The effect of international opportunity recognition processes on problem-solving competence: how does past negative entrepreneurial experience matter?

OeconomiA cope nicana Volume 13 Issue 2 June 2022 p-ISSN 2083-1277, e-ISSN 2353-1827 www.oeconomia.pl Copy igh © Ins y u Badań Gospoda czych / Ins i u e o Economic Resea ch (Poland) This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep o- duc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. ORIGINAL ARTICLE Ci a ion: Te án-Yépez, E., Jiménez-Cas illo, D., & Sánchez-Pé ez, M. (2022). The e ec o in e na ional oppo uni y ecogni ion p ocesses on p oblem-sol ing compe ence: how does pas nega i e en ep eneu ial expe ience ma e ? Oeconomia Cope nicana, 13(2), 541–579. doi: 10.24136/oc.2022.017 Con ac o co esponding au ho : Edua do Te án-Yépez, edua do. e [email protected] A icle his o y: Recei ed: 8.02.2022; Accep ed: 1.06.2022; Published online: 30.06.2022 Edua do Te án-Yépez Uni e si y o Alme ia (CIMEDES Resea ch Cen e and ceiA3), Spain o cid.o g/0000-0002-1260-2477 Da id Jiménez-Cas illo Uni e si y o Alme ia (CIMEDES Resea ch Cen e and ceiA3), Spain o cid.o g/0000-0002-2302-1002 Manuel Sánchez-Pé ez Uni e si y o Alme ia (CIMEDES Resea ch Cen e and ceiA3), Spain o cid.o g/0000-0003-3709-3389 The e ec o in e na ional oppo uni y ecogni ion p ocesses on p oblem-sol ing compe ence: how does pas nega i e en ep eneu ial expe ience ma e ? JEL Classi ica ion: D91; F23; L26; M16 Keywo ds: in e na ional oppo uni y ecogni ion; p oblem-sol ing speed; p oblem-sol ing c ea i i y; pas nega i e en ep eneu ial expe ience; ma ke oppo uni y iden i ica ion Abs ac Resea ch backg ound: Li le esea ch add esses how iden i ying an oppo uni y in he in e na- ional ma ke , whe he h ough ac i e sea ch o se endipi ous disco e y, may ha e implica ions o he subsequen ges a ion beha io s. An eme ging body o esea ch sugges s add essing his de iciency by ocusing on he cogni i e and expe ience-based ac o s ha in e na ional en ep e- neu s use o de elop an oppo uni y in he o eign ma ke once iden i ied. Pu pose o he a icle: The aim o his s udy is o explo e he ole o ac i e and se endipi ous in e na ional oppo uni y ecogni ion (IOR) in en ep eneu s’ p oblem-sol ing skills (p oblem- Oeconomia Cope nicana, 13(2), 541–579 542 sol ing speed and c ea i i y), as well as he mode a ing e ec o pas nega i e en ep eneu ial expe ience (PNE) on such ela ionships. Me hods: This s udy used he su ey da a collec ion me hod h ough an online sel -adminis e ed ques ionnai e. Pa ial leas squa e s uc u al equa ion modeling (PLS-SEM) me hod was used o analyze he esul s o e a sample o Spanish in e na ional en ep eneu s. Findings & alue added: The esul s show ha en ep eneu s ha ha e iden i ied an in e na ion- al oppo uni y (IO) ac i ely a e be e equipped o sol e p oblems speedily han hose ha ha e disco e ed an IO se endipi ously. In con as , en ep eneu s who iden i ied an IO se endipi ously a e be e equipped o sol e p oblems c ea i ely. Fu he mo e, he impac o ac i e sea ch on p oblem-sol ing speed is g ea e o en ep eneu s wi hou PNE, and he impac o ac i e sea ch on p oblem-sol ing c ea i i y is g ea e o en ep eneu s wi h PNE. Also, i is ound ha PNE does no mode a e se endipi ous IOR ela ionship wi h p oblem-sol ing compe ence. These indings a e ele an as hey indica e ha he IOR p ocesses ha e implica ions o he subsequen ges a ion beha io s. Thus, he alue added o his s udy is he combina ion o cogni i e and expe- ience-based ac o s o he in e na ional en ep eneu o en ich he link be ween IOR p ocesses and IO de elopmen . These a e wo in e wined and in e dependen p ocesses, which, howe e , ha e been sca cely s udied as a whole. The esul s o his esea ch help in e na ional en ep e- neu s o unde s and how and why he way hey ecognized an IO and ha ing o no PNE in luence hei p oblem-sol ing skills du ing he IO de elopmen phase. In oduc ion The iden i ica ion and de elopmen o in e na ional ma ke oppo uni ies a e c i ical o i ms’ in e na ional expansion since, wi hou ecogni ion and de elopmen , in e na ional oppo uni ies (IOs) canno be exploi ed (Chand a e al., 2009). As en ep eneu s o en play a c ucial ole in hese ac i i ies, one o he opics ha ha e ecen ly a ac ed he in e es o in e - na ional en ep eneu ship (IE) academics and p ac i ione s is o unde s and how en ep eneu s ecognize and de elop IOs (Che y e al., 2018; Di G ego io e al., 2021; Hilme sson e al., 2021a). D awing on oppo uni y iden i ica ion heo y (A dich ili e al., 2003), he li e a u e a gues ha en ep eneu s ecognize IOs h ough ac i e sea ch o se endipi ous disco e y (Taba es e al., 2021). Ac i e sea ch claims o he iden i ica ion o IOs h ough a sys ema ic sea ch o in o ma ion and egula analysis o he in e na ional ma ke (Hilme sson e al., 2021a; Tuomisalo, 2019), while in se endipi ous disco e y, IOs a e ecognized h ough en ep eneu s’ ale ness and sensi i i y du ing passi e sea ch si ua- ions (Kiss e al., 2020; Kon inen & Ojala, 2011). Al hough an e o has been made in ecen yea s o s udy he in e na ional oppo uni y ecogni ion (IOR) phenomenon in g ea e dep h, esea ch has mainly been ocused on how and why en ep eneu s iden i y IOs ei he by ac i e sea ch o se endip- i ous disco e y (Galan & To sein, 2021; Zae a ian e al., 2016). Howe e , he IOR is no hing mo e han he beginning o he IE p ocess and, a e ecognizing an IO h ough pu pose ul sea ch o o ui ous ci cums ance, Oeconomia Cope nicana, 13(2), 541–579 543 in e na ional en ep eneu s en e in o a de elopmen s age whe e he IO is e alua ed and whe e hey ypically ha e o sol e a whole se o p oblems ha esul om he in e na ionaliza ion challenges (Di G ego io e al., 2021; Hilme sson e al., 2021b; Taba es e al., 2021). The di e ences be ween ac i e sea ch and se endipi ous disco e y a ise a ques ion ega ding whe he he way en ep eneu s iden i y a new in e na- ional ma ke oppo uni y can a ec how hey ace he eme gence o p ob- lems and challenges ha a e common du ing in e na ional business ac i i- ies (Chand a e al., 2020; Ji e al., 2016) and can limi oppo uni y de el- opmen and exploi a ion. This issue is pa icula ly ele an conce ning en- ep eneu s’ p oblem-sol ing skills as hey a e seen as key compe i i e en ep eneu ial and manage ial weapons o success ully de elop and ex- ploi an IO (Mos a iz & Goh, 2018; Taba es e al., 2021). P oblem-sol ing skills, e e ed o as p oblem-sol ing speed and p oblem-sol ing c ea i i y (A uahene-Gima & Wei, 2011), help en ep eneu s o o e come he chal- lenges ha can hinde he oppo uni y o ope a e o e seas, as hey imply de eloping imely, no el, and cos -e ec i e solu ions ha lead o he ex- ploi a ion o he IO (Chand a e al., 2020; Mainela & Puhakka, 2009). Al - hough ecen s udies sugges ha p oblem-sol ing skills a e a c i ical con- e e ha can ansla e ecognized IOs in o aluable IOs o exploi (Mos a iz & Goh, 2018; Zucchella, 2021), he e is a lack o esea ch ha e i ies how each IOR app oach (ac i e IOR and se endipi ous IOR) is ela ed o p oblem-sol ing compe ence and why can he e be p oblem- sol ing compe ence he e ogenei y among in e na ional en ep eneu s. Acco dingly, his s udy analyzes whe he ac i e and se endipi ous IOR dis inc i ely de e mines p oblem-sol ing compe ence du ing in e na ional oppo uni y de elopmen (IOD) p ocess. The heo e ical logic o expec ing hese di e ences may be ound in p io li e a u e which sugges s ha di - e ences in IOR may subs an ially a ec IOD and en ep eneu s’ skills and beha io . Indeed, academics ha e a gued ha ac i e and se endipi ous IOR di e en ly impac s on oppo uni y implemen a ion quickness, in e na ional- iza ion speed, and isk- aking p opensi y (Ci a egna e al., 2014; De Cle cq e al., 2012; Hilme sson & Papaioannou, 2015). In his s udy, we heo ize ha ac i e in e na ional en ep eneu s — hose who iden i y IOs h ough a pu pose ul p ocess — end o sol e p oblems as e han hei passi e coun e pa s. On he con a y, we pos ula e ha passi e in e na ional en e- p eneu s — hose who iden i y IOs h ough a se endipi ous p ocess — end o sol e p oblems mo e c ea i ely han ac i e in e na ional en ep eneu s. In speci ying he model, we d aw on he en ep eneu ial cogni ion heo y (Mi chell e al., 2002) ollowing he ecen call o IE schola s o apply psy- chological heo ies o o e a iche unde s anding o indi iduals’ he e oge- Oeconomia Cope nicana, 13(2), 541–579 544 nei y, and hus o explo e he e ec o en ep eneu ial mindse (e.g., knowledge, expe ience, cogni ion) on hei beha io and skills (Wach & Głodowska, 2021; Zucchella, 2021). Also, p e ious li e a u e a gues ha ce ain indi idual ac o s would im- pac he IOD p ocess. In pa icula , he na u e o p e ious expe ience (posi- i e o nega i e) has been sugges ed as a c i ical ac o a ec ing he speed and c ea i i y o decisions (La uen e e al., 2019; Shephe d e al., 2015). Schola s highligh ha in e na ional en ep eneu s wi h pas nega i e en e- p eneu ial expe ience (PNE) di e om hose wi hou PNE since hey show di e en beha io s and skills (La uen e e al., 2019; Ucbasa an e al., 2013). In his s udy, we pay pa icula a en ion o PNE as an impo an expe ience-based ac o o he e ec i eness o he ela ionship be ween IOR and p oblem-sol ing skills. In pa icula , we pos ula e ha ega dless o he ype o IOR p ocess, he p oblem-sol ing speed o in e na ional en- ep eneu s wi h PNE is lesse han hose wi hou PNE. Also, we p opose ha he e e se migh be expec ed o p oblem-sol ing c ea i i y; ega d- less o he ype o IOR s yle, he p oblem-sol ing c ea i i y o in e na ional en ep eneu s wi h PNE is g ea e han hose wi hou PNE. The esul s o pa ial leas squa e s uc u al equa ion modeling (PLS- SEM) analysis using da a om Spanish in e na ional en ep eneu s la gely suppo ou hypo heses. The s udy p o ides wo main con ibu ions o he IE li e a u e. Fi s , by disen angling he ela ionship be ween ac i e and se endipi ous IOR and p oblem-sol ing skills, we con ibu e by demon- s a ing ha each IOR pe spec i e di e en ly impac s p oblem-sol ing speed and c ea i i y. Thus, we con i m ha he way an en ep eneu ecog- nizes an IO is a sou ce o he e ogenei y ha explains why some in e na- ional en ep eneu s sol e p oblems ha a ise du ing IOD as e and mo e c ea i ely han o he s. Second, we also con ibu e by p o ing ha he na- u e o pas expe iences, in ou case PNE, can be a ac o ha mode a es he ela ionship be ween ac i e IOR and p oblem-sol ing skills, bu no he ela ionship be ween se endipi ous IOR and p oblem-sol ing skills. Be- sides, we show ha PNE has a dicho omous e ec on ac i e in e na ional en ep eneu s ha ei he imp o es o hinde s hei p oblem-sol ing skills. The es o he pape is o ganized as ollows. In he ollowing sec ion, we ou line he heo e ical backg ound o he esea ch and e eal he hy- po heses de elopmen . Nex , we elucida e he esea ch me hodology and p esen he esul s o he s udy. Finally, we discuss hese esul s and con- clude he s udy by p o iding manage ial implica ions, limi a ions and u- u e esea ch di ec ions. Oeconomia Cope nicana, 13(2), 541–579 545 Li e a u e e iew and hypo heses de elopmen Rele ance o IOR pe spec i es How do in e na ional en ep eneu s disco e IOs? Based on oppo uni y iden i ica ion heo y (A dich ili e al., 2003), p e ious li e a u e sugges s wo undamen al possibili ies: ac i e IOR and se endipi ous IOR (C ick & Spence, 2005; Te án-Yépez e al., 2021; Zae a ian e al., 2016). In ac i e IOR, en ep eneu s iden i y IOs h ough sys ema ic pu sui and scanning he en i onmen o gaps in he in e na ional ma ke (Che y e al., 2018; Hilme sson & Papaioannou, 2015). This IOR s yle is based on a a ional, delibe a e, analy ic, and ule-based sea ch and ga he ing o in o ma ion ha conduc s o conscious easoning (Oyson & Whi ake , 2015; Tuomisalo, 2019). Ac i e IOR includes conduc ing s a egic planning, expo ma ke esea ch, goal se ing, in e na ional ma ke scanning, o compe i i e analy- sis, and he e o e i enables in e na ional en ep eneu s o inc ease hei in o ma ion and knowledge base (Ci a egna e al., 2014; Muzychenko & Liesch, 2015; Tuomisalo, 2019). Con e sely, in se endipi ous IOR, en ep eneu s ecognize IOs o ui- ously o acciden ally, h ough hei gu eeling in he cou se o o he ac i i- ies (Hilme sson e al., 2021a; Kiss e al., 2020). Since his IOR s yle is no based on any o m o planned sea ch, schola s poin ou ha i eases and as s IOR (C ick & Spence, 2005). Al hough he se endipi ous IOR is con- side ed o ui ous, p e ious esea ch s a es ha i is no pu e luck (Chand a e al., 2009; Galan & To sein, 2021) bu a he ha in e na ional en ep e- neu s ale ness in passi e sea ch si ua ions and he assimila ion ha he in o ma ion ecei ed by chance could be o alue a e he main d i e s o his p ocess (Che y e al., 2018; Kon inen & Ojala, 2011). These IOs a e mainly iden i ied in unexpec ed con e sa ions wi h iends and colleagues, when eading newspape s, o du ing aca ions (Kiss e al., 2020; Zae a ian e al., 2016). P e ious esea ch has sugges ed ha di e ences in IOR may ha e sub- s an ial e ec s bo h on IOD and o e en ep eneu s’ skills and beha io . Fo example, schola s ha e ound ha ac i e and se endipi ous IOR di e - en ly in luence oppo uni y implemen a ion quickness, in e na ionaliza ion speed, idea gene a ion, idea newness, isk- aking p opensi y, and knowledge accumula ion (Ci a egna e al., 2014; De Cle cq e al., 2012; Hilme sson & Papaioannou, 2015). In his ega d, schola s a gue ha di - e ences in he IOR p ocess impac he iming and speed o implemen a ion in he way ha IOs iden i ied o ui ously a e implemen ed mo e slowly han hose iden i ied pu pose ully (Ci a egna e al., 2014). Resea ch on Oeconomia Cope nicana, 13(2), 541–579 546 in e na ionaliza ion speed s a es simila beha io , as i s ipula es ha IOs ecognized h ough delibe a e sea ch lead o a quicke in e na ionaliza ion han hose IOs iden i ied o ui ously (De Cle cq e al., 2012). Conce ning idea gene a ion, schola s sugges ha en ep eneu s ha ely on se endipi ous IOR gene a e mo e ideas han en ep eneu s ha ely on ac i e IOR (Fie & Pa el, 2008). Rega ding idea newness, p e ious esea ch shows inconclusi e and opposi e esul s. Some au ho s a gue ha he delib- e a e iden i ica ion o IOs leads o a highe deg ee o idea newness because he sea ch inc eases he de ec ion o new ideas (Dahlq is e al., 2004). Howe e , o he au ho s sugges ha en ep eneu s wi h se endipi ous IOR p edisposi ion end o gene a e mo e c ea i e and no el ideas wi h many unique cha ac e is ics because you canno sea ch o wha you do no know (Hilme sson & Papaioannou, 2015). Also, when IOs a e iden i ied se en- dipi ously, en ep eneu s p e e o ake a lowe - isk, lowe -cos app oach o in e na ional expansion (Chand a e al., 2015). Fu he mo e, as ac i e IOR in ol es deep e lec ion, as i leads o high knowledge in ensi y and acili a es e ie al o ele an knowledge om memo y in con as o se endipi ous IOR, which elies on shallow ep esen- a ions and ew, supe icial knowledge (Jones & Casulli, 2014). Ex apola - ing om hese indings, i is easonable o expec ha di e ences in he IOR p ocess will also ma e conce ning p oblem-sol ing skills du ing IOD. P oblem-sol ing compe ence in IOD A uahene-Gima and Wei (2011) p oposed wo p oblem-sol ing skills: p oblem-sol ing speed and p oblem-sol ing c ea i i y, which a e ele an o new p oduc de elopmen and an e ec i e esponse o ma ke ends. As such, p oblem-sol ing compe ence desc ibes he abili y o p oduce no - el solu ions and implemen hem in a imely manne (A uahene-Gima & Wei, 2011; Mo is e al., 2013). In he IE p ocess, and speci ically in he IOD s age, p oblems ha a ise in he in e na ional ma ke a e ha d o o e- cas in ad ance and may equi e u gen , use ul, and cos -e ec i e solu ions (Hilme sson e al., 2021b). Thus, p oblem-sol ing speed and c ea i i y a e c i ical. as hey help o e come he challenges ha can hinde he oppo uni- y o ope a e in he in e na ional ma ke (Boso e al., 2019; Chand a e al., 2020). Mo eo e , hese skills ha e been ound o posi i ely in luence new en u e su i al (S enholm & Renko, 2016), and in e na ional en y mode pe o mance (Ji e al., 2016), and he lack o hese skills hinde he in e na- ionaliza ion p ocess (Chand a e al., 2020). Indeed, p e ious esea ch has elucida ed ha p oblem-sol ing is he mos demanded compe ence o new Oeconomia Cope nicana, 13(2), 541–579 547 employees by companies acing in e na ionaliza ion ac i i ies (Vibhaka & Smi h, 2005), and i is also a key compe ence ha should be de eloped by uni e si y s uden s who wan o engage in in e na ional business ac i i ies (Mo is e al., 2013). A uahene-Gima and Wei (2011) and Hilme sson e al. (2021b) sugges ha inding quickly many solu ions o he p oblems de ec ed in he in e na- ional ma ke is no enough, bu implemen ing hem speedily is ac ually a skill. This s udy concep ualizes p oblem-sol ing speed as he abili y o apidly ind and implemen an app op ia e numbe o solu ions o he mul i- ace ed in e na ionaliza ion- ela ed p oblems ha jeopa dize he exploi a- ion o an IO. Mo eo e , because he concep o c ea i i y in ol es no el y and cos -e iciency (A uahene-Gima & Wei, 2011; Mo is e al., 2013), hese wo aspec s a e included in ou concep ualiza ion o p oblem-sol ing c ea i i y, de ined as he abili y o ind and implemen o iginal, use ul, and cos -e ec i e solu ions o he in e na ionaliza ion- ela ed p oblems ha pu he exploi a ion o an IO a isk. Since hese skills de i e om knowledge and in o ma ion-based p ocesses and hus hey a e no homogeneously de eloped, i is di icul o in e na ional en ep eneu s o show he same p oblem-sol ing e ec i eness (G ube e al., 2015; Hilme sson e al., 2021b). The e o e, hey a e seen as a compe i i e ad an age o exploi ing oppo uni ies in new in e na ional ma ke s (Mos a iz & Goh, 2018). En ep eneu ial cogni ion heo y The en ep eneu ial cogni ion heo y (Mi chell e al., 2002) helps o un- de s and how in e na ional en ep eneu s hink, how hey disco e IOs, and why hey beha e di e en ly when aced wi h new IOs (Taba es e al., 2021; Zah a e al., 2005). Zucchella (2021, p. 4) a gues ha his heo y could shed ligh on he “non-di ec ly obse able elemen s in he mind o he en ep eneu s like knowledge o ma ion, judgmen and e alua ion, ea- soning, and p oblem-sol ing.” In his ein, en ep eneu ial cogni ion heo y cha ac e izes he en ep eneu as a collec ion o unique in o ma ion and knowledge ha signi ican ly impac s on his/he skills and beha io in he en ep eneu ial p ocess (Ku a ko e al., 2021). Acco dingly, based on en- ep eneu ial cogni ion heo y, schola s sugges ha an en ep eneu ’s p ob- lem-sol ing abili y a ies while abso bing in o ma ion and knowledge o ma ke s (Chen e al., 2018; Chen e al., 2020). As p e iously a gued, we know ha each IOR app oach en ails di e en beha io s o ga he ing ma ke in o ma ion. The e a e, in ac , di e ences in he knowledge accumula ion abou he IO and he acili y o acqui e new in o ma ion abou i depending on whe he he IO was ecognized pu pose- Oeconomia Cope nicana, 13(2), 541–579 548 ully o se endipi ously (Hilme sson & Papaioannou, 2015; Jones & Casulli, 2014; Oyson & Whi ake , 2015). Fo example, en ep eneu s who disco e an IO based on ac i e sea ch end o seek and acqui e mo e do- main-speci ic in o ma ion abou he IO (Oyson & Whi ake , 2015). The e- o e, a knowledge acquisi ion gap is c ea ed be ween IOs iden i ied ac i ely and hose iden i ied se endipi ously. As p oblem-sol ing skills a e knowledge and in o ma ion-based, p oblem-sol ing speed and c ea i i y in he IOD s age may be condi ional on he in o ma ion and knowledge he en ep eneu has gene a ed du ing he IOR phase. D awing on he en ep e- neu ial cogni ion heo y, we a gue ha he e ogenei y in how en ep eneu s iden i y IOs, especially he unique o m hey de elop he in o ma ion and knowledge s uc u es ela ed o he IO, ul ima ely de e mines hei abili y o sol e p oblems. Simila ly, Bai e al. (2019) ha e demons a ed ha di e - ences in he in o ma ion and knowledge acqui ed abou he IO lead o a i- a ion in p oduc -de elopmen skills. Ac i e IOR and p oblem-sol ing compe ence When en ep eneu s pu pose ully iden i y IOs, hey sys ema ically sea ch o in o ma ion ha en ails ha ing di e en knowledge sou ces, causing en ep eneu s o ha e ex ensi e in o ma ion necessa y o IOR (Che y e al., 2018; Tuomisalo, 2019). This sea ch o in o ma ion would help en ep eneu s o disco e he a eas in which he IO may all sho . Thus, ac i e IOR enables hem o an icipa e p oblems ha may a ise du ing he in e na ionaliza ion p ocess and ha e some po en ial p oblem-solu ion pai s al eady p ede ined, acili a ing he quicke implemen a ion o solu- ions (Hilme sson e al., 2021a; Vaghely & Julien, 2010). Fu he mo e, he in o ma ion ga he ing consis s ypically no only o egula analysis o he in e na ional en i onmen , bu may also in ol e in es iga ing he needs o in e na ional cus ome s, examining and ques ioning he compe i o s, and/o main aining echnical dialogues wi h supplie s (Chand a e al., 2015; Ci a egna e al., 2014). Such ex e nal knowledge in e ac ion may inc ease speed when sol ing p oblems du ing IOD since i educes misunde s and- ings abou cus ome p oblems, allows o ha e in mind a pool o possible solu ions al eady implemen ed by compe i o s, and builds capabili ies wi h- in speci ic a eas (Bai e al., 2018; De Cle cq & Zhou, 2014). As such, he in ense hough , planning, and in o ma ion exposu e and ga he ing expend- ed a he ac i e IOR p ocess can sa e ime la e on, suppo ing speedily de elopmen o he IO because in o ma ion gained h ough ac i e IOR can be ecalled and p ojec ed in o de o ace he in e na ionaliza ion p oblems (Chand a e al., 2009; Jones & Casulli, 2014). Oeconomia Cope nicana, 13(2), 541–579 549 Mo eo e , p e ious esea ch has emphasized he bond be ween acqui - ing in o ma ion and knowledge and being c ea i e (Giampaoli e al., 2017). Schola s emphasize ha access o in o ma ion and knowledge in he ele- an domain bene i s he gene a ion o c ea i e solu ions (Chen e al., 2018) The in o ma ion and knowledge acqui ed h ough he ac i e IOR can ensu e a comple e analysis and unde s anding o ac ions equi ed o de el- op and exploi an IO (Hilme sson e al., 2021a). Such ea ly analysis and unde s anding may inc ease he likelihood o inding and implemen ing cos -e ec i e and c ea i e solu ions (Chen e al., 2018; Mo is e al., 2013). Addi ionally, h ough sys ema ic sea ch and in e na ional en i on- men scanning, in e na ional en ep eneu s gain insigh in o he na u e and con ex o he in e na ional ma ke . This would allow hem o imp o e he unde s anding o he solu ion ends demanded by he in e na ional cus om- e s, enac ed by supplie s, and implemen ed by he compe i o s (Bai e al., 2018, 2019), which may heigh en he eme gence o a g ea a ie y o alu- able and esh solu ions. The e o e, we hypo hesize as ollows: H1: Ac i e IOR is posi i ely ela ed o (a) p oblem-sol ing speed and (b) p oblem-sol ing c ea i i y. Se endipi ous IOR and P oblem-Sol ing Compe ence Se endipi ous IOR usually occu s h ough he in e p e a ion and eason- ing o incomple e in o ma ion o unusual si ua ions in he in e na ional en i onmen , implying ha en ep eneu s ha e limi ed knowledge abou he IO (Hilme sson e al., 2021a; Kiss e al., 2020). Howe e , p e ious e- sea ch sugges s ha despi e he ac ha en ep eneu s wi h se endipi ous IOR p edisposi ion do no ha e s uc u ed in o ma ion abou he IO (C ick & Spence, 2005; Kiss e al., 2020), hey could be well-skilled o sol e p oblems ha may a ise du ing IOD since hey will be likely o connec knowledge acqui ed in di e en con ex s o implemen solu ions (Hilme sson e al., 2021b; Muelle & Shephe d, 2016). Mo eo e , passi e en ep eneu s a e likely o p edic some p oblems and imagine al e na i e solu ions ha hei new IOs may ace in ad ance (Chen e al., 2020). Rein- o cing his idea, p e ious esea ch has a gued ha i would be expec ed ha passi e in e na ional en ep eneu s use he a ailable dispe se in o - ma ion abou he IO o deli e imely solu ions (E e s & O’Go man, 2011) because hey a e amed in an impulsi e and amo phous p ocedu e ha leads hem o conside he sca ce a ailable in o ma ion o make a decision (Kon inen & Ojala, 2011). Oeconomia Cope nicana, 13(2), 541–579 556 coun s). As sugges ed by p e ious esea ch, he ques ionnai e was adminis- e ed o owne –manage s o ac i e in e na ionalized i ms (e.g., Bolzani & Foo, 2018). To ensu e ha he pa icipan s a e in e na ional en ep eneu s, we es ab- lished as sampling c i e ia ha we will only conside en ep eneu s who a e cu en ly pu suing (de eloping) an oppo uni y in he in e na ional ma ke . Acco dingly, ou inal a ge sample was 997 in e na ional en ep eneu s who we e asked by e-mail o comple e an online sel -adminis e ed ques- ionnai e de eloped wi h Google Fo ms so wa e. Two weeks and ou weeks a e he ini ial su ey mailing, a eminde e-mail was sen o non- esponden s. We ealized he da a collec ion be ween June 22 and Oc obe 11, 2020. We ecei ed 174 illed ques ionnai es. We nex e ined ou sam- ple by checking he knowledgeabili y abou he opics and consis ency o esponden s’ a i udes in he ques ionnai e by applying wo con ol ques- ions. We emo ed 2 esponden s om he sample, one because incon- sis ency in he esponses and he o he o bo h easons, i.e., low knowl- edgeabili y and inconsis ency in he esponses. Thus, we ob ained 172 usa- ble su eys, ep esen ing an e ec i e esponse a e o 17.3%. Fu he mo e, o ensu e ha non- esponse bias is no a p oblem in ou s udy, we com- pa ed he answe s o he 15 ea ly and he 15 la e esponden s by applying a se ies o - es s (A ms ong & O e on, 1977). The esul s con i med ha he e exis no signi ican di e ences and, he e o e, non- esponse bias is no a p oblem in his s udy. Fu he mo e, wi h he help o G*Powe so wa e (Memon e al., 2020), we assessed i ou ini ial sample size (n = 172) mee s he minimum e- qui ed o achie e he desi ed le el o s a is ical powe . Gi en ha he mos complex eg ession in ou hypo hesized model has wo p edic o s, and as- suming a medium e ec size o 0.15 and a powe o 0.80 being alpha 0.05, a minimum sample size o 68 indi iduals is equi ed. Consequen ly, ou inal sample highly exceeds he minimum equi ed. As o he esponden s’ cha ac e is ics, 102 indi iduals (59.3%) we e men and 70 (40.7%) women, and he a e age age o esponden s was 42 yea s. 6.9% o hose su eyed had a second-le el educa ion (seconda y school), 48.3% had a hi d-le el educa ion (uni e si y deg ee), and 44.8% had a ou h-le el educa ion (pos g adua e deg ee). I is also wo h no ing ha he en ep eneu s ep e- sen nine di e en indus ies and ha 42.4% (n = 73) o hose acknowl- edged ha hey had a PNE. Oeconomia Cope nicana, 13(2), 541–579 557 Measu es and ques ionnai e design The su ey consis ed o 36 ques ions di ided in h ee sec ions. The i s sec ion se ed as an in oduc ion o he su ey and es ablished a ques ion in which pa icipan s had o gi e hei consen o pa icipa e in he s udy. Sec- ion wo con ained 25 ques ions e e ed o he main a iables o he s udy (i.e., in e na ional oppo uni y ecogni ion, p oblem-sol ing skills) and o a a iable ha was used as ma ke a iable, as well as wo con ol ques- ions. Finally, he hi d sec ion includes 10 ques ions abou socio- demog aphic and p e ious expe ience ac o s o esponden s. The su ey was ini ially designed in English based on adap a ions o p e iously alida ed scales. Then, i was ansla ed in o Spanish, as his is he mo he ongue o he esponden s. Once ansla ed, he ques ionnai e was judged by an expe panel consis ing o i e in e na ional en ep eneu s and six schola s. Following he expe s’ sugges ions, mino adjus men s we e made o adap he ins umen o he s udy con ex . As a inal s ep, he su ey was back- ansla ed o English, and no di e ences ha a ec he na u e o he o iginal scales we e ound. The majo i y o i ems we e meas- u ed as closed ques ions using a se en-poin Like scale (1 = o ally disa- g ee o 7 = o ally ag ee) excep o hose ques ions whe e esponden s we e equi ed o espond ei he posi i ely o nega i ely (e.g., i hey ha e had any nega i e en ep eneu ial expe ience in he pas ) o whe e hey we e equi ed o exp ess a speci ic esponse (e.g., hei age in yea s). Table 1 con ains a de ailed lis o he main a iables and i ems used in his s udy. In e na ional oppo uni y ecogni ion. To measu e ac i e and se endipi- ous IOR, we adap ed p e ious scales de eloped by Kucke z e al. (2017), Lo enz e al. (2018), and Nicolaou e al. (2009) o ou con ex . Ac i e IOR was measu ed u ilizing six i ems ha cap u e he igid planning o ind ele an in o ma ion abou he in e na ional ma ke , he delibe a e in es - men in ime and esou ces, and he a emp o scan he in e na ional ma ke cons an ly. Se endipi ous IOR was measu ed wi h i e i ems ha cap u e he spon aneous eme gence o an IO, lack o in en ionali y o iden i y an IO, and he sensi i i y and ale ness o disco e an IO in momen s o in- sigh . P oblem-sol ing speed and P oblem-sol ing c ea i i y The measu e o hese wo cons uc s was assessed by adap ing he scales de eloped by A uahene Gima and Wei (2011). Focusing on p oblems ha in e na ional en ep eneu s encoun e du ing IOD, p oblem-sol ing speed was measu ed wi h ou i ems ha add ess he quickness in inding and Oeconomia Cope nicana, 13(2), 541–579 558 implemen ing solu ions, and p oblem-sol ing c ea i i y was measu ed wi h ou i ems, which e lec he cos -e ec i eness and no el y o he solu ions. Pas nega i e en ep eneu ial expe ience. As in p e ious s udies (e.g., La uen e e al. 2019), o measu e PNE esponden s epo ed whe he hey ha e (yes o no) any nega i e en ep eneu ial expe ience in he pas (e.g., closu e/sale/ ailu e o a business whose pe o mance was oo low o he en ep eneu ’s expec a ions). Con ol a iables. We included i e con ol a iables in he s udy; which a e, gende (1 = emale; 2 = male), educa ional le el ( i e le els), age (in yea s), in e na ional business expe ience (1 = yes; 2 = no), and PNE se e i y (numbe o imes he en ep eneu has had nega i e en ep eneu ial expe iences). Ou logic was ha hese i e a iables ha e p e iously been ecognized as ele an o explaining possible sou ces o a ia ion when s udying en ep eneu ial and in e na ionaliza ion beha io (Mu nieks e al., 2020; Shephe d e al., 2015; Wach & Głodowska, 2021). Da a analysis app oach PLS-SEM was used o conduc he da a analysis (Hensele e al., 2009). Se e al a gumen s jus i y he use o his echnique in ou s udy. Fi s , PLS is app op ia e when he esea ch is conce ned wi h p edic ing mo e han one dependen a iable (Hai e al., 2019). Second, since PLS is based on boo s apping echniques, i p o ides mo e ealis ic in e ences (Rigdon, 2016), making i pa icula ly p e e able o s udy p e iously unp o en ela- ionships. As he hi d a gumen , PLS-SEM, compa ed o co a iance-based s uc u al equa ion modeling (CB-SEM), usually p oduces smalle s uc u - al coe icien s, hus p o iding mo e secu e hypo hesis es ing in less s udied con ex s such as Andalusia (Reina z e al., 2009). Fo he es ima ion me h- od, Mode A ( e lec i e measu es) was selec ed since i s usage is sugges ed when co ela ed i ems a e expec ed (i.e., he i ems o a scale a e in e - changeable) (Ja is e al., 2003) and because a ela ionship is expec ed o occu om he cons uc o he indica o s (Diaman opoulos & Winklho e , 2001). The so wa e used o ca y ou he analysis was Sma PLS 3.3. Common me hod bias Common me hod bias (CMB) could be a p oblem in su ey-based s ud- ies when he same esponden answe s independen and dependen a ia- bles. The e o e, we ook some p ocedu es o a oid and con ol o CMB. Fi s , we applied a ious emedies ecommended by MacKenzie and Pod- sako (2012) o mi iga e CMB om he e y design phase o he su ey Oeconomia Cope nicana, 13(2), 541–579 559 (e.g., p e es he ques ionnai e, gua an ee pa icipan s anonymi y, ask pa - icipan s o hones y, ensu e pa icipan s knowledgeabili y). Second, once he da a was collec ed, we applied he measu ed la en ma ke a iable (MLMV) echnique o de ec and con ol CMB, which is ecognized as he only e ec i e s a is ical emedy o ha pu pose (Chin e al., 2013). The basis o he MLMV is ha i should be om a di e en uni o analysis and a di e en domain han he a iables included in he hypo hesized model. Thus, o con ol o CMB, we included he o ganiza ional exhibi ionism scale de eloped by Hamedoglu and Po as (2012) as MLMV. Ca ying ou he men ioned analysis, we can a gue ha he CMB is no a p oblem in ou esea ch, since he esul s ob ained a e he inclusion o he MLMV show ha (1) he model p esen s a wo se i han he o iginal model and (2) ha he pa h coe icien s do no show signi ican di e ences be ween hem and a e consis en wi h he o iginal es ima es. Resul s Measu emen model assessmen As we used Mode A ( e lec i e measu es) as es ima ion me hod, i s , well- es ablished adi ional measu es should be applied o e alua e he meas- u emen model’s in e nal consis ency, eliabili y, and alidi y (Hai e al., 2019). All indica o loadings a e abo e he minimum limi sugges ed in he li e a u e o 0.707 (Ca mines & Zelle , 1979), and hus, indi idual i em eliabili y is conside ed sa is ac o y. Rega ding C onbach’s Alpha and composi e eliabili y (CR), all a iables ha e alues be ween 0.70 and 0.95, which a e he minimum and maximum alues espec i ely o ake in o con- side a ion (Hai e al., 2019; Nunnally, 1978). Besides, he AVE o all a i- ables is abo e he 0.50 cu -o le el, which con i ms he con e gen alidi- y o ou measu emen s (Fo nell & La cke , 1981). Table 1 shows all hese alues. Disc iminan alidi y was e alua ed h ough he He e o ai - Mono ai Ra io (HTMT) 0.85 c i e ion (see Table 2), which is s ic e in compa ison wi h o he disc iminan alidi y measu es such as c oss- loadings and Fo nell-La cke c i e ion (Hensele e al., 2015). The esul s ob ained h ough all hese measu es con i m he alidi y o ou measu e- men model and he e o e we can p oceed o e alua e he s uc u al model. Oeconomia Cope nicana, 13(2), 541–579 560 S uc u al model assessmen To s a e alua ing he s uc u al model, we mus i s look a he global adjus men model calcula ed by he S anda dized Mean Squa e Residue (SRMR). Ou model a ains an SRMR o 0.063, a alue ha is below he ecommended maximum limi o 0.080, and he e o e means ha ou model has an app op ia e i (Hu & Ben le , 1999). F om he e, we e alua e he s uc u al model by examining he sign, size, and meaning o he s uc u al pa h coe icien s and he 2 , which allow he assessmen o he ela ion- ships’ s a is ical signi icance. 10,000 esamples (boo s apping echnique) we e u ilized o gene a e -s a is ics o e alua e he s a is ical signi icance o he pa h coe icien s. As displayed in Table 3, h ee o he i s ou hypo hesized ela ion- ships a e suppo ed, ha is, H1a, H1b, and H2b. Howe e , H2a is no sup- po ed, as se endipi ous IOR shows an ex emely low 2 , unde he mini- mum le el o 0.02. Thus, se endipi ous IOR and PS speed a e no associa - ed. I should also be no ed ha H1a and H2b show a s ong e ec size, while H1b shows a weak e ec size (Cohen, 1988). Addi ionally, o es i- ma e H3 and H4, we applied he app oach p oposed by Rod íguez-En ena e al. (2018), which examines whe he he es ima es o wo pa ame e s ha e signi ican di e ences be ween hem. In his ega d, H3 and H4 a e sup- po ed, which con i ms ou supposi ion ha ac i e IOR possesses a signi i- can ly g ea e e ec on p oblem-sol ing speed han se endipi ous IOR. On he con a y, se endipi ous IOR possesses a signi ican ly g ea e e ec on p oblem-sol ing c ea i i y han ac i e IOR. Fu he mo e, he R 2 and he Q 2 alues we e examined o assess he p e- dic i e ele ance o ou model. Ou hypo hesized model appea s o possess sui able p edic i e powe (in-sample p edic ion) o bo h endogenous a i- ables. Indeed, PS speed (R 2 = 0.433) and PS c ea i i y (R 2 = 0.384) show a mode a e R 2 signi icance (Hensele e al., 2009). Likewise, when analyz- ing ou esea ch model’s p edic i e pe o mance (ou -o -sample p edic- ion), ou model seems o possess an adequa e p edic i e pe o mance by calcula ing Q 2 alues h ough PLS p edic (PS Speed – Q 2 = 0.314; PS c e- a i i y – Q 2 = 0.268), sugges ing ha ou model can p edic exac alues o indi idual cases (E e mann & Ta e, 2016). Mo eo e , i should be no ed ha none o he i e con ol a iables ha e signi ican e ec s o e he de- penden a iables and he e o e a e no sou ces o a ia ion. Indeed, by including only he con ol a iables in a model, we con i m ha none o hem un old signi ican e ec s and gene a e a he weak R 2 alues (0.050 o PS speed and 0.056 o PS c ea i i y). Oeconomia Cope nicana, 13(2), 541–579 561 Finally, we conduc ed a mul i-g oup analysis (Hensele & Fasso , 2010) o es he mode a ing e ec o PNE on he ela ionships included in ou hypo hesized model. Consequen ly, we spli he sample in o wo g oups, in e na ional en ep eneu s wi h (n = 73) and wi hou (n = 99) PNE. Be o e compa ing pa h es ima es ac oss g oups, i is necessa y o exam- ine he measu emen in a iance o composi e models (MICOMs). In his way, i can be assu ed ha he e ec o he in ensi y o PNE as a mode a - ing a iable is limi ed o he pa h coe icien s o he s uc u al model and no o he pa ame e s o he measu emen model. Thus, we apply he h ee- s ep p ocedu e (i.e., (1) con igu al in a iance, (2) composi ional in a iance, and (3) an assessmen o equal a iances and means) o examine he MICOMs when using PLS (Ma hews, 2017). As Table 4 shows, he ull measu emen in a iance o bo h g oups was achie ed o all a iables. Consequen ly, we nex applied a non-pa ame ic app oach, he pe mu a- ion-based p ocedu e, o conduc mul i-g oup analyses h ough he e alua- ion o he di e ences be ween he pa h coe icien s o wo g oups (Chin & Dibbe n, 2010). The ad an age o his echnique is ha i is no based on dis ibu ion assump ions, bu on a ough andomiza ion es in which a sub- se o all possible da a pe mu a ions be ween g oups is made. In he pe mu- a ion-based me hod, only a p- alue o he di e ences be ween he pa h coe icien s lowe han 0.05 a he 5% le el indica es he exis ence o sig- ni ican di e ences in he wo g oups, and hus i is he mos eliable me h- od o conduc mul i-g oup analyses (Chin & Dibbe n, 2010; Hai e al., 2018). As Table 4 illus a es, he e exis signi ican di e ences in wo o he ou di ec e ec s on which mode a ion hypo heses we e pos ula ed. H5a and H5b a e s a is ically suppo ed, while H6a and H6b a e no sup- po ed. The e o e, PNE mode a es he ela ionships be ween ac i e IOR and p oblem-sol ing skills bu does no a ec he ela ionships be ween se endipi ous IOR and p oblem-sol ing skills. Pa h coe icien s, signi i- cance le el, and R 2 o each subsample a e displayed in Figu e 2. Discussion We unde ook his s udy o ad ance he IE li e a u e by examining he ela- ionship be ween IOR and p oblem-sol ing compe ence. In pa icula , we pay special a en ion o whe he he wo IOR app oaches (i.e., ac i e IOR and se endipi ous IOR) di e en ly in luence p oblem-sol ing speed and c ea i i y and, hus, i hey a e a sou ce o he e ogenei y ha explains why some in e na ional en ep eneu s sol e p oblems ha a ise du ing IOD as - e and mo e c ea i ely. Mo eo e , we analyzed whe he PNE may be Oeconomia Cope nicana, 13(2), 541–579 562 a sou ce o di e en ia ion ha mode a es he ela ionship be ween IOR s yles and p oblem-sol ing skills. Ou indings, apa om ha ing ele an heo e ical con ibu ions o he li e a u e, also ha e manage ial implica- ions. Theo e ical implica ions As he i s con ibu ion, his s udy adds empi ical e idence ha con- i ms ha he way an en ep eneu ecognizes an IO (ac i ely e sus se en- dipi ously) di e en ly a ec s he de elopmen o he IO (c . Ci a egna e al., 2014; Zhou, 2007). In pa icula , we o e ini ial insigh s on how ac i e and se endipi ous IOR in luence en ep eneu s’ p oblem-sol ing compe- ence o IOD. Ac i e IOR was ound o be posi i ely ela ed o bo h p ob- lem-sol ing speed and c ea i i y, while se endipi ous IOR has a posi i e ela ionship o p oblem-sol ing c ea i i y. Addi ionally, we ound ha se - endipi ous IOR is mo e highly ela ed o p oblem-sol ing c ea i i y han ac i e IOR, and on he con a y, ha ac i e IOR has a g ea e posi i e ela- ionship wi h p oblem-sol ing speed han se endipi ous IOR. Indeed, al - hough se endipi ous IOR does ha e a posi i e ela ionship wi h p oblem- sol ing speed, his ela ionship is no signi ican . The dynamic na u e o he o eign ma ke sugges s ha o compe e e - ec i ely is i al o o e come he challenge and c i icali y o apidly de el- oping he IO ma e ializa ion p ocess (Acedo & Jones, 2007). As such, he posi i e e ec o ac i e IOR on p oblem-sol ing speed b ings new e i- dence on he ele ance o in o ma ion ga he ing as a de e minan o he apid de elopmen o IOs. Thus, his esul aligns somewha wi h p e ious belie s ha IOs ecognized by ac i eness ha e a as e implemen a ion and in e na ionaliza ion (Ci a egna e al., 2014) han hose ecognized by se - endipi y which akes longe o be ma e ialized and in e na ionalized (De Cle cq e al., 2012; Zhou, 2007). Wi h ou indings, we o e a possible explana ion o his phenomenon, as i seems ha in se endipi ous IOR, he p oblems ha appea ake longe o be sol ed by in e na ional en ep e- neu s, which will esul in a la e implemen a ion o he IO. Indeed, ega d- ing he non- ela ionship be ween se endipi ous IOR and p oblem-sol ing speed, i appea s ha he limi ed knowledge abou he IO p e en s p ob- lems ha a ise om being sol ed quickly. This inding con adic s p e ious pos ula ions ha a i m ha en ep eneu s ha ecognized IOs se endipi- ously use he sca ce a ailable in o ma ion o make imely decisions (E e s & O’Go man, 2011; Kon inen & Ojala, 2011). Howe e , i is in line wi h p io esea ch sugges ing ha en ep eneu s ha ecognized IOs se endipi- ously need mo e hough ime o encoun e new in o ma ion om o eign Oeconomia Cope nicana, 13(2), 541–579 563 ma ke s be o e selec ing he igh solu ion (Chand a e al., 2009; Hohen hal e al., 2003) which slowdowns p oblem-sol ing p ocess. Thus, ou indings yield he pa adox ha while se endipi ous disco e y accele a es he ecogni ion o IOs (C ick & Spence, 2005), i appea s ha o IOs iden i ied in his way, in e na ional en ep eneu s will ake longe o ind and implemen solu ions o IO de elopmen and exploi a ion ela ed p oblems. Ins ead, while iden i ying an IO h ough ac i eness akes ime and e o , sol ing p oblems ha a ise in hese oppo uni ies will be as e . Ul ima ely, hese indings imply ha iden i ying IOs in one way o ano he condi ion he speed o p oblem esolu ion, impac ing in e na ional oppo - uni y exploi a ion and i m pe o mance (Giampaoli e al., 2017). Rega ding p oblem-sol ing c ea i i y, e en i we ound i o be posi i e- ly ela ed o bo h ac i e and se endipi ous IOR, se endipi ous IOR is a be - e p edic o o implemen ing c ea i e solu ions, which aligns wi h ou p io e lec ions. Se endipi ous IOR leads en ep eneu s o an open-ended app oach o p oblem-sol ing, which indeed is be e sui ed o his end by enabling analogical easoning p ocesses and he connec ion o knowledge o igina ing in di e en con ex s (Muelle & Shephe d, 2016). Al hough en ep eneu s wi h an ac i e IOR p edisposi ion may gene a e a lis o pos- sible “new o he i m” solu ions based on hei sys ema ic sea ch on in e - na ional ma ke s, he solu ions hey apply will usually be based on p ede- ined p oblem-solu ion pai s wi h limi ed c ea i i y (A uahene-Gima & Wei, 2011; Vaghely & Julien, 2010). Ou indings suppo he iew ha he elaxed s anda ds o se endipi ous disco e y will allow in e na ional en e- p eneu s o a i e a new p oblem-solu ion pai s, as hey possess an imagi- na i e and uncon en ional way o hinking ha leads o gene a ing o iginal and unique ideas (Hilme sson & Papaioannou, 2015; Vaghely & Julien, 2010). Likewise, c ea i e p oblem-sol ing does encompass no only he o iginali y and no el y o he implemen ed solu ions bu also he ac ha hese a e cos -e icien (A uahene-Gima & Wei, 2011). Ou indings add o he li e a u e ha sugges s ha en ep eneu s ha ha e ecognized an IO se endipi ously end o implemen lowe -cos ideas, as hey a e mo e likely o use he a ailable esou ces o ma e ialize hei business oppo uni ies (Hilme sson e al., 2021b; S enholm & Renko, 2016). Does PNE ma e ? As a second con ibu ion, ou indings e eal ha he na u e o he pas expe ience o in e na ional en ep eneu s plays a ole in some o he p oposed ela ionships. In pa icula , he p oblem-sol ing speed o ac i e en ep eneu s dec eases when hey ha e su e ed a PNE, bu hei p oblem-sol ing c ea i i y inc eases. These indings con i m ha PNE has dicho omous e ec s on en ep eneu s, ei he posi i e o nega i e (c . Dias & Teixei a, 2017; Walsh & Cunningham, 2017). Conce ning Oeconomia Cope nicana, 13(2), 541–579 564 p oblem-sol ing speed, ou esul s ag ee wi h Shephe d (2003) and La- uen e e al. (2019) as PNE will lead o a lack o con idence and isk- aking p opensi y ha inhibi he use o knowledge acqui ed du ing ac i e IOR limi ing he numbe o solu ions conside ed and hus leng hening solu ion implemen a ion. Rega ding p oblem-sol ing c ea i i y, he PNE will ha e he opposi e e ec as i will in ensi y he use o acqui ed knowledge o ind and ma e ialize mo e inno a i e solu ions (Ellis e al., 2006; La uen e e al., 2019). Hence, ou esul s sugges ha PNE does no ha e a unique im- pac on en ep eneu s’ skills. In his s udy, we can conclude ha PNE ha ms skills ha in ol e a lack o e lec ion, such as ha ing o ake ac ion quickly, and ha i will ha e a posi i e e ec on skills ha in ol e mo e e lec ion, such as c ea i i y. In he case o passi e in e na ional en ep eneu s, ha ing a PNE does no in luence he ela ionship be ween se endipi ous IOR and p oblem- sol ing compe ence. Al hough he e a e sligh di e ences be ween he wo g oups conside ed, he e ec is no s a ically signi ican . We o e an ex- pos explana ion o such a non- ela ionship. The esul may indica e ha in o ui ously iden i ied IOs, he impac o p io expe ience-based ac o s ega ding en ep eneu s’ skills is no sound (Yamakawa & Ca don, 2015). Thus, he inding could be explained om he poin o iew ha in se en- dipi ously disco e ed IOs, he e is al eady plen i ul unce ain y, ambigui y, and isk a e sion (Casillas e al., 2009; Chand a e al., 2020; Chand a e al., 2015) ha e en i en ep eneu s ha e su e ed a PNE, i does no add e en mo e u bulence o hei hinking. In his way, i appea s ha iden i ying IOs h ough se endipi y will make in e na ional en ep eneu s implemen solu ions slowly and cau iously enough and use di e gen hinking and a ailable esou ces o deli e no el and cos -e icien solu ions. Conclusions D awing on oppo uni y iden i ica ion and en ep eneu ial cogni ion heo- ies, his s udy p o ides new insigh s in o he IOR p ocess and i s e ec s by examining i he way en ep eneu s ecognize an IO (ac i ely o se endipi- ously) di e en ly in luences en ep eneu s’ p oblem-sol ing speed and c ea i i y. Mo eo e , we ha e analyzed i PNE plays a mode a ing ole in hese ela ionships. Using da a om 172 Spanish in e na ional en ep e- neu s, we ha e ound ha ac i e IOR is posi i ely ela ed o p oblem- sol ing speed and c ea i i y, while se endipi ous IOR has a posi i e ela- ionship solely wi h p oblem-sol ing c ea i i y. Fu he mo e, we ha e demons a ed ha en ep eneu s wi h an ac i e IOR p edisposi ion sol e Oeconomia Cope nicana, 13(2), 541–579 565 p oblems as e han hei passi e coun e pa s, bu ha on he con a y, en ep eneu s wi h a se endipi ous IOR p edisposi ion sol e p oblems mo e c ea i ely. Addi ionally, we ha e ound ha PNE can ac as a mode a o by dec easing he impac o ac i e IOR on p oblem-sol ing speed bu inc eas- ing he impac o ac i e IOR on p oblem-sol ing c ea i i y. Rega ding se - endipi ous IOR, ha ing o no a PNE does no mode a e i s ela ionship wi h ei he p oblem-sol ing speed o c ea i i y. In sho , ou indings sug- ges IOR s yles and PNE as sou ces o he e ogenei y o en ep eneu s’ p oblem-sol ing skills. Manage ial implica ions Ou s udy o e s se e al manage ial implica ions o in e na ional en e- p eneu s. They should be acknowledged ha iden i ying IOs ac i ely o se endipi ously will di e en ly in luence hei p oblem-sol ing compe- ence. Likewise, he ac o ha ing a PNE may in luence hei p oblem- sol ing compe ence. Rega ding p oblem-sol ing speed, hey should be awa e ha in ac i ely iden i ied IOs hey will be p one o sol e p oblems as e , which could ansla e in o as e implemen a ion o he IO and, he e o e, o being mo e compe i i e in he in e na ional ma ke . Those en ep eneu s who ac i ely iden i y IOs possess ele an ma ke in o - ma ion ha allows hem o an icipa e a se ies o solu ions o add ess u u e p oblems, hus speeding up hei esponse o hese di icul ies. Thus, ou indings jus i y he esou ces ha in e na ional en ep eneu s in es in ac- qui ing knowledge o iden i y an IO delibe a ely, o example, h ough in e na ional ma ke esea ch. Howe e , in e na ional en ep eneu s o he i ms whe e hey ope a e mus conside ha ha ing su e ed an en ep e- neu ial ailu e will dec ease he quickness o p oblem-sol ing since i will educe he con idence in e na ional en ep eneu s ha e in he in o ma ion hey hold. Likewise, en ep eneu s who ha e ecognized an IO se endipi- ously should be awa e ha hey will equi e mo e hough e o and ime o acqui e new in o ma ion be o e sol ing he p oblems ha a ise, i espec- i e o whe he o no hey ha e su e ed a PNE. The e o e, hey should bea in mind ha se endipi ous IOs will ake longe o ma e ialize. Indeed, his addi ional ime needed could jeopa dize new en u e su i al and/o company pe o mance. Rega ding p oblem-sol ing c ea i i y, in e na ional en ep eneu s o he i ms o which hey wo k should no ice ha whe he hey disco e IOs ac i ely o se endipi ously, hey will be able o sol e he p oblems ha a ise c ea i ely. Howe e , i IOs a e disco e ed o ui ously, he e will be a g ea e endency o implemen inno a i e solu ions. The openness in Oeconomia Cope nicana, 13(2), 541–579 572 Ma hews, L. (2017). Applying mul ig oup analysis in PLS-SEM: a s ep-by-s ep p ocess. In Pa ial leas squa es pa h modeling: basic concep s, me hodological issues and applica ions. Sp inge In e na ional Publishing, 219–243. doi: 10.1007/978-3-319-64069-3_10. Memon, M., Ting, H., Cheah, J.-H., Thu asamy, R., Chuah, F., & Huei Cham, T. (2020). Sample size o su ey esea ch: e iew and ecommenda ions. Jou nal o Applied S uc u al Equa ion Modeling, 4(2), 1–20. doi: 10.47263/JASEM. 4(2)01. Mi chell, R. K., Buseni z, L., Lan , T., McDougall, P. P., Mo se, E. A., & Smi h, J. B. (2002). Towa d a heo y o en ep eneu ial cogni ion: e hinking he people side o en ep eneu ship esea ch. En ep eneu ship Theo y and P ac ice, 27(2), 93–104. doi: 10.1111/1540-8520.00001. Mo al-Paja es, E., Mozas-Mo al, A., Be nal-Ju ado, E., & Medina-Vi uel, M. J. (2015). E iciency and expo s: e idence om Sou he n Eu opean companies. Jou nal o Business Resea ch, 68(7), 1506–1511. doi: 10.1016/j.jbus es.2015. 01.042. Mo is, M. H., Webb, J. W., Fu, J., & Singhal, S. (2013). A compe ency-based pe spec i e on en ep eneu ship educa ion: concep ual and empi ical insigh s. Jou nal o Small Business Managemen , 51(3), 352–369. doi: 10.1111/jsbm.1 2023. Mos a iz, I., & Goh, S. K. (2018). In e na ional women en ep eneu s and in e na ional oppo uni y ecogni ion skills o s a -up en u es. Jou nal o In e na ional Business and En ep eneu ship De elopmen , 11(3), 201–220. doi: 10.1504/jibed.2018.095171. Muelle , B. A., & Shephe d, D. A. (2016). Making he mos o ailu e expe iences: explo ing he ela ionship be ween business ailu e and he iden i ica ion o business oppo uni ies. En ep eneu ship Theo y and P ac ice, 40(3), 457–487. doi: 10.1111/e ap.12116. Mu nieks, C. Y., Ca don, M. S., & Haynie, J. M. (2020). Fueling he i e: examining iden i y cen ali y, a ec i e in e pe sonal commi men and gende as d i e s o en ep eneu ial passion. Jou nal o Business Ven u ing, 35(1), 105909. doi: 10.1016/j.jbus en .2018.10.007. Muzychenko, O., & Liesch, P. W. (2015). In e na ional oppo uni y iden i ica ion in he in e na ionalisa ion o he i m. Jou nal o Wo ld Business, 50(4), 704– 717. doi: 10.1016/j.jwb.2014.12.001. Nicolaou, N., Shane, S., Che kas, L., & Spec o , T. D. (2009). Oppo uni y ecogni ion and he endency o be an en ep eneu : a bi a ia e gene ics pe spec i e. O ganiza ional Beha io and Human Decision P ocesses, 110(2), 108–117. doi: 10.1016/j.obhdp.2009.08.005y. Nunnally, J. C. (1978). Psychome ic heo y. McG aw-Hill. Oyson, M. J., & Whi ake , H. (2015). En ep eneu ial cogni ion and beha io in he disco e y and c ea ion o in e na ional oppo uni ies. Jou nal o In e na ional En ep eneu ship, 13(3), 303–336. doi: 10.1007/s10843-015- 0156-6. Oeconomia Cope nicana, 13(2), 541–579 573 Reina z, W., Haenlein, M., & Hensele , J. (2009). An empi ical compa ison o he e icacy o co a iance-based and a iance-based SEM. In e na ional Jou nal o Resea ch in Ma ke ing, 26(4), 332–344. doi: 10.1016/j.ij esma .2009.08.001. Rigdon, E. E. (2016). Choosing PLS pa h modeling as analy ical me hod in Eu opean managemen esea ch: a ealis pe spec i e. Eu opean Managemen Jou nal, 34(6), 598–605. doi: 10.1016/j.emj.2016.05.006. Rod íguez-En ena, M., Schube h, F., & Gelha d, C. (2018). Assessing s a is ical di e ences be ween pa ame e s es ima es in pa ial leas squa es pa h modeling. Quali y and Quan i y, 52(1), 57–69. doi: 10.1007/s11135-016-0400- 8. Shephe d, D. A. (2003). Lea ning om business ailu e: p oposi ions o g ie eco e y o he sel -employed. Academy o Managemen Re iew, 28(2), 318– 328. doi: 10.5465/AMR.2003.9416377. Shephe d, D. A., Williams, T. A., & Pa zel , H. (2015). Thinking abou en ep eneu ial decision making: e iew and esea ch agenda. Jou nal o Managemen , 41(1), 11–46. doi: 10.1177/0149206314541153. Spanish Ins i u e o Fo eign P omo ion. (2020). Andalusia’s mon hly o eign ade epo . Decembe 2019. Re ie ed om h ps://www.icex.es/icex/es/na egaci on-p incipal/que-es-icex/donde-es amos/ ed- e i o ial-de-come cio/na egacion -p incipal/in o mes-y-documen os/index.h ml?idComunidad=15. S enholm, P., & Renko, M. (2016). Passiona e b icoleu s and new en u e su i al. Jou nal o Business Ven u ing, 31(5), 595–611. doi: 10.1016/j.jbus en .2016. 05.004. Sung, S. Y., & Choi, J. N. (2012). E ec s o eam knowledge managemen on he c ea i i y and inancial pe o mance o o ganiza ional eams. O ganiza ional Beha io and Human Decision P ocesses, 118(1), 4–13. doi: 10.1016/j.obhdp. 2012.01.001. Taba es, A., Chand a, Y., Al a ez, C., & Escoba -Sie a, M. (2021). Oppo uni y- ela ed beha io s in in e na ional en ep eneu ship esea ch: a mul ile el analysis o an eceden s, p ocesses, and ou comes. In e na ional En ep eneu ship and Managemen Jou nal, 17(1), 321–368. doi: 10.1007/s11 365-020-00636-3. Te án-Yépez, E., Jiménez-Cas illo, D., & Sánchez-Pé ez, M. (2021). In e na ional oppo uni y ecogni ion: a comp ehensi e bibliome ic e iew. Jou nal o In e na ional En ep eneu ship, 19(1), 18–52. doi: 10.1007/s10843-020-00276- 2. Tuomisalo, T. (2019). Eme gence o an en ep eneu ial oppo uni y: a case wi hin a Finnish elecommunica ion in e na ional new en u e. Jou nal o In e na ional En ep eneu ship, 17(3), 334–354. doi: 10.1007/s10843-019-00 247-2. Ucbasa an, D., Shephe d, D. A., Locke , A., & Lyon, S. J. (2013). Li e a e business ailu e: he p ocess and consequences o business ailu e o en ep eneu s. Jou nal o Managemen , 39(1), 163–202. doi: 10.1177/0149206 312457823. Oeconomia Cope nicana, 13(2), 541–579 574 Ucbasa an, D., Wes head, P., & W igh , M. (2009). The ex en and na u e o oppo uni y iden i ica ion by expe ienced en ep eneu s. Jou nal o Business Ven u ing, 24(2), 99–115. doi: 10.1016/j.jbus en .2008.01.008. Vaghely, I. P., & Julien, P. A. (2010). A e oppo uni ies ecognized o cons uc ed? An in o ma ion pe spec i e on en ep eneu ial oppo uni y iden i ica ion. Jou nal o Business Ven u ing, 25(1), 73–86. doi: 10.1016/j.jbus en .2008.06 .004. Vaillan , Y., & La uen e, E. (2019). The inc eased in e na ional p opensi y o se ial en ep eneu s demons a ing ambidex ous s a egic agili y. In e na ional Ma ke ing Re iew, 36(2), 239–259. doi: 10.1108/IMR-01-2018-0015. Vibhaka , A. P., & Smi h, R. K. (2005). De eloping he app op ia e employee skill se and deg ee o small in e na ional businesses. Jou nal o Teaching in In e na ional Business, 16(1), 51–66. doi: 10.1300/J066 16n01_04. Wach, K., & Głodowska, A. (2021). How do demog aphics and basic ai s o an en ep eneu impac he in e na ionaliza ion o i ms? Oeconomia Cope nicana, 12(2), 399–424. doi: 10.24136/OC.2021.014. Walsh, G. S., & Cunningham, J. A. (2017). Regene a i e ailu e and a ibu ion: examining he unde lying p ocesses a ec ing en ep eneu ial lea ning. In e na ional Jou nal o En ep eneu ial Beha iou and Resea ch, 23(4), 688– 707. doi: 10.1108/IJEBR-03-2015-0072. Yamakawa, Y., & Ca don, M. S. (2015). Causal asc ip ions and pe cei ed lea ning om en ep eneu ial ailu e. Small Business Economics, 44(4), 797–820. doi: 10.1007/s11187-014-9623-z. Zae a ian, R., Eng, T. Y., & Tasa o i, M. (2016). An explo a o y s udy o in e na ional oppo uni y iden i ica ion among amily i ms. In e na ional Business Re iew, 25(1), 333–345. doi: 10.1016/j.ibus e .2015.06.002. Zah a, S. A., Ko i, J. S., & Yu, J. F. (2005). Cogni ion and in e na ional en ep eneu ship: implica ions o esea ch on in e na ional oppo uni y ecogni ion and exploi a ion. In e na ional Business Re iew, 14(2), 129–146. doi: 10.1016/j.ibus e .2004.04.005. Zhou, L. (2007). The e ec s o en ep eneu ial p ocli i y and o eign ma ke knowledge on ea ly in e na ionaliza ion. Jou nal o Wo ld Business, 42(3), 281–293. doi: 10.1016/j.jwb.2007.04.009. Zucchella, A. (2021). In e na ional en ep eneu ship and he in e na ionaliza ion phenomenon: aking s ock, looking ahead. In e na ional Business Re iew, 30(2), 101800. doi: 10.1016/j.ibus e .2021.101800. Acknowledgmen s This publica ion is pa o he R&D p ojec PID2020-119994RB-I00, inanced by MCIN/AEI/10.13039/501100011033/. This esea ch was also unded by he Anda- lusian Regional Go e nmen [Resea ch P ojec PAIDI2018 P18-RT-4663] and he Uni e si y o Alme ia. Annex Table 1. Su ey i ems and measu emen model e alua ion Cons uc /i ems Weigh Loading α CR AVE Please indica e you le el o ag eemen wi h each o he ollowing s a emen s ega ding he iden i ica ion o he business oppo uni y ha you a e cu en ly pu suing (de eloping) in he in e na ional ma ke . Se endipi ous in e na ional oppo uni y ecogni ion (SIOR) ( e lec i e) 0.903 0.928 0.721 I had a “s a e o ale ” o special sensi i i y o de ec he oppo uni y in he in e na ional ma ke 0.260* 0.879* I is di icul o explain how I came up wi h he in e na ional business idea; i jus came o me spon aneously 0.234* 0.828* I go he in e na ional business idea om con e sa ions wi h o he people 0.192* 0.774* I had he in ui ion o iden i y he oppo uni y in he in e na ional ma ke 0.250* 0.886* Al hough i was no my in en ion, I iden i ied he oppo uni y o ini ia e a new business in he in e na ional ma ke 0.238* 0.874* Ac i e in e na ional oppo uni y ecogni ion (AIOR) ( e lec i e) 0.916 0.933 0.704 I esea ched in e na ional ma ke s o iden i y he business oppo uni y h ough an o ganized/me hodical p ocess 0.181* 0.795* I in en ionally sea ched o cus ome needs o which I ha e de eloped a solu ion subsequen ly 0.195* 0.834* In he sea ch o he in e na ional oppo uni y, I ook my ime 0.164* 0.770* I examined and ques ioned exis ing solu ions (own and/o compe i o s’) in a pu pose ul way o gene a e a new in e na ional business idea 0.218* 0.884* I looked o in o ma ion o gene a e new ideas o p oduc s o se ices ocused on he in e na ional ma ke 0.208* 0.847* I egula ly analyzed he in e na ional en i onmen o ind a business oppo uni y 0.220* 0.898* Table 1. Con inued Cons uc /i ems Weigh Loading α CR AVE Re e o hose momen s when you ha e expe ienced in e na ional ma ke - ela ed p oblems (i.e., p oblems du ing he in e na ionaliza ion p ocess), du ing he de elopmen o he business oppo uni y ha you a e cu en ly pu suing in he o eign ma ke (i.e.,, a e ha ing iden i ied he oppo uni y in he in e na ional ma ke ) and indica e you deg ee o ag eemen wi h he ollowing s a emen s P oblem-sol ing speed (PSS) ( e lec i e) 0.885 0.921 0.744 I belie e I was quick o de ine, ind, and implemen solu ions o he p oblems ha a ose in he in e na ional ma ke 0.297* 0.847* I quickly came up wi h di e en al e na i e solu ions o each p oblem ha a ose in he in e na ional ma ke 0.293* 0.892* I usually ound and implemen ed a app op ia e imes an app op ia e numbe o al e na i e solu ions o he p oblems ha a ose in he in e na ional ma ke 0.293* 0.876* Fo e e y p oblem ound in he in e na ional ma ke , I had many solu ions 0.276* 0.836* P oblem-sol ing c ea i i y (PSC) ( e lec i e) 0.874 0.913 0.724 The solu ions I ound and implemen ed o he p oblems ha a ose in he in e na ional ma ke we e cos - e icien 0.249* 0.816* The solu ions o he p oblems ha a ose in he in e na ional ma ke we e usually inno a i e and o iginal 0.343* 0.871* I usually ound solu ions o hese p oblems ha had a lowe cos han expec ed 0.284* 0.862* In gene al, I ound and applied c ea i e solu ions o he p oblems ha a ose in he in e na ional ma ke 0.295* 0.855* No e: * Signi ican a he 1% le el. Table 2. Disc iminan alidi y (HTMT) 1. 2. 3. 4. 1. Se endipi ous IOR 2. Ac i e IOR 0.392 3. P oblem - sol ing speed 0.213 0.724 4. P oblem - sol ing c ea i i y 0.657 0.075 0.264 Table 3. S uc u al model esul s Hypo hesis Impac o On Pa h coe icien - alue 2 Hypo hesis suppo ed H1a+ Ac i e IOR PS speed 0.675 12.571*** 0.702 Yes H1b+ Ac i e IOR PS c ea i i y 0.183 2.565** 0.047 Yes H2a+ Se endipi ous IOR PS speed 0.055 0.875 0.005 No H2b+ Se endipi ous IOR PS c ea i i y 0.661 11.486*** 0.620 Yes H3 Ac i e IOR > Se endipi ous IOR PS speed 0.620 a 9.493*** -- Yes H4 Se endipi ous IOR > Ac i e IOR PS c ea i i y 0.478 a 5.526*** -- Yes No es: Two- ailed es . a Coe icien di e ences. ** Signi ican a he 5% le el. *** Signi ican a he 1% le el. Table 4. Resul s o he MICOMs p ocedu e and Mul i-g oup analysis based on pe mu a ion es S ep 1 S ep 2 S ep 3a S ep 3b Composi ional In a iance Equal Va iances Equal Means Cons uc Con igu al In a iance O iginal Co ela ion 5% Pa ial MIE Va iance- O iginal Di e ence (PNEW- PNEN) 2.5% 97.5% Equal Mean- O iginal Di e ence (PNEW- PNEN) 2.5% 97.5% Equal Full MIE SIOR Yes 0.999 0.997 Yes - 0.108 - 0.283 0.269 Yes - 0.084 - 0.297 0.300 Yes Yes AIOR Yes 1.000 0.998 Yes - 0.256 - 0.306 0.282 Yes 0.176 - 0.310 0.292 Yes Yes PSS Yes 0.998 0.998 Yes - 0.321 - 0.365 0.350 Yes 0.190 - 0.305 0.303 Yes Yes PSC Yes 0.998 0.996 Yes - 0.093 - 0.397 0.367 Yes - 0.196 - 0.302 0.299 Yes Yes Di ec e ec s on Endogenous Va iables PNEW PNEN Pe mu a ion Signi ican R 2 Di ec E ec p- alue R 2 Di ec E ec p- alue p- alue PSS 0.301 0.533 H5a: AIOR 0.547 0.000 0.777 0.000 0.030 Yes H6a: SIOR 0.076 0.435 0.094 0.302 0.892 No PSC 0.478 0.375 H5b: AIOR 0.363 0.000 0.077 0.442 0.044 Yes H6b: SIOR 0.603 0.000 0.651 0.000 0.680 No No es: SIOR: Se endipi ous in e na ional oppo uni y ecogni ion; AIOR: Ac i e in e na ional oppo uni y ecogni ion; PSS: P oblem-sol ing speed; PSC: P oblem- sol ing c ea i i y; PNEW: In e na ional en ep eneu s wi h pas nega i e en ep eneu ial expe ience subsample; PNEN: In e na ional en ep eneu s wi hou pas nega i e en ep eneu ial expe ience subsample; Pa ial MIE: Pa ial Measu emen In a iance Es ablished; Full MIE: Full Measu emen In a iance Es ablished. Figu e 1. Concep ual model Figu e 2. The model’s pa h coe icien s and R 2 s (PNEW g oup and PNEN g oup) No es: PNEW: In e na ional en ep eneu s wi h pas nega i e en ep eneu ial expe ience subsample; PNEN: In e na ional en ep eneu s wi hou pas nega i e en ep eneu ial expe ience subsample. ***Signi ican a he 1% le el.