Self-confidence predicts entrepreneurship and entrepreneurial success

This is a sel -a chi ed e sion o an o iginal a icle. This e sion may di e om he o iginal in pagina ion and ypog aphic de ails. Au ho (s): Ti le: Yea : Ve sion: Copy igh : Righ s: Righ s u l: Please ci e he o iginal e sion: CC BY 4.0 h ps://c ea i ecommons.o g/licenses/by/4.0/ Sel -con idence p edic s en ep eneu ship and en ep eneu ial success © 2023 The Au ho s. Published by Else ie Inc. Published e sion Maczulskij, Te hi; Viinikainen, Ju a Maczulskij, T., & Viinikainen, J. (2023). Sel -con idence p edic s en ep eneu ship and en ep eneu ial success. Jou nal o Business Ven u ing Insigh s, 19, A icle e00382. h ps://doi.o g/10.1016/j.jb i.2023.e00382 2023 Jou nal o Business Ven u ing Insigh s 19 (2023) e00382 A ailable online 8 Ma ch 2023 2352-6734/© 2023 The Au ho s. Published by Else ie Inc. This is an open access a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/). Sel -con idence p edic s en ep eneu ship and en ep eneu ial success Te hi Maczulskij a , b , Ju a Viinikainen c , * a ETLA Economic Resea ch, A kadianka u 23 B, 00100, Helsinki, Finland b IZA Ins i u e o Labo Economics, Schaumbu g-Lippe-S asse 5–9, 53113, Bonn, Ge many c Jy ¨ askyl¨ a Uni e si y School o Business and Economics, Uni e si y o Jy ¨ askyl¨ a, P.O. Box 35, 40100, Finland ARTICLE INFO Keywo ds: Pe sonali y En ep eneu ship En ep eneu ial success Unobse ed he e ogenei y 1. In oduc ion En ep eneu ship p omo es employmen , p oduc i i y g ow h, inno a ion ac i i y, and he comme cializa ion o inno a ions (e.g., Van P aag and Ve sloo , 2007). Howe e , only a small numbe o i ms in Finland (6%) c ea e a signi ican po ion o new jobs, and a la ge sha e o i ms expe ience limi ed o dec easing g ow h. While new jobs a e ypically c ea ed in new i ms, only a small p opo ion o hese i ms su i e he “ alley o dea h” du ing he i s yea s a e hei ma ke en y (Vanhala and Vi ´ en, 2016). Because o he economic bene i s o success ul en ep eneu ship, he ques ions o who becomes and wha makes a success ul en ep eneu a e o in e es o bo h academics and policymake s. En ep eneu ship di e s om paid employmen in se e al ways. Fi s , en ep eneu s bea subs an ial isk, which is ypically poo ly di e si ied (Ve eshchagina and Hopenhayn, 2009). Second, en ep eneu s a e cha ac e ized as “jacks-o -all- ades” who mus be compe en in many skills (Lazea , 2005). Thi d, conce ning job asks, an en ep eneu needs o de ec and exploi oppo uni ies and make apid decisions in unce ain condi ions (Rauch and F ese, 2007). Due o hese cha ac e is ics, en ep eneu ship, pa icula ly when success ul, equen ly equi es “en ep eneu ial capi al,” consis ing o inancial, social, and human capi al dimensions (Blumbe g and P ann, 2016). Pe sonali y ai s can be conside ed pa o human capi al. P e ious esea ch has ound ha en ep eneu ship is ela ed o pe sonali y ai s such as high openness o expe iences (in he Big Fi e sense), need o achie emen , sel -e icacy, in e nal locus o con ol, and isk ole ance ( o e iews, see Zhao and Seibe , 2006; Ke e al., 2018; K i ikos, 2022). Howe e , he pe sonali y ai s ha d i e people o open new businesses a e no always he same ones ha make en ep eneu s success ul (Caliendo e al., 2014; Benne e al., 2016; Hamil on e al., 2019). Fo example, ounde CEOs a e no necessa ily bes sui ed o managemen (Benne e al., 2016), and ai s such as openness o expe ience, which a e ela ed o en y in o en ep eneu ship, may no p edic i m su i al (Caliendo e al., 2014). The e is e idence ha amily backg ound and gene ics may also explain en ep eneu ial ou comes (Nicolaou e al., 2008; Ca della e al., 2020). Howe e , as amily backg ound and gene ics may also a ec pe sonali y ai s (K uege and Johnson, 2008), i emains unclea whe he * Co esponding au ho . E-mail add ess: [email p o ec ed] (J. Viinikainen). Con en s lis s a ailable a ScienceDi ec Jou nal o Business Ven u ing Insigh s jou nal homepage: www.else ie .com/loca e/jb i h ps://doi.o g/10.1016/j.jb i.2023.e00382 Recei ed 29 No embe 2022; Recei ed in e ised o m 22 Feb ua y 2023; Accep ed 3 Ma ch 2023 Jou nal o Business Ven u ing Insigh s 19 (2023) e00382 2 pe sonali y ai s a e ela ed o en ep eneu ship a e accoun ing o hese po en ial con ounding ac o s. Acqui ing a be e unde - s anding o he mechanisms behind he di e en s ages o he en ep eneu ial p ocess can o e use ul guidance in de eloping mo e e ec i e policy in e en ions. In his s udy, we examine how pe sonali y ai s (neu o icism, ex a e sion, sociabili y, hos ili y, sel -con idence, and conscien- iousness) a e ela ed o he p opensi y o become and succeed as an en ep eneu . We go beyond he exis ing li e a u e by p o iding wo impo an con ibu ions. Fi s , we use su ey da a on pe sonali y ai s linked wi h long- e m adminis a i e egis e da a, which enables us o ack en ep eneu ial ac i i y and success o e a 20-yea pe iod (1990–2009). Second, we u ilize win da a me hods o con ol o unobse ed amily backg ound and gene ic e ec s. Ou esul s show ha sel -con idence is an impo an p edic o o bo h en y in o en ep eneu ship and be e subsequen en ep eneu ial pe o mance. 2. Ma e ials and me hods 2.1. Da a We use da a ha combine in o ma ion om wo da a sou ces: he Olde Finnish Twin Coho S udy om he Depa men o Public Heal h a he Uni e si y o Helsinki and he Finnish Longi udinal Employe -Employee Da a (FLEED) om S a is ics Finland. These wo da ase s we e linked using unique pe sonal iden i ie s; hus, he ma ching was exac . The Olde Finnish Twin Coho S udy includes same-sex wins bo n be o e 1958 (Kap io e al., 1979). The subjec s o he su ey we e collec ed om he Cen al Popula ion Regis y in 1975. The esponse a e in he 1975 su ey was 89% (25,004 pa icipan s aged ≥18), and subsequen ollow-ups we e conduc ed in 1981 (84% esponse a e) and 1990 (77% esponse a e). The Finnish Coho S udy is ep esen a i e o he o e all Finnish popula ion in he same age coho s. The su ey da a we e linked o egis e in o ma ion on educa ion, demog aphic and amily cha ac e is ics, socioeconomic s a us, and employmen s a us om FLEED o e he 1990–2009 pe iod. The in o ma ion in FLEED is egis e -based, which elimina es he isks o non esponse and measu emen e o s ela ed o sel - epo ed measu es. Da a on pe sonali y ai s we e d awn om he 1981 su ey o he Finnish Coho S udy. This in o ma ion was linked wi h egis e - based da a om FLEED, which o med a sample o 16,397 indi iduals. Res ic ing he analyses o wo king indi iduals o whom we ha e in o ma ion on o he con ol a iables educed he sample o 13,755 indi iduals, wi h a o al o 170,283 yea ly obse a ions. The younges coho was 33 yea s old in 1990 and 52 yea s old in 2009. 2.2. Measu es 2.2.1. Dependen a iables: en ep eneu ial ou comes To measu e so ing in o en ep eneu ship, we use wo indica o s. The i s is en ep eneu ship s a us in yea ( =1990,…, 2009), which equals 1 i he pe son is an en ep eneu and 0 i he pe son is in paid employmen . The second indica o cap u es en y in o en ep eneu ship. The a iable equals 1 o hose who en e ed en ep eneu ship be ween yea s -1 and and 0 o hose who we e no en ep eneu s in ei he yea (excluding e i ed pe sons). The in o ma ion on en ep eneu ship s a us is based on S a is ics Finland’s egis e s, whe e en ep eneu s a e de ined as pe sons who, du ing he las week o he yea , had a sel -employed pe son’s pension insu ance and we e no unemployed. I he pe son is also in an employmen ela ionship, hei en ep eneu ial income has o exceed hei wage income o he pe son o be de ined as an en ep eneu . Fu he mo e, people whose en ep eneu ial incomes exceed a speci ic le el o ea nings belong o he ca ego y o en ep eneu s ( o a de ailed de ini ion, see Appendix A). We use h ee indica o s o en ep eneu ial success. The i s ou come a iable is an indica o cap u ing exi om en ep eneu ship. The a iable equals 1 o hose who exi ed om en ep eneu ship be ween yea s -1 and and 0 o hose who we e en ep eneu s in bo h yea s. Those who exi ed en ep eneu ship because o e i emen we e excluded om he analyses. The second a iable is he loga i hm o he sum o capi al and ea ned income om en ep eneu ship in yea . As he hi d and las indica o o en ep eneu ial success, we use he loga i hm o he i m’s o al sales di ided by he numbe o employees in yea . 1 This is a p oxy o he i m’s labo p oduc i i y. Va iables o income and labo p oduc i i y we e de la ed o 2009 p ices using S a is ics Finland’s cos -o -li ing index. All he dependen a iables we e based on he FLEED egis e da a. 2.2.2. P edic o a iables: pe sonali y ai s In o ma ion on pe sonali y ai s was d awn om he Olde Finnish Twin Coho S udy. Two pe sonali y ai s (neu o icism and ex a e sion) a e based on he sho - o m Eysenck Pe sonali y In en o y d awn om he 1981 su ey. Fu he mo e, he 1981 su ey con ains 15 i ems ha desc ibe di e en dimensions o pe sonali y (e.g., “lazy e sus ha dwo king” o “insecu e e sus sel -con i- den ”), which we e assessed based on a 5-poin scale (Langin ainio e al., 1982). The Kaise -Meye -Olkin (KMO) measu e o sampling adequacy (KMO =0.76) and Ba le ’s es o sphe ici y ( χ 2 =1.29e +06, d =105, p <0.001) bo h suppo ed he ac o abili y o he 15-i em pe sonali y ma ix. P incipal componen analysis e ained ou la en ac o s o pe sonali y, simila o Langin ainio e al. 1 In o ma ion on he i m’s o al sales by he numbe o pe sonnel is a ailable only o he yea s 1998–2009. The sales in o ma ion is based on da a om he Finnish T ade Regis e , a public egis e ha con ains in o ma ion on p i a e ade s and businesses. The ade egis e co e s he en i e coun y and includes all ade s, ega dless o hei co po a e o m o na u e o ac i i y. A limi ed company mus always epo i s inancial s a emen s o he ade egis e , bu p i a e en ep eneu s a e no equi ed o do so. The limi ed epo ing obliga ion, along wi h missing in o ma ion, limi s he co e age o sales in o ma ion, which is app oxima ely 26% in ou sample o en ep eneu s in 1998–2009. T. Maczulskij and J. Viinikainen Jou nal o Business Ven u ing Insigh s 19 (2023) e00382 3 (1982). The la en ac o s we e labeled as sociabili y, hos ili y, sel -con idence, and conscien iousness. 2 A de ailed lis o he 15 i ems and hei o a ed ac o loadings is p esen ed in Appendix Table B1. Each ac o was s anda dized such ha he dis ibu ion had a mean o 0 and a s anda d de ia ion o 1. 2.2.3. Con ol a iables As con ol a iables, we use sex, age, age squa ed, an indica o o ha ing unde age child en, ma i al s a us (1 =ma ied, 0 = o he wise), le el o educa ion ( i e ca ego ies), and ield o educa ion (nine ca ego ies). The in o ma ion on he le el and ield o educa ion is based on he In e na ional S anda d Classi ica ion o Educa ion. These con ol a iables we e d awn om FLEED in yea ( =1990,…, 2009). We also include sel - epo ed in o ma ion on occupa ion, including 12 indica o s, which in he case o en e- p eneu s indica es he b anch o indus y. We con ol o occupa ion in 1975 as i is likely o be associa ed wi h bo h pe sonali y ai s and en ep eneu ial ou comes and may hus con ound he ela ionship o in e es . To condi ion ou po en ial wo-way causali y, we also con ol o en ep eneu ial s a us (1 =en ep eneu , 0 =o he wise) ha o igina es om he 1975 su ey (i.e., be o e he pe - sonali y ai s we e measu ed). 2.3. Models and me hods To s udy he ela ionship be ween pe sonali y ai s and en ep eneu ial ou comes, we es ima e models using he ollowing o m (Equa ion (1)): Yi = α ′Pi,1981 +β′Xi +δ′Oi,1975 + τ + ε i (1) In Equa ion (1), i and e e o an indi idual and yea , espec i ely; Y is he ou come a iable; P i,1981 is a ec o o six pe sonali y ai s o indi idual i measu ed in 1981; and X i is a ec o o con ol a iables measu ed in yea . The model also includes yea in- dica o s ( τ ) and a ec o ha includes occupa ion and en ep eneu ship s a us measu ed in 1975 (O i,1975 ). Indi idual-le el clus e ed s anda d e o s a e used o ake in o accoun he wi hin-pe son co ela ion o e ime. 2.3.1. Baseline model A pooled logi model is used o s udy whe he pe sonali y ai s a e ela ed o he p obabili y o being an en ep eneu and en y in o and exi om en ep eneu ship. When he loga i hm o en ep eneu ial income and he i m’s p oduc i i y we e used as dependen a iables, he models we e es ima ed using o dina y leas squa es (OLS). 2.3.2. Models u ilizing win da a Pe sonali y ai s de elop o e childhood and you h and become inc easingly s able in adul hood. Changes in pe sonali y ai s in adul hood a e mos ly ela ed o ma u a ion, o example, people end o become mo e emo ionally s able o e ime (e.g., So o and Tacke , 2015), bu majo li e e en s may s ill shape pe sonali y in adul hood (Spech e al., 2011). 3 P e ious s udies ha e shown ha app oxima ely one-hal o he a ia ion in pe sonali y ai s is explained by gene ic ac o s, wi h a small con ibu ion o sha ed en i onmen (Boucha d and Loehlin, 2001; K uege and Johnson, 2008). The emaining a ia ion is a ibu ed pa icula ly o he non-sha ed en i onmen , esul ing om, o example, pee g oup in luences, amily in e ac ions, and andom li e e en s, and wins expe iencing he same en i onmen di e en ly (e.g., Bake and Daniels, 1990; To gensen and Janson, 2002). I sha ed en i onmen and gene ic ac o s a ec bo h pe sonali y and en ep eneu ial ou comes, he indi idual-le el es ima es in he baseline model a e likely o be biased. To con ol o he a o emen ioned unobse ed cha ac e is ics, we u ilize in o ma ion on he zygosi y o wins, whe he hey a e dizygo ic (DZ) and ha e he same amily backg ound and app oxima ely 50% o hei genes o monozygo ic (MZ) and ha e bo h he same amily backg ound and 100% o genes. Fo speci ica ions ha we es ima e using OLS (loga i hm o en ep eneu ial income and he i m’s p oduc i i y), we apply he adi ional win-pai di e ence model. Speci ica ions whe e he dependen a iable is bina y ( he p obabili y o being an en ep eneu , en y in o o exi om en ep eneu ship), we add win-speci ic ixed e ec s in he model and apply a condi ional ( ixed-e ec s) logi eg ession o wins (e.g., Magnac, 2004). The s anda d e o s in bo h models a e clus e ed by win pai s, and he models use a sample in which bo h wins a e obse ed in yea . Addi ionally, o be able o exploi he win dimension o he da a o iden i ica ion, he indi idual membe s o he win pai mus ha e di e en employmen s a uses, wi h, o example, only one o he wins being an en ep eneu . This dec eases he sample size signi ican ly and hus inc eases he isk o a Type II e o (i.e., non ejec ion o a alse null hypo hesis). The models a e es ima ed sepa a ely o DZ and MZ wins, o which he o me speci ica ion con ols o he sha ed amily backg ound and pa ly o gene ics, and he la e con ols addi ionally o he whole gene ic inhe i ance. The e o e, as a as di - e ences in pe sonali y ai s a e no a ibu able o measu emen e o s, he esul s based on MZ wins iden i y he link be ween pe sonali y ai s and en ep eneu ial ou comes, whe e he a ia ion in pe sonali y ai s s ems om he non-sha ed en i onmen , such as di e en li e expe iences. 2 The sociabili y and ex a e sion ai s a e highly co ela ed ( =0.63). We he e o e e-es ima ed all he models epo ed in Tables 1 and 2 by including hese wo a iables in he models sepa a ely. The main conclusions emained in ac . 3 Appendix C p o ides addi ional e idence on how emo ional closeness be ween amily membe s and di e en li e e en s a e ela ed o di e ences in pe sonali y ai s among monozygo ic wins in ou sample. T. Maczulskij and J. Viinikainen Jou nal o Business Ven u ing Insigh s 19 (2023) e00382 4 3. Resul s 3.1. Desc ip i e s a is ics Fig. 1 p esen s he scales and dis ibu ions o he six s anda dized pe sonali y ai sco es by employmen s a us. The dis ibu ions o he pe sonali y ai s a e app oxima ely no mal, excep o hos ili y, o which he dis ibu ion is le skewed. The dis ibu ions o pe sonali y ai s by employmen s a us show ha en ep eneu s a e sligh ly mo e sel -con iden and conscien ious, on a e age, compa ed o paid employees. The a e age sco es o he s anda dized pe sonali y measu es and he key demog aphic cha ac e is ics sepa a ely o en ep eneu s and paid employees a e epo ed in Appendix Table B2. App oxima ely 14% o all he employed in- di iduals in ou da a a e en ep eneu s. This sha e is compa able wi h he o al employed wo k o ce in Finland in he same age dis ibu ion as ou sample (indi iduals o e 33 yea s old). As Fig. 1 indica es, en ep eneu s’ sco es in sel -con idence and consci- en iousness a e, on a e age, highe han hose o paid employees. They also sco e lowe in neu o icism and hos ili y on a e age. Compa ed o he paid employees in ou sample, he en ep eneu s a e also olde , less educa ed, and mo e likely o be males han emales. Las ly, he means o he absolu e alues o he win di e ences show ha MZ wins a e much mo e simila o each o he in e ms o pe sonali y ai s han DZ wins (Appendix Table B3). Howe e , di e ences in en ep eneu ship s a us and pe sonali y ai s also exis be ween MZ wins, which is necessa y o model iden i ica ion. 3.2. Pe sonali y and en ep eneu ship The ma ginal e ec s indica ing he links be ween pe sonali y ai s and he p obabili y o being an en ep eneu a e epo ed in Table 1 4 . The esul s show ha a one-s anda d de ia ion inc ease in sel -con idence is associa ed wi h a 1.6 pe cen age poin inc ease (β =0.016) in he p obabili y o being an en ep eneu (Column 1). This ma ginal e ec is equi alen o an inc ease o app oxima ely 12% in he same p obabili y e alua ed a a mean o 0.139. The poin es ima e o sel -con idence is 0.040 (~8% inc ease in p oba- bili y) when we ocus on he DZ sample, which con ols o unobse ed amily backg ound and pa o he gene ic e ec s (Column 2), and 0.067 (~13% inc ease in p obabili y) when he esul s a e based on he MZ win design, which condi ion ou all he unobse ed amily backg ound and gene ic e ec s ha iden ical wins sha e (Column 3). Howe e , sel -con idence in he DZ sample is no s a- is ically signi ican a he con en ional le el (p =0.108), whe eas in he MZ sample we also obse e ha neu o icism is posi i ely ela ed o en ep eneu ial p obabili y. The esul s and e ec sizes conce ning en y in o en ep eneu ship (Columns 4–6) a e in line wi h he esul s abo e. The di e ences in he esul s conce ning en ep eneu ial s a us (pe cen inc eases in p obabili ies) be ween he baseline, DZ, and MZ models sugges s ha omission o amily backg ound and gene ics may bias he esul s in opposi e di ec ions and o se each o he in ou model. 3.3. Pe sonali y and en ep eneu ial success Table 2 shows how he pe sonali y ai s in ou da a a e ela ed o di e en measu es o en ep eneu ial success. The esul s indica e ha sel -con idence is nega i ely associa ed wi h an exi om en ep eneu ship (β = − 0.006, p <0.050, Column 1). The ma ginal e ec is oughly equi alen o an o 8% dec ease in p obabili y. We also ind ha a one-s anda d de ia ion inc ease in sel - con idence is ela ed o an app oxima ely 6% highe en ep eneu ial income (Column 4). The posi i e link be ween sel -con idence and en ep eneu ial success emains obus and he poin es ima es become la ge when con olling o bo h amily and gene ic e - ec s using he sample o MZ wins. Fo example, a one s anda d de ia ion inc ease in sel -con idence is ela ed o a nea ly 20% highe en ep eneu ial income in Column 6. A compa ison o he baseline and MZ esul s sugges s ha omi ing bo h amily backg ound and gene ic ac o s may lead o downwa d biased es ima es when examining he ela ionship be ween sel -con idence and en ep eneu ial success. Using he sample o MZ wins, we also ind ha sociabili y is posi i ely ela ed o an exi om en ep eneu ship (β =0.068, p <0.050). The esul s ega ding he ela ionship be ween pe sonali y ai s and he i m’s labo p oduc i i y a e shown in Column 7. Because in o ma ion on i m sales is a ailable only o a subsample o en ep eneu s, he sample dec eased o 2,779 yea ly obse a ions. The wi hin- win di e encing me hod o MZ wins u he dec eases he sample size o only 137 obse a ions; he e o e, we epo only he esul s using da a as an indi idual panel. Howe e , he indings based on his smalle sample also indica e ha sel -con idence is ela ed o be e en ep eneu ial success as a one-s anda d de ia ion inc ease in sel -con idence is ela ed o an app oxima ely 6% highe labo p oduc i i y among en ep eneu s. 4. Discussion Using longi udinal da a ha combine pe sonali y ai s wi h en ep eneu ial ou comes, we ound ha highe sel -con idence was associa ed wi h a highe p obabili y o becoming an en ep eneu and be e en ep eneu ial pe o mance measu ed by exi om en ep eneu ship, en ep eneu ial income, and he i m’s p oduc i i y. The link be ween sel -con idence and en ep eneu ial ou - comes emained posi i e when he po en ial con ounde s om sha ed amily and gene ic backg ound we e con olled o using he MZ win sample. P e ious s udies examining sel -con idence and en ep eneu ship ha e ocused on en ep eneu ial sel -e icacy (ESE). This speci ic 4 The ma ginal e ec s o he o he con ol a iables a e shown in Appendix Table B4. These esul s show, o example, ha emales a e less likely o be en- ep eneu s han males. P e ious s udies ha e ound ha sex di e ences in pe sonali y ai s may pa ly explain he di e ences in he p e alence o en ep eneu ship be ween he sexes (Ki kwood, 2009). Based on he esul s shown in Table B4, sex di e ences in en ep eneu ship s ill exis e en i pe sonali y di e ences a e aken in o accoun . T. Maczulskij and J. Viinikainen Jou nal o Business Ven u ing Insigh s 19 (2023) e00382 5 domain o sel -con idence e e s o a pe son’s belie in hei abili y o success ully launch an en ep eneu ial en u e (McGee e al., 2009). In line wi h ou esul s, hese s udies ha e shown ha ESE is posi i ely ela ed o en ep eneu ial in en ion, en u e c ea ion, and en ep eneu ial pe o mance ( o a e iew, see Newman e al., 2019). In addi ion, gene al sel -e icacy, which is no speci ically ela ed o he en ep eneu ial domain, has been ound o p edic s a -up pe o mance and su i al du ing he i s yea s a e business c ea ion (Caliendo e al., 2023). In ela ion o sel -con idence, p e ious esea ch has also ound ha en ep eneu s, pa icula ly Fig. 1. Dis ibu ions o s anda dized pe sonali y ai sco es o paid employees (ligh -b own ba s) and en ep eneu s (black-lined ba s). Table 1 Ma ginal e ec s o pe sonali y ai s and en ep eneu ship. En ep eneu ship (1) En ep eneu ship wi hin DZ wins (2) En ep eneu ship wi hin MZ wins (3) En y in o en ep eneu ship (4) En y in o en ep eneu ship wi hin DZ wins (5) En y in o en ep eneu ship wi hin MZ wins (6) Neu o icism 0.002 (0.003) 0.001 (0.026) 0.074 (0.028) *** 0.0002 (0.0003) −0.028 (0.028) 0.062 (0.037) * Ex a e sion −0.001 (0.004) 0.010 (0.027) 0.050 (0.038) 0.0002 (0.0004) 0.001 (0.031) 0.036 (0.044) Sociabili y 0.000 (0.004) −0.023 (0.025) −0.002 (0.036) 0.0004 (0.0003) −0.007 (0.030) 0.028 (0.051) Hos ili y 0.003 (0.003) 0.016 (0.020) 0.053 (0.030) * 0.0001 (0.0003) 0.028 (0.024) −0.010 (0.032) Sel -con idence 0.016 (0.003) *** 0.040 (0.025) 0.067 (0.029) ** 0.0009 (0.0003) *** 0.046 (0.031) 0.066 (0.034) * Conscien iousness 0.002 (0.003) −0.009 (0.020) −0.004 (0.028) −0.0000 (0.0003) 0.015 (0.026) 0.014 (0.037) Family ixed e ec s No Yes Yes No Yes Yes Gene ic ixed e ec s No No Yes No No Yes O he con ols Yes Yes Yes Yes Yes Yes Mean o he ou come 0.139 0.50 0.50 0.008 0.50 0.50 No. o obse a ions 170,283 27,890 12,707 166,195 10,685 5,340 No es: S anda dized coe icien s. DZ =dizygo ic wins; MZ =monozygo ic wins. O he con ols include ma i al s a us, ha ing unde age child en, ield and le el o educa ion, and p io occupa ion and en ep eneu ship s a us. The models in Columns 1 and 4 also include age, age squa ed, sex, and yea indica o s as con ols. Pe sonali y ai s we e measu ed in 1981, p io occupa ion and en ep eneu ship s a us we e measu ed in 1975, and o he con ol a iables we e measu ed in yea ( = 1990,…, 2009). S anda d e o s a e clus e ed a he indi idual le el (Columns 1 and 4) o a he win-pai le el (Columns 2–3 and 5–6). Es ima ions we e pe o med using he logi model (Columns 1 and 4) o he condi ional ( ixed e ec s) logi model (Columns 2–3 and 5–6). *** (p <0.01), ** (p <0.05), * (p <0.10). T. Maczulskij and J. Viinikainen Jou nal o Business Ven u ing Insigh s 19 (2023) e00382 6 Table 2 Pe sonali y ai s and en ep eneu ial success. Exi om en ep eneu ship (1) Exi om en ep eneu ship wi hin DZ wins (2) Exi om en ep eneu ship wi hin MZ wins (3) Log o en ep eneu ial income (4) Log o en ep eneu ial income wi hin DZ wins (5) Log o en ep eneu ial income wi hin MZ wins (6) Log o i m p oduc i i y (7) Neu o icism 0.001 (0.003) −0.015 (0.041) 0.023 (0.028) 0.043 (0.026) −0.035 (0.064) 0.060 (0.095) −0.037 (0.028) Ex a e sion 0.005 (0.003) * 0.015 (0.055) −0.031 (0.049) −0.020 (0.032) 0.051 (0.070) 0.093 (0.089) −0.015 (0.038) Sociabili y 0.003 (0.003) 0.016 (0.056) 0.068 (0.034) ** 0.003 (0.029) −0.059 (0.076) −0.060 (0.088) 0.004 (0.034) Hos ili y 0.002 (0.002) −0.023 (0.042) 0.025 (0.038) 0.002 (0.024) 0.038 (0.058) 0.080 (0.074) 0.013 (0.026) Sel -con idence −0.006 (0.003) ** 0.058 (0.049) −0.072 (0.042) * 0.059 (0.026) ** 0.002 (0.070) 0.188 (0.090) ** 0.062 (0.034) * Conscien iousness −0.002 (0.002) −0.097 (0.032) 0.026 (0.027) −0.018 (0.023) −0.078 (0.076) −0.048 (0.084) −0.007 (0.025) Family ixed e ec s No Yes Yes No Yes Yes No Gene ic ixed e ec s No No Yes No No Yes No O he con ols Yes Yes Yes Yes Yes Yes Yes Mean o he ou come 0.074 0.50 0.50 9.27 … … 11.26 No. o obse a ions 22,009 4,999 2,574 19,007 4,079 2,276 2,779 No es: S anda dized coe icien s. O he con ols include ma i al s a us, ha ing unde age child en, ield and le el o educa ion, and p io occupa ion and en ep eneu ship s a us. The models in Columns 1, 4, and 7 also include age, age squa ed, sex and yea indica o s as con ols. Pe sonali y ai s we e measu ed in 1981, p io occupa ion and en ep eneu ship s a us we e measu ed in 1975, and o he con ol a iables we e measu ed in yea ( =1990,…, 2009). S anda d e o s a e clus e ed a he indi idual le el (Columns 1, 4, and 7) o a he win-pai le el (Columns 2–3 and 5–6). Es ima ions we e pe o med using he logi model (Column 1), condi ional ( ixed e ec s) logi model (Columns 2–3), o OLS (Columns 4–7). ** (p <0.05), * (p <0.10). T. Maczulskij and J. Viinikainen Jou nal o Business Ven u ing Insigh s 19 (2023) e00382 7 ounde -manage s, may e en be o e con iden (Fo bes, 2005). Fu he mo e, sex di e ences in sel -con idence may pa ly explain why mo e males han emales ha e en ep eneu ial ca ee s (Ki kwood, 2009). The win design has limi a ions. In pa icula , he MZ win design allows us o es ablish associa ions ha a e no induced by a sha ed amily en i onmen o gene ics. Howe e , he e may also be o he a e unobse able ac o s ha a ec bo h pe sonali y and en ep eneu ial ou comes; he e o e, esul s based on he win app oach may no e eal causal e ec s (e.g., Boa dman and Fle che , 2015). Ano he limi a ion o his app oach is ha po en ial measu emen e o s in pe sonali y ai s may yield es ima es ha a e mo e biased han hose om o dina y c oss-sec ional analyses a he indi idual le el (e.g., Bound and Solon, 1999). Finally, he possibili y o e e se causali y canno be uled ou . To con ol o po en ial e e se causali y, we used pe sonali y ai s ha we e measu ed be o e he ou comes, and we also con olled o en ep eneu ial s a us ob ained om he 1975 su ey (i.e., be o e he pe sonali y ai s we e measu ed). Howe e , i is possible ha en ep eneu ial expe iences be o e he measu emen o pe sonali y a ec ed sel -con idence (Newman e al., 2019). The main s eng h o his s udy is longi udinal da a ha we e linked o egis e in o ma ion on en ep eneu ial ou comes. The longi udinal da a, which co e ed 20 yea s, allowed us o examine long- e m en ep eneu ial success. Fu he mo e, egis e in o - ma ion elimina es po en ial biases ela ed o sel - epo ed measu es. We also u ilized he win dimension o he da a, which enabled us o con ol o unobse ed sha ed gene ic and amily backg ounds ha may o he wise ha e con ounded he esul s. 5. Conclusion High sel -con idence may inc ease bo h he p obabili y o being an en ep eneu and en ep eneu ial pe o mance. Ea lie s udies ha e shown ha sel -con idence, which is speci ically ela ed o en ep eneu ship (i.e., ESE), may no be s able and can change (Newman e al., 2019). Fo example, Wilson e al. (2007) ound ha en ep eneu ship educa ion signi ican ly inc eased he le el o ESE in hei s udy, pa icula ly among emales. The e o e, p o iding suppo ha enhances people’s sel -con idence may p omo e bo h en ep eneu ship and en ep eneu ial pe o mance in socie y. Au ho con ibu ion Te hi Maczulskij: Concep ualiza ion, Me hodology, Fo mal analysis, W i ing – o iginal d a ; W i ing – e iew & edi ing. Ju a Viinikainen: Concep ualiza ion, Me hodology, W i ing – o iginal d a , W i ing – e iew & edi ing. Decla a ion o compe ing in e es The au ho s decla e ha hey ha e no known compe ing inancial in e es s o pe sonal ela ionships ha could ha e appea ed o in luence he wo k epo ed in his pape . Acknowledgemen s The au ho s a e g a e ul o an anonymous e e ee and he Edi o o use ul commen s and sugges ions. Financial suppo om he Helsinki Bou se Club is g a e ully acknowledged. Appendix A De ini ion o en ep eneu ship In S a is ics Finland’s egis e s, en ep eneu s a e de ined as pe sons aged 18–74 who ha e a alid en ep eneu ’s pension in- su ance in he las week o he yea and who a e no unemployed, consc ip s, o ci il se an s a he ime. I a pe son has a alid employmen ela ionship in addi ion o an en ep eneu ’s pension insu ance, he en ep eneu ial income mus be highe han he sala y income o a pe son o be de ined as an en ep eneu . In addi ion, hose pe sons whose en ep eneu ial incomes exceed he de ined income limi a e classi ied as en ep eneu s i hey a e no e i ed. The income limi is de ined annually by S a is ics Finland using da a on en ep eneu s om he Labo Fo ce Su ey. Appendix B Table B1 Ro a ed ac o loadings. Va iable Fac o 1 Fac o 2 Fac o 3 Fac o 4 Talka i e .724 .187 −.042 −.084 Exp esses eelings .684 −.053 −.180 .158 Open .787 .076 −.064 .137 Social .748 .062 −.101 −.052 De e mined .089 .718 −.136 .111 Sel -con iden .162 .744 −.114 −.003 (con inued on nex page) T. Maczulskij and J. Viinikainen Jou nal o Business Ven u ing Insigh s 19 (2023) e00382 8 Table B1 (con inued ) Va iable Fac o 1 Fac o 2 Fac o 3 Fac o 4 Dominan .091 .681 .177 −.178 Ambi ious −.054 .587 .118 −.179 Qua els o en .027 −.029 .806 −.014 O en ang y .010 −.023 .889 .045 Easily i i able −.053 −.017 .814 .013 Ha dwo king −.110 .378 −.007 .418 Always on ime −.243 .345 .034 .474 Hones .032 −.051 .004 .801 Always ells he u h .164 −.141 .034 .784 Fac o name Sociabili y Sel -con idence Hos ili y Conscien iousness No e: High ac o loadings a e bold aced. Table B2 Summa y s a is ics o he basic cha ac e is ics by employmen s a us, 1990–2009. En ep eneu s Paid employees - es Pe sonali y ai s Neu o icism −0.145 −0.044 4.09 *** Ex a e sion 0.028 0.018 0.41 Sociabili y −0.014 0.001 0.59 Hos ili y −0.035 0.007 1.70 * Sel -con idence 0.175 0.021 6.32 *** Conscien iousness 0.011 −0.054 2.69 *** Demog aphic cha ac e is ics Female (indica o ) 0.39 0.54 12.00 *** Age 49.4 48.5 7.67 *** Ha ing child en (<18 yea s old) (indica o ) 0.42 0.40 2.98 *** Ma ied (indica o ) 0.78 0.69 9.55 *** Yea s o educa ion 11.5 12.2 12.01 *** No. o obse a ions 23,685 146,598 No es: S anda dized pe sonali y sco es. The pe sonali y ai sco es a e based on he 1981 su ey, and he means o he demog aphic cha ac e is ics a e based on he egis e da a co e ing he yea s 1990–2009. The - es s o equal g oup means by employmen s a us a e clus e ed a he indi idual le el. ***p <0.01, *p <0.10. Table B3 Wi hin- win di e ences in pe sonali y and en ep eneu ship in absolu e alues. Wi hin DZ wins Wi hin MZ wins En ep eneu 0.203 0.157 Neu o icism 0.985 0.831 Ex a e sion 1.007 0.777 Sociabili y 1.080 0.798 Hos ili y 1.047 0.893 Sel -con idence 14.003 0.844 Conscien iousness 1.010 0.871 Table B4 Full esul s o he model p esen ed in Table 1, column 1 (ma ginal e ec s). En ep eneu ship Neu o icism 0.002 (0.003) Ex a e sion −0.001 (0.004) Sociabili y 0.000 (0.004) Hos ili y 0.003 (0.003) Sel -con idence 0.016 (0.003) *** Conscien iousness 0.002 (0.003) Female −0.037 (0.007) *** Age 0.003 (0.002) Age 2 /100 −0.003 (0.002) Ma ied 0.034 (0005) *** Ha ing child en (<18 yea s old) 0.021 (0.005) *** Educa ion le el Seconda y educa ion −0.024 (0.020) Lowes le el e ia y −0.018 (0.023) Lowe deg ee le el e ia y −0.081 (0.021) *** Highe deg ee le el e ia y −0.048 (0.023) ** Field o educa ion Educa ion −0.062 (0.024) ** A s and humani ies −0.009 (0.028) Social sciences and business −0.003 (0.019) (con inued on nex page) T. Maczulskij and J. Viinikainen