The dashed ela ionships mean ha hey we e no s a is ically signi ican and bes p edic o s in ha pe iod. Mo eo e , he pe o mance and pe cei ed beha iou con ol on con inuance in en ion we e s a is ically signi ican and bes p edic o s only wi h all he da ase (model A). 25 Doc o al P og amme in In o ma ion Managemen Figu e 2.3 - Theo e ical models based on me a- and weigh analysis (*** p < 0.01; ** p < 0.05; * p < 0.10). No e: The dashed ela ionships ep esen ed abo e we e no s a is ically signi ican and bes p edic o s in he pe iod o analysis. 32 Doc o al P og amme in In o ma ion Managemen The esul s unde sco e ha small sample size, high le els o indi idualism, low le els o unce ain y a oidance, and low le els o long- e m o ien a ion posi i ely mode a e he ela ionship be ween pe cei ed use ulness and con inuance in en ion. Rega ding indi idualism, manage s should ocus on he ask and au onomy o he indi idual. Indi idual managemen is mo e impo an han g oup managemen (Ho s ede & Minko , 2010). Fo unce ain y a oidance, he esul s a e impo an o manage s o p omo e he usabili y o he IS, conside ing ha hey a e mo e ecep i e o new echnology (Bap is a & Oli ei a, 2015). Fo long e m o ien a ion, he esul s a e pi o al o manage s because people wi h a low le el o long- e m o ien a ion ollow ins uc ions, and espec ules and adi ions. Howe e , he esul s de e mine ha a low le el o powe dis ance, a high le el o masculini y, and a high le el o indulgence posi i ely mode a e he ela ionship be ween sa is ac ion and con inuance in en ion. Conce ning powe dis ance, independence is a ca dinal cha ac e is ic, and powe should be dis ibu ed unequally (Bap is a & Oli ei a, 2015; Ho s ede & Minko , 2010). Fo masculini y, manage s should p omo e compe i ions and aining o mo i a e people. Fo indulgence, impulses and desi es a e mo e impo an . Manage s should sa is y people acco ding o hei desi es (Ho s ede & Minko , 2010). 2.5.4. Limi a ions and ecommenda ion The esea ch con ains se e al limi a ions. Fi s , we did no include ce ain s udies because o he una ailabili y o hei quan i a i e da a, o because hey we e quali a i e. Fo example, some s udies did no epo he e ec size when he ela ionships we e s a is ically non-signi ican . In eg a ing hese s udies could gene a e ele an in o ma ion in e ms o he signi icance o he cons uc s. Secondly, in he p e ious esea ch i he da a a e biased on IS con inuance in en ion, hen he e ec o he mean p esen ed h ough me a-analysis will also e lec his bias, and in his s udy, he handling publica ion biases a e missing. Mo eo e , we used me a-essen ials, a me a- analysis ool. This ool has limi a ions, conside ing ha me a-essen ials is no able o pe o m mo e ad anced analyses, like linea model o s uc u ed equa ion model (Rhee e al., 2015), u u e esea ch should conside a mo e ad anced ool o p o ide mo e insigh s and a di e en app oach o esea ch. 33 Doc o al P og amme in In o ma ion Managemen Fu u e esea ch should conside ela ionships like se ice quali y on con i ma ion, hedonic ou come expec a ion, and in insic mo i a ion on con inuance in en ion, sa is ac ion on habi , us , hedonic bene i s, and pe o mance on sa is ac ion as hey we e classi ied as p omising p edic o s and we e s a is ically signi ican wi h a high a e age co ela ion coe icien . Cul u e has an impo an impac on IS con inuance in en ion, and u u e esea ch should inco po a e a cul u e a iable such as subjec i e no ms, habi , o a i ude, o p o ide addi ional insigh s. Conside ing ha some IS such as mobile paymen s and gami ica ion a e g owing exponen ially, u u e esea ch could con empla e a me a-analysis o syn hesise and p o ide mo e indings in hese subjec s. 2.6. Conclusions The goal o ou esea ch was o collec and analyse di e en s udies on con inuance in en ion o use IS, by using me a-analysis combined wi h weigh analysis. The sys ema ic e iew o exis ing s udies comp ised 115 pape s, which cons i u ed he basis o he analysis o ou esea ch. The mos used echnology was e-lea ning, ollowed by social ne wo k se ices. In he p ocess o collec ing da a o ou analysis, we ound mo e han 600 ela ionships and selec ed he ela ionships ha had been examined h ee o mo e imes, o alling 60 di e en ela ionships. 34 ou o he 60 mos used ela ionships ha e been examined i e o mo e imes. We iden i ied he mos used a iables in he li e a u e and highligh ed hei ele ance, con ibu ing o he cu en s a e o he a (Hama i & Ke onen, 2017; K. Wu e al., 2011; Y. Zhao e al., 2018). Conce ning mode a ion analysis, we p esen ed no ewo hy esul s. The ela ionship pe cei ed use ulness o con inuance in en ion was mode a ed wi h sample size, indi idualism, unce ain y a oidance and long e m o ien a ion. Fu he mo e, he ela ionship sa is ac ion o con inuance in en ion was mode a ed wi h powe dis ance, masculini y and indulgence. Acco ding o ou empi ical esul , he e olu ion o he heo e ical models was p esen ed, using he bes p edic o s and he signi ican ela ionships o p edic con inuance in en ion o use an IS. This s udy wo ks as a e e ence basis o u u e esea ch ha seek o de elop he a ea o con inuance in en ion o use any ype o IS. Mo eo e , among he mos used cons uc s, we ound cons uc s 34 Doc o al P og amme in In o ma ion Managemen om he expec a ion con i ma ion model (A Bha ache jee, 2001), echnology accep ance model (Da is, 1989), he heo y o planned beha iou (Ajzen, 1985), and he DeLone and McLean IS success model (DeLone & McLean, 1992). Acco ding o Jeya aj e al., (2006) c i e ia, 31 ela ionships we e classi ied as bes p edic o s (examined i e o mo e imes, and weigh ≥ 0.8), and 24 ela ionships we e classi ied as p omising p edic o s (examined ewe han i e imes, and weigh = 1), needing mo e es s o quali y as he bes p edic o s. 35 Doc o al P og amme in In o ma ion Managemen 36 Doc o al P og amme in In o ma ion Managemen Chap e 3 – Unde s anding he ac o s o mobile paymen con inuance in en ion: empi ical es in an A ican con ex 3.1. In oduc ion In ecen yea s he numbe o mobile phone use s has been g owing exponen ially, mo i a ing companies o deli e se ices ia mobile phones (Ka jaluo o e al., 2019; Pe saud and Azha , 2012; J. Wu e al., 2017). Mobile paymen (m-paymen ) is one o he many se ices ha can be used ia mobile phones. M-paymen is a imbu semen me hod ha uses mobile phones o make inancial ansac ions such as paying o goods o se ices, ans e ing money, and wi hd awing money (Fan e al., 2018a; T. Zhou, 2013). M-paymen echnology was a dis up i e e olu ion ha a ec ed paymen ecosys ems, o igina ing in he Uni ed S a es and sp eading h oughou he wo ld (Fan e al., 2018a). In A ica, m-paymen was launched in Kenya, and was quickly adop ed in o he coun ies. Mozambique is one o he A ican coun ies ha adop ed m- paymen , helping u al people who do no ha e banking in as uc u e nea hem (Ba is a and Vicen e, 2018). When we compa e he use ulness o m-paymen in de eloped and de eloping economies, i appea s ha he impac on people's li es is mos no iceable in de eloping economies, conside ing ha inancial se ices do no each mos o he popula ion ye , and mos people a el long dis ances o access hem (Asamoah e al., 2020; Humbani and Wiese, 2018; Iman, 2018). M-paymen has a majo impac on hese communi ies because i p o ides basic inancial se ices such as money ans e , paymen o goods and se ices, and/o wi hd awing money, he eby imp o ing people's li es (Iman, 2018; Rahman e al., 2020). Much esea ch has been unde aken o unde s and m-paymen in di e en con ex s (e.g., Lu e al. (2017) in China; Lin e al. (2017) in he Uni ed S a es; Sinha e al. (2019) in India; Oli ei a e al. (2016) in Po ugal). Howe e , ew s udies ha e been conduc ed in he A ican con ex (Chen and Li, 2017; Lin e al., 2017). P e ious li e a u e has used di e en heo e ical models o unde s and con inuance in en ion o use m-paymen . Shao e al. (2019) used us and inno a ion di usion heo y; Lu e al. (2017) used expec a ion-con i ma ion heo y, mobili y, p i acy p o ec ion, social in luence, and cul u al alues; Chen and Li (2017) used IT con inuance, isk- us , and a ec - 37 Doc o al P og amme in In o ma ion Managemen cogni ion li e a u e. Wi h he excep ion o wo s udies ha in eg a ed quali y ac o s o unde s and con inuance in en ion o m-paymen (T. Zhou, 2013, 2014b), ou s udy shows how impo an i is o combine DeLone and McLean in o ma ion sys em (D&M IS) success model, and expec a ion-con i ma ion model (ECM). Each model has s eng hs and weaknesses, and hese a e o se and complemen ed by combining hese wo models. Despi e he bene i s o m-paymen , he e a e s ill ba ie s o he con inuance use. Many use s emain conce ned abou indi idual pe o mance and he quali y o se ices, since he m-paymen se ice in ol es ansac ion in o ma ion ha a ec s use p i acy. I is impo an ha he use s eel con iden abou m-paymen , ealize ha he se ice is o quali y, ha i con ains use ul in o ma ion and ha hey eel he need o use mo e and mo e. Ea lie s udies add essed simila issues (Fan e al., 2018a; Shao e al., 2019; T. Zhou, 2011b, 2013), bu did no in eg a e a model ha can explain di e en quali ies o m- paymen , sa is ac ion, and pe cei ed indi idual pe o mance o unde s and con inuance in en ion o m-paymen . Conside ing he impac o m-paymen in he A ican con ex , due o he lack o access o echnology in he same p opo ion, a se ice ha can be used any ime and anywhe e, ega dless o he educa ion and economic le el, educing he need o use banks is o g ea impo ance (Pal e al., 2020). The con ex can challenge he heo e ical models o explain m-paymen . Based on hese easons, we app oached he esea ch ques ion (RQ): How do he indi idual pe o mance d i e s in luence he con inuance in en ion o use m–paymen in an A ican con ex ? In his sense, i is in ou in e es o unde s and he e ec s o indi idual pe o mance d i e s combined wi h he ECM on m-paymen . We joined wo well-es ablished models, he D&M IS success model (DeLone and Mclean, 2003) and ECM (A Bha ache jee, 2001) o in es iga e con inuance in en ion o use m-paymen and gain a holis ic iew o he quali y o se ice and indi idual pe o mance on con inuance in en ion. The cu en esea ch con ibu es o he li e a u e i s ly by combining he D&M IS success model wi h he ECM model wi h he aim o imp o ing he unde s anding o con inuance in en ion o use m–paymen , iden i ying impo an de e minan s. As pe p e ious esea ch, his would be he i s s udy o combine all he ac o s o D&M IS success model and ECM model wi h he pu pose o unde s and con inuance in en ion 38 Doc o al P og amme in In o ma ion Managemen o use m–paymen . Secondly, conside ing ha he A ican ma ke is de eloping, his esea ch will bene i people and companies ha a e de eloping IT ela ed o m-paymen by iden i ying he mos impo an ac o s ha can lead o end-use ’s long- e m usage. Thi dly, by add essing he ac o s o indi idual’s con inuance in en ion o use m- paymen , he s udy deepens knowledge, abou wha is impo an o he long– e m usage o an IS (A Bha ache jee, 2001). The nex sec ion p esen s he bibliog aphic e iew. Sec ion 3 ou lines he hypo heses and he esea ch model. Sec ion 4 desc ibes he esea ch me hodology. Da a analyses and esul s o esea ch a e p esen ed in Sec ion 5. Finally, he discussion and conclusions a e de ailed in Sec ions 6 and 7, espec i ely. 3.2. Li e a u e e iew 3.2.1. Mobile Paymen M-paymen is a ype o paymen ha can be pe o med by mobile de ices (such as mobile phones, sma phones, e c.), o pay o goods, se ices, and bills. They use wi eless echnologies (mobile phone ne wo ks, NFI, Blue oo h, RFID, e c.) o pe o m ansac ions (Kau e al., 2020; Kujala e al., 2017; Liébana-cabanillas e al., 2014, 2018; F. Liébana-Cabanillas and La a-Rubio, 2017; J. K. Pa k e al., 2019). O he au ho s e e o m-paymen as a se ice o ca y ou paymen , check balances, and ans e money in a simple way, any ime and anywhe e (Pal e al., 2020; T. Zhou, 2013, 2014b). The e a e di e en ways o conduc a ansac ion using m-paymen , he simples is based on using sho -messages wi h a simple mobile phone whe eby he use can check balances o conduc paymen s using sho messages (Luna e al., 2019; Zhou, 2013). Ano he me hod is by using NFC (nea ield communica ion), communica ion is es ablished by he p oximi y o wo de ices, and he ansac ion is made (de Luna e al., 2019; Kujala e al., 2017; F ancisco Liébana-Cabanillas e al., 2019). The mos sophis ica ed means is by using a mobile applica ion (app), he use downloads he app, ins alls i on a sma phone, and egis e s o s a using he app (Singh e al., 2020; Ve kijika, 2020). 39 Doc o al P og amme in In o ma ion Managemen M-paymen con inuance in en ion was s udied by Lu e al. (2017), who applied mobili y, p i acy p o ec ion, and social in luence, and concluded ha pos -usage p i acy p o ec ion and social in luence belie impac use s’ in en ions o con inue using m-paymen . Yu e al. (2018) applied us ans e heo y, pe cei ed simila i y, and en i a i i y, and de e mined ha sa is ac ion is an impo an p edic o in luencing m- paymen con inuance in en ion, and ha he us ans e p ocess posi i ely in luences con inuance in en ion h ough sa is ac ion. Zhou (2013) used in o ma ion sys ems success and low heo y, and posi ed ha low, sa is ac ion, and us in luence con inuance in en ion; and m-paymen p o ide s need o o e a quali y sys em, in o ma ion, and se ice o gua an ee long- e m usage. Chen and Li (2017) applied IT con inuance, isk- us , and a ec -cogni ion, and concluded ha sa is ac ion and pos adop ion pe cei ed use ulness posi i ely in luence con inuance in en ion o m- paymen . In gene al, mos p e ious s udies ocused on us and sa is ac ion o he m- paymen con inuance in en ion (J. Lu, Wei, e al., 2017; Tam e al., 2020; T. Zhou, 2012, 2014b), bu quali y and pe o mance also play an impo an ole in m-paymen con inuance usage. 3.2.2. Mobile paymen in an A ican con ex In an A ican con ex , elecommunica ion companies (Sa a icom) launched m-paymen (M-Pesa) in Kenya in 2007 (Jack and Su i, 2011; Wenne e al., 2018). Kenyans consolida ed he use ulness o his echnology (Omigie e al., 2017; Uwama iya and Loebbecke, 2020). The echnology hen g ew exponen ially and sp ead ac oss he con inen , and is now being used in mo e han i e A ican coun ies and has mo e han 29 million ac i e use s (Voda one G oup, 2016; Wenne e al., 2018). M-paymen was adop ed in Mozambique and g ew apidly, conside ing ha his se ice is an al e na i e o bank-based sys ems o he popula ion o access inancial se ices. Using simple sho messages, use s can pe o m a ansac ion o ans e money, and pay o goods and se ices. This se ice was in oduced by he Mozambican Telecommunica ion Company MCel (mKesh) and by Vodacom (M-Pesa) (O igão e al., 2015). Ba is a and Vicen e (2018) ound ha m-paymen is e y impo an in u al a eas o inc ease inancial inclusion. Jack and Su i (2011) a gued ha m-paymen 40 Doc o al P og amme in In o ma ion Managemen sp ead e y quickly because i is an al e na i e banking se ice and has subs an ial impac on people in low economic condi ions. Tobbin and Kuwo nu (2011) a gued ha pe cei ed ease o use and pe cei ed use ulness a e he mos impo an ac o s o beha iou al in en ion o use m-paymen in Ghana. Humbani and Wiese (2018) show ha con enience and compa ibili y posi i ely in luence he adop ion o m-paymen . Addi ionally, hey demons a ed ha only gende could mode a e he ela ionship be ween con enience and he adop ion o m-paymen . 3.2.3. Theo e ical models 3.2.3.1. DeLone and McLean IS success model The DeLone and McLean IS success model has been b oadly used o explain indi idual and o ganiza ional pe o mance (DeLone and McLean, 1992). Howe e , in his s udy, he ocus is on he indi idual le el. The D&M IS model explains ha (1) bo h quali y sys em and in o ma ion quali y signi ican ly in luence he use o IS and use sa is ac ion, (2) he use o IS in luences he use ’s sa is ac ion and ice e sa, (3) bo h use and sa is ac ion signi ican ly in luence indi idual pe o mance, and (4) indi idual pe o mance signi ican ly in luences o ganiza ional impac . Se e al s udies con i m ha his model is powe ul o explain indi idual pe o mance and can be used wi h o he models o a iables (Baabdullah e al., 2019; Sha ma and Sha ma, 2019). Tam and Oli ei a (2016) employed i o unde s and he impac o mobile banking; Hsu e al. (2014) used he model o explain he epu chase in en ion on online g oup-buying; Wang (2008) applied i o explain he impac o e-comme ce sys em success. A e en yea s DeLone and Mclean (2003) e iewed se e al pape s ha alida e, challenge, and p opose imp o emen s o he o iginal model, and p oposed an upda ed model. In he upda ed model hey include he ac ha se ice quali y signi ican ly in luences he use o he sys em and sa is ac ion. They ealized ha wi h he g ow h o IS, use s s a ed paying a en ion o he quali y o se ices (Tam and Oli ei a, 2016). Conce ning he impac , in he upda ed model, hey ealized ha o he s udies ha e p oposed se e al ypes o impac s and decided o join all he impac s in o a single 41 Doc o al P og amme in In o ma ion Managemen impac called ne bene i s. Conside ing ha he p oposed model is based on he indi idual le el, he indi idual pe o mance will also be used. 3.2.3.2. In o ma ion Sys em Con inuance Model The e a e se e al heo ies used in s udies ela ed o in o ma ion sys ems, such as he uni ied heo y o accep ance and usage o echnology (UTAUT), he echnology accep ance model (TAM), and he ask- echnology i (TTF). Ou in e es is based on he pos -adop ion heo e ical model o IS, a he indi idual le el. The e is a di e ence be ween adop ion and con inuance in en ion o IS. Adop ion e e s o ac o s ha explain why an indi idual adop s o ejec s a echnology (Humbani and Wiese, 2019; S aub, 2009). A his s age, he use s ha e hei i s con ac wi h he echnology, and depending on hei expe ience, hey decide whe he o use i o o ejec i . Con inuance in en ion e e s o ac o s ha explain why an indi idual uses a echnology o a long ime, hus con ibu ing o he con inued use o he echnology (F anque e al., 2020; X. Lin e al., 2017). I in ol es unde s anding he long- e m ac o s ha con ibu e o he success o he IS (A Bha ache jee, 2001; X. Lin e al., 2017). Despi e he me i o p e ious s udies ha applied adop ion heo ies such as UTAUT and TAM o explain con inuance in en ion (Hadji and Degoule , 2016; Joo e al., 2016; Wu and Chen, 2017), he applica ion o hese models may su e om some limi a ions, leading o misunde s andings and misapplica ions o hese heo ies (Bha ache jee and Ba a , 2011; F anque e al., 2020; Naba i e al., 2016). The ECM model was based on he expec a ion-con i ma ion heo y o Oli e (1986). The model explains ha (1) con inuance in en ion o use IS was s ongly an icipa ed by use sa is ac ion, ollowed by use s’ pe cei ed use ulness o he sys em, (2) use sa is ac ion was p edic ed by use s’ con i ma ion o pe cei ed use ulness and expec a ion, and (3) use con i ma ion o expec a ion was a signi ican p edic o o use s’ pe cei ed use ulness. The model was ex ensi ely es ed in IS esea ch and con i med o be a good model o explain con inuance in en ion (Ca illo e al., 2017; Ryu, 2018; Talwa e al., 2020; Wang e al., 2019; Zheng, 2019). Use s’ sa is ac ion is he bes ac o o imp o e he con inuance 48 Doc o al P og amme in In o ma ion Managemen 3 Se ice Quali y (SERQ) SERQ1. The esponsible se ice pe sonnel a e always highly willing o help whene e I need suppo wi h he M-Paymen . SERQ2. The esponsible se ice pe sonnel p o ide pe sonal a en ion when I expe ience p oblems wi h he M-Paymen . SERQ3. The esponsible se ice pe sonnel p o ide se ices ela ed o M-Paymen a he p omised ime. SERQ4. The esponsible se ice pe sonnel ha e enough knowledge o answe my ques ions wi h espec o M-Paymen . (Tam & Oli ei a, 2016) 4 Use (U) U1. I use M-Paymen . U2. I use M-Paymen o buy p oduc s and se ices. U3. I use M-Paymen o make ans e s. U4. I use M-Paymen o wi hd aw money. (Venka esh, 2003) 5 Sa is ac ion (S) S1. I am e y pleased o use M-Paymen . S2. I am e y happy wi h M-Paymen . S3. I am deligh ed wi h M-Paymen . (A Bha ache jee, 2001) 6 Con i ma ion (C) C1. My expe ience wi h using M-Paymen was be e han I expec ed. C2. The se ice le el p o ided by M-Paymen was be e han I expec ed. C3. O e all, mos o my expec a ions om using M-Paymen we e con i med. (A Bha ache jee, 2001) 7 Pe cei ed Use ulness (PU) PU1. Using M-Paymen imp o es my pe o mance. PU2. Using M-Paymen inc eases my p oduc i i y. PU3. Using M-Paymen enhances my e ec i eness. PU4. I ind M-Paymen o be use ul o my wo k. (A Bha ache jee, 2001) 8 Indi idual Pe o mance (IP) IP1: M-Paymen enables me o accomplish asks mo e quickly. IP2: M-Paymen makes i easie o accomplish asks. IP3: M-Paymen is use ul o my job. (Tam & Oli ei a, 2016) 9 Con inuance In en ion (CI) CI1. I in end o con inue using M-Paymen a he han discon inue i s use. CI2. My in en ions a e o con inue using M-Paymen a he han manual p ocessing o o he al e na i e means. CI3. I plan o con inue using M-Paymen in my job. (A Bha ache jee, 2001) 3.4.2. Da a Da a we e collec ed om June 2018 o Oc obe 2018. The i ems we e assessed on a se en-poin scale, anging om one ( o ally disag ee) o se en ( o ally ag ee). The ques ion was c ea ed and managed in English and e ised o con en alidi y by a language expe . Ne e heless, a p o essional ansla o ansla ed he ques ionnai e in o Po uguese o adjus i o he Mozambican con ex . The ques ionnai e was e e se ansla ed o English by a di e en ansla o o ensu e equi alence (B islin, 1970). To alida e he ins umen s, a pilo es was conduc ed on a g oup o 40 s uden s, who we e excluded om he main sample. Gi en ha he goal o his s udy is o in es iga e con inuance in en ion o use m-paymen , he a ge esponden s should ha e expe ience in using m-paymen . To ensu e his, he alid esponden s in he Mozambican con ex we e con ined o M-Pesa and Mkesh use s. The su ey was sen o he esponden s, p o iding a hype link o he ques ionnai e. We ecei ed 338 alid esponses by he end o Oc obe 2018 om he 900 e-mails sen , which co esponds o a 37.5% esponse a e. We es ed he sample dis ibu ed o he i s and second esponden g oups using 49 Doc o al P og amme in In o ma ion Managemen he Kolmogo o -Smi no (K-S) es and con i med ha hey do no di e s a is ically (Ryans, 1974), showing ha non- esponse bias was no p esen . The common me hod bias was also examined using Ha man’s es (Podsako e al., 2003) con i ming no signi ican common me hod bias in he da a. The cha ac e is ics o he sample a e shown in Table 3.2; 60% o he esponden s we e men, 42% o he esponden s had used m-paymen one (1) o ou (4) imes du ing he las 3 mon hs. Table 3.2 - Sample cha ac e is ics. Age < 25 119 35% 25 - 30 96 28% 31 - 40 76 22% 41 - 50 38 11% > 50 9 3% Gende Female 134 40% Male 204 60% Educa ion High school o below 83 25% Bachelo ’s deg ee 151 45% Mas e ’s deg ee o highe 104 31% Employmen S uden s 95 28% Wo king p o essionals 203 60% Re i ed 1 0% Unemployed 39 12% Ma i al s a us Single 162 48% Ma ied 74 22% Di o ced 23 7% Widowed 10 3% Common-law ma iage (cohabi a ion) 67 20% Do no know answe s 2 1% M-paymen usage equency ( ime / 3 mon hs) 1 - 4 143 42% 5 - 10 92 27% > 10 103 30% 50 Doc o al P og amme in In o ma ion Managemen 3.5. Da a analysis and esul s S uc u al equa ion modeling (SEM) wi h pa ial leas squa e (PLS) was used o es and assess he alidi y o he heo e ical model. P e ious esea ch has ecognized he po en ial o SEM o measu ing s uc u al models (Al aimi e al., 2015; Tam and Oli ei a, 2016). SEM is a se o s a is ical models used o assess he alidi y o heo ies wi h empi ical da a (Ringle e al., 2005). Addi ionally, he Kolmogo o –Smi no (K- S es ) was implemen ed, as i is used in cases whe e da a a e no usually dis ibu ed, and he esea ch model is complex and has no been es ed in he li e a u e. Thus, PLS is he app op ia e me hod o his esea ch. Sma PLS 3 so wa e (Ringle e al., 2015) was used o analyse he heo e ical model ela ionships. 3.5.1. Measu emen model The esul s o he measu emen model a e p esen ed in Tables 3.3 and 3.4. The esul s o composi e eliabili y (CR) a e g ea e han 0.70, indica ing ha he model has good in e nal consis ency. To assess he indica o eliabili y, we conside ed a loading g ea e han 0.70. The ins umen s p esen a good indica o eliabili y. A e age a iance ex ac ed (AVE) was used o es con e gen alidi y. AVE should be g ea e han 0.50 so ha he la en a iables explain mo e han hal o he a iance o hei indica o s (Fo nell and La cke , 1981; Hai J e al., 2016; Hensele e al., 2009). 51 Doc o al P og amme in In o ma ion Managemen Table 3.3 - Measu emen model. Cons uc AVE Composi e Reliabili y C onbach's Alpha I em Loadings - alue In o ma ion Quali y (INFQ) 0.652 0.882 0.822 INFQ1 0.829 38.958 INFQ2 0.851 44.922 INFQ3 0.809 40.380 INFQ4 0.737 17.910 Sys em Quali y (SYSQ) 0.628 0.871 0.801 SYSQ1 0.736 17.964 SYSQ2 0.839 32.255 SYSQ3 0.837 42.697 SYSQ4 0.753 20.061 Se ice Quali y (SERQ) 0.617 0.865 0.792 SERQ1 0.725 15.049 SERQ2 0.824 30.969 SERQ3 0.818 31.505 SERQ4 0.771 23.059 Use (U) 0.595 0.855 0.773 U1 0.781 27.522 U2 0.745 21.760 U3 0.792 22.677 U4 0.767 23.880 Sa is ac ion (S) 0.690 0.870 0.776 S1 0.850 48.015 S2 0.833 40.310 S3 0.809 27.531 Con i ma ion (C) 0.623 0.832 0.697 C1 0.777 21.874 C2 0.824 32.682 C3 0.765 24.587 Indi idual pe o mance (IP) 0.630 0.836 0.705 IP1 0.782 27.527 IP2 0.846 39.725 IP3 0.751 18.229 Pe cei ed Use ulness (PU) 0.575 0.844 0.753 PU1 0.729 17.764 PU2 0.813 35.602 PU3 0.778 25.344 PU4 0.709 15.089 Con inuance In en ion (CI) 0.525 0.812 0.695 CI1 0.757 22.151 CI2 0.769 20.020 CI3 0.807 41.005 No e: AVE: A e age a iance ex ac ed. 52 Doc o al P og amme in In o ma ion Managemen Table 3.4 - La en cons uc co ela ions and squa e oo s o AVEs. Mean STDEV INFQ SYSQ SERQ U S C IP PU CI INFQ 4.869 1.123 0.807 SYSQ 4.813 1.123 0.608 0.793 SERQ 4.247 1.166 0.247 0.284 0.785 U 4.693 1.154 0.509 0.395 0.349 0.771 S 4.464 1.165 0.524 0.432 0.361 0.566 0.831 C 4.330 1.102 0.498 0.413 0.396 0.510 0.640 0.789 IP 4.545 1.055 0.384 0.390 0.257 0.415 0.333 0.397 0.794 PU 4.464 1.085 0.446 0.327 0.329 0.495 0.530 0.566 0.485 0.758 CI 4.481 0.986 0.389 0.342 0.288 0.496 0.426 0.439 0.481 0.386 0.724 No es: Values in bold a e he squa e oo o he a e age a iance ex ac ed; STDEV: S anda d de ia ion; INFQ: In o ma ion Quali y; SYSQ: Sys em Quali y; SERQ: Se ice Quali y; U: Use; S: Sa is ac ion; C: Con i ma ion; IP: Indi idual pe o mance; PU: Pe cei ed Use ulness; CI: Con inuance In en ion. As seen in Table 3.3, all he cons uc s mee hese c i e ia, ensu ing con e gence. This shows ha he cons uc s can be used o assess he heo e ical model. Disc iminan alidi y was measu ed using Fo nell-La cke c i e ion (Table 3.4), c oss-loadings c i e ion (Table 3.5), and he he e o ai -mono ai a io o co ela ions (HTMT) (Table 3.6). Table 3.4 epo s he squa e oo o he AVE in bold along he diagonal, and he co ela ions be ween he cons uc s. Based on Fo nell and La cke (1981) c i e ion, he squa e oo o he AVE should be g ea e han he co ela ion be ween he cons uc s, and hus he cons uc s ul il he c i e ion. To ensu e he disc iminan alidi y, each i em p esen s a highe loading on i s co esponding ac o han in he c oss-loading (Chinn, 1998; Gö z e al., 2010). Based on HTMT (Table 3.6) i can be seen ha all he alues a e below 0.90, and i he e o e can be concluded ha he e is disc iminan alidi y (Hensele e al., 2015). The measu emen model indings indica e ha he model has a good in e nal consis ency, eliabili y indica o , con e gence alidi y, and disc iminan alidi y, illus a ing ha he cons uc s a e s a is ically di e en and can be used o assess he s uc u al model. 53 Doc o al P og amme in In o ma ion Managemen Table 3.5 - C oss loadings. INFQ SYSQ SERQ U S C IP PU CI INFQ1 0.829 0.427 0.185 0.476 0.461 0.428 0.359 0.411 0.329 INFQ2 0.851 0.541 0.166 0.440 0.440 0.385 0.364 0.369 0.354 INFQ3 0.809 0.516 0.222 0.379 0.434 0.420 0.312 0.408 0.327 INFQ4 0.737 0.490 0.234 0.334 0.346 0.371 0.183 0.231 0.235 SYSQ1 0.469 0.736 0.230 0.314 0.328 0.277 0.269 0.199 0.198 SYSQ2 0.477 0.839 0.255 0.306 0.385 0.339 0.326 0.264 0.293 SYSQ3 0.483 0.837 0.242 0.302 0.360 0.343 0.332 0.321 0.312 SYSQ4 0.499 0.753 0.171 0.332 0.293 0.346 0.306 0.247 0.277 SERQ1 0.245 0.278 0.725 0.251 0.289 0.319 0.174 0.275 0.216 SERQ2 0.172 0.208 0.824 0.295 0.283 0.331 0.229 0.271 0.265 SERQ3 0.176 0.168 0.818 0.282 0.287 0.297 0.170 0.232 0.195 SERQ4 0.183 0.239 0.771 0.268 0.275 0.296 0.235 0.253 0.226 U1 0.426 0.325 0.316 0.781 0.402 0.357 0.346 0.353 0.438 U2 0.386 0.292 0.219 0.745 0.406 0.381 0.308 0.421 0.425 U3 0.354 0.248 0.255 0.792 0.436 0.408 0.308 0.337 0.318 U4 0.400 0.346 0.282 0.767 0.498 0.426 0.316 0.414 0.349 S1 0.495 0.415 0.332 0.585 0.850 0.502 0.315 0.445 0.368 S2 0.432 0.348 0.267 0.435 0.833 0.545 0.259 0.469 0.367 S3 0.372 0.307 0.299 0.378 0.809 0.552 0.252 0.407 0.325 C1 0.341 0.294 0.294 0.402 0.561 0.777 0.352 0.456 0.341 C2 0.446 0.369 0.349 0.417 0.497 0.824 0.341 0.456 0.375 C3 0.390 0.312 0.294 0.387 0.454 0.765 0.241 0.426 0.321 IP1 0.309 0.286 0.234 0.356 0.318 0.354 0.782 0.369 0.389 IP2 0.295 0.333 0.181 0.330 0.249 0.278 0.846 0.375 0.403 IP3 0.312 0.309 0.198 0.301 0.223 0.314 0.751 0.414 0.351 PU1 0.388 0.227 0.230 0.347 0.420 0.468 0.322 0.729 0.249 PU2 0.323 0.234 0.260 0.410 0.375 0.466 0.380 0.813 0.325 PU3 0.312 0.283 0.275 0.344 0.431 0.395 0.395 0.778 0.335 PU4 0.333 0.252 0.230 0.402 0.386 0.381 0.377 0.709 0.258 CI1 0.310 0.261 0.231 0.389 0.312 0.339 0.380 0.337 0.757 CI2 0.276 0.271 0.175 0.367 0.334 0.317 0.319 0.274 0.769 CI3 0.359 0.316 0.248 0.431 0.373 0.364 0.462 0.296 0.807 No es: Indica o loading (in bold) g ea e han all i s c oss-loadings; INFQ: In o ma ion Quali y; SYSQ: Sys em Quali y; SERQ: Se ice Quali y; U: Use; S: Sa is ac ion; C: Con i ma ion; IP: Indi idual pe o mance; PU: Pe cei ed Use ulness; CI: Con inuance In en ion. 54 Doc o al P og amme in In o ma ion Managemen Table 3.6 - He e o ai -mono ai a io o co ela ions (HTMT). INFQ SYSQ SERQ U S C IP PU CI INFQ SYSQ 0.755 SERQ 0.311 0.357 U 0.631 0.500 0.444 S 0.648 0.544 0.460 0.721 C 0.657 0.551 0.533 0.694 0.871 IP 0.497 0.519 0.345 0.561 0.447 0.564 PU 0.561 0.422 0.426 0.650 0.695 0.778 0.670 CI 0.487 0.433 0.386 0.658 0.563 0.623 0.656 0.527 No es: INFQ: In o ma ion Quali y; SYSQ: Sys em Quali y; SERQ: Se ice Quali y; U: Use; S: Sa is ac ion; C: Con i ma ion; IP: Indi idual pe o mance; PU: Pe cei ed Use ulness; CI: Con inuance In en ion. 3.5.2. S uc u al model A e he alida ion o he measu emen model, he s uc u al model was analysed o hypo heses and cons uc s es ing. Figu e 3.2 p esen s he esea ch esul s. The s uc u al model assessmen used 5000 boo s ap esamples o es ima e he pa h signi icance le el o he model (Hensele e al., 2009). The VIF ( a iance in la ion ac o ) was also es ed o assess he mul icollinea i y. All o he cons uc s a e below he h eshold o 5, indica ing he absence o mul icollinea i y be ween he cons uc s (Hai J . e al., 2016). Figu e 3.2 - Resea ch model. No es: In eg a ion o Expec a ion Con i ma ion Model (ECM) and DeLone and McLean in o ma ion sys em success (D&M IS); (*** p < 0.01; ** p < 0.05; * p < 0.10). 55 Doc o al P og amme in In o ma ion Managemen The model explains 32% o he a ia ion in he use o m-paymen . The in o ma ion quali y ( " # = 0.400, p < 0.01) and se ice quali y ( " # = 0.225, p < 0.01) a e s a is ically signi ican in explaining use, hus con i ming H1a and H3a. The sys em quali y is no s a is ically signi ican in explaining use, and hus hypo hesis H2a is no con i med. The model explains 53% o he a ia ion in sa is ac ion in using m-paymen . The in o ma ion quali y ( " # = 0.152, p < 0.01), con i ma ion ( " # = 0.389, p < 0.01), use ( " # = 0.240, p < 0.01), and pe cei ed use ulness ( " # = 0.136, p < 0.05) a e s a is ically signi ican in explaining sa is ac ion, hus con i ming H1b, H4a, H5a, and H7a. Sys em quali y and se ice quali y a e no s a is ically signi ican in explaining sa is ac ion, and hus hypo heses H2b and H3b a e no con i med. The model explains 34% o he a ia ion in con i ma ion o m-paymen . In o ma ion quali y ( " # = 0.359, p < 0.01) and se ice quali y ( " # = 0.275, p < 0.01) a e s a is ically signi ican in explaining con i ma ion, hus con i ming H1c and H3c. Sys em quali y is no s a is ically signi ican , and hus hypo hesis H2c is no con i med. The model explains 32% o a ia ion in pe cei ed use ulness o m-paymen . Con i ma ion ( " # = 0.566, p < 0.01) is s a is ically signi ican in explaining pe cei ed use ulness, hus con i ming H4b. The model explains 19% o a ia ion in indi idual pe o mance. Use ( " # = 0.416, p < 0.01) and sa is ac ion ( " # = 0.146, p < 0.1) a e s a is ically signi ican in explaining indi idual pe o mance, he e o e con i ming H5b and H6a. 36% o he a ia ion is explained by he model in con inuance in en ion o use m- paymen . Use ( " # = 0.272, p < 0.01), indi idual pe o mance ( " # = 0.310, p < 0.01), and sa is ac ion ( " # = 0.270, p < 0.01) a e s a is ically signi ican in explaining con inuance in en ion, he e o e con i ming H5c, H6b, and H8. Pe cei ed use ulness is no s a is ically signi ican , and hus hypo hesis H7b is no con i med. 56 Doc o al P og amme in In o ma ion Managemen The s onges ela ionships we e con i ma ion on pe cei ed use ulness ( " # = 0.566), use on indi idual pe o mance ( " #= 0.416), in o ma ion quali y on use ( " #= 0.400), and con i ma ion on sa is ac ion ( " # = 0.389). 3.6. Discussion The model p oposed is a combina ion o he D&M IS success model (DeLone and Mclean, 2003) and he ECM (A Bha ache jee, 2001), o explain con inuance in en ion o use m-paymen . Based on he indings (see Table 3.7), o 19 hypo heses 14 we e con i med and 5 we e no . The e o e, we can a gue ha mos o he hypo hesized ela ionships we e con i med. Indi idual pe o mance is he s onges p edic o o con inuance in en ion, ollowed by use and sa is ac ion. In o ma ion quali y and se ice quali y de e mine con i ma ion. In o ma ion quali y, use, con i ma ion, and pe cei ed use ulness de e mine use s’ sa is ac ion. Use is explained by in o ma ion and se ice quali y. Indi idual pe o mance is explained by use and sa is ac ion o m-paymen . Pe cei ed use ulness is explained by con i ma ion. Su p isingly, se ice quali y does no explain sa is ac ion, pe cei ed use ulness does no explain con inuance in en ion, and sys em quali y explains none o he p oposed ela ionships. Table 3.7 - Hypo heses esul s. Hypo hesis Independen Cons uc → Dependen cons uc Findings (β) P alue Conclusion H1a In o ma ion Quali y → Use 0.400 0.000 Suppo ed H1b In o ma ion Quali y → Sa is ac ion 0.152 0.008 Suppo ed H1c In o ma ion Quali y → Con i ma ion 0.359 0.000 Suppo ed H2a Sys em Quali y → Use 0.088 0.235 No suppo ed H2b Sys em Quali y → Sa is ac ion 0.065 0.261 No suppo ed H2c Sys em Quali y → Con i ma ion 0.116 0.140 No suppo ed H3a Se ice Quali y → Use 0.225 0.001 Suppo ed H3b Se ice Quali y → Sa is ac ion 0.067 0.163 No suppo ed H3c Se ice Quali y → Con i ma ion 0.275 0.000 Suppo ed H4a Con i ma ion → Sa is ac ion 0.389 0.000 Suppo ed H4b Con i ma ion → Pe cei ed Use ulness 0.566 0.000 Suppo ed H5a Use → Sa is ac ion 0.240 0.000 Suppo ed H5b Use → Indi idual pe o mance 0.416 0.000 Suppo ed H5c Use → Con inuance In en ion 0.272 0.000 Suppo ed 57 Doc o al P og amme in In o ma ion Managemen H6a Sa is ac ion → Indi idual pe o mance 0.146 0.071 Suppo ed H6b Sa is ac ion → Con inuance In en ion 0.162 0.018 Suppo ed H7a Pe cei ed Use ulness → Sa is ac ion 0.136 0.025 Suppo ed H7b Pe cei ed Use ulness → Con inuance In en ion 0.016 0.834 No suppo ed H8 Indi idual pe o mance → Con inuance In en ion 0.310 0.000 Suppo ed The esul s indica e ha in o ma ion quali y posi i ely in luences use (H1a), sa is ac ion (H1b), and con i ma ion (H1c) o m-paymen . The esul s o H1a and H1b a e consis en wi h hose o Tam and Oli ei a (2016) and Cid al e al. (2018), who asse ha in o ma ion quali y posi i ely in luences use and sa is ac ion. This means ha when m-paymen p o ides in o ma ion wi h quali y, i posi i ely impac s he usage, sa is ac ion, and con i ma ion o he expec a ion o m-paymen (Chang e al., 2014; Cheng, 2014; Cid al e al., 2018). The use s may always expec o access comp ehensi e, accu a e, and up- o-da e in o ma ion on he m-paymen sys em. As use s unde s and ha he m-paymen p esen s quali y in o ma ion, hey will unde s and ha he m-paymen p o ide is main aining he in o ma ion up o da e, and hey will he e o e con inue using m-paymen . I he in o ma ion is ou o da e o inaccu a e, i will nega i ely in luence he usage, sa is ac ion, and expec a ions (Gao e al., 2015). Ou esul s indica e ha sys em quali y did no a ec use (H2a), sa is ac ion (H2b), o con i ma ion (H2c) o m-paymen , indica ing ha in he pos -accep ance phase he quali y o he sys em is no impo an o he usage, use sa is ac ion, o he con i ma ion o he expec a ions ega ding con inuance in en ion o m-paymen . This esul is no consis en wi h he indings epo ed in se e al s udies (Budia djo e al., 2017; Cid al e al., 2018; Gao e al., 2015). A possible explana ion may eside in he ac ha use s al eady assume ha m-paymen wo ks well, and ha i is a ma u e echnology. Ano he eason may be ha ou s udy was conduc ed in a de eloping economy con ex , and ha he o he s udies we e conduc ed in de eloped economies. A hi d eason may ha e o do wi h he echnology s udied. The o he s udies applied di e en echnologies such as web-based lea ning, social ca aloguing si es, online lea ning, e c. (Cid al e al., 2018; Daʇhan and Akkoyunlu, 2016; Gao e al., 2015). Fu he mo e, ou esul s indica e ha se ice quali y posi i ely in luenced use (H3a) and con i ma ion (H3c), bu no sa is ac ion (H3b). The esul o H3a is con adic o y wi h ha o Cid al e al. (2018) and Tam and Oli ei a (2016), whe e he ela ionship 64 Doc o al P og amme in In o ma ion Managemen Chap e 4 – Con inuance in en ion o mobile paymen : TTF model wi h us in an A ican con ex : case s udy o Mozambique 4.1. In oduc ion How do us and he alignmen o ask and echnology cha ac e is ics a ec he con inuance use o mobile paymen (m-paymen )? M-paymen has become one o he p ominen se ices in use oday (Gao & Waech e , 2015; T. Zhou, 2014b), by which use s can make paymen s o goods, se ices, and bills; check balances; and ans e money any ime and anywhe e (Kujala e al., 2017; Liébana-cabanillas e al., 2018; J. Lu, Wei, e al., 2017; Oli ei a e al., 2016; T. Zhou, 2014b). M-paymen is de ined as any paymen in which a mobile de ice (mobile phone o able ) is used o pe o m inancial alue exchanges (ini ia e, au ho ize, and con i m) in e u n o goods and se ices (Shao e al., 2019). The e a e di e en ways o conduc a ansac ion using m- paymen . The simples way is sho -message-based, by which using a simple mobile phone he use can check balances o conduc paymen using sho ex messages (Singh e al., 2020; T. Zhou, 2013). The m-paymen se ice is g owing exponen ially all o e he wo ld and is b inging bene i s o use s and he p o ide s (Humbani & Wiese, 2018). Conside ing i s bene i s, companies a e p o iding i in di e en ways a ound he globe (Fan e al., 2018a; Singh e al., 2020). The echnological sec o in he A ican con ex is s ill unde de elopmen , which ep esen s a challenge o he en i e economy. Financial se ices like banks a e no a ailable in he same p opo ions in all egions o a coun y, e.g. Mozambique (Humbani & Wiese, 2018; INE, 2019), so people li ing in u al a eas a e o ced o a el a he o gain access o banks (Ba is a & Vicen e, 2014; Humbani & Wiese, 2019). Due o he lack o access o echnology in he same p opo ion; a se ice ha can be used by he popula ion, ega dless o hei li e a y and economic le el, educing he need o use banks, is o g ea impo ance (Pal e al., 2020). Fo m-paymen , he echnological ac o s in he A ican con ex acili a e i s accessibili y (e.g., only need elephone ne wo k) (Jack & Su i, 2011), he eby b inging he se ice o many people. 65 Doc o al P og amme in In o ma ion Managemen Much esea ch has been di ec ed o he subjec o m-paymen (J. Lu, Wei, e al., 2017; Oli ei a e al., 2016; Sinha e al., 2019). P e ious li e a u e epo s he use o di e en heo e ical models o in es iga e con inuance in en ion o use m-paymen . Zhou, (2013a) used in o ma ion sys em success and low heo y o examine con inuance in en ion o m-paymen se ices. Shao e al., (2019) used us and inno a ion di usion heo y o unde s and m-paymen pla o ms. Lu e al., (2017) used expec a ion- con i ma ion heo y, mobili y, p i acy p o ec ion, social in luence, and cul u al alues o unde s and m-paymen con inua ion. Chen & Li, (2017) used IT con inuance, isk- us , and a ec -cogni ion li e a u e o unde s and con inuance in en ion o m-paymen se ices. To he bes o ou knowledge, he e a e no published s udies ha examine he in en ion o con inue using m-paymen in Mozambique. To ill he gap, we in es iga e he ac o s ha may in luence he con inued use o m-paymen in ha coun y. Conside ing ha a e he change om adi ional paymen me hod (cash) o m- paymen , use s may ha e unce ain y and mis us when using m-paymen . Knowing ha he ansac ions in ol e cash, as well as he impo ance o unc ionali ies and echnological ac o s ha may be e help he use o m-paymen , we analyse he ask- echnology i (TTF) (Goodhue and Thompson, 1995) and o e all us (Oli ei a e al., 2017) heo ies and e alua e hei ela ionships o con inuance in en ion. Fu he mo e, no p e ious s udy has joined he TTF, o e all us , and ECM cons uc s in o a single model, as we do in he p esen s udy (Da ison and Ma insons, 2016). The TTF model s a es ha IT can ha e a posi i e impac on an indi idual's ask pe o mance i he IT unc ionali ies ma ch he equi emen s o he asks ha he use needs o pe o m (Goodhue and Thompson, 1995). Rega ding he TTF, he model was used in di e en echnological con ex s such as: mobile banking (Tam and Oli ei a, 2016), MOOCs (Wu and Chen, 2017), social media sea ch sys em (Dang e al., 2020), and in e ne banking (Rahi e al., 2020), and has been combined wi h di e en heo e ical models such as: echnology con inuance heo y (Rahi e al., 2020), echnology accep ance model (TAM) (Wu and Chen, 2017), DeLone and McLean in o ma ion sys ems success model (D&M ISSM) (Tam and Oli ei a, 2016), men al wo kload and uni ied heo y o accep ance and use o echnology (UTAUT) (Dang e al., 2020). O e all us e e s o he combina ion o ac o s such as compe ence, in eg i y, and bene olence in o de o unde s and he use ’s con idence o use IT (Oli ei a e al., 2017). Many s udies ha e 66 Doc o al P og amme in In o ma ion Managemen used o e all us in combina ions wi h di e en models and in di e en con ex s. Oli ei a e al. (2017) used o e all us wi h consume cha ac e is ics, i ms’ cha ac e is ics, websi e in as uc u e, and in e ac ions o examine pu chase in en ion in e-comme ce. Tam e al. (2019) used o e all us wi h D&M ISSM o examine indi idual pe o mance in e-comme ce. Conside ing ha cybe secu i y is oday conside ed a undamen al challenge o any coun y o o ganiza ion, and especially in mobile paymen s con ex s, i is impo an o ha e a sense o secu i y and us . Conside ing ha ac , we expec ha he e is a posi i e link be ween o e all us and use and con inuance in en ion. In addi ion o us , he asymme y o in o ma ion and communica ion echnology (ICT) ac oss A ican coun ies is ano he challenge. The cons ain s o ha asymme y could a ec he long- e m o iabili y o mobile paymen , and o ha eason i would be aluable o unde s and how he ask- echnology- i o mobile paymen may explain he use and indi idual pe o mance in Mozambique. The con ibu ion o his s udy is wo old. Fi s , i con ibu es o he li e a u e on con inuance in en ion, since many s udies ha e been ca ied ou in he con ex o echnology adop ion (Khalilzadeh e al., 2017; Ve kijika, 2020). Howe e , s udying con inuance in en ion has an impac on he long- e m su i al o echnology (A Bha ache jee, 2001). Fo his eason, his s udy aims o expand he knowledge on his opic by joining he wo heo ies men ioned abo e: ask- echnology cha ac e is ics and expec a ion-con i ma ion model. Second, his s udy was based on he ECM model (A Bha ache jee, 2001), which has been es ed in di e en con ex s such as m-paymen pla o ms (Shao e al., 2019), MOOCs (Gao e al., 2015) and mobile apps (Tam e al., 2020). Howe e , in o ma ion sys ems (IS) ha e di e en cha ac e is ics and unc ionali ies (Nascimen o e al., 2018). In his sense, we seek o ex end he model o unde s and he impac o echnological aspec s (TTF) and us in con inuance in en ion o use m-paymen . In addi ion, we in end o unde s and he impac o sa is ac ion as a mode a o . To he bes o ou knowledge, his s udy is he i s o combine he TTF, ECM, and he us dimension o in es iga e con inuance in en ion. The plan o he pape is as ollows. We begin wi h a li e a u e e iew o he ele an s udies ega ding m-paymen , TTF, and he dimension o us . Second, we p esen he 67 Doc o al P og amme in In o ma ion Managemen esea ch model, ollowed by he hypo heses. Thi d, we p esen he me hodologies used o es he hypo heses. We hen show he esul s, ollowed by a discussion and implica ions o his s udy, and sugges ions o u u e esea ch. 4.2. Li e a u e e iew 4.2.1. Mobile Paymen in an A ican con ex In A ica m-paymen has o e 29 million ac i e use s in mo e han i e coun ies (Jack and Su i, 2011; Voda one G oup, 2016). In some coun ies i is called mobile money (Koloseni and Manda i, 2017). M-paymen e e s o se ices ha enable use s o ans e money, pay se ices and goods, and wi hd aw money ia a mobile phone (Koloseni and Manda i, 2017; Shao e al., 2019). I is some imes con used wi h online paymen , bu i is no he same because online paymen uses any mobile de ice connec ed o he in e ne , such as a able , mobile phone, o lap op (T. Zhou, 2015), while m-paymen uses only mobile phones wi h o wi hou in e ne . In Mozambique 26.4% o he popula ion uses a mobile phone and 11.7% uses m-paymen (INE, 2019). This exponen ial g ow h is occu ing because his se ice is becoming an al e na i e solu ion o u al and u ban people o access inancial se ices (Humbani and Wiese, 2019). As mos bank b anches a e dis an om people, hey na u ally wish o a oid a elling long dis ances o access he bank’s se ices. M-paymen hus o e s subs an ial bene i s by ha ing an app op ia e accoun and being able o use i any ime and anywhe e (Jack and Su i, 2011). Ea lie s udies ha e poin ed ou ha m-paymen o igina ed in de eloping coun ies ia SMSs and ha i sp ead quickly due o limi ed cash al e na i es, such as bank accoun s and c edi ca ds, he eby helping communi ies ha we e o he wise excluded om he inancial sys em (Humbani and Wiese, 2019; Makina, 2017). The alue o conduc ing s udies in he A ican con ex is he e o e e iden (Humbani and Wiese, 2019). Once a new echnology such as m-paymen is de eloped and eleased o he socie y, i is impo an o unde s and he accep ance ac o s (Venka esh e al., 2003). When he echnology has been in use o some ime, i is impo an o in es iga e he ac o s ha in luence he con inui y o use om he use s' poin o iew (A Bha ache jee, 2001). 68 Doc o al P og amme in In o ma ion Managemen Rega ding m-paymen , conside ing he impac o mis us o unce ain y, he unc ionali ies, and he echnological d i e s o m-paymen use, i is o g ea impo ance and necessi y o examine he use s' con inued use in en ion and pe cep ions owa d m-paymen . Fu he mo e, i is essen ial o o e e ec i e m-paymen unc ionali ies ha can be e handle use ansac ions (T. Zhou, 2014b). The e o e, we decided o use TTF and o e all us as ou base heo ies and examine hei impac on he con inuance in en ion. Ea lie s udies on con inuance in en ion o use m-paymen ha e been published (X. Chen and Li, 2017; Lu e al., 2017; Pa k e al., 2017; Yu e al., 2018). Conside ing ha ou aim is o unde s and he in en ion o con inue using m-paymen , we e iewed p e ious s udies o unde s and wha has al eady been done in he con ex o m-paymen . Zhou (2014) in es iga ed he con inued use o m-paymen based on us , low, sys em quali y, and in o ma ion quali y. Chen and Li (2017) in es iga ed he in en ion o con inue o use m-paymen se ices using IT con inuance, isk- us , and a ec - cogni ion. Koloseni and Manda i (2017) used he heo y o planned beha iou , pe cei ed cos , pe cei ed us , and sa is ac ion o examine con inuance usage o mobile money se ices. Yu e al. (2018) in es iga ed he in en ion o con inue using m- paymen based on us ans e heo y. Humbani and Wiese (2019) used he echnology eadiness index o p edic adop ion and con inuance in en ion, wi h he goal o explo ing he eadiness o he m-paymen app echnology. Shao e al. (2019) in es iga ed an eceden s o us and con inuance in en ion in m-paymen pla o ms based on us and inno a ion di usion heo y. Raman and Aashish (2021) in es iga ed he an eceden s o use s' willingness o con inue using m-paymen se ices, based on us , con enience, social alue, sa is ac ion, se ice quali y, a i ude, isk, and e o expec ancy. Odoom and Kosiba (2020) used UTAUT o in es iga e con inuance in en ion o m-money. Liébana-Cabanillas e al. (2021) in es iga ed he de e minan s o in en ion o con inue using and he mode a ing e ec o gende and age o NFC m- paymen use s. They applied cons uc s om di e en models such as heo y o easoned ac ion, pe cei ed alue heo y, UTAUT2, pe sonal inno a ion in in o ma ion echnology, mobile paymen echnology accep ance model, and ECM. Wi h ou analysis o he li e a u e, we ound ha he e a e di e en models applied o s udy m- paymen . Rega ding he ICT asymme y ac oss A ican coun ies and gene al 69 Doc o al P og amme in In o ma ion Managemen pe cep ion o us o gi e anspa ency and secu i y o mobile paymen , we joined TTF and o e all us o he ECM. 4.2.2. In o ma ion sys em con inuance model “IS con inuance” was p esen ed by Bha ache jee (2001) o explain he in en ion o con inue using IS. The ECM model ocuses on h ee cogni i e eelings (expec a ion o con i ma ion, pe cei ed use ulness, and sa is ac ion). The model p oposes ha sa is ac ion is he s onges in luence o con inuance in en ion, on he g ounds ha i esul s om he con i ma ion o expec a ions and he pe cep ion o pe o mance. This means ha a e he use adop s he IS, and uses i o a while, (s)he will ealize i he expec a ions ha e been con i med and i he IS is/a e use ul. I he esul is posi i e, he le el o sa is ac ion inc eases and, consequen ly, he in en ion o con inue using he IS inc eases (A Bha ache jee, 2001). The model has been used in p e ious s udies, has been in eg a ed wi h o he models, and has been applied in di e en con ex s (F anque e al., 2020). Wu & Chen (2017) in eg a ed TAM, TTF, MOOCs ea u es and social mo i a ion o unde s and con inuance in en ion o use MOOCs. Gao e al. (2015) joined he in o ma ion success model, low heo y, and us o in es iga e mobile pu chase. Humbani & Wiese (2019) in eg a ed he echnology eadiness index o explo e he use o mobile apps. In he p esen s udy we combine TTF and o e all us in o de o unde s and m-paymen . The model will help us o assess he e ec o TTF and o e all us ac o s on con inuance in en ion. 4.2.3. Task echnology i (TTF) model The TTF model in oduced by Goodhue & Thompson (1995) is applied in IS esea ch o explain he pe o mance impac o IS. The heo y a gues ha he abili y o an IS o pe o m an ac i i y ask cha ac e is ic easily and well, and he echnology cha ac e is ics ha suppo such ac i i ies, signi ican ly in luence he use and pe o mance impac o an IS. When he ask and echnology i oge he , he use s’ ac i i ies a e acili a ed, imp o ing he use o he echnology, and he eby imp o ing he pe cep ion o pe o mance impac (Goodhue & Thompson, 1995). The be e alignmen o he ask and echnology cha ac e is ics makes i possible o encou age he 70 Doc o al P og amme in In o ma ion Managemen use and pe o mance impac o m-paymen . Se e al esea ch pape s apply he TTF model combined wi h o he heo ies. Oli ei a e al. (2014) combined TTF, UTAUT, and ini ial us model (ITM) o in es iga e mobile banking adop ion. The au ho s used TTF o p edic he pe o mance expec ancy and adop ion in he UTAUT model. Tam & Oli ei a (2016b) combined TTF and he IS Success model o unde s and he in luence o mobile banking on indi idual pe o mance. They used TTF o p edic usage and indi idual pe o mance (pe o mance impac ) cons uc s and as a mode a o o he ela ionship be ween use sa is ac ion and indi idual pe o mance. Dang e al. (2018) combined men al wo kload (MWL), TTF, and UTAUT o examine he impac s o men al wo kload and TTF on accep ance o he social media sea ch sys em. They used TTF o p edic he pe o mance expec ancy, e o expec ancy, and acili a ing condi ions o he o iginal UTAUT model. Wu & Chen (2017) combined TTF and TAM o unde s and con inuance in en ion o use MOOCs, using TTF o p edic he cons uc s’ pe cei ed use ulness and pe cei ed ease o use o he TAM model. La sen e al. (2009) combined TTF and pos -accep ance model (PAM) o unde s and use s’ mo i a ion o con inue he use o IS. They used TTF o p edic he cons uc s’ pe cei ed use ulness and u iliza ion in he PAN model. A shan & Sha i (2016) combined TTF, UTAUT, and ITM o in es iga e mobile banking accep ance in Pakis an. They used TTF o p edic he beha iou al in en ion cons uc o he UTAUT model. We hus ind ha mos s udies ha used TTF had beha iou al in en ion and adop ion as he ou come. In his sense, he in eg a ion o TTF wi h ECM will p o ide some insigh in o m- paymen esea ch. 4.2.4. T us T us has been concep ualized in se e al ways acco ding o he con ex in which i is applied (Ge en, Ka ahanna, & S aub, 2003; Oli ei a, Alhinho, Ri a, & Dhillon, 2017). Acco ding o ea ly li e a u e, in a gene al iew us e lec s he abili y o IS o ul il he asks co ec ly, i.e., he IS p o ide keeps hei p omise and does no decei e use s, and he bene i s o he IS need o be pe cei ed by he use s and he p o ide s (Zhou, 2014). In m-paymen , us e lec s use s’ belie s in he eliabili y o he ansac ions made h ough m-paymen . I an m-paymen p o ide ensu es secu e ansac ions, ul ils he asks co ec ly, and does no decei e he use s, i will be possible o imp o e he 71 Doc o al P og amme in In o ma ion Managemen pe cep ion o eliabili y amongs he m-paymen use s (Chen & Li, 2017; Zhou, 2014). The e a e se e al esea ch a icles ha combine us wi h o he heo e ical models in IS esea ch, such as Zhou & Li (2014) who in es iga e mobile social ne wo k se ices; Zhou (2013) o unde s and m-paymen con inuance in en ion; Gao e al. (2015) o pe cei e consume s’ mobile pu chase con inuance in en ion. In ou esea ch we use he o e all us and us dimension. Bene olence, compe ence, and in eg i y comp ise he dimension o us (Oli ei a e al., 2017). Compe ence e lec s he abili y o an IS p o ide o en o ce hei p omises o use s. In eg i y e lec s he IS p o ide ’s capaci y o ac consis en ly, eliably, and hones ly while keeping i s p omises. Bene olence e lec s on he p obabili y o an IS p o ide o main ain use s’ in e es s and o show since e conce n wi h he well-being o he use s (Chen & Dhillon, 2003; Oli ei a e al., 2017; Pal ia, 2009); 4.3. Resea ch model The pu pose o ou s udy is o unde s and he con inuance in en ion, which we use as he basis o he ECM model (A Bha ache jee, 2001). Conside ing ECM an axioma ic heo y, which is accep able and uly sel -e iden (J. K. Lee e al., 2021), in he cu en s udy we adop ed o ou model wo cons uc s, sa is ac ion and con inuance in en ion (A Bha ache jee, 2001), conside ing ha he dependen cons uc o ou s udy is con inuance in en ion. Following he pa simony app oach and educing model complexi y o make i easie o g asp, he cons uc s pe cei ed use ulness and con i ma ion we e no added. The addi ion o o he models and a iables o e s a be e unde s anding o con inuance in en ion o use m-paymen . We he e o e join he TTF model, which di ec ly in luences use and indi idual pe o mance, wi h us , which is a di ec de e minan o use and con inuance in en ion. The heo e ical model (Fig. 4.1) is designed o examine he con inuance in en ion o use m-paymen in he A ican con ex . The model asse s ha : 1. TTF de e mines he use o m-paymen and he pe cei ed indi idual pe o mance: 72 Doc o al P og amme in In o ma ion Managemen 2. he use o m-paymen can ha e an indi ec o di ec in luence on con inuance in en ion, and a di ec in luence on indi idual pe o mance; 3. indi idual pe o mance de e mines he con inuance in en ion di ec ly; 4. he us dimension can ha e an indi ec o di ec in luence on con inuance in en ion, and a di ec in luence on he use; 5. use sa is ac ion may mode a e he impac o indi idual pe o mance, use, and he us dimension on m-paymen con inuance in en ion. The ollowing sec ion p esen s he p oposed hypo heses. Figu e 4.1 - P oposed esea ch model. F om a echnical pe spec i e, asks a e he ac i i ies ca ied ou by use s when hey a e using m-paymen . Howe e ,, i he ask cha ac e is ics o he m-paymen a e easy o use, app op ia e and unde s andable (Wu & Chen, 2017), he use s will be com o able wi h hei use and con inue using i , on he o he hand, use s acing di icul ies in using m-paymen will p e e op ing adi ional paymen me hod a he han m-paymen (Rahi e al., 2020). Task cha ac e is ics a e ele an cons uc s o in luence posi i ely TTF (Rahi e al., 2020; Tam & Oli ei a, 2019). 73 Doc o al P og amme in In o ma ion Managemen Technology cha ac e is ics a e physical and logical ools (ha dwa e and so wa e), i.e., he look, eel, and speed o echnology. An e ec i e echnology makes m-paymen mo e a ac i e and use ul o he use s (Rahi e al., 2020). This ac o can a ec he long- e m usage o he echnology (Tam & Oli ei a, 2016b, 2019). Fo m-paymen , he echnology cha ac e is ics play an impo an ole, as use s a e ca ying ou mone a y ansac ions, he speed o he ope a ion and he esponse ime a e impo an ac o s. Wi h he minimum echnology cha ac e is ics, he use s need o unde s and he ac ual wo king o asks o m-paymen (Tam & Oli ei a, 2019; B. Wu & Chen, 2017). The TTF is he i be ween ask and echnology cha ac e is ics (Goodhue & Thompson, 1995; Wu & Chen, 2017). This a ibu e means ha when m-paymen use s pe cei e a ma ch be ween ask and echnology (unde s and ha he ea u es and echnological cha ac e is ics a e sui able o ca ying ou ansac ions) i will be possible o imp o e he usage and he con inued usage o m-paymen (Tam & Oli ei a, 2016b). The e o e, i is expec ed ha he m-paymen use s will pe o ms he asks e icien ly (Rahi e al., 2020). when he m-paymen asks a e easy o use, as , and p o ided any ime and anywhe e, use s will eel he use ulness o m-paymen , he eby inc easing he wo k pe o mance o he indi idual. Thus, we hypo hesise: H1a: TTF posi i ely in luences use. H1b: TTF posi i ely in luences indi idual pe o mance. When use s s a o use any sys em, hey s a o pe cei e bene i s. When using m- paymen , use s will pe cei e i s bene i s and he le el o pe cei ed indi idual pe o mance will inc ease. Ea lie s udies epo empi ical suppo o he ela ionship be ween use and indi idual pe o mance (Tam & Oli ei a, 2016a, 2016b). Thus, he equen use o m-paymen o check balances, make ans e s, pay o goods, e c. will in luence he indi idual pe o mance, and inc ease he in en ion o con inue using m- paymen . When use s pe cei e e o less use, and when hey s a pe cei ing pe o mance ou comes (Chang, Liu, & Chen, 2014), usage will become mo e equen . When he use o m-paymen se ices s a s o become au oma ic and use s use i mo e o en, we expec ha he con inuance usage o m-paymen will inc ease (Tam & Oli ei a, 2016b). Thus, we hypo hesize: 80 Doc o al P og amme in In o ma ion Managemen 4.5.2. S uc u al model The esul s o he s uc u al model we e examined by he pa h coe icien s ha p esen he s eng h o he cons uc s’ ela ionships, a iance in la ion ac o (VIF), -s a is ic alue, and a iance explained (R2) o alida e he hypo heses and cons uc s (see Fig. 4.2). The s uc u al model examina ion used 5000 boo s ap esamples o es ima e he pa hs’ signi icance (Hensele e al., 2009). We es ed he VIF o assess he mul icollinea i y and all he cons uc s a e below he h eshold o 5, hus indica ing he absence o mul icollinea i y (Hai e al., 2016). Figu e 4.2 - Resea ch model Rega ding R2 (see de ail in Fig. 4.2), he esul s o he PLS s uc u al model explain 29.5% o he a ia ion in use. The ask echnology i , and o e all us a e signi ican in explaining use. Thus, con i ming H1a and H7a. The esea ch model explains 45.2% o he a ia ion in indi idual pe o mance and a e explained by ask echnology i and use, con i ming H1b and H2a. The esea ch model explains 47.5% o o e all us a ia ion and a e explained by bene olence, compe ence, and in eg i y, con i ming H4, H5, and H6. The esea ch model explains 47.8% o he con inuance in en ion, explained by use, indi idual pe o mance, and o e all us , con i ming H2b, H3, and H7b. Addi ionally, h ee mode a ing e ec s (H8a, H8b, and H8c) we e examined. The 81 Doc o al P og amme in In o ma ion Managemen indings showed ha H8b and H8c we e signi ican , and H8a was no s a is ically signi ican . Howe e , he mode a ion e ec o H8c is nega i e, meaning ha g ea e use sa is ac ion will be weake in he ela ionship be ween o e all us o con inuance in en ion, hus con i ming H8b and H8c. 4.5.3. Media ing ole o use and indi idual pe o mance The indings e eal ha he e a e media ion e ec s on some cons uc s. Media ion e ec (indi ec e ec ) is p esen ed by a hi d in e ening a iable be ween an independen and a dependen cons uc (Hai J . e al., 2016). We pe o med a media ion analysis and he esul s (Table 4.4) show ha indi idual pe o mance is a pa ial media o be ween use and con inuance in en ion. Also, use is a pa ial media o be ween o e all us and con inuance in en ion. Table 4.4 - Media ion analysis. Be a -Tes p-Value conclusion H9a: OT → U → CI 0.103 3.922 0.000 Pa ial media ion H9b: U → IP → CI 0.096 3.635 0.000 Pa ial media ion No es: Use (U); indi idual pe o mance (IP); o e all us (OT), and con inuance in en ion (CI). 4.6. Discussion We de eloped and alida ed a concep ual model ha explains he impo ance o he TTF model and o e all us owa d con inuance in en ion o use m-paymen . The indings show ha 13 o he 15 hypo heses we e con i med. O e all us can in luence use and con inuance in en ion o use m-paymen . O e all us is suppo ed by bene olence, compe ence, and in eg i y. This means ha m-paymen se ice p o ide s should ensu e he bes in e es s o hei end-use s, se ice suppo should do i s bes o assis use s, and end-use s should eel suppo ed and con iden wi h he se ices. Addi ionally, he se ice p o ide mus be hones wi h use s, keeping i s commi men s wi h he end-use s. When he m-paymen se ice p o ide has enough expe ise o suppo use s, hey will s a us ing he p o ide and he echnology, he eby mo i a ing use s o use m-paymen , and boos ing hei in en ion o con inue using m- paymen (Oli ei a e al., 2017; Yu e al., 2018; T. Zhou, 2011b). By imp o ing use s’ 82 Doc o al P og amme in In o ma ion Managemen o e all us , i will be possible o imp o e use and he in en ion o con inue using m- paymen (Zhou, 2013, 2014). TTF posi i ely impac s use and indi idual pe o mance. The ask and echnology cha ac e is ics a e undamen al o he use o m-paymen , conside ing ha use s ha e di e en expe iences. Use s mus ealize ha he ea u es a e objec i e, easy o use, wi h pe cep ible in o ma ion, and ha he cha ac e is ics o he echnology a e adequa e o use he unc ionali ies. Fo example, in Mozambique, se ice p o ide s should ensu e a g ea e i o ask and echnology cha ac e is ics o enhance he usage o m-paymen by use s (Tam & Oli ei a, 2016b). Conside ing he limi a ions o banking in as uc u e, in e ms o space and ime, m- paymen is a use ul al e na i e, as i p o ides eal- ime se ices any ime and anywhe e and is an a ac i e al e na i e o people who li e a om banks (Yu e al., 2018). By imp o ing he ask and echnology cha ac e is ics, i will be possible o imp o e he use and pe cei ed indi idual pe o mance o end-use s, and consequen ly, imp o e he in en ion o con inue using m-paymen . Use posi i ely a ec s indi idual pe o mance and con inuance in en ion o use m-paymen . Howe e , when he use s sense us wo hiness and m-paymen p o ide s deli e adequa e se ices wi h good cha ac e is ics o end-use s, hey will eel sa is ied and mo i a ed o use m-paymen , consequen ly pe cei ing he pe o mance o m-paymen and being sa is ied o con inue using i (Fan, Shao, Li, & Huang, 2018; Liébana-Cabanillas e al., 2018). Indi idual pe o mance posi i ely in luences con inuance in en ion o use m-paymen . When end- use s unde s and ha m-paymen is use ul o hei daily inancial ac i i ies, helps o accomplish hei asks easily, and enables hem o do asks mo e quickly, use s pe cei e indi idual pe o mance, and con inue using m-paymen . The m-paymen p o ide should ensu e good m-paymen cha ac e is ics ha a e easy o use, easy o in e p e , as , and a ailable anywhe e and any ime. Doing so will allow he use s o pe cei e he bene i s, gua an eeing pe cei ed indi idual pe o mance (La sen e al., 2009; Tam & Oli ei a, 2016b). By imp o ing indi idual pe o mance, i will be possible o imp o e he in en ion o con inue using m-paymen . Table 4.5 - Resul s o he hypo heses. Hypo heses Independen cons uc → Dependen cons uc Findings (β) P- alue Suppo H1a Task echnology i → Indi idual pe o mance 0.500 0.000 Yes H1b Task echnology i → Use 0.342 0.000 Yes 83 Doc o al P og amme in In o ma ion Managemen H2a Use → Indi idual pe o mance 0.275 0.000 Yes H2b Use → Con inuance in en ion 0.328 0.000 Yes H3 Indi idual pe o mance → Con inuance in en ion 0.350 0.000 Yes H4 Bene olence → O e all us 0.333 0.000 Yes H5 Compe ence → O e all us 0.246 0.000 Yes H6 In eg i y → O e all us 0.247 0.000 Yes H7a O e all us → Use 0.315 0.000 Yes H7b O e all us → Con inuance in en ion 0.099 0.071 Yes H8a Indi idual pe o mance * Sa is ac ion → Con inuance in en ion 0.017 0.778 No H8b Use * Sa is ac ion → Con inuance in en ion 0.099 0.033 Yes H8c O e all us * Sa is ac ion → Con inuance in en ion -0.126 0.025 No Addi ionally, sa is ac ion mode a es he ela ionships among use and o e all us o m- paymen o explain con inuance in en ion (Susan o e al., 2016; Yu e al., 2018). When he use o m-paymen is mode a ed by he exis ence o sa is ac ion, i is obse ed ha he impac will be high o explain con inuance in en ion o use m-paymen . In con as , he mode a ing e ec o sa is ac ion on o e all us o explain con inuance in en ion will be weake . In his sense, i m-paymen use s ha e a high le el o sa is ac ion, use will gain s eng h, and o e all us will lose s eng h in explaining m-paymen con inuance in en ion. The indings also e eal ha indi idual pe o mance is a pa ial media o be ween use and con inuance in en ion, and use is a pa ial media o be ween o e all us and con inuance in en ion. When use s ha e con idence o m-paymen , in he se ices, and alue he eliabili y p ope ies, hey inc ease hei in en ion o con inue using m-paymen . Wi h use media ion and equen use o m-paymen , us imp o es, and consequen ly in en ion o con inue using m-paymen inc eases. This means ha when he use s use m-paymen in hei daily li es o buy p oduc s o se ices and ans e o wi hd aw money, he in en ion o con inue using i inc eases. Du ing m- paymen use, pe cei ed indi idual pe o mance such as accomplishing asks quickly, easily, and pe cei ing he use ulness o m-paymen in e e yday li e, enhances he posi i e impac o con inuance in en ion o use m-paymen . Thus, m-paymen p o ide s should ensu e as ask comple ion, ease o use, and p o ide se ices ha a e use ul o he use . We plo ed he mode a ions o sa is ac ion o be e unde s and i s beha iou (Fig. 4.3) (Aiken e al., 1991). Figu e 4.3 illus a es ha use o m-paymen has a mo e signi ican 84 Doc o al P og amme in In o ma ion Managemen impac on m-paymen con inuance in en ion when use sa is ac ion is highe . Thus, wi h a highe le el o use sa is ac ion, he use o m-paymen will inc ease he in en ion o con inue using m-paymen . Addi ionally, o e all us has a low signi ican impac on m-paymen con inuance in en ion when use sa is ac ion is highe . The e o e, he impo ance o o e all us o m-paymen con inuance in en ion is impo an when use sa is ac ion is low. Figu e 4.3 - Mode a ion e ec o use sa is ac ion 4.6.1. Theo e ical implica ions The cu en s udy in es iga es he con inuance in en ion o use m-paymen . The e o e, we in eg a ed TTF, o e all us and ECM. Ou esul s indica e ha bo h TTF and o e all us a e impo an and should be conside ed when e alua ing he con inuance in en ion o use m-paymen o simila sys ems. This means ha use s' pe cep ions associa ed wi h he le el o i be ween ea u es and echnology, as well as use s' con idence in using m-paymen may ul ima ely lead o he in en ion o con inue using m-paymen . The s udy joins he TTF, o e all us , and ECM models o e alua e m- paymen , which has no been epo ed in p e ious li e a u e. TTF and o e all us we e combined o explain con inuance in en ion o he ECM, which is, ne e heless, a well- known model and is one o he mos popula and widely accep ed models (A Bha ache jee, 2001). The indings sugges ha he in eg a ion o TTF and o e all us p esen p edic i e powe o explain con inuance in en ion. Indi idual pe o mance, use, and o e all us explain 47.8% o he a ia ion in con inuance in en ion. This esul indica es ha ou model pe o ms well compa ed o p e ious s udies such as he o iginal ECM (A Bha ache jee, 2001), which explained R2=41%, Idemudia e al. 85 Doc o al P og amme in In o ma ion Managemen (2018), which explained R2=46%, Albash awi and Mo iwalla (2019), which explained R2=45.9%, and Gong e al. (2020), which explained R2=41.6%. Thus, indica ing ha TTF and o e all us a e impo an an eceden s o m-paymen con inuance in en ion. Second, he model was applied in he A ican con ex o m-paymen , add essing he concep o con inuance in en ion (Humbani & Wiese, 2019). To he bes o ou knowledge, e y ew s udies ha e add essed con inuance in en ion in his con ex . The e o e, wi h he p oposed model, esea che s in he IS ield can adap i o sui o he si ua ions in he u u e. Thi d, he indings sugges ha indi idual pe o mance and use a e he s onges p edic o s o con inuance in en ion in he con ex o m-paymen . Ne e heless, he esul s show ha he cons uc s o o e all us and sa is ac ion mus be aken in o conside a ion when add essing con inuance in en ion (A. Bha ache jee & Lin, 2014; T. Zhou, 2014b). In he Bha ache jee (2001) model sa is ac ion is he s onges p edic o o con inuance in en ion. In ou model sa is ac ion was explo ed as a mode a o o use, indi idual pe o mance, and o e all us in con inuance in en ion. The esul s show ha sa is ac ion mode a es he ela ionship be ween use and o e all us on con inuance in en ion. In e es ingly, he mode a ion e ec in he ela ionship o o e all us on con inuance in en ion is nega i e, sugges ing ha when he le el o sa is ac ion is high, us is no an impo an ac o in luencing con inuance in en ion, showing ha use us is impo an only when he le el o sa is ac ion is low. The esul s also sugges ha he cons uc s o indi idual pe o mance and use a e pa ial media o s o use and con inuance in en ion, and o e all us and con inuance in en ion, espec i ely. This s udy demons a es ha he p oposed model p o ides suppo o he impo ance o he added cons uc s, such as he us dimension o explain con inuance in en ion. The s udy shows ha TTF is an impo an p edic o o use o m-paymen and pe cei ed indi idual pe o mance (Oli ei a, Fa ia, e al., 2014; Tam & Oli ei a, 2016b). Use is an impo an p edic o o pe cei ed indi idual pe o mance and m- paymen con inuance in en ion. O e all us , use, and pe cei ed indi idual pe o mance a e impo an p edic o s o con inuance in en ion. 86 Doc o al P og amme in In o ma ion Managemen 4.6.2. P ac ical implica ions Ou s udy has se e al p ac ical implica ions o m-paymen decision-make s and p o ide s. Ou esul s sugges ha m-paymen p o ide s seeking long- e m usage should ocus on eal- ime accessibili y, eal- ime se ices, and se ices ha a e quick and secu e in o de o enhance he ask and echnology i o m-paymen (Ouyang e al., 2017). This inding is e y impo an o decision- make s and p o ide s because when hey p o ide se ices wi h a be e i be ween ask and echnology (e.g., p o iding asks ha a e easy o use, enhancing sys em speed, educing sys em down ime, e c.), i will a ec he use o m-paymen , inc ease he pe cei ed indi idual pe o mance, and consequen ly enhance m-paymen con inuance in en ion. In his sense, i m-paymen p o ide s wan hei ac i e cus ome s (use s) o con inue using m- paymen , hey should p o ide adequa e se ices o hem. This s udy implies ha bene olence, compe ence, and in eg i y ha e a signi ican impac on o e all us (Oli ei a e al., 2017; Tam e al., 2019). This, in u n, sugges s ha m-paymen p o ide s should handle m-paymen ansac ions, be u h ul o use s, ac genuinely wi h use s, keep hei commi men s, and do hei bes o help use s, especially when i in ol es aud, o pending ansac ions. Doing hese, i will be possible o inc ease use s’ us in m-paymen . Pe cei ed bene olence occu s when he use belie es ha he m-paymen p o ide ac s in hei bes in e es , and i needed he p o ide will do hei bes o help. Pe cei ed compe ence is when he use belie es ha he m-paymen p o ide has he abili y o handle m-paymen ansac ions. Pe cei ed in eg i y occu s when he use belie es ha he m-paymen p o ide ac s since ely, is hones , and keeps o hei commi men s. The e o e, in o de o imp o e he eliabili y o use s, he m-paymen p o ide should be hones and use -o ien ed o c ea e a good image, because use s need o belie e ha he m-paymen p o ide will always ul il hei p omises. This migh encou age use s o use and con inue using m-paymen (Oli ei a e al., 2017; Pal ia, 2009). Howe e , o enhance he con inuance in en ion o use m-paymen , he p o ide should ensu e us wo hiness and p o ide quick and easy o use asks ha a e easy o unde s and in o de o ensu e pe cei ed indi idual pe o mance o use s. Conside ing 87 Doc o al P og amme in In o ma ion Managemen he mode a ing e ec s o sa is ac ion, he m-paymen p o ide should ensu e a high le el o sa is ac ion o imp o e he use o m-paymen . When he use s a e sa is ied, hey will use and also in i e o he s o use m-paymen . Based on hese indings, i is ecommended ha he m-paymen p o ide base hei ac ion plans on he de e minan s ha in luence m-paymen use s, such as ask- echnology i , us , pe cei ed indi idual pe o mance, and he use o m-paymen . 4.6.3. Limi a ions and u u e esea ch Some limi a ions exis in ou s udy. The da a we e collec ed in Mozambique, and o gene alize he applicabili y o he s udy, i is sugges ed ha u u e s udies could be conduc ed in o he A ican coun ies. The sample ep esen s a highly educa ed popula ion because we collec ed he da a a uni e si ies, bu mos m-paymen use s a e om he u al a eas o Mozambique. Fu u e s udies may es ou model in a di e en pa o he coun y and/o in ano he A ican coun y. Conside ing ha gende equali y is an in e es ing and impo an opic in he A ican con ex (Humbani and Wiese, 2018), u u e esea ch may examine he di e ences be ween gende s. This s udy is ela ed o a single ype o echnology (m-paymen ); a compa ison wi h ano he echnology (e.g., m-banking) migh e eal o he insigh s and enhance gene aliza ion. Conside ing ha cul u al ac o s play an impo an ole in an A ican con ex , u u e esea ch migh include some cul u al ac o s, such as unce ain y a oidance o indi idualism. 4.7. Conclusions The ise o m-paymen in A ica has b ough many oppo uni ies o people, and banks a e changing he way ha hey p o ide se ices o local communi ies (Ba is a & Vicen e, 2018; Humbani & Wiese, 2018). Ou s udy empi ically assesses TTF and he dimension o us o c ea e he in en ion o con inue using m-paymen amongs use s. This s udy con ibu es o he li e a u e by p o iding a heo e ical model o explain con inuance in en ion o use m-paymen . I p o ides a baseline o decision-make s and p o ide s o how echnological ac o s and us a e impo an o ensu e long- e m usage o m-paymen . The esul s demons a e ha he mos impo an ac o s o explain m-paymen con inuance in en ion a e indi idual pe o mance, use, and o e all us . 88 Doc o al P og amme in In o ma ion Managemen Indi idual pe o mance and use play impo an oles as pa ial media o s be ween use – con inuance in en ion and o e all us – con inuance in en ion. Ou esul s show ha sa is ac ion has signi ican impo ance as a mode a o be ween use and o e all us on con inuance in en ion. 89 Doc o al P og amme in In o ma ion Managemen 96 Doc o al P og amme in In o ma ion Managemen Figu e 5.1 - P oposed model The con i ma ion e e s o he expec a ions ha he use s ha e in using he IS. The use s make hei e alua ion when compa ing hei ini ial bene i s wi h he expec ed bene i s. The ECM shows ha he posi i e e ec s o con i ma ion will impac he sa is ac ion and pe cei ed use ulness (Al aimi e al., 2015; Oghuma e al., 2016). When he ini ial expec a ion o use s is con i med, i a ec s he le el o he use sa is ac ion and pe cei ed bene i s o IS (Susan o e al., 2016). Fo m-paymen , he use who con i med he expec a ions can ealize he bene i s and in luence he sa is ac ion. The e o e, we hypo hesize: H1. Con i ma ion posi i ely impac s he m-paymen pe cei ed use ulness. H2. Con i ma ion posi i ely impac s he m-paymen sa is ac ion. The de e minan ha a ec s he use s o conside ha IS enhances hei e ec i eness, pe o mance, o p oduc i i y is pe cei ed use ulness (C. M. Chiu & Wang, 2008; Da is, 1989). This means, when use s pe cei e he bene i s o he IS, he long- e m usage is ein o ced (Lee, 2010; Rez ani e al., 2017). ECM pos ula es ha when he expec ed bene i s a e con i med, he use ealizes he ad an age, which consequen ly in luences posi i ely hei sa is ac ion and long- e m usage. In ou s udy, when he m- 97 Doc o al P og amme in In o ma ion Managemen paymen use eels ha using m-paymen is use ul and enhances his o he pe o mance, he o she will be mo e sa is ied and will con inue using i (Cho, 2016; Joo e al., 2018; Shin e al., 2017). We hypo hesize: H3. Pe cei ed use ulness posi i ely impac s he m-paymen sa is ac ion. H4. Pe cei ed use ulness posi i ely impac s he m-paymen con inuance in en ion. Sa is ac ion is he ex en o which a use acqui es a posi i e eeling by using m-paymen esul ing om he usage expe iences and pe o mance ou comes (A. Bha ache jee & Lin, 2014). When he IS lea es he use sa is ied, he long- e m ela ionships become s onge (Yu e al., 2018). The ECM pos ula es ha use sa is ac ion is he cen al eason o he con inuance in en ion o IS (A Bha ache jee, 2001). Bha ache jee (2001) alida ed empi ically ha he ela ionship sa is ac ion on con inuance in en ion was he s onges . O he esea ch also demons a e ha sa is ac ion o he use s ongly in luence in en ion o con inue o use IS (Cho, 2016; Joo e al., 2018; Shin e al., 2017). In ou s udy, i he m-paymen use eels sa is ied, he long- e m usage will be gua an eed. The e o e, we hypo hesize: H5. Sa is ac ion posi i ely impac s he m-paymen con inuance in en ion. 5.3.1. Mode a ion ole o unce ain y a oidance Con i ma ion e e s o he use ’s e alua ions owa d a p oduc , se ice, o echnology, whe he i is posi i e o nega i e. The posi i e con i ma ion is when he use eaches he ini ial expec a ions, while he nega i e con i ma ion is when he use does no each he ini ial expec a ions (Al aimi e al., 2015; Oghuma e al., 2016). Use s make hei assessmen s when compa ing hei ini ial expec a ions wi h he e iciency o he p oduc , se ice, o echnology. The con i ma ion o expec a ion is when m-paymen se ices enhance he use ’s pe cei ed use ulness and sa is ac ion o he se ice (Susan o e al., 2016). Gi en he cul u al aspec o he use s, hey will ha e a i udes ha a y acco ding o he con ex . 98 Doc o al P og amme in In o ma ion Managemen Unce ain y a oidance is he a i ude ha he use akes o a oid ambiguous o unknown si ua ions (Ho s ede, 1984). A low le el o unce ain y a oidance means ha people a e no a e se o aking isks. Thus, he e is a g ea e deg ee o accep ance o new echnologies. The low-le el o unce ain y a oidance o use s o a gi en cul u al con ex is a ac o ha suppo s he ela ionship be ween he use sa is ac ion and con i ma ion. Unce ain y a oidance also in luences he ela ionship among pe cei ed use ulness and con i ma ion. The in luence o he unce ain y a oidance on hese ela ionships inc eases use ’s sa is ac ion and pe cei ed use ulness. We hypo hesize: H6a. Unce ain y a oidance posi i ely mode a es he impac o he con i ma ion on pe cei ed use ulness o m-paymen . H6b. Unce ain y a oidance posi i ely mode a es he impac o he con i ma ion on sa is ac ion o m-paymen . The ac o ha may help an indi idual o s a o unde s and he ad an ages in e ms o u iliza ion o he IS is pe cei ed use ulness (Da is, 1989). This means, when he use s pe cei e he imp o emen o using he se ices and sys em, he long- e m ela ionship is ein o ced (Lee, 2010; Rez ani e al., 2017). Howe e , o m-paymen , pe cei ed use ulness is impo an because i enables he equen use o m-paymen . Gi en he cul u al con ex o he use s, he unce ain y a oidance le el can be low o high. Low le el is an indica o ha use s will use m-paymen se ices wi h li le hesi a ion, while high-le el o unce ain y a oidance is an indica o ha use s will use m-paymen se ices wi h much hesi a ion. The use unce ain y a oidance low le el o a gi en cul u al con ex is a ac o ha a ou s he ela ionship among he sa is ac ion and pe cei ed use ulness. The unce ain y a oidance also in luences he ela ionship among pe cei ed use ulness and he in en ion o con inue o use m-paymen . We hypo hesize: H6c. Unce ain y a oidance has a posi i e impac on he mode a ion o pe cei ed use ulness on sa is ac ion o m-paymen . H6d. Unce ain y a oidance posi i ely impac s he mode a ion o pe cei ed use ulness on m-paymen con inuance in en ion. 99 Doc o al P og amme in In o ma ion Managemen Sa is ac ion is he deg ee o which an indi idual is com o able using m-paymen se ices due o usage expe iences and pe o mance ou comes, meaning ha sa is ac ion s a s o become s onge a e he use s adop he se ice o sys em (A. Bha ache jee & Lin, 2014). When use s a e sa is ied wi h he se ice, long- e m ela ionships become s onge (Yu e al., 2018). The le el o unce ain y a oidance may in luence he inc ease o dec ease o he impac on he ela ionship be ween sa is ac ion and he m- paymen con inuance in en ion. Gi en he cul u al con ex o he use s, he low le el o unce ain y a oidance will posi i ely in luence he ela ionship be ween sa is ac ion and he m-paymen con inuance in en ion. We hypo hesize: H6e. Unce ain y a oidance posi i ely impac s he mode a ion o sa is ac ion and he m-paymen con inuance in en ion. 5.4. Resea ch me hods The cu en esea ch employs a mixed-me hods app oach (So e & Hada , 2007; Venka esh e al., 2013). Fi s , was applied he quan i a i e me hod, in which he main me hod o da a collec ion was an online su ey (Al aimi e al., 2015; Tam & Oli ei a, 2016b). A ques ionnai e was cons uc ed o he su ey using a iables and i ems om he esea ch (Appendix A). The measu emen i ems o he model we e adop ed om published s udies and o unce ain a oidance we adop ed om S i e and Ka ahanna (2006). I ems o con inuance in en ion, sa is ac ion, pe cei ed use ulness, and con i ma ion we e adap ed om Bha ache jee (2001). Secondly, we applied a quali a i e me hod o iangula e and ob ain addi ional imp essions ega ding he indings, and we hus employed ield in e iews (Venka esh e al., 2013). 5.4.1. Da a Fo he quan i a i e me hod, a se en-poin scale was applied o assess he i ems, anging om 1 ( o ally disag ee) o 7 ( o ally ag ee). The ques ionnai e was managed in English and e iewed o con en alidi y by a p o essional. A p o essional ansla o ansla ed 100 Doc o al P og amme in In o ma ion Managemen o he Po uguese language, aking in o conside a ion ha he su ey was adminis e ed in Mozambique. The ques ionnai e was e e se ansla ed o he o iginal language (B islin, 1970). To alida e he ins umen s, we conduc ed a pilo es on a g oup o 30 s uden s ( hese da a we e excluded om he inal analysis). To ele a e he esponse a e, a ious s a egies we e used. Fi s , we applied he “key in o man ” p ocedu e o collec da a (Oli ei a, Thomas, e al., 2014; Pinsonneaul & K aeme , 1993), which helped in he iden i ica ion o quali ied esponden s. To boos esponses o he su ey, a ollow-up message was sen wo mon hs a e he ini ial con ac . A o al o 384 usable esponses we e ob ained om which 272 we e pa o he i s esponden s and 112 we e pa o he esponden s eminded by he ollow-up email. Compa ing he i s and las esponden g oups using Kolmogo o –Smi no (K–S), he es s indica ed a lack o non- esponse bias (Ryans, 1974). Two es s we e made o de ec he common me hod bias. Ha man’s one- ac o es (Podsako e al., 2003). The i s cons uc explains 36.7% o a iance, i acknowledges ha any o he cons uc s indi idually accoun o g ea e a iance. Second, using a ma ke a iable p ocedu e (Lindell & Whi ney, 2001), adding a heo e ically i ele an ma ke a iable in he s udy’s model, ob aining 0.040 (4.0%) as he maximum sha ed a iance wi h o he a iables; he alue can be conside ed as low (Johnson e al., 2011). We he e o e ound no signi ican common me hod bias. The da a we e collec ed be ween July 2018 and Janua y 2019. S a is ics show ha 64% o he esponden s we e p o essional wo ke s, o which 59% we e men, and ha 39% had used m-paymen one o ou imes du ing he p eceding h ee mon hs (see Table 5.1). Fo he quali a i e me hod we employed ield in e iews o explain he mo i a ions o use s ega ding he in en ion o con inue o use m-paymen (So e and Hada , 2007). To collec di e en pe spec i es o m-paymen , we in e iewed se en people, selec ed andomly, he i s and second in e iewees (I1 and I2) we e in o ma ion echnology (IT) s uden s, I3 a bank o ice , I4 an IT echnician, I5 an en ep eneu , and I6 and I7 we e uni e si y p o esso s. M-paymen use s a e iden i ied as indi iduals who use m- paymen se ices (e.g., M-Pesa o Mkesh in Mozambique). The in e iews we e conduc ed using he i ems o he heo e ical model p oposed in he s udy. 101 Doc o al P og amme in In o ma ion Managemen Table 5.1 - Sample cha ac e is ics Age < 25 129 34% 25 - 30 122 32% 31 - 40 85 22% 41 - 50 39 10% > 50 9 2% Gende Female 158 41% Male 226 59% Educa ion High school o below 91 24% Bachelo ’s deg ee 179 47% Mas e 's deg ee o highe 114 30% Employmen S uden s 99 26% P o essional wo ke s 244 64% Re i ed 1 0% Unemployed 40 10% Ma i al s a us Single 187 49% Ma ied 86 22% Di o ced 23 6% Widowed 10 3% Ma iage in ac (cohabi a ion) 75 20% Do no know answe s 3 1% M-paymen usage equency ( ime / 3 mon hs) 1 - 4 149 39% 5 - 10 98 26% > 10 137 36% 5.4.2. Da a analysis and esul s To es and assess he esea ch hypo heses o he model, we applied pa ial leas squa es - s uc u al equa ion modelling (PLS-SEM), Sma PLS 3 (Ringle e al., 2015). Ea lie esea ch has ecognized he po en ial o PLS-SEM o heo y de elopmen (Al aimi e al., 2015; Cô e-Real e al., 2020; Tam & Oli ei a, 2016b). Addi ionally, he da a do no ha e no mal dis ibu ion, he esea ch model is complex, and has no been es ed in p e ious esea ch. The e o e, PLS-SEM is app op ia e o he cu en esea ch. 5.4.3. Measu emen model The measu emen model assessmen was conduc ed by applying (1) indica o eliabili y (conside ing good loading g ea e han 0.70, and hus excluding CI4), (2) cons uc eliabili y (using composi e eliabili y (CR) indica o , good CR g ea e han 0.70), (3) 102 Doc o al P og amme in In o ma ion Managemen con e gen alidi y (using a e age a iance ex ac ed (AVE), good AVE g ea e han 0.50) (Fo nell & La cke , 1981; Hai J . e al., 2016; Hensele e al., 2009), and (4) disc iminan alidi y. The indings a e shown in Tables 5.2 and 5.3. Table 5.2 - Cons uc Reliabili y and Validi y Cons uc s AVE Composi e Reliabili y C onbach's Alpha I em Loadings - alue Con i ma ion 0.666 0.857 0.749 C1 0.820 35.782 C2 0.839 35.733 C3 0.787 27.602 Pe cei ed use ulness 0.615 0.865 0.791 PU1 0.767 25.937 PU2 0.805 31.442 PU3 0.810 37.051 PU4 0.755 22.124 Sa is ac ion 0.721 0.886 0.807 S1 0.859 55,392 S2 0.858 52.386 S3 0.832 37,340 Con inuance in en ion 0.567 0.834 0.739 CI1 0.829 44.778 CI2 0.815 30.272 CI3 0.831 45.300 Unce ain y a oidance 0.631 0.873 0.806 UA1 0.788 33.868 UA2 0.809 34.681 UA3 0.807 28,310 UA4 0.774 24,918 Table 5.3 - Fo nell-La cke C i e ion Mean STDEV C PU S CI UA Con i ma ion 4.449 1.198 0.816 Pe cei ed use ulness 4.597 1.175 0.631 0.784 Sa is ac ion 4.610 1.250 0.676 0.600 0.849 Con inuance in en ion 4.635 1.144 0.469 0.489 0.462 0.753 Unce ain y a oidance 4.923 1.186 0.346 0.357 0.281 0.459 0.794 All cons uc s me he abo e-desc ibed s anda ds, he eby ensu ing con e gence. This shows ha he ac o s can be applied o es he heo e ical model. The Fo nell-La cke es was used o assess (4) disc iminan alidi y. The AVE squa e oo o each ac o should be g ea e han he co ela ion among he ac o s (Fo nell and La cke , 1981) (see Table 5.3). Fu he , we examined c oss-loadings c i e ia (Appendix B), and he he e o ai -mono ai a io o co ela ions (HTMT) (Appendix C). The i ems show highe loading on hei co esponding cons uc s han he c oss-loadings, he eby ensu ing disc iminan alidi y (Chinn, 1998; Gö z e al., 2010). All o he HTMT alues a e below 0.90, hus concluded o disc iminan alidi y (Hensele e al., 2015). The 103 Doc o al P og amme in In o ma ion Managemen esul s indica e ha he ac o s a e s a is ically di e en and can be assessed in he s uc u al model. The measu emen model indings show a good indica o eliabili y, cons uc eliabili y, con e gence alidi y, and disc iminan alidi y. 5.4.4. S uc u al model The s uc u al model was examined a e he con i ma ion o he measu emen model, using he pa h coe icien s, a iance in la ion ac o (VIF), -s a is ic alue, and a iance explained (R2) o es he hypo heses and he cons uc s (see Figu e 5.2). The s uc u al model used 5,000 boo s ap esamples o e alua e he pa hs’ signi icance (Hensele e al., 2009). VIF was used o assess he mul icollinea i y, all cons uc s me he c i e ia wi h alues below 5, and hus i can be concluded ha he e is an absence o mul icollinea i y (Hai J . e al., 2016). Figu e 5.2 - Resea ch model (*=p<0.10; **=p<0.05; ***=p<0.01). The s uc u al model explains 43.0% o he a ia ion in pe cei ed use ulness. The con i ma ion ( " # = 0.556, p < 0.01) is s a is ically signi ican in explaining pe cei ed use ulness, suppo ing hypo hesis H1. Mo eo e , he esea ch model explains 52.4% o he a ia ion in sa is ac ion. Con i ma ion ( " # = 0.479, p < 0.01) and pe cei ed 104 Doc o al P og amme in In o ma ion Managemen use ulness ( " # = 0.261, p < 0.01) a e s a is ically signi ican in explaining sa is ac ion, suppo ing hypo heses H2 and H3. The esea ch model explains 38.2% o a ia ion in con inuance in en ion, explained by pe cei ed use ulness ( " # = 0.201, p < 0.01) and sa is ac ion ( " # = 0.224, p < 0.01), suppo ing hypo heses H4 and H5. Addi ionally, i e mode a ing models we e examined (H6a, H6b, H6c, H6d, and H6e). The esul s show ha H6a ( " # = 0.081, p < 0.05), H6b ( " # = 0.101, p < 0.10), and H6d ( " # = 0.127, p < 0.05) we e s a is ically signi ican , hus suppo ing hypo heses H6a, H6b, and H6d. The hypo heses H6c ( " # = 0.007, p < 0.10) and H6e ( " # = -0.025, p < 0.10) we e no s a is ically signi ican . 5.5. Discussion This s udy assesses he impac o cul u al dimension unce ain y a oidance on he ECM model o m-paymen . As a as we know i is he i s empi ical esea ch ha examines ECM aking in o conside a ion he mode a ion e ec o unce ain y a oidance. By applying a mixed-me hods app oach we we e able o explo e he quali a i e iew o he an eceden s o he in en ion o con inue o use m-paymen . Table 5.4 shows a summa y o he indings o he hypo hesis’s conclusions. As expec ed, all o he model ela ionships we e suppo ed (A Bha ache jee, 2001). This inding is in line wi h hose epo ed in o he s udies (X. Lin e al., 2017; Susan o e al., 2016). Fu he mo e, ou indings indica e ha con i ma ion o expec a ion posi i ely impac s he sa is ac ion and pe cei ed use ulness o m-paymen . This means ha he expe iences using m- paymen we e posi i e and use expec a ions we e con i med (Cheng, 2014; Tam e al., 2020). In a quali a i e iew, he in e iewees I1, I2, and I5 (IT s uden s and en ep eneu ) highligh ed he pe cep ion o use ulness and he eeling o sa is ac ion wi h m-paymen , s a ing ha hey can easily pay o school and ood expenses. The I5 (en ep eneu ) highligh ed he ease o selling p oduc s, because mos cus ome s chose o pay wi h m-paymen because o he acili ies, and hey did no use he adi ional me hod ia cash, mainly because o he COVID-19. They also epo ed he acili ies o ca y ou ansac ions, anywhe e and any ime, o e ing g ea e a ailabili y, as and sa e, simpli ying hei daily li es, especially o hose who li e a om u ban cen es. 105 Doc o al P og amme in In o ma ion Managemen The pe cei ed use ulness posi i ely impac s he sa is ac ion and con inuance in en ion. Meaning ha when use s pe cei e pe o mance, e ec i eness, o bene i s on m- paymen , sa is ac ion will be con i med, consequen ly in luencing con inuance in en ion. All he in e iewees ema ked ex ensi ely abou he use ulness o m- paymen , in di e en aspec s; o he IT s uden s (I1 and I2) i is use ul o paying ees and pu chasing ood; o banke s i is use ul because i educes he numbe o people needing o use bank se ices; and also, o pe sonal expenses, goods, and se ices such as gas, wa e , and elec ici y. Mo eo e , he in e iewees a gued ha m-paymen has g ea u ili y, especially nowadays wi h he COVID-19 pandemic, acili a ing ansac ions om home o anywhe e and any ime, educing dis ances, and sa ing ime – hus, allowing o alloca e ime o o he ac i i ies and helping o educe he isk o con amina ion by he co ona i us. The esul s indica e ha sa is ac ion posi i ely in luences con inuance in en ion. This means ha when he use is happy and deligh ed, hey will con inue o use m-paymen . The in e iewees I1, I2 and I5 (IT s uden s and en ep eneu ) g ea ly emphasized hei sa is ac ion wi h m-paymen and a gued ha i is a e y impo an ac o o he in en ion o con inue o use m-paymen . They men ioned ease o use and he abili y o use i anywhe e and any ime, especially a his ime o he COVID-19 pandemic. They indica ed hei sa is ac ion because hey can pay basic expenses such as wa e , ene gy, and TV om home, he eby educing a el and he numbe o co ona i us in ec ions. The m-paymen makes use ’s li es much easie . I is widely used a he na ional le el. The pandemic accele a ed use, mainly because o go e nmen -imposed con ainmen ules and because he use s should a oid ouching objec s such as ATM, POS, doo knobs, e c. Table 5.4 - Hypo heses esul s Hypo heses Independen Cons uc → Dependen cons uc Findings (β) P alue Conclusion H1 Con i ma ion → Pe cei ed use ulness 0.556 < 0.01 Suppo ed H2 Con i ma ion → Sa is ac ion 0.479 < 0.01 Suppo ed H3 Pe cei ed use ulness → Sa is ac ion 0.261 < 0.01 Suppo ed H4 Pe cei ed use ulness → Con inuance in en ion 0.201 < 0.01 Suppo ed H5 Sa is ac ion → Con inuance in en ion 0.224 < 0.01 Suppo ed 112 Doc o al P og amme in In o ma ion Managemen Chap e 6 – Conclusions 6.1. Summa y o esul s The main objec i e o his pape is o de e mine he main de e minan s o he con inuance in en ion o use m-paymen . We conduc ed ou quan i a i e s udies (Chap e 2 o Chap e 5), one li e a u e e iew and h ee empi ical s udies ha analyse he e ec s o di e en ac o s and models on he in en ion o con inue using m- paymen . Table 6.1 summa izes he s a is ical esul s o he ela ionships o he di e en models analysed in he s udies. Table 6.1 - Rela ionships analyses in all he s udies Independen Cons uc Dependen cons uc Chap e 2 Chap e 3 Chap e 4 Chap e 5 Con inuance in en ion Con inuance beha iou 0.375 A i ude Con inuance in en ion 0.441 Flow Con inuance in en ion 0.358 Subjec i e no ms Con inuance in en ion 0.179 Hedonic alue Con inuance in en ion 0.437 U ili a ian alue Con inuance in en ion 0.242 A ec i e commi men Con inuance in en ion 0.556 T us Con inuance in en ion 0.239 Pe cei ed enjoymen Con inuance in en ion 0.187 Pe o mance Con inuance in en ion 0.241 Habi Con inuance in en ion 0.255 Pe cei ed beha iou con ol Con inuance in en ion 0.296 Use Con inuance in en ion 0.272 0.328 Sa is ac ion Con inuance in en ion 0.416 0.162 0.224 Pe cei ed use ulness Con inuance in en ion 0.285 0.016 (ns) 0.201 Indi idual pe o mance Con inuance in en ion 0.310 0.350 O e all us Con inuance in en ion 0.099 Se ice quali y Sa is ac ion 0.067 (ns) Sys em quali y Sa is ac ion 0.279 0.065 (ns) Con i ma ion Sa is ac ion 0.431 0.389 0.479 Use Sa is ac ion 0.240 Pe cei ed use ulness Sa is ac ion 0.248 0.136 0.261 In o ma ion quali y Sa is ac ion 0.248 0.152 Discon i ma ion Sa is ac ion 0.576 Pe cei ed ease o use Sa is ac ion 0.188 113 Doc o al P og amme in In o ma ion Managemen Pe cei ed enjoymen Sa is ac ion 0.251 Con i ma ion Pe cei ed use ulness 0.504 0.566 0.556 Discon i ma ion Pe cei ed use ulness 0.174 Pe cei ed ease o use Pe cei ed use ulness 0.327 Sys em quali y Con i ma ion 0.116 (ns) Se ice quali y Con i ma ion 0.275 In o ma ion quali y Con i ma ion 0.359 Use Indi idual pe o mance 0.416 0.275 Sa is ac ion Indi idual pe o mance 0.146 Task echnology i Indi idual pe o mance 0.500 Sys em quali y Use 0.088 (ns) Se ice quali y Use 0.225 In o ma ion quali y Use 0.400 Task echnology i Use 0.342 O e all us Use 0.315 Bene olence O e all us 0.333 Compe ence O e all us 0.246 In eg i y O e all us 0.247 Rela ional capi al A ec i e commi men 0.258 U ili a ian alue A ec i e commi men 0.112 Hedonic alue A ec i e commi men 0.339 Sa is ac ion A i ude 0.481 Pe cei ed use ulness A i ude 0.408 Con i ma ion Pe cei ed enjoymen 0.622 Con i ma ion Pe cei ed ease o use 0.458 No e: ns (no s a is ically signi ican ) The esul s o he li e a u e e iew show ha he mos used echnologies we e e- lea ning and social ne wo k se ices. The mos used heo e ical models o s udy he in en ion o con inue o use we e ECM, TAM, and ECT. Addi ionally, ew s udies used a single heo y, mos s udies in eg a ed mo e han one heo y. A heo e ical model was p esen ed based on he signi ican cons uc s om me a-analysis and bes p edic o om weigh analysis. O he 600 ela ionships collec ed om 115 s udies, 60 ela ionships we e analysed h ee o mo e imes. O hese ela ionships, 31 we e classi ied as “bes p edic o s” and 24 we e classi ied as “p omising p edic o s”. The mos s udied egions we e Eas Asia, No h Ame ica, Eu ope, Middle Eas , Sou h Ame ica, and Sou hwes Asia. No s udies we e ound o he A ican egion. 114 Doc o al P og amme in In o ma ion Managemen In he i s empi ical s udy, he in en ion o con inue using m-paymen in an A ican con ex was analysed, combining ECM wi h D&M ISSM. The esul s show ha he use o m-paymen , sa is ac ion, and pe cei ed indi idual pe o mance a e he mos impo an ac o s o explain he con inuance in en ion o use m-paymen . In o ma ion quali y and se ice quali y posi i ely impac he use and con i ma ion o he expec a ions. In o ma ion quali y, use, and con i ma ion o expec a ions posi i ely impac use sa is ac ion. In he second empi ical s udy, in en ion o con inue using m-paymen was analysed by in eg a ing TTF, o e all us , and ECM. The esul s show ha he mos impo an ac o s explaining he in en ion o con inue using m-paymen a e indi idual pe o mance, use, and o e all us . Indi idual pe o mance and use play impo an oles as pa ial media o s be ween use-con inuance in en ion and o e all us - con inuance in en ion. Ou esul s show ha sa is ac ion has signi ican impo ance as a mode a o be ween use and o e all us on con inuance in en ion. In he las empi ical s udy, he impac o cul u e on he in en ion o con inue using m- paymen was analysed using a mixed-me hods app oach based on quan i a i e da a and ield in e iews. The esul s show ha he cul u al ac o unce ain y a oidance mode a es he ela ionships be ween con i ma ion on sa is ac ion and con i ma ion on pe cei ed use ulness, and pe cei ed use ulness on con inuance in en ion. In a quali a i e iew, mos o he in e iewees highligh ed he impo ance o m-paymen in hei daily li es, especially o hose who li e a om u ban cen es. Thus, i was ound ha he cul u e plays a signi ican ole in ensu ing use ulness, sa is ac ion, and he con inuance in en ion o use m-paymen . 6.2. Con ibu ions 6.2.1. Implica ions o heo y This disse a ion p o ides se e al con ibu ions o esea ch. The quan i a i e app oach o he li e a u e e iew con ibu es o esea ch by p o iding a mo e concise, clea e , and ex ensi e image o he cons uc s e alua ed in p e ious s udies in a ious subjec a eas on con inuance in en ion o use IS om yea s 2001 o 2017, se ing as he basis 115 Doc o al P og amme in In o ma ion Managemen o u u e esea ch and con ibu ing o esea che s o accu a ely selec he cons uc s o be included in esea ch models o assess m-paymen con inuance in en ion. Mo eo e , he esul s show ha he e is a wide a ie y o cons uc s coming om di e en heo ies and con ex s ha can signi ican ly in luence con inuance in en ion, hus demons a ing ha di e en heo ies and sel -cons uc s can be in eg a ed in o he ISCI con ex . In Chap e 3 we p oposed he joining o D&M ISSM and he ECM model, wi h he aim o iden i ying an eceden s ha ocus on sa is ac ion, indi idual pe o mance, and con inuance in en ion. The indings p o ide suppo o he impo ance o he added cons uc s om DeLone and Mclean (2003) in use s’ con inuance in en ion o use m- paymen . As ano he heo e ical implica ion he model alida es IS con inuance in en ion heo y o he case o m-paymen use in Mozambique, unlike p e ious s udies ha ocused on de eloped economies (X. Chen & Li, 2017; Fan e al., 2018a). In Chap e 4 we ad anced he body o knowledge o m-paymen by p oposing he in es iga ion o con inuance in en ion o use m-paymen in eg a ing ECM, TTF, and o e all us . The esul s indica e ha bo h TTF and o e all us a e impo an and should be conside ed when e alua ing he con inuance in en ion o use m-paymen . To he bes o ou knowledge e y ew s udies ha e add essed con inuance in en ion in he A ican con ex . Thus, wi h he p oposed model esea che s in he IS ield can adap i o sui o he si ua ions in he u u e. In Chap e 5 we p esen ed he e ec s o he unce ain y a oidance cul u al dimension in he ECM model using he m-paymen case, applying mixed-me hods, and we e able o o e a holis ic iew o he an eceden s o he con inuance in en ion. The cu en esea ch en iches he body o li e a u e on ECM by ex ending he scope o IS con inuance in en ion o he m-paymen con ex and e eals he mode a ing ole o unce ain y a oidance in p edic ing con inuance in en ion. Conside ing ha Mozambique is s ill consolida ing he adop ion phase o m-paymen (Ba is a & Vicen e, 2014, 2018), he cul u al ac o plays a signi ican ole in he mode a ion o he con inuous use o m-paymen . 116 Doc o al P og amme in In o ma ion Managemen 6.2.2. Implica ions o p ac ice The esul s o his disse a ion ha e aluable p ac ical implica ions o manage s and decision make s o ensu e he use s’ e en ion and long- e m usage o m-paymen . Fi s , acco ding o he indings o ou esea ch i is c ucial o ecognize he bes p edic o s o con inuance in en ion o use an IS, o be e design and implemen a ion o he IS (A. Bha ache jee, 2001; Shao, 2018; Yu e al., 2018). The con inuance in en ion o use an IS has been s udied in di e en coun ies wi h di e en cul u es. The e o e, manage s should ha e di e en managing s a egies o ensu e he sa is ac ion o he use s and long- e m usage o he IS (L. Zhang e al., 2012). Manage s and decision make s should p o ide all he necessa y in o ma ion, such as use guides, ad e isemen s, and lye s ha explain he se ices and unc ionali ies o he IS, o accele a e he unde s anding o he se ices. Second, he m-paymen p o ide s should ensu e ha he in o ma ion a ailable is co ec , up- o-da e, and use ul o he use . In addi ion, m-paymen p o ide s should help he use s whene e hey need help, should ensu e ha use s ha e a good expe ience wi h m-paymen , should p o ide se ices ha exceed use s’ expec a ions, and ensu e ha he use s can use m-paymen easily. This sugges s ha he m-paymen p o ide s should cons an ly imp o e he m-paymen in ea u es ela ed o sa e y, ease o use, and in o ma ion, o p o ide a well-s uc u ed sys em ha is easy o na iga e and has use ul in o ma ion. Thi d, o p omo e long- e m usage o m-paymen , p o ide s should ocus on eal- ime accessibili y, eal- ime se ices, and se ices ha a e quick and secu e in o de o enhance he ask and echnology i o m-paymen (Ouyang e al., 2017). In o he wo ds, i he p o ide ’s goals a e o in luence he ac i e cus ome s (use s) o con inue using m-paymen , hey should p o ide adequa e se ices o hem. Mo eo e , he m-paymen p o ide s should handle m-paymen ansac ions, be u h ul o use s, ac genuinely wi h use s, keep hei commi men s, and do hei bes o help use s, especially in si ua ions in ol ing aud, o pending ansac ions. Fou h, he indings ega ding he impac o cul u e on con inuance in en ion o use m- paymen indica e ha manage s should p o ide use s wi h guidelines including all he ins uc ions, p ocedu es, and egula ions o acili a e and help he use s o ope a e wi h 117 Doc o al P og amme in In o ma ion Managemen m-paymen . Wi h he suppo o ins uc ion manuals, use s will mo e easily and equen ly use m-paymen , and o e ime will come o pe cei e i s bene i s and he inc ease o e iciency i p o ides. Unde s anding cul u e could be impo an in he de elopmen and managemen o solu ions o m-paymen . Fo example, wi h he high le el o unce ain y a oidance, manage s could ocus on mi iga ing unce ain y and isk, and ensu ing he comp ehension o he se ice o use s in o de o posi i ely in luence con inuous use. 6.3. Limi a ions and u u e esea ch The p esen disse a ion con ains se e al limi a ions. Rega ding he li e a u e e iew, we excluded ce ain s udies because o he una ailabili y o hei quan i a i e da a, o because hey we e quali a i e. Including hese s udies could gene a e ele an in o ma ion in e ms o he signi icance o he cons uc s. We used me a-essen ials, a me a-analysis ool. This ool has limi a ions, and conside ing ha me a-essen ials is no able o pe o m mo e ad anced analyses wi h linea models o s uc u ed equa ion modeling (Rhee e al., 2015), u u e esea che s should conside a mo e ad anced ool o p o ide mo e insigh s and a di e en app oach o esea ch. Rega ding he h ee empi ical s udies p esen ed in Chap e s 2, 3, 4, and 5, he da a we e collec ed in uni e si ies, he eby a ac ing a highly educa ed popula ion. Howe e , mos o he m- paymen use s in Mozambique a e no uni e si y s uden s, and u u e s udies could collec he da a in di e en en i onmen s such as ma ke s, companies, and communi ies, and in a di e en pa o he coun y o in ano he A ican coun y. This esea ch is ela ed o a single ype o echnology, m-paymen , a compa ison wi h ano he echnology (e.g., m-banking) migh e eal o he insigh s and enhance gene aliza ion. We used he cul u al dimension o unce ain y a oidance, conside ing ha he e a e o he cul u al dimensions, i is ecommended in u u e s udies o use all he dimensions, wi h he pu pose o b inging mo e insigh . Ou models we e p oposed o and alida ed in Mozambique, and o gene alize he applicabili y o he s udy, and o ully ep esen all po en ial m-paymen s use s in he A ican con ex he samples o u u e esea ch could be conduc ed in o he A ican coun ies. 118 Doc o al P og amme in In o ma ion Managemen 119 Doc o al P og amme in In o ma ion Managemen Chap e 7 - Bibliog aphy Abbas, H. A., & Hamdy, H. I. (2015). De e minan s o con inuance in en ion ac o in Kuwai communica ion ma ke : Case s udy o Zain-Kuwai . Compu e s in Human Beha io , 49, 648–657. h ps://doi.o g/10.1016/j.chb.2015.03.035 A shan, S., & Sha i , A. (2016). Accep ance o mobile banking amewo k in Pakis an. Telema ics and In o ma ics, 33(2), 370–387. h ps://doi.o g/10.1016/j. ele.2015.09.005 Ahmad, W., & Sun, J. (2018). An eceden s o SMMA con inuance in en ion in wo cul u ally di e se coun ies: An empi ical examina ion. Jou nal o Global In o ma ion Technology Managemen , 21(1), 45–68. h ps://doi.o g/10.1080/1097198X.2018.1423840 Aiken, L. S., Wes , S. G., & Reno, R. R. (1991). Mul iple eg ession: Tes ing and in e p e ing in e ac ions. Sage. Ajzen, I. (1985). F om in en ions o ac ions: A heo y o planned beha io . In Ac ion Con ol (pp. 11–39). h ps://doi.o g/10.1007/978-3-642-69746-3_2 Al-Debei, M. M., Al-Lozi, E., & Papaza ei opoulou, A. (2013). Why people keep coming back o Facebook: Explaining and p edic ing con inuance pa icipa ion om an ex ended heo y o planned beha iou pe spec i e. Decision Suppo Sys ems, 55(1), 43–54. h ps://doi.o g/10.1016/j.dss.2012.12.032 Albash awi, M., & Mo iwalla, L. (2019). P i acy and pe sonaliza ion in con inued usage in en ion o mobile banking: An in eg a i e pe spec i e. In o ma ion Sys ems F on ie s, 21(5), 1031–1043. h ps://doi.o g/10.1007/s10796-017-9814-7 Albuque que, J. P. de, Diniz, E. H., & Ce ne , A. K. (2016). Mobile paymen s: a scoping s udy o he li e a u e and issues o u u e esea ch. In o ma ion De elopmen , 32(3), 527–553. h ps://doi.o g/10.1177/0266666914557338 Al aimi, K. M., Zo, H., & Ciganek, A. P. (2015). Unde s anding he MOOCs con inuance: The ole o openness and epu a ion. Compu e s and Educa ion, 80, 28–38. h ps://doi.o g/10.1016/j.compedu.2014.08.006 Amo oso, D., & Lim, R. (2017). The media ing e ec s o habi on con inuance in en ion. In e na ional Jou nal o In o ma ion Managemen , 37(6), 693–702. h ps://doi.o g/10.1016/j.ijin omg .2017.05.003 Asamoah, D., Takieddine, S., & Amedo u, M. (2020). Examining he e ec o mobile money ans e (MMT) capabili ies on business g ow h and de elopmen impac . In o ma ion Technology o De elopmen , 26(1), 146–161. h ps://doi.o g/10.1080/02681102.2019.1599798 Baabdullah, A. M., Abdallah, A., Rana, N. P., & Kizgin, H. (2019). Consume use o mobile banking (M-Banking) in Saudi A abia: Towa ds an in eg a ed model. In e na ional Jou nal o In o ma ion Managemen , 44, 38–52. h ps://doi.o g/10.1016/j.ijin omg .2018.09.002 120 Doc o al P og amme in In o ma ion Managemen Bankole, F. O., & Bankole, O. O. (2017). The e ec s o cul u al dimension on ICT inno a ion: Empi ical analysis o mobile phone se ices. Telema ics and In o ma ics, 34(2), 490–505. h ps://doi.o g/10.1016/j. ele.2016.08.004 Bap is a, G., & Oli ei a, T. (2015). Unde s anding mobile banking: The uni ied heo y o accep ance and use o echnology combined wi h cul u al mode a o s. Compu e s in Human Beha io , 50, 418–430. h ps://doi.o g/10.1016/j.chb.2015.04.024 Bap is a, G., & Oli ei a, T. (2016). A weigh and a me a-analysis on mobile banking accep ance esea ch. Compu e s in Human Beha io , 63, 480–489. h ps://doi.o g/10.1016/j.chb.2016.05.074 Ba nes, S., & Böh inge , M. (2011). Modeling use con inuance beha iou in mic oblogging se ices: The case o wi e . Jou nal o Compu e In o ma ion Sys ems, 51(4), 1–10. h ps://doi.o g/10.1080/08874417.2011.11645496 Ba nes, S. J. (2011). Unde s anding use con inuance in i ual wo lds: Empi ical es o a esea ch model. In o ma ion and Managemen , 48(8), 313–319. h ps://doi.o g/10.1016/j.im.2011.08.004 Ba is a, C., & Vicen e, P. C. (2014). In oducing mobile money in u al Mozambique: Ini ial e idence om a ield expe imen . No a A ica, 1301. h ps://doi.o g/10.2139/ss n.2384561 Ba is a, C., & Vicen e, P. C. (2018). Imp o ing access o sa ings h ough mobile money: Expe imen al e idence om smallholde a me s in Mozambique. No a A ica, 1705. Bay amus a, M., & Nasi , V. A. (2016). A ad o u u e o IT?: A comp ehensi e li e a u e e iew on he cloud compu ing esea ch. In e na ional Jou nal o In o ma ion Managemen , 36(4), 635–644. h ps://doi.o g/10.1016/j.ijin omg .2016.04.006 Belanche, D., Casaló, L. V., Fla ián, C., & Schepe s, J. (2014). T us ans e in he con inued usage o public e-se ices. In o ma ion and Managemen , 51(6), 627– 640. h ps://doi.o g/10.1016/j.im.2014.05.016 Bha ache jee, A., & Ba a , A. (2011). In o ma ion Technology Con inuance Resea ch: Cu en S a e and Fu u e Di ec ions. Asia Paci ic Jou nal o In o ma ion Sys ems, 21(2), 1–18. h ps://doi.o g/10.2307/3250921 Bha ache jee, A., & Lin, C.-P. (2014). A uni ied model o IT con inuance: Th ee complemen a y pe spec i es and c osso e e ec s. Eu opean Jou nal o In o ma ion Sys ems, 24(4), 1–10. h ps://doi.o g/10.1057/ejis.2013.36 Bha ache jee, A., & P emkuma , G. (2004). Unde s anding changes in belie and a i ude owa d in o ma ion echnology usage: A heo e ical model and longi udinal es . MIS Qua e ly, 28(2), 229–254. h ps://doi.o g/10.2307/25148634 Bha ache jee, A. (2001). Unde s anding in o ma ion sys ems con inuance: An 121 Doc o al P og amme in In o ma ion Managemen expec a ion con i ma ion model. MIS Qua e ly, 25(3), 351–370. h ps://doi.o g/10.2307/3250921 Bha ache jee, Anol, Pe ols, J., & San o d, C. (2008). In o ma ion echnology con inuance: A heo e ic ex ension and empi ical es . Jou nal o Compu e In o ma ion Sys ems, 49(1), 17–26. h ps://doi.o g/10.1080/08874417.2008.11645302 Bøe, T., Gulb andsen, B., & Sø ebø, O. (2015). How o s imula e he con inued use o ICT in highe educa ion: In eg a ing in o ma ion sys ems con inuance heo y and agency heo y. Compu e s in Human Beha io , 50, 375–384. h ps://doi.o g/10.1016/j.chb.2015.03.084 B islin, R. W. (1970). Back- ansla ion o c oss-cul u al esea ch. Jou nal o C oss- Cul u al Psychology, 1(3), 185–216. h ps://doi.o g/10.1177/135910457000100301 Budia djo, E. K., Pamenan, G., Hidayan o, A. N., Meyliana, & Co iyan i, E. (2017). The impac o knowledge managemen sys em quali y on he usage con inui y and ecommenda ion in en ion. Knowledge Managemen & E-Lea ning, 9(2), 200– 224. Budne , P., Fische , M., Rosenk anz, C., Bas en, D., & Te lecki, L. (2017). In o ma ion sys em con inuance in en ion in he con ex o ne wo k e ec s and eemium business models: A eplica ion s udy o cloud se ices in Ge many. AIS T ansac ions on Replica ion Resea ch, 3(Decembe ), 1–13. h ps://doi.o g/10.17705/1a .00019 Ca illo, K., Sco na acca, E., & Za, S. (2015). The ole o media dependency in p edic ing con inuance in en ion o use ubiqui ous media sys ems. In o ma ion and Managemen , 54(3), 317–335. h ps://doi.o g/10.1016/j.im.2016.09.002 Ca illo, K., Sco na acca, E., & Za, S. (2017). The ole o media dependency in p edic ing con inuance in en ion o use ubiqui ous media sys ems. In o ma ion and Managemen , 54(3), 317–335. h ps://doi.o g/10.1016/j.im.2016.09.002 Chang, I.-C., Liu, C.-C., & Chen, K. (2014). The e ec s o hedonic/u ili a ian expec a ions and social in luence on con inuance in en ion o play online games. In e ne Resea ch, 24(1), 21–45. h ps://doi.o g/10.1108/In R-02-2012-0025 Chang, Y. P., & Zhu, D. H. (2012). The ole o pe cei ed social capi al and low expe ience in building use s’ con inuance in en ion o social ne wo king si es in China. Compu e s in Human Beha io , 28(3), 995–1001. h ps://doi.o g/10.1016/j.chb.2012.01.001 Chen, C. P., Lai, H. M., & Ho, C. Y. (2015). Why do eache s con inue o use eaching blogs? he oles o pe cei ed olun a iness and habi . Compu e s and Educa ion, 82(1), 236–249. h ps://doi.o g/10.1016/j.compedu.2014.11.017 Chen, I. Y. L. (2007). The ac o s in luencing membe s’ con inuance in en ions in p o essional i ual communi ies a longi udinal s udy. Jou nal o In o ma ion 128 Doc o al P og amme in In o ma ion Managemen I inedo, P. (2017). Examining s uden s’ in en ion o con inue using blogs o lea ning: Pe spec i es om echnology accep ance, mo i a ional, and social-cogni i e amewo ks. Compu e s in Human Beha io , 72, 189–199. h ps://doi.o g/10.1016/j.chb.2016.12.049 Iman, N. (2018). Is mobile paymen s ill ele an in he in ech e a? Elec onic Comme ce Resea ch and Applica ions, 30(05), 72–82. h ps://doi.o g/10.1016/j.ele ap.2018.05.009 INE. (2019). Resul ados de ini i os do VI ecenseamen o ge al da população e habi ação 2017. Ins i u o Nacional de Es a ís ica (INE). h p://www.ine.go .mz/ Jack, W., & Su i, T. (2011). Mobile money: The economics o M-PESA. Jeya aj, A., Ro man, J. W., & Laci y, M. C. (2006). A e iew o he p edic o s, linkages, and biases in IT inno a ion adop ion esea ch. Jou nal o In o ma ion Technology, 21(1), 1–23. h ps://doi.o g/10.1057/palg a e.ji .2000056 Jin, X. L., Lee, M. K. O., & Cheung, C. M. K. (2010). P edic ing con inuance in online communi ies: Model de elopmen and empi ical es . Beha iou and In o ma ion Technology, 29(4), 383–394. h ps://doi.o g/10.1080/01449290903398190 Johns, G. (2006). The essen ial impac o con ex on o ganiza ional beha io . Academy o Managemen Jou nal, 31(2), 386–408. h ps://doi.o g/10.5465/am .2006.20208687 Johnson, R. E., Rosen, C. C., & Dju dje ic, E. (2011). Assessing he impac o common me hod a iance on highe o de mul idimensional cons uc s. Jou nal o Applied Psychology, 96(4), 744–761. h ps://doi.o g/10.1037/a0021504 Joo, Y. J., Kim, N., & Kim, N. H. (2016). Fac o s p edic ing online uni e si y s uden s’ use o a mobile lea ning managemen sys em. Educa ional Technology Resea ch and De elopmen , 64(4), 611–630. h ps://doi.o g/10.1007/s11423-016-9436-7 Joo, Y. J., So, H. J., & Kim, N. H. (2018). Examina ion o ela ionships among s uden s’ sel -de e mina ion, echnology accep ance, sa is ac ion, and con inuance in en ion o use K-MOOCs. Compu e s and Educa ion, 122(01), 260–272. h ps://doi.o g/10.1016/j.compedu.2018.01.003 Kaewki ipong, L., Chen, C. C., & Rac ham, P. (2016). Using social media o en ich in o ma ion sys ems ield ip expe iences: S uden s’ sa is ac ion and con inuance in en ions. Compu e s in Human Beha io , 63, 256–263. h ps://doi.o g/10.1016/j.chb.2016.05.030 Kang, S. (2014). Fac o s in luencing in en ion o mobile applica ion use. In e na ional Jou nal o Mobile Communica ions, 12(4), 360–379. h ps://doi.o g/10.1504/IJMC.2014.063653 Kang, Y. J., & Lee, W. J. (2015). Sel -cus omiza ion o online se ice en i onmen s by use s and i s e ec on hei con inuance in en ion. Se ice Business, 9(2), 321– 342. h ps://doi.o g/10.1007/s11628-014-0229-y 129 Doc o al P og amme in In o ma ion Managemen Ka i, T., Salo, M., & F ank, L. (2020). Role o si ua ional con ex in use con inuance a e c i ical exe gaming inciden s. In o ma ion Sys ems Jou nal, 30(3), 596–633. h ps://doi.o g/10.1111/isj.12273 Ka jaluo o, H., Shaikh, A. A., Saa ijä i, H., & Sa aniemi, S. (2019). How pe cei ed alue d i es he use o mobile inancial se ices apps. In e na ional Jou nal o In o ma ion Managemen , 47, 252–261. h ps://doi.o g/10.1016/j.ijin omg .2018.08.014 Kau , P., Dhi , A., Singh, N., Sahu, G., & Almo ai i, M. (2020). An inno a ion esis ance heo y pe spec i e on mobile paymen solu ions. Jou nal o Re ailing and Consume Se ices, 55(June 2019), 102059. h ps://doi.o g/10.1016/j.j e conse .2020.102059 Khalilzadeh, J., Oz u k, A. B., & Bilgihan, A. (2017). Secu i y- ela ed ac o s in ex ended UTAUT model o NFC based mobile paymen in he es au an indus y. Compu e s in Human Beha io , 70, 460–474. h ps://doi.o g/10.1016/j.chb.2017.01.001 Kim, B. (2010). An empi ical in es iga ion o mobile da a se ice con inuance: Inco po a ing he heo y o planned beha io in o he expec a ion-con i ma ion model. Expe Sys ems wi h Applica ions, 37(10), 7033–7039. h ps://doi.o g/10.1016/j.eswa.2010.03.015 Kim, B. (2011). Unde s anding an eceden s o con inuance in en ion in social- ne wo king se ices. Cybe psychology, Beha io and Social Ne wo king, 14(4), 199–205. h ps://doi.o g/10.1089/cybe .2010.0009 Kim, G., Shin, B., & Lee, H. G. (2009). Unde s anding dynamics be ween ini ial us and usage in en ions o mobile banking. In o ma ion Sys ems Jou nal, 19(3), 283– 311. h ps://doi.o g/10.1111/j.1365-2575.2007.00269.x Kim, Y., & Pe e son, R. A. (2017). A Me a-analysis o online us ela ionships in E- comme ce. Jou nal o In e ac i e Ma ke ing, 38, 44–54. h ps://doi.o g/10.1016/j.in ma .2017.01.001 Koksal, M. H. (2016). The in en ions o Lebanese consume s o adop mobile banking. In e na ional Jou nal o Bank Ma ke ing, 34(3), 327–346. h ps://doi.o g/10.1108/IJBM-03-2015-0025 Koloseni, D., & Manda i, H. (2017). Why mobile money use s keep inc easing? in es iga ing he con inuance usage o mobile money se ices in Tanzania. 26(2), 117–145. K asno a, H., Vel i, N. F., Eling, N., & Buxmann, P. (2017). Why men and women con inue o use social ne wo king si es: The ole o gende di e ences. Jou nal o S a egic In o ma ion Sys ems, 26(4), 261–284. h ps://doi.o g/10.1016/j.jsis.2017.01.004 K oebe , A. L., & Pa sons, T. (1958). The concep s o cul u e and o social sys em. Ame ican Sociological Re iew, 23(5), 582–583. 130 Doc o al P og amme in In o ma ion Managemen Ku, E. C. S., & Chen, C.-D. (2015). Cul i a ing a elle s’ e isi in en ion o e- ou ism se ice: The mode a ing e ec o websi e in e ac i i y. Beha iou and In o ma ion Technology, 34(5), 465–478. h ps://doi.o g/10.1080/0144929X.2014.978376 Kujala, S., Mugge, R., & Mi on-Sha z, T. (2017). The ole o expec a ions in se ice e alua ion: A longi udinal s udy o a p oximi y mobile paymen se ice. In e na ional Jou nal o Human Compu e S udies, 98, 51–61. h ps://doi.o g/10.1016/j.ijhcs.2016.09.011 Lagna, A., & Ra ishanka , M. N. (2021). Making he wo ld a be e place wi h in ech esea ch. In o ma ion Sys ems Jou nal, 1–42. h ps://doi.o g/10.1111/isj.12333 Lank on, N., & McKnigh , H. (2012). Examining wo expec a ion discon i ma ion heo y models: Assimila ion and asymme y e ec s. Jou nal o he Associa ion o In o ma ion Sys ems, 13(2), 88–115. La sen, T. J., Sø ebø, A. M., & Sø ebø, Ø. (2009). The ole o ask- echnology i as use s’ mo i a ion o con inue in o ma ion sys em use. Compu e s in Human Beha io , 25(3), 778–784. h ps://doi.o g/10.1016/j.chb.2009.02.006 Laukkanen, T. (2007). In e ne s mobile banking: compa ing cus ome alue pe cep ions. Business P ocess Managemen Jou nal, 13(6), 788–797. h ps://doi.o g/10.1108/14637150710834550 Lee, I., Choi, B., Kim, J., & Hong, S.-J. (2007a). Cul u e- echnology i : E ec s o cul u al cha ac e is ics on he pos -adop ion belie s o mobile in e ne use s. In e na ional Jou nal o Elec onic Comme ce, 11(4), 11–51. h ps://doi.o g/10.2753/JEC1086-4415110401 Lee, I., Choi, B., Kim, J., & Hong, S.-J. (2007b). Cul u e-Technology Fi : E ec s o Cul u al Cha ac e is ics on he Pos -Adop ion Belie s o Mobile In e ne Use s. In e na ional Jou nal o Elec onic Comme ce, 11(4), 11–51. h ps://doi.o g/10.2753/JEC1086-4415110401 Lee, J. K., Pa k, J., G ego , S., & Yoon, V. (2021). Axioma ic heo ies and imp o ing he ele ance o in o ma ion sys ems esea ch. In o ma ion Sys ems Jou nal, 32(1), 147–171. h ps://doi.o g/10.1287/is e.2020.0958 Lee, M.-C. (2010). Explaining and p edic ing use s’ con inuance in en ion owa d e- lea ning: An ex ension o he expec a ion-con i ma ion model. Compu e s and Educa ion, 54(2), 506–516. h ps://doi.o g/10.1016/j.compedu.2009.09.002 Lee, Y., & Kwon, O. (2011). In imacy, amilia i y and con inuance in en ion: An ex ended expec a ion-con i ma ion model in web-based se ices. Elec onic Comme ce Resea ch and Applica ions, 10(3), 342–357. h ps://doi.o g/10.1016/j.ele ap.2010.11.005 Leh o, T., & Oinas-Kukkonen, H. (2015). Explaining and p edic ing pe cei ed e ec i eness and use con inuance in en ion o a beha iou change suppo sys em o weigh loss. Beha iou & In o ma ion Technology, 34(2), 176–189. h ps://doi.o g/10.1080/0144929X.2013.866162 131 Doc o al P og amme in In o ma ion Managemen Leidne , D. E., & Kaywo h, T. (2006). Re iew: A e iew o cul u e in in o ma ion sys ems esea ch: Towa d a heo y o in o ma ion echnology cul u e con lic . MIS Qua e ly, 30(2), 357–399. Liang, T.-P., Ling, Y.-L., Yeh, Y.-H., & Lin, B. (2013). Con ex ual ac o s and con inuance in en ion o mobile se ices. In e na ional Jou nal o Mobile Communica ions, 11(4), 313–329. h ps://doi.o g/10.1504/IJMC.2013.055746 Liao, C., Pal ia, P., & Lin, H.-N. (2006). The oles o habi and websi e quali y in e- comme ce. In e na ional Jou nal o In o ma ion Managemen , 26(6), 469–483. h ps://doi.o g/10.1016/j.ijin omg .2006.09.001 Liao, Z., & Shi, X. (2017). Web unc ionali y, web con en , in o ma ion secu i y, and online ou ism se ice con inuance. Jou nal o Re ailing and Consume Se ices, 39, 258–263. h ps://doi.o g/10.1016/j.j e conse .2017.06.003 Liébana-Cabanillas, F., & La a-Rubio, J. (2017). P edic i e and explana o y modeling ega ding adop ion o mobile paymen sys ems. Technological Fo ecas ing and Social Change, 120, 32–40. h ps://doi.o g/10.1016/j. ech o e.2017.04.002 Liébana-cabanillas, F., Sánchez-Fe nández, J., & Muñoz-Lei a, F. (2014). The mode a ing e ec o expe ience in he adop ion o mobile paymen ools in i ual social ne wo ks: The m-paymen accep ance model in i ual social ne wo ks (MPAM-VSN). In e na ional Jou nal o In o ma ion Managemen , 34, 151–166. h ps://doi.o g/10.1016/j.ijin omg .2013.12.006 Liébana-cabanillas, F., Sánchez-Fe nández, J., & Muñoz-Lei a, F. (2018). A global app oach o he analysis o use beha io in mobile paymen sys ems in he new elec onic en i onmen . Se ice Business, 12(1), 25–64. h ps://doi.o g/10.1007/s11628-017-0336-7 Liébana-Cabanillas, F ancisco, Molinillo, S., & Ruiz-Mon añez, M. (2019). To use o no o use, ha is he ques ion: Analysis o he de e mining ac o s o using NFC mobile paymen sys ems in public anspo a ion. Technological Fo ecas ing and Social Change, 139, 266–276. h ps://doi.o g/10.1016/j. ech o e.2018.11.012 Liébana-Cabanillas, F ancisco, Singh, N., Kalinic, Z., & Ca ajal-T ujillo, E. (2021). Examining he de e minan s o con inuance in en ion o use and he mode a ing e ec o he gende and age o use s o NFC mobile paymen s: a mul i-analy ical app oach. In o ma ion Technology and Managemen , 22(2), 133–161. h ps://doi.o g/10.1007/s10799-021-00328-6 Limayem, M., & Cheung, C. M. K. (2008). Unde s anding in o ma ion sys ems con inuance: The case o In e ne -based lea ning echnologies. In o ma ion and Managemen , 45(4), 227–232. h ps://doi.o g/10.1016/j.im.2008.02.005 Limayem, M., & Cheung, C. M. K. (2011). P edic ing he con inued use o In e ne - based lea ning echnologies: he ole o habi . Beha iou & In o ma ion Technology, 30(1), 91–99. h ps://doi.o g/10.1080/0144929X.2010.490956 Limayem, M., Hi , S. G., & Cheung, C. M. K. (2007). How habi limi s he p edic i e 132 Doc o al P og amme in In o ma ion Managemen powe o in en ion: The case o in o ma ion sys ems con inuance. MIS Qua e ly, 31(4), 705–737. Lin, C.-P., & Bha ache jee, A. (2010). Ex ending echnology usage models o in e ac i e hedonic echnologies: A heo e ical model and empi ical es . In o ma ion Sys ems Jou nal, 20(2), 163–181. h ps://doi.o g/10.1111/j.1365- 2575.2007.00265.x Lin, H., Fan, W., & Chau, P. Y. K. (2014). De e minan s o use s’ con inuance o social ne wo king si es: A sel - egula ion pe spec i e. In o ma ion and Managemen , 51(5), 595–603. h ps://doi.o g/10.1016/j.im.2014.03.010 Lin, K.-M. (2016). Unde s anding unde g adua es’ p oblems om de e minan s o Facebook con inuance in en ion. Beha iou & In o ma ion Technology, 35(9), 693–705. h ps://doi.o g/10.1080/0144929X.2016.1177114 Lin, K. M. (2011). E-Lea ning con inuance in en ion: Mode a ing e ec s o use e- lea ning expe ience. Compu e s and Educa ion, 56(2), 515–526. h ps://doi.o g/10.1016/j.compedu.2010.09.017 Lin, K. M., Chen, N. S., & Fang, K. (2011). Unde s anding e-lea ning con inuance in en ion: A nega i e c i ical inciden s pe spec i e. Beha iou and In o ma ion Technology, 30(1), 77–89. h ps://doi.o g/10.1080/01449291003752948 Lin, X., Fea he man, M., & Sa ke , S. (2017). Unde s anding ac o s a ec ing use s’ social ne wo king si e con inuance: A gende di e ence pe spec i e. In o ma ion and Managemen , 54(3), 383–395. h ps://doi.o g/10.1016/j.im.2016.09.004 Lindell, M. K., & Whi ney, D. J. (2001). Accoun ing o common me hod a iance in c oss-sec ional esea ch designs. Jou nal o Applied Psychology, 86(1), 114–121. h ps://doi.o g/10.1037/0021-9010.86.1.114 Lipsey, M., & Wilson, D. (2001). P ac ical Me a-analysis. Cen e s o Teaching and Technology - Book Lib a y, 49. Low y, P. B., Gaskin, J. E., & Moody, G. D. (2015). P oposing he mul imo i e in o ma ion sys ems con inuance model ( MISC ) o be e explain end-use sys em e alua ions and con inuance in en ions. Jou nal o he Associa ion o In o ma ion, 16(7), 515–579. Lu, H.-P., & Lee, M.-R. (2011). Expe ience di e ences and con inuance in en ion o blog sha ing. Beha iou & In o ma ion Technology, 31(11), 1–15. h ps://doi.o g/10.1080/0144929X.2011.611822 Lu, J., Liu, C., & Wei, J. (2017). How impo an a e enjoymen and mobili y o mobile applica ions? Jou nal o Compu e In o ma ion Sys ems, 57(1), 1–12. h ps://doi.o g/10.1080/08874417.2016.1181463 Lu, J., Wei, J., Yu, C., & Liu, C. (2017). How do pos -usage ac o s and espoused cul u al alues impac mobile paymen con inua ion? Beha iou & In o ma ion Technology, 36(2), 140–164. h ps://doi.o g/10.1080/0144929X.2016.1208773 133 Doc o al P og amme in In o ma ion Managemen Makina, D. (2017). In oduc ion o he inancial se ices in A ica special issue. A ican Jou nal o Economic and Managemen S udies, 8(1), 2–7. h ps://doi.o g/10.1108/AJEMS-03-2017-149 Mohammadya i, S., & Singh, H. (2015). Unde s anding he e ec o e-lea ning on indi idual pe o mance: The ole o digi al li e acy. Compu e s & Educa ion, 82, 11–25. h ps://doi.o g/h ps://doi.o g/10.1016/j.compedu.2014.10.025 Mo osan, C., & DeF anco, A. (2016). I ’s abou ime: Re isi ing UTAUT2 o examine consume s’ in en ions o use NFC mobile paymen s in ho els. In e na ional Jou nal o Hospi ali y Managemen , 53, 17–29. h ps://doi.o g/10.1016/j.ijhm.2015.11.003 Mouakke , S. (2015). Fac o s in luencing con inuance in en ion o use social ne wo k si es: The Facebook case. Compu e s in Human Beha io , 53, 102–110. h ps://doi.o g/10.1016/j.chb.2015.06.045 Mouakke , S. (2018). The ole o pe sonali y ai s in mo i a ing use s’ con inuance in en ion owa ds Facebook: Gende di e ences. Jou nal o High Technology Managemen Resea ch, 29(1), 124–140. h ps://doi.o g/10.1016/j.hi ech.2016.10.003 Muñoz-Lei a, F., Mayo-Muñoz, X., & De la Hoz-Co ea, A. (2018). Adop ion o homesha ing pla o ms: a c oss-cul u al s udy. Jou nal o Hospi ali y and Tou ism Insigh s, 1(3), 220–239. h ps://doi.o g/10.1108/jh i-01-2018-0007 Naba i, A., Tagha i-Fa d, M. T., Hana izadeh, P., & Tagh a, M. R. (2016). In o ma ion Technology Con inuance In en ion: A Sys ema ic Li e a u e Re iew. In e na ional Jou nal o E-Business Resea ch, 12(1), 58–95. h ps://doi.o g/10.4018/IJEBR.2016010104 Nascimen o, B., Oli ei a, T., & Tam, C. (2018). Wea able echnology: Wha explains con inuance in en ion in sma wa ches? Jou nal o Re ailing and Consume Se ices, 43, 157–169. h ps://doi.o g/h ps://doi.o g/10.1016/j.j e conse .2018.03.017 Odoom, R., & Kosiba, J. P. (2020). Mobile money usage and con inuance in en ion among mic o en e p ises in an eme ging ma ke – he media ing ole o agen c edibili y. Jou nal o Sys ems and In o ma ion Technology, 22(4), 97–117. h ps://doi.o g/10.1108/JSIT-03-2019-0062 Oghuma, A. P., Libaque-Saenz, C. F., Wong, S. F., & Chang, Y. (2016). An expec a ion-con i ma ion model o con inuance in en ion o use mobile ins an messaging. Telema ics and In o ma ics, 33(1), 34–47. h ps://doi.o g/10.1016/j. ele.2015.05.006 Oli ei a, T., Alhinho, M., Ri a, P., & Dhillon, G. (2017). Modelling and es ing consume us dimensions in e-comme ce. Compu e s in Human Beha io , 71, 153–164. h ps://doi.o g/10.1016/j.chb.2017.01.050 Oli ei a, T., Fa ia, M., Thomas, M. A., & Popo ič, A. (2014). Ex ending he 134 Doc o al P og amme in In o ma ion Managemen unde s anding o mobile banking adop ion: When UTAUT mee s TTF and ITM. In e na ional Jou nal o In o ma ion Managemen , 34(5), 689–703. h ps://doi.o g/10.1016/j.ijin omg .2014.06.004 Oli ei a, T., Thomas, M., Bap is a, G., & Campos, F. (2016). Mobile paymen : Unde s anding he de e minan s o cus ome adop ion and in en ion o ecommend he echnology. Compu e s in Human Beha io , 61, 404–414. h ps://doi.o g/10.1016/j.chb.2016.03.030 Oli ei a, T., Thomas, M., & Espadanal, M. (2014). Assessing he de e minan s o cloud compu ing adop ion: An analysis o he manu ac u ing and se ices sec o s. In o ma ion and Managemen , 51(5), 497–510. h ps://doi.o g/10.1016/j.im.2014.03.006 Oli e , R. L. (1986). A cogni i e model o he an eceden s and consequences o sa is ac ion decisions. Jou nal o Ma ke ing Resea ch, 17(4), 460–469. h ps://doi.o g/10.2307/3150499 Omigie, N. O., Zo, H., Rho, J. J., & Ciganek, A. P. (2017). Cus ome p e-adop ion choice beha io o M-PESA mobile inancial se ices: Ex ending he heo y o consump ion alues. Indus ial Managemen and Da a Sys ems, 117(5), 910–926. h ps://doi.o g/10.1108/IMDS-06-2016-0228 O igão, M., Macome, E., & Vicen e, P. (2015). Elec onic paymen in Mozambique: A baseline on hei adop ion in Mapu o and Ma ola. No a A ica, 1503. Ouyang, Y., Tang, C., Rong, W., Zhang, L., Yin, C., & Xiong, Z. (2017). Task- echnology i awa e expec a ion-con i ma ion model owa ds unde s anding o MOOCs con inued usage. 50 h Hawaii In e na ional Con e ence on Sys em Sciences, 174–183. Pal, A., He a h, T., De’, R., & Rao, H. R. (2020). Con ex ual acili a o s and ba ie s in luencing he con inued use o mobile paymen se ices in a de eloping coun y: insigh s om adop e s in India. In o ma ion Technology o De elopmen , 26(2), 394–420. h ps://doi.o g/10.1080/02681102.2019.1701969 Pal ia, P. (2009). The ole o us in e-comme ce ela ional exchange: A uni ied model. In o ma ion and Managemen , 46(4), 213–220. h ps://doi.o g/10.1016/j.im.2009.02.003 Pa k, J. K., Ahn, J., Tha isay, T., & Ren, T. (2019). Examining he ole o anxie y and social in luence in mul i-bene i s o mobile paymen se ice. Jou nal o Re ailing and Consume Se ices, 47, 140–149. h ps://doi.o g/10.1016/j.j e conse .2018.11.015 Pa k, M., Jun, J., & Pa k, H. (2017). Unde s anding mobile paymen se ice con inuous use in en ion: An expec a ion - Con i ma ion model and ine ia. Quali y Inno a ion P ospe i y, 21(3), 78–94. h ps://doi.o g/10.12776/QIP.V21I3.983 Pe ei a, F. A. de M., Ramos, A. S. M., And ade, A. P. V. de, & Oli ei a, B. M. K. de. (2015). Con inued usage o e-lea ning: Expec a ions and pe o mance. In Jou nal o In o ma ion Sys ems and Technology Managemen (Vol. 12, Issue 2, pp. 333– 135 Doc o al P og amme in In o ma ion Managemen 350). scielo. h ps://doi.o g/10.4301/S1807-17752015000200008 Pe saud, A., & Azha , I. (2012). Inno a i e mobile ma ke ing ia sma phones: A e consume s eady? Ma ke ing In elligence & Planning, 30(4), 418–443. h ps://doi.o g/10.1108/02634501211231883 Pinsonneaul , A., & K aeme , K. L. (1993). Su ey esea ch me hodology in managemen in o ma ion sys ems: An assessmen . Jou nal o Managemen In o ma ion Sys ems, 10(2), 75–105. h ps://doi.o g/10.1080/07421222.1993.11518001 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. h ps://doi.o g/10.1037/0021-9010.88.5.879 P emkuma , G., & Bha ache jee, A. (2008). Explaining in o ma ion echnology usage: A es o compe ing models. Omega, 36(1), 64–75. h ps://doi.o g/10.1016/j.omega.2005.12.002 Rahi, S., Khan, M. M., & Alghizzawi, M. (2020). Ex ension o echnology con inuance heo y (TCT) wi h ask echnology i (TTF) in he con ex o In e ne banking use con inuance in en ion. In e na ional Jou nal o Quali y and Reliabili y Managemen , ahead-o -p(ahead-o -p in ). h ps://doi.o g/10.1108/IJQRM-03- 2020-0074 Rahman, S. A., Alam, M. M. D., & Taghizadeh, S. K. (2020). Do mobile inancial se ices ensu e he subjec i e well-being o mic o-en ep eneu s? An in es iga ion applying UTAUT2 model. In o ma ion Technology o De elopmen , 26(2), 421–444. h ps://doi.o g/10.1080/02681102.2019.1643278 Raman, P., & Aashish, K. (2021). To con inue o no o con inue: a s uc u al analysis o an eceden s o mobile paymen sys ems in India. In e na ional Jou nal o Bank Ma ke ing, 39(2), 242–271. h ps://doi.o g/10.1108/IJBM-04-2020-0167 Rana, N. P., Dwi edi, Y. K., & Williams, M. D. (2015). A me a-analysis o exis ing esea ch on ci izen adop ion o e-go e nmen . In o ma ion Sys ems F on ie s, 17(3), 547–563. h ps://doi.o g/10.1007/s10796-013-9431-z Rez ani, A., Khos a i, P., & Dong, L. (2017). Mo i a ing use s owa d con inued usage o in o ma ion sys ems: Sel -de e mina ion heo y pe spec i e. Compu e s in Human Beha io , 76, 263–275. h ps://doi.o g/10.1016/j.chb.2017.07.032 Rhee, H. an, Suu mond, R., & Hak, T. (2015). Use manual o me a-essen ials: Wo kbooks o me a-analyses. The Ne he lands: E asmus Resea ch Ins i u e o Managemen , 1–49. Ringle, C. M., Wende, S., & Becke , J.-M. (2015). Sma PLS 3. Bönnings ed : Sma PLS. h p://www.sma pls.com Ringle, C. M., Wende, S., & Will, A. (2005). Sma PLS 2.0. M3. Hambu g: Sma PLS. 136 Doc o al P og amme in In o ma ion Managemen Roca, J. C., Chiu, C. M., & Ma ínez, F. J. (2006). Unde s anding e-lea ning con inuance in en ion: An ex ension o he Technology Accep ance Model. In e na ional Jou nal o Human Compu e S udies, 64(8), 683–696. h ps://doi.o g/10.1016/j.ijhcs.2006.01.003 Rosen hal, R., & Rubin, D. B. (1982). A simple, gene al pu pose display o magni ude o expe imen al e ec . Jou nal o Educa ional Psychology, 74(2), 166–169. h ps://doi.o g/10.1037/0022-0663.74.2.166 Ryans, A. B. (1974). Es ima ing consume p e e ences o a new du able b and in an es ablished p oduc class. Jou nal o Ma ke ing Resea ch, 11(4), 434–443. h ps://doi.o g/10.2307/3151290 Ryu, H. S. (2018). Wha makes use s willing o hesi an o use Fin ech?: he mode a ing e ec o use ype. Indus ial Managemen and Da a Sys ems, 118(3), 541–569. h ps://doi.o g/10.1108/IMDS-07-2017-0325 Salehan, M., Kim, D. J., & Lee, J. N. (2018). A e he e any ela ionships be ween echnology and cul u al alues? A coun y-le el end s udy o he associa ion be ween in o ma ion communica ion echnology and cul u al alues. In o ma ion and Managemen , 55(6), 725–745. h ps://doi.o g/10.1016/j.im.2018.03.003 Sällbe g, H., & Beng sson, L. (2016). Compu e and sma phone con inuance in en ion: A mo i a ional model. Jou nal o Compu e In o ma ion Sys ems, 56(4), 321–330. h ps://doi.o g/10.1080/08874417.2016.1164007 San ini, F. D. O., Ladei a, W. J., Sampaio, C. H., Pe in, M. G., & Dolci, P. C. (2019). A me a-analy ical s udy o echnological accep ance in banking con ex s. In e na ional Jou nal o Bank Ma ke ing, 37(3), 755–774. h ps://doi.o g/10.1108/IJBM-04-2018-0110 San ini, F. de O., Ladei a, W. J., Sampaio, C. H., Pe in, M. G., & Dolci, P. C. (2019). P opensi y o echnological adop ion: an analysis o e ec s size in he banking sec o . Beha iou and In o ma ion Technology, 1–15. Schmid ., F. L., & Hun e , J. E. (2004). Me hods o me a-analysis: Co ec ing e o and bias in esea ch indings. In Sage. Schmid , F. L., & Hun e , J. E. (2016). Me hods o Me a-Analysis: Co ec ing E o and Bias in Resea ch Findings. In Sage. Seol, S., Lee, H., Yu, J., & Zo, H. (2016). Con inuance usage o co po a e SNS pages: A communica i e ecology pe spec i e. In o ma ion & Managemen , 53(6), 740– 751. h ps://doi.o g/h ps://doi.o g/10.1016/j.im.2016.02.010 Shaikh, A. A., & Ka jaluo o, H. (2015). Making he mos o in o ma ion echnology & sys ems usage: A li e a u e e iew, amewo k and u u e esea ch agenda. Compu e s in Human Beha io , 49, 541–566. h ps://doi.o g/10.1016/j.chb.2015.03.059 Shao, Z. (2018). Examining he impac mechanism o social psychological mo i a ions on indi iduals’ con inuance in en ion o MOOCs-The mode a ing e ec o 137 Doc o al P og amme in In o ma ion Managemen gende . In e ne Resea ch, 28(1), 232–250. h ps://doi.o g/10.1108/In R-11-2016- 0335 Shao, Z., Zhang, L., Li, X., & Guo, Y. (2019). An eceden s o us and con inuance in en ion in mobile paymen pla o ms: The mode a ing e ec o gende . Elec onic Comme ce Resea ch and Applica ions, 33(11), 100823. h ps://doi.o g/10.1016/j.ele ap.2018.100823 Sha ma, K. S., & Sha ma, M. (2019). Examining he ole o us and quali y dimensions in he ac ual usage o mobile banking se ices: An empi ical in es iga ion. In e na ional Jou nal o In o ma ion Managemen , 44, 65–75. h ps://doi.o g/10.1016/j.ijin omg .2018.09.013 Shiau, W. L., Yuan, Y., Pu, X., Ray, S., & Chen, C. C. (2020). Unde s anding in ech con inuance: pe spec i es om sel -e icacy and ECT-IS heo ies. Indus ial Managemen and Da a Sys ems, 120(9), 1659–1689. h ps://doi.o g/10.1108/IMDS-02-2020-0069 Shin, D. H., & Biocca, F. (2017). Heal h expe ience model o pe sonal in o ma ics: The case o a quan i ied sel . Compu e s in Human Beha io , 69, 62–74. h ps://doi.o g/10.1016/j.chb.2016.12.019 Shin, D. H., Lee, S., & Hwang, Y. (2017). How do c edibili y and u ili y play in he use expe ience o heal h in o ma ics se ices? Compu e s in Human Beha io , 67, 292–302. h ps://doi.o g/10.1016/j.chb.2016.11.007 Singh, N., Sinha, N., & Liébana-cabanillas, F. J. (2020). De e mining ac o s in he adop ion and ecommenda ion o mobile walle se ices in India : Analysis o he e ec o inno a i eness, s ess o use and social in luence. In e na ional Jou nal o In o ma ion Managemen , 50, 191–205. h ps://doi.o g/10.1016/j.ijin omg .2019.05.022 Sinha, M., Maj a, H., Hu chins, J., & Saxena, R. (2019). Mobile paymen s in India: The p i acy ac o . In e na ional Jou nal o Bank Ma ke ing, 37(1), 192–209. h ps://doi.o g/10.1108/IJBM-05-2017-0099 Smi h, M. L. (2006). O e coming heo y-p ac ice inconsis encies: C i ical ealism and in o ma ion sys ems esea ch. In o ma ion and O ganiza ion, 16(3), 191–211. h ps://doi.o g/10.1016/j.in oando g.2005.10.003 So e , P., & Hada , I. (2007). Applying on ology-based ules o concep ual modeling: A e lec ion on modeling decision making. Eu opean Jou nal o In o ma ion Sys ems, 16(5), 599–611. h ps://doi.o g/10.1057/palg a e.ejis.3000683 Song, J., Kim, J., & Cho, K. (2017). Unde s anding use s’ con inuance in en ions o use sma -connec ed spo s p oduc s. Spo Managemen Re iew, 21(5), 477–490. h p://dx.doi.o g/10.1016/j.sm .2017.10.004 Sø ebø, Ø., Hal a i, H., Gulli, V. F., & K is iansen, R. (2009). The ole o sel - de e mina ion heo y in explaining eache s’ mo i a ion o con inue o use e- lea ning echnology. Compu e s and Educa ion, 53(4), 1177–1187.