The role of regional differences in immigration: the case of OECD countries

190 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 THE ROLE OF REGIONAL DIFFERENCES IN IMMIGRATION: THE CASE OF OECD COUNTRIES Domicián Má é, Uni e si y o Deb ecen, Facul y o Economics and Business, Deb ecen, Hunga y, E-mail: [email p o ec ed] Im an Sa ihasan, Uni e si y o Deb ecen, Ká oly Ih ig Doc o al School Deb ecen, Hunga y, E-mail: [email p o ec ed] Józse Popp, Uni e si y o Deb ecen, Facul y o Economics and Business, Deb ecen, Hunga y, E-mail: [email p o ec ed] Judi Oláh, Uni e si y o Deb ecen, Facul y o Economics and Business, Deb ecen, Hunga y, E-mail: [email p o ec ed] Recei ed: Ap il, 2018 1s Re ision: May, 2018 Accep ed: Augus , 2018 DOI: 10.14254/2071- 789X.2018/11-3/12 ABSTRACT. Despi e physical, cul u al, economic and o he obs acles, millions o people ha e ecen ly emig a ed om one coun y o ano he in sea ch o a be e li e. Consequen ly, he impo ance o his esea ch opic has g own o e ime. Analysis o his egional app oach is based on he OECD censuses, and i s unique Da abase on Immig an s (DIOC), which makes i possible o gene a e a wide a ie y o c oss- abula ions on he cha ac e is ics o asylum-seeke s by he coun y o hei bi h. The aim o his s udy is o con ibu e o li e a u e by analysing he ole o egional di e ences in mig a ion mo e closely in e ms o he impo ance o mig a ion egula ion policies in he OECD coun ies. Acco ding o he esul s based on Bina y Logis ic (Logi ) and Linea (OLS) Reg ession models, he educa ional a ainmen , age, sex and he place o bi h o mig an s a e ela ed in di e en ways o hei egion o o igin. Mo eo e , male mig an s a e mo e skilled han emales, mig an s who a e mo e educa ed a e olde han he less skilled, and o eign-bo n mig an s seemed o be olde han na i e-bo n mig an s. The e is also addi ional e idence sugges ing ha di e en egions and coun ies migh ollow di e en policies and no ms o admission o mig an s. The e o e, o hcoming mig a ion go e nance p og ams ha aim o acili a e he labou ma ke in eg a ion o mig an s should also ake in o conside a ion hei egional cha ac e is ics. JEL Classi ica ion : K37, J21, J15 Keywo ds: mig a ion; egionalism; educa ional a ainmen , aging, OECD. Má é, D., Sa ihasan I., Popp, J., & Oláh, J. (2018). The Role o Regional Di e ences in Immig a ion: The Case o OECD Coun ies. Economics and Sociology, 11(3), 190-206. doi:10.14254/2071-789X.2018/11-3/12 191 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 In oduc ion Regions o he wo ld accoun o di e en p opo ions in he global numbe o mig an s. The o igin o in e na ional mig an s has become inc easingly di e si ied o e he pas wo decades (Uni ed Na ions, 2017). Asia and Eu ope hos ed o e 60% o all in e na ional mig an s wo ldwide in 2017, wi h nea ly 80 million li ing in Asia and 78 million in Eu ope. No h Ame ica hos ed he hi d la ges numbe o in e na ional mig an s (58 million), ollowed by A ica (25 million), La in Ame ica and he Ca ibbean (10 million), and Oceania (8 million) (UN, 2017). Asia, Eu ope and No h Ame ica accoun ed o o e 85% o he inc ease in he numbe o in e na ional mig an s be ween 1990 and 2017. Thus, be ween 2010 and 2017, A ica expe ienced he second as es annual g ow h, wi h he a e age o 1.1 million pe annum (Figu e 1). Figu e 1. A e age annual change in he numbe o in e na ional mig an s by majo egions o o igin, 1990-2017 (millions) Sou ce: Au ho s’ own compila ion based on Uni ed Na ions (UN, 2017). The la es wa e o immig a ion in he pos -c isis pe iod (2010-2015) is unlike p e ious wa es in e ms o skill and o igin s uc u e. Bu zynski, Docquie , & Rapopo (2018) ha e shown ha he wel a e e ec s o immig a ion a e e y he e ogeneous ac oss skill g oups, coun ies o des ina ion, and immig a ion wa es. Fo ins ance, he la es wa e pe iod o immig a ion is less skilled and b ings ewe bene i s. The wel a e impac o ecen mig a ion lows ha e been bene icial o 69% o he non-mig an OECD esiden s, and o 83% o non- mig an ci izens in 22 iches OECD coun ies. The leading bene icia ies a e mainly adi ional immig a ion coun ies; hei gains a e ex ensi e due o he en y o immig an s om non-OECD coun ies (Aub y, Bu zyński, & Docquie , 2016). The e a e a ious economic, social and physical easons why people emig a e and hey can usually be classi ied in o push (nega i e incen i es) and pull (posi i e) ac o s (Bod a sson & Van den Be g, 2013). The main economic eason o choosing o be a mig an is o seek an imp o ed li e by inding a job, and he d i ing o ces o mig a ion a e highe wages, human secu i y and wo k oppo uni ies (Cas les, 2013). The economic push ac o s end o be he exac opposi e o he pull ac o s ha a e associa ed wi h he a ea o des ina ion (o e popula ion, insu icien numbe o jobs, low wages e c.). Howe e , o he social easons, i.e. ci il and poli ical igh s, eligious eedom, law and o de , social mobili y, pe sonal 192 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 sa e y, and peace, oge he wi h o he a ac i e en i onmen al ac o s (clima e, seaside loca ion e c.) all make a des ina ion coun y a ac i e o po en ial immig an s. These economic, poli ical and social incen i es exis wo ldwide, hus in luencing in e na ional mig a ion o e ecen decades (Ozgu & Deniz, 2014). The economic cha ac e is ics o mig an s and hei mo i a ions o mo ing a y by age g oup and by mig an s a us (Philip, Macleod, & S ockdale, 2013). Se e al s udies (McInnis 1971; Milling on, 2000 e c.) demons a e he e is a posi i e ela ionship be ween he age o esponsi eness and he economic gains om mig a ion. Hence, younge mig an s ha e g ea adap abili y o new challenges when en e ing he labou ma ke o a des ina ion coun y. Mo eo e , egional di e ences - including wage and age dispa i ies e c. - can a ec mig a ion as well (Hun , 2012). Essen ially, people mig a e om low-income o high-income egions, which inc eases he mobili y cos o mig a ion. This p esump ion ocuses almos exclusi ely on he size and di ec ion o popula ion lows ac oss egions (Bo jas, B ona s, & T ego, 1992). Thus, he ela ionship be ween mig a ion, age, educa ion, seg ega ion and dis ance ha e been discussed in many economic s udies (Schwa z, 1976;, Muese , 1989; Iceland, 2017 e c.). Al hough mig an s ypically ans e om lowe o highe income egions, mig a ion lows inc ease wi h highe educa ion and educe wi h aging. Ne e heless, he economic impac o mig a ion depends on he numbe and he educa ion le el o mig an s and also on how well educa ion o mig a ing indi iduals ma ches hei u u e occupa ions (Quinn & Rubb, 2005). The skill- and o igin-mix o mig a ion o he OECD has e ol ed o e ime. Meanwhile, many OECD coun ies ha e been a emp ing o a ac quali ied human esou ces om ab oad, since knowledge-based and p og essi e economies need o sus ain hei economic g ow h (Dumon & Lamai e, 2005). I is bene icial o hese coun ies o acqui e mig an s who ha e solid educa ion and well-de eloped skills in o de o imp o e hei economic pe o mance, since i has been widely acknowledged ha educa ion and skills oge he a e he majo sou ces o economic p ospe i y (Flisi, Me oni, & Ve a-Toscano, 2016). These in en ions a e highly dependen on he ime ame in ol ed, i.e. educa ion has a g ea e in luence on long‐ e m mig a ion decisions (Williams e al., 2018). Mo eo e , he in luence o mig a ion may lead o changes in he a e age le els o educa ion in des ina ion coun ies and gene a e educa ional ex e nali ies and new incen i es o human capi al in es men s (Dus mann & Gli z, 2011). Fu he mo e, each mig an is a sou ce o pe sonal consump ion expendi u e, which is also ans e ed among he egions. Di Gio anni, Le chenko, & O ega (2015) compa ed wel a e wi h le els o mig a ion om he c oss-coun y global analysis pe spec i e, and ound ha na i es in hose coun ies (Canada, Aus alia) ha ecei ed high le els o mig a ion a e be e o due o a g ea e p oduc a ie y a ailable in consump ion and as in e media e inpu s. Meanwhile, in he sho un, skilled and unskilled na i es end o expe ience wel a e changes wi h opposi e signs. Ini ial s udies ha e discussed he p ocess o unde s anding he ela ionships be ween ade, mig a ion, and egional changes (Fou nie , 1989). Meanwhile, educa ion sys ems and social policies a e no simila ac oss egions, while socioeconomic s a us and e hnocul u al a angemen s o immig an s di e ge conside ably. Such cha ac e is ics o mig a ion may be associa ed wi h egional di e ences in he educa ion ou comes. Mig a ion also has economic, social and cul u al consequences o bo h he o igin and he des ina ion coun ies. This phenomenon is signalled by he absence o social ba ie s, pa icipa ion in mains eam ins i u ions, in e ma iage wi h hos coun y na i es, and deg ees o sel -iden i ica ion by immig an s as na i es (Beck, Co ak, & Tienda, 2012). The aim o his s udy is o con ibu e o li e a u e by analysing he ole o egional di e ences o mig a ion mo e closely in e ms o he impo ance o mig a ion egula ion policies in he OECD coun ies. The signi icance o ou egional pe spec i e is o explain he complexi y o mig a ion cha ac e is ics in di e en socie ies, aking in o accoun spa ial and 193 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 geog aphic aspec s. Conside ing he consequences o popula ion dis ibu ion, age, and gende s uc u e e c. can be compa ible wi h he in eg a ion pu poses o mig a ion policies (Findlay & Mulde , 2015). Assuming also ha common mig a ion policies do no wo k p ope ly, he e is no unique solu ion o social and economic p oblems o mig a ion in e ms o inequali ies. The e ec s o hese inequali ies in mig a ion on he places o o igin and des ina ion a e also in luenced by a p ocess called mig a ion selec i i y. Hence, ce ain indi iduals a e mo e likely o mig a e, based on hei pe sonal and sociodemog aphic cha ac e is ics. The es o his pape is o ganized as ollows. The nex (1) sec ion summa izes he b ie concep ual amewo k o mig a ion heo ies wi h a egional pe spec i e. Then, in sec ion 2, logis ics (Logi ) and linea (OLS) eg ession analyses a e ca ied ou wi h c oss-coun y s a is ics in o de o in es iga e how gende , aging, place o bi h, and educa ional a ainmen di e ences a e ela ed o egional cha ac e is ics o mig an s. Finally, based on hese esul s, he pape ends wi h some policy implica ions and conclusions (Sec ion 4). Howe e , ou mo i a ion is no only o sugges a easible poin o e e ence o esea che s, o enhance he e ec i eness o policies, o educe he nega i e consequences o mode n sla e y, bu also o ou line u he esea ch di ec ions in his in e na ional pe spec i e ela ed o he quali y o mig a ion egula ion ac oss coun ies in he in e es s o human de elopmen o e ime. 1. Li e a u e e iew The no ion o mig a ion is commonly unde s ood as a mo emen om one loca ion o ano he o he pu pose o ei he empo a y o pe manen se lemen . This may in ol e long- dis ance a eling, ac oss o wi hin he bo de s o a coun y om one egion o ano he wi h a dis inc ly di e en cul u al, poli ical o social e c. en i onmen (Biswas, McHa dy, & Nolan, 2005). The de elopmen o mig a ion occu s when an indi idual decides ha i is p e e able o mo e a he han o s ay, and whe e he di icul ies o mo ing seem o be mo e han o se by he expec ed ewa ds. Popula ion mo emen s a e esponses o dis inc condi ions and ela ed inciden s, which people encoun e bo h in places o o igin and des ina ion. The ea ly heo ies o mig a ion we e mos ly ocused on domes ic mig a ion, and we e closely ela ed o loca ion models om egional economics and he economics o geog aphy. E. G. Ra ens ein i s de eloped he ‘law o mig a ion’, which classi ied mig an s by he dis ance o eloca ion (Ra ens ein, 1885). Acco ding o Ra ens ein’s laws o mig a ion, mos mig an s mo e only a sho dis ance and usually o la ge ci ies ha g ow apidly and end o be popula ed by mig an s om nea by u al a eas. Thus women a e mo e likely o mig a e han men, he ou -mig a ion is in e sely ela ed o in-mig a ion, and a majo mig a ion wa e will gene a e a compensa ing coun e wa e (G eenwood, 1997). In his ea ly app oach, local mig an s mo ed wi hin he coun y, o only om he coun y o hei bi h o bo de coun ies (Kose & Sal , 1997). The e o e, a se ies o analyses a emp ed o explain and p edic mig a ion pa e ns bo h wi hin and be ween na ions. The basic and addi ional laws o mig a ion subsequen ly eme ged o se e as he s a ing poin o all se ious models o mig a ion pa e ns. One o hese laws ela ing o he gende gap asse ed ha sho -dis ance mig an s a ou numbe ed longe dis ance ones, and women domina ed wi hin coun ies. The ou come o hese endencies led o an in e se ela ionship be ween he olume o mig a ion and he dis ance be ween he home and des ina ion coun ies. The heo e ical backg ound o in e na ional mig a ion is mainly g ounded in he consump ion app oach a ou ed by u ban and egional economis s and which ocuses p ima ily on domes ic mig a ion. G eenwood (1993) poin s ou ha by he ea ly 1980s, he ‘equilib ium’ pe spec i e on mig a ion sugges ed people mig a e o ake ad an age o egional income di e ences. The elemen a y idea behind his pe spec i e is ha people mig a e o adjus hei consump ion o equen changes in incomes, p ices, he supply o 194 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 goods and se ices and hei u ili y unc ions e c. (Rosen, 1974). The no ion ha people mig a e in esponse o spa ial di e ences in ameni ies was also p oposed by Tiebou (1956), who a gued ha an impo an ac o explaining why people mo e om one locali y o ano he is di e ences in he quali y o public goods, such as police and i e p o ec ion, educa ion, hospi als, cou s and o he acili ies. Ne e heless, he idea has no been applied o he s udies o in e na ional mig a ion in e ms o i s egula ed na u e and he ela i ely highe cos s o in e na ional mo emen . Thus, he empi ical esul s om na ional analyses only con ibu ed o he e i ica ion o mig a ion heo ies, and o iden i ying po en ial di e ences be ween (in e )na ional de e minan s o mig a ion, so as o become he g oundwo k o egional economic policies (Jando á & Pale a, 2015). Ne e heless, acco ding o he ‘push and pull’ gene al heo y o mig a ion, immig a ion is de e mined by a summa ized compa ison o a wide ange o posi i e ac o s in des ina ion coun ies and a ious nega i e ones in he o iginal place o esidence (Rahmandous , 2011). In o he wo ds, mig a ion is selec i e and ends o conside posi i e (pull) ac o s which occu in he mig a ion des ina ion a ac ing mig an s o he coun y o des ina ion (Kain h, 2009). Mo eo e , nega i e (push) easons also occu , which in some ci cums ances d i e mig an s o lea e hei coun ies o o igin o des ina ion egions. Meanwhile, in he sociological li e a u e on mig a ion, he ole o he communi y o amily and iends a he des ina ion is o en e e ed o as a ela ionship. The banding oge he o p e ious mig an s om a simila e hnic o egional backg ound is also e e ed o as a mig an ne wo k, based on he linguis ic simila i y be ween speci ic a eas (Chiswick & Mille , 2014). E idence also suppo s he idea ha immig an s end o concen a e whe e ea lie na ionals ha e se led because he cos o adap ing o a no el socie y is mi iga ed by he p esence o esiden s amilia wi h bo h he sou ce and des ina ion coun y cul u es. The le el o spa ial mobili y is la ge , ce e is pa ibus, when he e bal and cul u al en i onmen in he des ina ion coun y is amilia (Rephann & Venca asawmy, 2000). In ou opinion, no unique immig a ion heo y has ye been de ised which can comple ely explain egional mig a ion dispa i ies. Thus, he e is s ill a big gap be ween heo y and empi ical e idence, and much needs o be done on he heo e ical side o his li e a u e, as well. One o he g ea es challenges o mig a ion heo is s is o o ganize all hypo he ically ele an ac o s in o one cohe en amewo k ha will speci y hei in e ac ion wi h each o he in an empi ically es able o m and he eby se e as a guide o u u e esea ch. Consequen ly, mo e s udies a e needed using many econome ic models, da a se s, ime pe iods, coun ies, and eplica ions o exis ing s udies be o e economis s can ha e enough con idence in he esul s o hei own s a is ical s udies o immig a ion. 2. Da a and me hodological app oach This esea ch uses he Da abase on Immig an s in OECD Coun ies (DIOC) which p o ides de ailed in o ma ion on he census da a o immig an s (OECD, 2018). This unique da abase makes i possible o gene a e an ex ensi e a ie y o c oss- abula ions on he cha ac e is ics o he mig an inhabi an s in OECD coun ies by hei coun y o bi h. The la es a ailable elease includes a speci ic a ea co e ing age, gende , na ionali y, place o bi h and du a ion o s ay e c. The es ima ions a e mainly based on bina y logis ic (Logi ) eg ession models (see Equa ion 1, 2, 3, 4, 5 and 6) ha a e equen ly used o apply o a bina y dependen a iable o di e en egions. The Logi eg ession me hod was i s de eloped by Cox (1958) o es ima e he p obabili y o a bina y esponse based on one o mo e p edic o (o independen ) a iables. In o de o analyse whe he educa ional a ainmen , age, sex and he place o bi h 195 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 o mig an s ela e di e en ly o hei egion o bi h he ollowing models a e es ed in each case o he mig an s obse ed i: 𝐷𝐴𝑓𝑟𝑖𝑐𝑎𝑖= 𝛽𝑜+ 𝛽1𝐸𝑑𝑢𝑖+ 𝛽2𝐴𝑔𝑒𝑖+ 𝛽3𝐷𝑆𝑒𝑥𝑖+ 𝛽4𝐷𝐹𝑏𝑜𝑟𝑛𝑖+𝑖 (1) 𝐷𝐴𝑠𝑖𝑎𝑖= 𝛽𝑜+ 𝛽1𝐸𝑑𝑢𝑖+ 𝛽2𝐴𝑔𝑒𝑖+ 𝛽3𝐷𝑆𝑒𝑥𝑖+ 𝛽4𝐷𝐹𝑏𝑜𝑟𝑛𝑖+𝑖 (2) 𝐷𝐸𝑢𝑟𝑜𝑝𝑒𝑖= 𝛽𝑜+ 𝛽1𝐸𝑑𝑢𝑖+ 𝛽2𝐴𝑔𝑒𝑖+ 𝛽3𝐷𝑆𝑒𝑥𝑖+ 𝛽4𝐷𝐹𝑏𝑜𝑟𝑛𝑖+𝑖 (3) 𝐷𝑁_𝐴𝑚𝑒𝑟𝑖𝑐𝑎𝑖= 𝛽𝑜+ 𝛽1𝐸𝑑𝑢𝑖+ 𝛽2𝐴𝑔𝑒𝑖+ 𝛽3𝐷𝑆𝑒𝑥𝑖+ 𝛽4𝐷𝐹𝑏𝑜𝑟𝑛𝑖+𝑖 (4) 𝐷𝑂𝑐𝑒𝑎𝑛𝑖𝑎𝑖= 𝛽𝑜+ 𝛽1𝐸𝑑𝑢𝑖+ 𝛽2𝐴𝑔𝑒𝑖+ 𝛽3𝐷𝑆𝑒𝑥𝑖+ 𝛽4𝐷𝐹𝑏𝑜𝑟𝑛𝑖+𝑖 (5) 𝐷𝑆&𝐶_𝐴𝑚𝑒𝑟𝑖𝑐𝑎𝑖= 𝛽𝑜+ 𝛽1𝐸𝑑𝑢𝑖+ 𝛽2𝐴𝑔𝑒𝑖+ 𝛽3𝐷𝑆𝑒𝑥𝑖+ 𝛽4𝐷𝐹𝑏𝑜𝑟𝑛𝑖+𝑖 (6) The dependen a iables a e dummies dis inguished by he mig an s’ di e en egion o bi h, i.e. DA ica is = 1 i he mig an is bo n in A ica, 0 = o he wise. Respec i ely, DAsia e e s o Asia, DEu ope o Eu ope, DN_Ame ica o No h Ame ica, DOceania o Oceania, and DS&C_Ame ica indica es Sou h and Cen al Ame ica. The i s independen a iable is (Edu), whe e he de ailed educa ional le els (1-6) a e based on he In e na ional S anda d Classi ica ion o Educa ion (ISCED) as de ined by UNESCO (2011). The nex con ol a iable is (Age), which is di ided in o he ollowing age g oup ca ego ies: 1 = 0-14; 2 = 15-24; 3 = 25-34; 4 = 35-44; 5 = 45-54; 6 = 55-64; 7 = 65+. The (DSex) dummy no es he gende di e ences; namely, 1 i he mig an is emale, 0 i male. The subsequen con ol dummy a iable (DFbo n) indica es whe he he mig an is o eign- o na i e-bo n. Acco ding o he OECD (OECD, 2017), o eign-bo n immig an s include all hose who ha e e e mig a ed om hei coun y o bi h o hei cu en coun y o esidence. Thus, a na i e-bo n ci izen o a coun y is one whose mig an pa en s’ place o bi h is he hos coun y. (ε) is he e o e m. Based on he p e iously epo ed (1-6) Equa ions, he cu en s udy o ms he ollowing hypo heses: H1: A ican mig an s a e less likely o be educa ed, olde , o eign-bo n and emale han o he mig an s. H2: Asian mig an s a e less likely o be educa ed, olde , o eign-bo n and emale han o he mig an s. H3: Eu opean mig an s a e less likely o be educa ed, olde , o eign-bo n and emale han o he mig an s. H4: No h Ame ican mig an s a e less likely o be educa ed, olde , o eign-bo n and emale han o he mig an s. H5: Oceanian mig an s a e less likely o be educa ed, olde , o eign-bo n and emale han o he mig an s. H6: Sou h and Cen al Ame ican mig an s a e less likely o be educa ed, olde , o eign-bo n and emale han o he mig an s. In o de o exempli y he alidi y o ou esul s, and e i y ha educa ional a ainmen , and age o mig an s ela e di e en ly o hei egion o bi h, addi ional me hodologies a e needed o cla i y he egional cha ac e is ics o mig a ion. Fo ins ance, linea OLS eg essions. Consequen ly, a linea (OLS) eg ession model is used o highligh he egional di e ences among ou e alua ions. 196 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 𝐸𝑑𝑢𝑖= 𝛽𝑜+ 𝛽1𝐴𝑔𝑒𝑖+ 𝛽2𝐷𝑆𝑒𝑥𝑖+ 𝛽3𝐷𝐹𝑏𝑜𝑟𝑛𝑖+ 𝛽4𝐷𝐴𝑓𝑟𝑖𝑐𝑎𝑖+ 𝛽5𝐷𝐴𝑠𝑖𝑎𝑖+ 𝛽6𝐷𝐸𝑢𝑟𝑜𝑝𝑒𝑖+ 𝛽7𝐷𝑁_𝐴𝑛𝑒𝑟𝑖𝑐𝑎𝑖+ 𝛽8𝐷𝑂𝑐𝑒𝑎𝑛𝑖𝑎𝑖+𝑖 (7) 𝐴𝑔𝑒𝑖= 𝛽𝑜+ 𝛽1𝐸𝑑𝑢𝑖+ 𝛽2𝐷𝑆𝑒𝑥𝑖+ 𝛽3𝐷𝐹𝑏𝑜𝑟𝑛𝑖+ 𝛽4𝐷𝐴𝑓𝑟𝑖𝑐𝑎𝑖+ 𝛽5𝐷𝐴𝑠𝑖𝑎𝑖+ 𝛽6𝐷𝐸𝑢𝑟𝑜𝑝𝑒𝑖+ 𝛽7𝐷𝑁_𝐴𝑛𝑒𝑟𝑖𝑐𝑎𝑖+ 𝛽8𝐷𝑂𝑐𝑒𝑎𝑛𝑖𝑎𝑖+𝑖 (8) No e ha he Sou h and Cen al Ame ican egion a iable was omi ed in o de o con ol and a oid he ‘dummy- ap’ p oblem. The dummy a iable ap is a phenomenon in which he independen a iables a e mul icollinea and wo o mo e a iables a e highly co ela ed, i.e. one a iable can be p edic ed om he o he s. Mo eo e , based on Equa ions 7 and 8, he ollowing o ms o he hypo heses can be also e i ied: H7: Female mig an s a e less educa ed han males. H8: Fo eign-bo n mig an s a e less educa ed han na i e-bo n mig an s. H9: Mig an s ha a e mo e educa ed a e olde han less educa ed mig an s. H10: Male mig an s a e olde han emales. H11: Fo eign-bo n mig an s a e younge han na i e-bo n mig an s. 3. Conduc ing esea ch and esul s Table 1, 2 and 3 ep esen he co esponding esul s o ou es ima ions in each model. In o de o ake accoun o he e ec o egional di e ences we ocus sepa a ely on mig an s by egion o bi h and assume ha hey ha e di e en educa ional, age, gende and o igin o bi h cha ac e is ics compa ed o he o he egions. In all obse ed logis ic (Logi ) eg ession models (see Tables 1 and 2), he dependen a iable indica es he mig an s’ egion o bi h. A he bo om sec ions o hese ables a e he p opo ion o a iances explained by he p edic o s (measu ed by Cox and Shell’s, and Nagelke ke’s pseudo R2). Howe e , he R2 s a is ics o hese models a e ela i ely small, hanks o a la ge numbe o obse a ions. Acco ding o he Omnibus (F- es ) and HL- es s, he explained a iances in a se o da a a e signi ican ly g ea e han he unexplained a iance. In he case o A ica and Eu ope, we ound a signi ican nega i e ela ionship be ween he educa ional a ainmen and he egion o bi h. Al hough he e ec o educa ion does no seem o be la ge in No h, Sou h and Cen al Ame ica and Oceania, in hese models he e was a posi i e co ela ion. In o he wo ds, A ican, Asian and Eu opean mig an s end o be less educa ed han o he s. In he case o A ica, Asia, Oceania, and Sou h and Cen al Ame ica age nega i ely associa ed wi h egion o bi h a iables, which indica es ha in hese egions mig an s a e mo e likely o be younge han he o he egions. Ne e heless, i he e a e wo mig an s and one o hem comes om Eu ope o No h Ame ica, he/she will be signi ican ly olde han he o he . Mo eo e , gende di e ences can be ound in only a ew cases. A ican mig an s end o be male, wi h Eu opean and Sou h and Cen al Ame icans ending o be emale. Measu ing he o igin o bi h di e ences, we can also claim ha Asian, and Sou h and Cen al Ame ican mig an s a e signi ican ly mo e likely o be o eign-bo n. F om he esul s i can be claimed ha No h Ame icans asylum seeke s and mig an s om Oceania end o be na i e-bo n. Al hough, he s ong H1, H2, H3, H4, H5 and H6 hypo heses should be ejec ed, based on he esul s i can be claimed ha he educa ional a ainmen , age, sex and he place o bi h o mig an s ela e di e en ly o hei egion o bi h. 197 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 Table 1. Resul s o he bina y logis ic (logi ) eg essions in he examined OECD coun ies in A ica, Asia and Eu ope Dependen DA ica DAsia DEu ope Independen Be a Wald EXP(B) Be a Wald EXP(B) Be a Wald EXP(B) Cons an -20.98 0.001 0 -4.281 1072.6*** 0.013 1.493 1259.2*** 4.452 Edu -0.011 31.6*** 0.989 0.001 0.238 1.001 -0.008 21.2*** 0.991 Age -0.027 215.9*** 0.974 -0.012 40.8*** 0.988 0.038 539.3*** 1.039 DSex -0.048 50.4*** 0.953 -0.005 0.471 0.995 0.016 6.7*** 1.016 DFbo n 20.011 0.001 0 3.138 582.6*** 23.04 -2.503 3893.2 0.081 Obse a ions 491813 Cox and Shell R2 0.06 0.04 0.13 Nagaike R2 0.08 0.06 0.18 Omnibus- es 2766.5*** 1984.6*** 6346.1*** HL- es 103.2*** 32.87*** 212.6*** Sou ce: Au ho s’ own compila ion, based on (OECD, 2018) No es: He e oscedas ici y obus Wald-s a is ics a e in pa en heses. Le e s in he uppe index e e o signi icance: ***: signi icance a 1 pe cen , **: 5 pe cen , *: 10 pe cen . P- alues wi hou an index mean ha he coe icien is no signi ican e en a he 10 pe cen le el. HL: Hosme and Lemeshow χ2 es . Table 2. Resul s o he bina y logis ic (logi ) eg essions in he examined OECD coun ies in No h Ame ica, Oceania and Sou h and Cen al Ame ica Dependen DN_Ame ica DOceania DS&C_Ame ica Independen Be a Wald EXP(B) Be a Wald EXP(B) Be a Wald EXP(B) Cons an -3.417 1423.1*** 0.033 -2.755 1719.3*** 0.064 -3.075 1870.3*** 0.046 Edu 0.017 6.6** 1.017 0.029 56.4*** 1.030 0.013 33.2*** 1.013 Age 0.034 33.1*** 1.035 -0.010 8.2*** 0.990 -0.010 23.6*** 0.990 DSex -0.022 0.995 0.978 -0.001 0.003 0.999 0.046 35.6*** 1.047 DFbo n -0.834 113.9*** 0.434 -0.246 16.3*** 0.782 1.390 402.1*** 4.016 Obse a ions 491813 Cox and Shell R2 0.01 0.01 0.01 Nagaike R2 0.02 0.02 0.02 Omnibus- es 134.3*** 76.98*** 711.5*** HL- es 9.123* 97.25*** 7.854* Sou ce: Au ho s’ own compila ion, based on (OECD, 2018) No es: He e oscedas ici y obus Wald-s a is ics a e in pa en heses. Le e s in he uppe index e e o signi icance: ***: signi icance a 1 pe cen , **: 5 pe cen , *: 10 pe cen . P- alues wi hou an index mean ha he coe icien is no signi ican e en a he 10 pe cen le el. HL: Hosme and Lemeshow χ2 es . 198 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 Measu ing he e ec s o aging, gende , and egion o bi h di e ences on educa ional a ainmen o mig an s (in Models 1 and 2), based on Equa ion 7, can also p o ide e icien empi ical ools o egional mig a ion policy e o ms. Thus, using Equa ion 8 in ou eg ession models (3 and 4), he in luence o hese examined a iables on he age o immig an s is also es ed in a egional con ex . In he case o Models 2 and 4 we only add he a iables which we e signi ican in Models 1 and 3; he o he s a e omi ed. Howe e , he adjus ed R2 alues a e qui e low because o he la ge numbe o obse a ions (491,813); he signi ican F- es s a is ics sugges ha ou linea eg ession speci ica ion should be p e e ed in all models. The mul i-collinea i y amongs he independen and con ol a iables is es ed by he a iance in la ion ac o (VIF) in each case. The maximum alues o VIF o each eg ession coe icien ange om a low o 1.181 o a high o 1.971. This sugges s ha he VIF alues a e a accep able (less han 10) le els (Hai , 2010). A he bo om sec ion o Table 3, as a goodness o i (GOF) es o ou eg ession he Kolmogo o –Smi no (K-S) no mali y es s o he non-s anda dized esiduals a e also epo ed, o check one o he assump ions o he linea eg ession. Table 3. Resul s o he OLS eg essions o Equa ions 7 and 8 in he OECD coun ies examined Independen a iables Model 1 Model 2 Model 3 Model 4 Cons an 3.163*** 3.172*** 3.885*** 3.894*** 117.981 306.573 181.981 136.407 Edu 0.081*** 0.081*** 50.744 50.737 Age 0.064*** 0.064*** 50.744 50.857 DSex -0.034*** -0.034*** 0.007 (-7.115) (-7.148) 1.229 DFbo n 0.008 -0.101*** -0.101*** 0.321 (-3.612) (-3.611) DA ica -0.057*** -0.058*** -0.043*** -0.041*** (-7.508) (-7.891) (-4.946) (-6.002) DAsia -0.031*** -0.031*** -0.002 (-4.029) (-4.293) (-0.236) DEu ope -0.049 -0.051*** 0.125*** 0.126*** (-6.831) (-7.278) 15.293 19.981 DN_Ame ica 0.014 0.148*** 0.150*** -0.71 6.861 7.13 DOceania 0.045*** 0.044*** -0.003 3.706 -3.669 (-0.252) Obse a ions 491813 Adj. R2 0.06 0.06 0.07 0.07 F- es 344.18*** 460.77*** 407.21*** 651.22*** VIF 1.971 1.821 1.971 1.181 No mali y es o he non-s anda dized esiduals K-S es 0.123*** 0.123*** 0.083*** 0.083*** Sou ce: own compila ion based on (OECD, 2018) No es: He e oscedas ici y obus -s a is ics a e in pa en heses. Le e s in he uppe index e e o signi icance: ***: signi icance a 1 pe cen , **: 5 pe cen , *: 10 pe cen . P- alues wi hou an index mean ha he coe icien is no signi ican e en a he 10 pe cen le el. F om he egional pe spec i e, he age and educa ional a ainmen ela ed di e en ly o he egion o bi h. Fo example, as we ound p e iously, A ican, Asian and Eu opean 205 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 UN. (2017). Uni ed Na ions S a is ics Di ision - Classi ica ions Regis y, ISIC Re . 4. UNESCO. (2011). In e na ional S anda d Classi ica ion o Educa ion (UIS/2012/INS/10/REV). UNESCO Ins i u e o S a is ics. Uni ed Na ions. (2017). In e na ional Mig a ion Repo 2017. an Dalen, H. P., & Henkens, K. (2012). Explaining low in e na ional labou mobili y: he ole o ne wo ks, pe sonali y, and pe cei ed labou ma ke oppo uni ies. Popula ion, Space and Place, 18(1), 31-44. h ps://doi.o g/10.1002/psp.642 Williams, A. M., Jephco e, C., Jan a, H., & Li, G. (2018). The mig a ion in en ions o young adul s in Eu ope: A compa a i e, mul ile el analysis. Popula ion, Space and Place, 24(1), e2123. h ps://doi.o g/10.1002/psp.2123 Zaice a, A. (2014). The impac o aging on he scale o mig a ion (99). IZA Wo kd o Labo . h ps://doi.o g/doi: 10.15185/izawol.99 206 Domicián Má é, Im an Sa ihasan, Józse Popp, Judi Oláh ISSN 2071-789X RECENT ISSUES IN SOCIOLOGICAL RESEARCH Economics & Sociology, Vol. 11, No. 3, 2018 Appendix Table A.1. Desc ip i e s a is ics o he examined a iables Minimum Maximum Mean S d. De ia ion Skewness Ku osis S a is ic S a is ic S a is ic S a is ic S a is ic S d. E o S a is ic S d. E o Age 1 7 4.13 1.863 -.013 .003 -1.116 .006 Edu 1 6 3.35 1.667 .095 .003 -1.239 .007 DSex 1 2 1.50 .500 -.002 .003 -2.000 .006 DFbo n 0 1 .99 .098 -9.988 .003 97.751 .006 DA ica 0 1 .23 .423 1.266 .003 -.398 .006 DAsia 0 1 .23 .421 1.281 .003 -.359 .006 DEu ope 0 1 .30 .458 .875 .003 -1.234 .006 DN_Ame ica 0 1 .02 .127 7.635 .003 56.298 .006 DOceania 0 1 .05 .220 4.084 .003 14.683 .006 DS&C_Ame ica 0 1 .16 .370 1.822 .003 1.318 .006 Sou ce: own compila ion based on (OECD, 2018) Table A.2. Pea son co ela ion ma ix o he examined a iables Edu Age DSex DFbo n DA ican DAsian DEu opean DN_A. DOceanian Edu 1.000 Age 0.072*** 1.000 DSex -0.01** 0.001 1.000 DFbo n 0.000*** -0.009*** 0.000 1.000 DA ica -0.009*** -0.022*** -0.01*** 0.054*** 1.000 DAsia 0.000 -0.01*** 0.000 0.05*** -0.3*** 1.000 DEu ope -0.004*** 0.034*** 0.004*** -0.114*** -0.362*** -0.361*** 1.000 DN_Ame ica 0.004*** 0.009*** 0.000 -0.016*** -0.072*** -0.071*** -0.086*** 1.000 DOceania 0.01*** -0.003** 0.001 -0.006*** -0.126*** -0.126*** -0.152*** -0.03*** 1.000 Sou ce: own compila ion based on (OECD, 2018) No es: Le e s in he uppe index e e o signi icance: ***: signi icance a 1 pe cen , **: 5 pe cen , *: 10 pe cen . P- alues wi hou an index mean ha he coe icien is no signi ican e en a he 10 pe cen le el.