Individual-level factors influencing voter turnout in the 2014 European Parliament election

Individual-level factors influencing voter turnout in the 2014 European Parliament election

The participation of EU citizens in the European Parliament elections is much lower than in the national elections and differs significantly across the EU member states. The articled is aimed at finding out the factors that influence the participation of EU citizens in the elections to the European...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Datum:2019
1. Verfasser: Shapovalov, S.
Format: Artikel
Sprache:English
Veröffentlicht: Iнститут соціології НАН України 2019
Schriftenreihe:Социология: теория, методы, маркетинг
Online Zugang:http://dspace.nbuv.gov.ua/handle/123456789/182304
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Zitieren:Individual-level factors influencing voter turnout in the 2014 European Parliament election / S. Shapovalov // Социология: теория, методы, маркетинг. — 2019. — № 3. — С. 142-158. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id irk-123456789-182304
record_format dspace
spelling irk-123456789-1823042021-12-29T01:26:01Z Individual-level factors influencing voter turnout in the 2014 European Parliament election Shapovalov, S. The participation of EU citizens in the European Parliament elections is much lower than in the national elections and differs significantly across the EU member states. The articled is aimed at finding out the factors that influence the participation of EU citizens in the elections to the European Parliament. The study outlines the theoretical approaches to understanding the phenomenon of the European Parliament elections and the assumptions of researchers about the factors that may influence the electoral activity of citizens. According to second-order elections theory by Karlheinz Reif and Hermann Schmitt citizens perceive European Parliament elections as less im portant than national elections that results in lower participation. However, the behavior of voters may also be affected by a number of factors which may be regarded as individual-level motivations (trust in national and European authorities, attitudes towards EU institutions etc.). With use of binary logistic regression method it was defined which factors influenced the participation of citizens in the European Parliament election 2014 and explained the differences in the electoral activity of citizens of different EU countries. The nature of the identified factors that influence the participation of citizens in the elections to the European Parliament suggests that the second-order elections theory is still valid. 2019 Article Individual-level factors influencing voter turnout in the 2014 European Parliament election / S. Shapovalov // Социология: теория, методы, маркетинг. — 2019. — № 3. — С. 142-158. — англ. 1563-4426 http://dspace.nbuv.gov.ua/handle/123456789/182304 316:722 en Социология: теория, методы, маркетинг Iнститут соціології НАН України
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection DSpace DC
language English
description The participation of EU citizens in the European Parliament elections is much lower than in the national elections and differs significantly across the EU member states. The articled is aimed at finding out the factors that influence the participation of EU citizens in the elections to the European Parliament. The study outlines the theoretical approaches to understanding the phenomenon of the European Parliament elections and the assumptions of researchers about the factors that may influence the electoral activity of citizens. According to second-order elections theory by Karlheinz Reif and Hermann Schmitt citizens perceive European Parliament elections as less im portant than national elections that results in lower participation. However, the behavior of voters may also be affected by a number of factors which may be regarded as individual-level motivations (trust in national and European authorities, attitudes towards EU institutions etc.). With use of binary logistic regression method it was defined which factors influenced the participation of citizens in the European Parliament election 2014 and explained the differences in the electoral activity of citizens of different EU countries. The nature of the identified factors that influence the participation of citizens in the elections to the European Parliament suggests that the second-order elections theory is still valid.
format Article
author Shapovalov, S.
spellingShingle Shapovalov, S.
Individual-level factors influencing voter turnout in the 2014 European Parliament election
Социология: теория, методы, маркетинг
author_facet Shapovalov, S.
author_sort Shapovalov, S.
title Individual-level factors influencing voter turnout in the 2014 European Parliament election
title_short Individual-level factors influencing voter turnout in the 2014 European Parliament election
title_full Individual-level factors influencing voter turnout in the 2014 European Parliament election
title_fullStr Individual-level factors influencing voter turnout in the 2014 European Parliament election
title_full_unstemmed Individual-level factors influencing voter turnout in the 2014 European Parliament election
title_sort individual-level factors influencing voter turnout in the 2014 european parliament election
publisher Iнститут соціології НАН України
publishDate 2019
url http://dspace.nbuv.gov.ua/handle/123456789/182304
citation_txt Individual-level factors influencing voter turnout in the 2014 European Parliament election / S. Shapovalov // Социология: теория, методы, маркетинг. — 2019. — № 3. — С. 142-158. — англ.
series Социология: теория, методы, маркетинг
work_keys_str_mv AT shapovalovs individuallevelfactorsinfluencingvoterturnoutinthe2014europeanparliamentelection
first_indexed 2025-07-16T00:37:49Z
last_indexed 2025-07-16T00:37:49Z
_version_ 1837761846736584704
fulltext Serhii Shapovalov Individual-level factors influencing voter turnout in the 2014 European Parliament election UDC 316:722 SERHII SHAPOVALOV, Political analyst at Ilko Kucheriv Democratic Ini - tia tives Foundation, Kyiv Individual-level factors influencing voter turnout in the 2014 European Parliament election Since the first elec tions to the Eu ro pean Par lia ment held in 1979, there has been a steady ten dency of de clin ing voter turn out in each sub se quent elec tion. More over, EU mem ber states dif fer sig nif i cantly in terms of the elec toral par ti c i - pa tion of their cit i zens. In Bel gium and Lux em bourg 89,6% and 85,5% of cit i zens re spec tively took part in the 2014 Eu ro pean Par lia ment elec tions, while in the Czech Re pub lic and Slovakia only 18,2% and 13% of the cit i zens took part in the elec tion of MEPs. Figure 1. Temporal decline of voter turnout in each subsequent election to the European Parliament. Source: European Parliament 142 Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3  S.Shapovalov, 2019 An other im por tant ob ser va tion is that the low level of cit i zen par tic i pa tion in elec tions to the Eu ro pean Par lia ment is not a sign of a gen eral de tach ment of the pop u la tion from pol i tics. The com par i son of cit i zens’ par tic i pa tion in the 2014 EP elec tions and in the tem po rally clos est na tional par lia men tary elec tions shows that the voter turn out rates in na tional par lia men tary elec tions are much higher (the term “Euro-Gap” is used to re fer to such a gap [Rose, 2004]). In the Czech Re pub lic and Slovakia, where the low est voter turn out was ob served in the Eu ro pean elec tions, about 60% of cit i zens par tic i pated in na tional par lia men - tary elec tions [IDEA Voter Turn out Da ta base, 2019]. Tem po ral de cline of voter turn out, cross-coun try dif fer ences and ‘Euro-Gap’ en cour age re search ers to de fine the fac tors in flu enc ing voter be hav ior. This re - search aims to find out the fac tors caus ing cross-coun try dif fer ences in voter turn out for EP elec tions us ing the data of 2014 EP elec tions. Possible factors influencing voter turnout One of the first ex pla na tions for these trends was the sec ond-or der elec tions the ory. Back in 1980, the au thors of the the ory em pha sized that the “sec ond-or der” elec tions are com mon for de moc ra cies. The most im por tant elec tions (first- or der elec tions) are the elec tions of na tional par lia ments in par lia men tary sys tems and the elec tion of a pres i dent in pres i den tial po lit i cal sys tems [Reif, Schmitt, 1980]. Such elec tions de ter mine who will hold the high est po si tions, for mu late and im ple - ment pub lic pol i cies. The sec ond-or der elec tions are mu nic i pal, re gional elec tions, elec tions to the sec ond cham ber of par lia ment. It is as sumed that for EU cit i zens the elec tions of MEPs are sec ond-or der elec tions [Marsh, 1998]. The the ory fo cuses more on the shifts in elec toral pref er ences in sec ond-or der elec tions. Lower voter turn out is ex plained by the the ory in a sim ple way — cit i - zens are more likely to ab stain when they con sider the elec tions not im por tant (elec tions re sult will not change the sit u a tion in their coun try) [Schmitt, 2004]. It is worth men tion ing that real pow ers of Eu ro pean Par lia ment and pow ers as per ceived by EU cit i zens may dif fer sig nif i cantly, but voter be hav ior is driven mainly by in di vid ual per cep tions. The the ory ex plains lower turn out in EP elec tions com pared to na tional par - lia men tary elec tions, but cross-coun try dif fer ences re main un ex plained. There - fore, it is nec es sary to in clude ad di tional fac tors in the anal y sis. Fac tors af fect ing cit i zens’ par tic i pa tion in elec tions may be di vided into mi - cro-level fac tors (in di vid ual mo ti va tions) and macro-lev els (ex ter nal con di - tions). It is worth men tion ing also that the voter be hav ior in the EP elec tions is also in flu enced by both fac tors that gen er ally in flu ence elec toral be hav ior (trust in na tional au thor i ties, in ter est in pol i tics, feel ing the abil ity to in flu ence de ci - sion-mak ing), as well as fac tors spe cific to the EP elec tions (trust in EU in sti tu - tions, eval u a tions of the EU, at ti tudes to wards EU in te gra tion). The re search ers of voter be hav ior sug gest com pul sory vot ing, in ten sity of po - lit i cal com pe ti tion, elec toral sys tem, num ber of par ties, tim ing of elec tion to be the most im por tant macro-level fac tors in flu enc ing voter turn out. For the re - search of cross-coun try dif fer ences in the EP elec tion voter turn out these fac tors are hardly ap pli ca ble. The strong ef fect of com pul sory vot ing on the voter turn out in both na tional and EP elec tions is un doubted and gained enough re search ers’ at ten tion [Jack - Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 143 Individual-level factors influencing voter turnout in the 2014 European Parliament election man, 1987; Blais, 2006; Fiorino, 2016; Blais, 2003]. Tim ing of elec tion is the same in all coun tries of the EU so there will be no dis tri bu tion of data for this fac tor [Mattila, 2003]. Num ber of par ties which par tic i pate in EP elec tion in each coun - try is hardly count able. For ex am ple, in Bul garia vot ers are al lowed to vote for in - di vid ual can di dates as well as po lit i cal par ties. Great Brit ain and Bel gium are di - vided into sev eral elec toral con stit u en cies in ev ery of which dif fer ent num ber of par ties takes part in elec tion [Oelbermann, 2015]. What about elec toral sys tem, in all EU mem ber states some kind of pro por tional sys tem is used while con duct - ing Eu ro pean elec tions [Oelbermann, 2015]. The sig nif i cant im pact of elec toral sys tem was found only on the gen eral level: voter turn out is higher in the coun - tries which use pro por tional sys tems in com par i son with coun tries where the majoritarian sys tem is used [Blais, Aarts, 2006; Blais, Carty, 1990]. The only fac tor left is the in ten sity of po lit i cal com pe ti tion. The in ten sity of po lit i cal com pe ti tion is as sumed to in crease voter turn out. Close com pe ti tion and ab sence of a clear leader be fore elec tions in creases the per ceived weight of cit i zen’s in di vid ual vote and per ceived abil ity to in flu ence elec tion re sult that mo ti vates a cit i zen to par tic i pate in elec tion. In the ab sence of a clear leader, the logic that the ma jor ity still elects a can di date with out the par tic i pa tion of a par - tic u lar cit i zen ceases to work, so the cit i zen feels more strongly about his or her abil ity to in flu ence the elec tion re sult. Close com pe ti tion en cour ages also po lit i - cal par ties to make more mo bi li za tion ef forts [Blais, 2006; Sцderlund, Wass, Blais, 2011]. How ever, this logic works in na tional elec tions when elec tions out - come is per ceived as im por tant by both cit i zens and par ties. Out comes of sec - ond-or der elec tion are less im por tant and ab sence of clear win ner does not mo ti - vate cit i zens to par tic i pate. Re search ers also sug gest the im pact of eco nomic de vel op ment on voter turn - out. It may be as sumed that there is a cor re la tion be tween turn out in EP elec tion and eco nomic in di ca tors in EU mem ber states (the low est turn out is ob served in East ern Eu ro pean states which have lower level of eco nomic de vel op ment). How ever, the dif fer ences in turn out are hardly ex plain able by eco nomic vari ables in case of EP elec tions and EU mem ber states be cause on the the o ret i cal level eco nomic de vel op ment af fects voter turn out to the cer tain ex tent [Powell, 2009; Blais, Dobrzynska, 1998]. All EU mem ber state have reached the level of eco - nomic de vel op ment af ter which eco nomic vari ables be come not so im por tant pre dic tors of elec toral par tic i pa tion. Mi cro-level fac tors are in di vid ual mo ti va tions and at ti tudes which en cour - age or dis cour age a cit i zen’s par tic i pa tion in the elec tions. Some of the fac tors which are used in the cur rent re search were used in pre vi ous works. Nev er the less, such fac tors are in cluded in the cur rent re search as the goal of re search is to de fine the ag gre gate ef fect of sev eral fac tors and the ef fect of these fac tors was not stud - ied on the case of 2014 EP elec tion. The main in di vid ual mo ti va tions likely to in flu ence the cit i zens’ par tic i pa - tion in the EP elec tions are: — trust in in sti tu tions (trust in the EU in sti tu tions is as sumed to en cour age loy alty and there fore elec toral par tic i pa tion, but trust in na tional au thor i - ties is also as sumed to shape cit i zens’ po lit i cal be hav ior which will in flu - ence turn out in EP elec tion) [Mattila, 2003; Gronlund, 2007]; — per cep tion of EU mem ber ship as a good or bad thing for one’s coun try (was as sumed to in flu ence elec toral par tic i pa tion by the au thors of sec - 144 Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 Serhii Shapovalov ond-or der elec tions the ory, but ap peared to be in sig nif i cant vari able in sev eral stud ies) [Frank lin, Marsh, van der Eijk, 1996; Reif, Schmitt, 1980; Rose, 2004]; — pos i tive or neg a tive at ti tudes to wards the fu ture EU in te gra tion [Hobolt, Spoon, Til ley, 2008]; — one’s iden ti fi ca tion with the EU/Eu rope is as sumed to fos ter par tic i pa - tion in the EP elec tions [Frank lin, Bernhard, 2009]; — one’s aware ness of EP’s func tions, ac tiv i ties, pow ers etc. [Mattila, 2003; Clark, 2013]; — low per ceived im por tance of ques tions which be long to the EP com pe - tences [Clark, 2013]; — one’s in ter est in pol i tics [Blais, 2006; Sцderlund, Wass, Blais, 2011]; — sense of po lit i cal ef fi cacy (the in di vid ual’s per cep tion of his or her abil ity to in flu ence the pol i tics) [Dyck, Lascher, 2009]; — ha bit ual vot ing and re lated fac tor of per cep tion of vot ing as the civic duty [Schmitt, Mannheimer, 1991; Frank lin, 2004]. The use of bi nary lo gis tic re gres sion to mea sure the ef fect of in di vid ual-level fac tors on the voter turn out in the 2014 EP elec tions To an a lyze the in flu ence of the listed in di vid ual-level fac tors on the voter turn out the Eu ro pean Elec tion Study 2014 (EES) data are used. The EES-2014 post-elec tion sur vey was con ducted in col lab o ra tion with the Eu ro pean Par lia - ment. The field re search was con ducted by TNS Opin ion, which also con ducts Eurobarometer polls. The sam ple for the EES-2014 post-elec tion sur vey in - cluded about 1,100 re spon dents in each EU Mem ber State (ex cept for Malta and Lux em bourg, where 500 cit i zens we polled, and United King dom with 1300 re - spon dents of which 300 rep re sented North ern Ire land). The to tal sam ple size was 30,000 re spon dents [EES, 2014]. The Eu ro pean Elec tion Study par tially du pli - cates the Eurobarometer ques tions re gard ing cit i zens’ at ti tudes to wards the EU, in ter est in pol i tics, trust in in sti tu tions etc. How ever, the EES poll does in clude the im por tant ques tion “Did you vote in the last Eu ro pean Par lia ment elec tions?” which is not in cluded in the Euro - barometer polls. An swer op tions to this ques tion are coded as a di chot o mous vari able (“yes” or “no”). Ob vi ously, re spon dents’ re ported turn out level may dif - fer from ac tual ones, but the in clu sion of this ques tion in the sur vey broad ens the range of pos si ble sta tis ti cal meth ods for anal y sis (for ex am ple, it be comes pos si ble to use bi nary lo gis tic re gres sion to find out the in flu ence of many fac tors on the re sult ing di chot o mous vari able). One of the most ap pro pri ate ways to find out in di vid ual-level fac tors in flu - enc ing the par tic i pa tion of EU cit i zens in the EP elec tion is to use the bi nary lo - gis tic re gres sion method. The bi nary lo gis tic re gres sion al lows us to find out the ef fect of pre dic tors ex pressed in nom i nal and in ter val scales on the re sult ing di - chot o mous vari able [Field, 2009]. In our case, the Eu ro pean Elec tion Sur vey post-elec tion poll in cludes the ques tion “Did you par tic i pate in the last Eu ro pean Par lia ment elec tions?” with the “Yes” and “No” op tions avail able to re spon dents. Since the an swer to this ques tion in volves two pos si ble op tions, this vari able can be con sid ered di chot o mous and used as a de pend ent vari able when con struct ing a bi nary lo gis tic re gres sion model. Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 145 Individual-level factors influencing voter turnout in the 2014 European Parliament election Fac tors likely to in flu ence re spon dents’ de ci sion to vote or re fuse to vote are mea sured with ques tions about re spon dents’ trust in EU in sti tu tions, eval u at ing their own coun try’s mem ber ship in the EU as a good or bad thing, and other fac - tors se lected for anal y sis in the cur rent sur vey. Since pre dic tors in the bi nary lo - gis tic re gres sion model can be in ter val and nom i nal (di chot o mous), the nom i nal vari ables avail able in the sur vey must be dichotomized by recoding pos si ble an - swers. Some of the vari ables al ready avail able in the sur vey are di chot o mous in their ini tial form and do not re quire fur ther pro cess ing. For ex am ple, the dis tri bu tion of re spon dents’ an swers to the ques tion of whether they are in ter ested in pol i tics or not is as shown in ta ble 1. Table 1 EU citizens’ reported interest for politics. Data: EES-2014 “For each of the following statements, please tell me to what extent it corre - sponds or not to your attitude or opinion: You are very interested in politics” Frequency Percent Valid Percent Cumulative Percent Yes, definitely 4179 13,9 14,0 14,0 Yes, to some extent 10420 34,7 34,9 48,9 No, not really 8601 28,6 28,8 77,7 No, not at all 6659 22,1 22,3 100,0 Total 29859 99,3 100,0 According to the data, 48,6% of respondents are interested in politics, of which 13,9% are “definitely interested” and 34,7% are “to some extent” inte - rested. For the use of the binary logistic regression model, four possible answers should be reduced to two categories. In our case, the most appropriate approach is grouping in the “interested” and “not interested” groups, which embrace groups of ‘very’ and ‘rather’ interested respondents and ‘rather not’ and ‘not at all’ interested respondents, respectively. Af ter dichotomizing the ‘in ter est in pol i tics’ vari able, the fre quency dis tri bu - tion of the re spon dents’ an swers looks like ta ble 2. Table 2 ‘Interest for politics’ variable after dichotomization. Data: EES-2014 “For each of the following statements, please tell me to what extent it corre - sponds or not to your attitude or opinion: You are very interested in politics” Frequency Percent Valid Percent Cumulative Percent Interested 14599 48,6 48,9 48,9 Not interested 15260 50,8 51,1 100,0 Total 29859 99,3 100,0 Non-dichotomized data (in their initial form) can be used for cross- tabu la - tion analysis to visually identify differences between groups of respondents. For example, the contingency table for the parameters of ‘voting in the European Parliament elections’ and ‘interest in politics’ for the whole sample without division by country would look like table 3. 146 Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 Serhii Shapovalov Table 3 Contingency table for ‘voting in the last EP elections’ and ‘interest for politics’ variables. Data: EES-2014 Are you interested in politics? Total Yes, definitely Yes, to some extent No, not really No, not at all Voted 3324 79,6% 7365 70,8% 4601 53,7% 1844 27,8% 7134 57,5% Did not vote 853 20,4% 3037 29,2% 3974 46,3% 4795 72,2% 12659 42,5% Total 4177 100,0% 10402 100,0% 8575 100,0% 6639 100,0% 29793 100,0% It is ob served that among the group of re spon dents who are most in ter ested in pol i tics, al most 80% par tic i pated in the last elec tions to the Eu ro pean Par lia - ment, while about 20% did not par tic i pate. The pro por tion of re spon dents who par tic i pated in the Eu ro pean elec tions grad u ally di min ishes with the de crease in the in ter est of these re spon dents in pol i tics: among those who are “rather in ter - ested” in pol i tics, about 71% have par tic i pated in the elec tions; among “rather not in ter ested”, 54% took part in the elec tions; and among those who are not in - ter ested in pol i tics at all, only 28% par tic i pated in the elec tions. The bi nary lo gis tic re gres sion has a dif fer ent fo cus com pared to mul ti ple lin - ear re gres sion. For the bi nary lo gis tic re gres sion in di vid ual cit i zens are the units of anal y sis. The goal of bi nary lo gis tic re gres sion is to pre dict the value of a de - pend ent di chot o mous vari able (voted in the EP elec tion or did not vote) based on re spon dents’ an swers to other ques tions in the sur vey. These ques tions are used as in de pend ent vari ables and dichotomized to be come us able for lo gis tic re - gres sion anal y sis. While bi nary lo gis tic re gres sion uses in di vid ual cit i zens as the units of anal y - sis, for mul ti ple lin ear re gres sion the units of anal y sis are the 28 EU mem ber states. For each mem ber state, the value of the de pend ent vari able is the ac tual level of voter turn out in the 2014 EP elec tion, ex pressed as the per cent age of reg is tered vot ers. The val ues of in de pend ent vari ables for each coun try are ex pressed in per - cent ages and rep re sent the pro por tion of cit i zens in each coun try who trust the gov ern ment, who con sider their coun try’s mem ber ship in the EU a good thing, etc. The ad van tage of bi nary lo gis tic re gres sion is its abil ity to an a lyze in di vid ual re spon dents, who re port them selves as vot ing or non-vot ing and at the same time ex press their at ti tudes while an swer ing other sur vey ques tions. In com par i son, mul ti ple lin ear re gres sion com pares the num ber of peo ple in each coun try who took part in the elec tion and who share cer tain at ti tudes. In this case, we es tab lish con nec tions be tween some phe nom ena based on a com par i son of per cent ages of cit i zens, but we are not able to say, for ex am ple, whether 60% of the cit i zens in a cer tain coun try who voted in the elec tion and 60% in the same coun try who trust the gov ern ment are the same peo ple. The lim i ta tion of lo gis tic re gres sion is that re ported voter be hav ior may dif - fer from the ac tual one. In par tic u lar, in the EES sur vey, the re ported level of par - tic i pa tion in elec tions to the Eu ro pean Par lia ment is of ten higher than the ac tual fig ures (Fig. 2). Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 147 Individual-level factors influencing voter turnout in the 2014 European Parliament election 148 Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 Serhii Shapovalov An other lim i ta tion is that re search ers are lim ited in the se lec tion of pre dic tors. As ob ser va tions in lo gis tic re gres sion are in di vid ual re spon dents, we are able to use only ques tions asked within the same sur vey as in de pend ent vari ables; the data from other sur veys are not com pa ra ble be cause ev ery sur vey has a dif fer ent sam ple. The re sult of lo gis tic re gres sion anal y sis is the prob a bil ity of a cer tain event. In our case, we are try ing to de ter mine the prob a bil ity that a re spon dent with a par tic u lar set of an swers to other sur vey ques tions will choose ‘yes’ when an swer - ing the ques tion ‘Did you vote in the last elec tions to the Eu ro pean Par lia ment?’. The prob a bil ity is cal cu lated by us ing the bi nary lo gis tic re gres sion for mula: p = 1 / 1 + e–z in which z = b1x1 + b2x2 + … bnxn + a, which is the or di nary lin ear re gres sion equa - tion; x1, x2, etc. are the val ues of the in de pend ent vari ables; b1, b2 are the re gres sion co ef fi cients for the re spec tive in de pend ent vari ables; the value of a is a con stant for the model. The re gres sion co ef fi cients and con stants are cal cu lated by the SPSS sta tis ti cal pack age when pro cess ing the lo gis tic re gres sion model [Field, 2009]. Variables The fol low ing ques tions from the EES sur vey were used for anal y sis (ta - ble 4). Table 4 Questions in the EES-2014 survey used as predictors of voter turnout Questions Answer choices Coding in EES-2014 database European Parliament elections were held on the ... (date according to country). For one reason or another, some people in ... (country) did not vote in these elections. Did you yourself vote in the recent European Parliament elections? 1 — “voted” 2 — “did not vote” qp1 For each of the following statements, please tell me to what extent it corresponds or not to your attitude or opinion. 1. You had all the necessary information in order to choose who to vote for in the recent European elections 2. You trust the institutions of the EU 3. You feel you are a citizen of the EU 4. The European Parliament takes into consideration the concerns of European citizens 5. You feel attached to ... (country) 6. You feel attached to Europe 7. It is very important for you which particular candidates have been elected as MEPs in the European Parliament elections in ... (country) 8. You are very interested in politics 1 — “Yes, totally” 2 — “Yes, somewhat” 3 — “No, not really” 4 — “No, not at all” Recoded: 1 — “Yes” 2 — “No” qp6_1 — qp6_9 (except for qp6_8, which duplicates qp6_7) Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 149 Individual-level factors influencing voter turnout in the 2014 European Parliament election Questions Answer choices Coding in EES-2014 database Generally speaking, do you think that ... (OUR COUNTRY)’s membership of the EU is...? 1 — “good thing” 2 — “bad thing” 3 — “neither good nor bad thing” qp7 Personally, do you remember having seen on TV, in the Internet or on posters, read in newspapers or heard on the radio a campaign encouraging people to vote in the European elections? 1 — “yes” 2 — “no” qp8 How often did you do any of the following during the four weeks before the recent European elections? 1. Watch a programme about the European elections on television 2. Read about the European elections in a newspaper 3. Talk to friends or family about the European elections 4. Attend a public meeting or rally about the European elections 5. Read about the European elections on the Internet (websites, social media, etc.) 1 — “Often” 2 — “Some times” 3 — “Never” Recoded: 1 — “Have done” 2 — “Have not done” qp11_1 — qp11_5 Did anyone from one of the national political parties contact you regarding your vote in the recent European elections? 1 — “yes” 2 — “no” qp12 And at which level do you think the most important issue or problem facing your country at the moment should be dealt with? 1 — “At local or regional level ”2 — “At national level” 3 — “At European level” Recoded: 1 — “At national level” 2 — “At Euro pean level” qpp3 For each of the following statements, please tell me to what extent it corresponds or not to your attitude or opinion: 1. You trust the national parliament of your country 2. Your national parliament takes the concerns of citizens of your country into consideration 3. Sometimes politics and government seem so complicated that a person like you can’t really understand what’s going on 1 — “Yes, totally” 2 — “Yes, somewhat” 3 — “No, not really” 4 — “No, not at all” Recoded: 1 — “Yes ”2 — “No” qpp9_1 — qpp9_3 150 Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 Serhii Shapovalov Table 4 continued Questions Answer choices Coding in EES-2014 database Some say European unification should be pushed further. Others say it already has gone too far. What is your opinion? 10-point scale where ‘0’means unification “has already gone too far” and ‘10’ means it “should be pushed further”. Recoded: values 0-5 and 6-10 united in two categories: “unification should not be pushed further” and “should go further”. qpp18 Do you consider yourself to be close to any particular political party? If so, which party do you feel close to? Recoded: 1 — “No” 2 — “Yes”, embraces all respondents who stated one of the parties qpp21 Please tell me to what extent you agree or disagree with each of the following statements 1. My voice counts in the EU 2. My voice counts in my country 1 — “Yes, totally” 2 — “Yes, some - what” 3 — “No, not really” 4 — “No, not at all” Recoded: 1 — “Yes” 2 — “No” d72_1 — d72_2 When you get together with friends or relatives, would you say you discuss frequently, occasionally or never about: 1. national political matters 2. European political matters 1 — “Often” 2 — “Sometimes” 3 — “Never” Recoded: 1 — “Yes” 2 — “No” d71_1 — d71_2 Results Most of the se lected pre dic tors ap pear to have sta tis ti cally sig nif i cant ef fects on the de pend ent vari able. The fol low ing vari ables were found to be in sig nif i cant: Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 151 Individual-level factors influencing voter turnout in the 2014 European Parliament election End of Table 4 — qpp9_2 (Your na tional par lia ment takes the con cerns of cit i zens of your coun try into con sid er ation) — d71 (When you get to gether with friends or rel a tives, would you say you fre quently, oc ca sion ally or never dis cuss na tional/Eu ro pean po lit i cal mat ters) — qp6_5 (Feel ing at tached to re spon dent’s coun try) — d72_2 (Feel ing that re spon dent’s voice counts in his or her coun try) — qp6_4 and qp6_2 (Trust ing EU in sti tu tions and feel ing that the Eu ro pean Par lia ment takes into con sid er ation the con cerns of Eu ro pean citizens) — qp11_5 (Read ing about the Eu ro pean elec tions on the Internet (websites, so cial me dia, etc.) The re sult of bi nary lo gis tic re gres sion anal y sis is as fol lows (ta ble 5). Table 5 Regression coefficients for the variables which have statistically significant impact on the dependent variable B S.E. Wald df Sig. qp6_1_binary –,555 ,042 175,259 1 ,000 qp6_3_binary ,142 ,052 7,574 1 ,006 qp6_6_binary ,172 ,052 11,074 1 ,001 qp6_7_binary –,947 ,041 529,687 1 ,000 qp6_9_binary –,342 ,042 66,808 1 ,000 qp7_binary –,377 ,044 73,110 1 ,000 qp8 ,250 ,045 30,641 1 ,000 qp11_1_binary –,416 ,045 84,081 1 ,000 qp11_2_binary –,213 ,042 26,307 1 ,000 qp11_3_binary –,469 ,044 113,555 1 ,000 qp11_4_binary –,435 ,077 31,592 1 ,000 qpp3_binary ,127 ,041 9,807 1 ,002 qpp9_1_binary –,158 ,042 14,319 1 ,000 qpp9_3_binary ,080 ,041 3,850 1 ,050 qpp18_binary –,092 ,041 4,997 1 ,025 qpp21_binary ,565 ,040 199,223 1 ,000 d72_1_binary –,621 ,042 216,330 1 ,000 qp12 –,525 ,057 84,112 1 ,000 Constant 6,344 ,250 642,477 1 ,000 Given the main characteristics of the binary logistic regression model, we have the necessary data for calculating the probability of a respondent’s decision to vote. Imagine the respondent who: — has all the nec es sary in for ma tion in or der to choose who to vote for in the re cent Eu ro pean elec tions (qp6_1 = “1”); — feels he or she is a cit i zen of the EU (qp6_3 = “1”); 152 Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 Serhii Shapovalov — feels at tached to Eu rope (qp6_6 = “1”);-for whom it is very im por tant which par tic u lar can di dates have been elected as MEPs in their coun try (qp6_7 = “1”); — is in ter ested in pol i tics (qp6_9 = “1”); — thinks that his/her coun try’s mem ber ship of the EU is good thing (qp7 = “1”); — has seen a me dia cam paign en cour ag ing peo ple to vote in the Eu ro pean elec tions (qp8 = “1”); — watched a programme about the Eu ro pean elec tions on tele vi sion, read about the elec tions in a news pa per, talked to friends or fam ily about the Eu ro pean elec tions, at tended a pub lic meet ing or rally about the Eu ro - pean elec tions (qp11_1 — qp11_4 = “1”); — be lieves that the main prob lem fac ing his or her coun try should be dealt at the Eu ro pean level (qpp3 = “1”); — trusts na tional par lia ment (qpp9_1 = “1”); — does not con sider pol i tics too com pli cated (qpp9_3 = “2”); — be lieves that the Eu ro pean in te gra tion should be pushed fur ther (qpp18 = “2”); — feels close to cer tain po lit i cal party (qpp21 = “2”); — be lieves that his or her voice counts at the Eu ro pean level (d72_1 = “1”); — was con tacted by na tional po lit i cal par ties re gard ing his or her vote in the re cent Eu ro pean elec tions (qp12 = “1”) We as sume that such a re spon dent has a high prob a bil ity to take part in the elec tions. Af ter at trib ut ing these val ues to in de pend ent vari ables, it is pos si ble to cal cu late the prob a bil ity of choos ing the an swer ‘voted’ to the ques tion ‘Did you vote in the re cent EP elec tions?’ by such a re spon dent. Firstly, it is nec es sary to cal cu late the value of z in the bi nary lo gis tic re gres - sion for mula us ing at trib uted val ues to the in de pend ent vari able, re gres sion co ef - fi cients, and the con stant. z = b1x1 + b2x2 + … bnxn + a = (–0,555 × 1) + (0,142 × 1) + (0,172 × 1) + (–0,947 × 1) + (–0,342 × 1) + (–0,377 × 1) + (0,25 × 1) + (–0,416 × 1) + (–0,213 × 1) + (–0,469 × 1) + (–0,435 × 1) + (0,127 × 1) + (–0,158 × 1) + (0,08 × 2) + (–0,092 × 2) + (0,565 × 2) + (–0,621 × 1) + (–0,525 × 1) + 6,344 = 3,083 The re sult of the bi nary lo gis tic re gres sion for mula cal cu la tion is as fol lows: p = 1 / 1 + e–z = 1 / 1 + 0,046 = 0,956 So, the re spon dent with the de scribed char ac ter is tics has the 95,6% prob a bil - ity to take part in the EP elec tion, Ob vi ously, we do not have many such real re - spon dents, There are only 25 re spon dents who an swered these sur vey ques tions in the above de scribed way from a gen eral sam ple of 30064 per sons in the EES study, Among them, 23 re spon dents took part in the Eu ro pean elec tion, 2 per sons re ported prob lems with health and be ing far from home as rea sons for non-vot ing. To check the re sults of the re search, the list of in de pend ent vari ables may be re duced to the most im por tant fac tors with the high est re gres sion co ef fi cients (more than 0,5). In this case, the imag i nary re spon dent: Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 153 Individual-level factors influencing voter turnout in the 2014 European Parliament election — has all the necessary information in order to choose who to vote for in the recent European elections (qp6_1 = “1”); — for whom it is very important which particular candidates have been elected as MEPs in their country (qp6_7 = “1”); — feels close to certain political party (qpp21 = “2”); — believes that his or her voice counts at the European level (d72_1 = “1”); — was contacted by national political parties regarding his or her vote in the recent European elections (qp12 = “1”); — communicated with friends or family about European elections (qp11_3 = “1”); The same cal cu la tions (but with up dated re gres sion co ef fi cients and con - stant) of the lo gis tic re gres sion for mula have the fol low ing re sult. p = 1 / 1 + e–z = 1 / 1 + 0,083 = 0,923 This re sult shows that the re spon dent who an swered the se lected 6 ques tions in the above sug gested way has 92,3% prob a bil ity to par tic i pate in the elec tions. In the EES dataset, we have 1024 such re spon dents, 954 of whom (93,2%) an - swered ‘yes’ to the ques tion about their par tic i pa tion in the elec tions. 70 re spon - dents who re ported non-vot ing state the fol low ing rea sons (ta ble 6). Table 6 Reasons of non-voting as reported by absentees Reason Numbers Sick or health problem at the time 10 On holiday or away from home 17 Too busy or no time or at work 15 Involved in a family or leisure activity 10 Registration or voting card problems 3 Lack of trust in or dissatisfaction with politics in general 8 Not interested in politics as such 5 Not interested in European matters 1 Not really satisfied with the European Parliament as an institution 3 Opposed to the EU 4 Do not know much about the EU or the European Parliament or the European Parliament elections 1 wrapdefaultVote has no consequences or vote does not change anything 6 Other 5 Do not know 1 Some re spon dents who were in clined to vote did not do this be cause of ex ter - nal rea sons (health prob lems, be ing away from home). Other re spon dents ex plain non-vot ing by lack of in ter est or trust in pol i tics, be ing op posed to the EU, etc. Such re spon dents ap pear be cause the cor re spond ing vari ables (which ex cluded such re spon dents) were not con sid ered af ter re duc ing the num ber of vari ables. Gen er ally, the bi nary lo gis tic re gres sion pre dicted the par tic i pa tion of re spon - 154 Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 Serhii Shapovalov dents in the EP elec tions with a high level of ac cu racy and has shown the main in - di vid ual mo ti va tions which make a per son in clined to vote. Discussion The iden ti fied fac tors which in flu ence the re spon dent’s de ci sion to vote in their es sence cor re spond to the as sump tions of sec ond-or der elec tions the ory. One of the stron gest fac tors is the im por tance for the re spon dent of who is elected as MEP in his or her coun try. This shows cit i zen’s per cep tion of the Eu ro pean elec tions as im por tant that mo ti vates a cit i zen to par tic i pate in the elec tions. This con clu sion di rectly cor re sponds to the sec ond-or der elec tions the ory, which ex plains low voter turn out by low per ceived im por tance of the EP elec tions [Reif, Schmitt, 1980]. Im por tant fac tor is the in for ma tion mes sages con cern ing Eu ro pean elec tions in the pub lic space. Cit i zens who stated that they had re ceived in for ma tion about the Eu ro pean elec tions from the me dia or who had been con tacted by party rep re - sen ta tives, ap peared to be in clined to par tic i pate in the Eu ro pean Par lia ment elec tions. The ac tu al iza tion of the Eu ro pean elec tions in pub lic space de pends to a large ex tent on the mo bi li za tion ef forts made by po lit i cal par ties. How ever, if na tional po lit i cal par ties them selves do not con sider the elec tions to the Eu ro - pean Par lia ment as im por tant, they make the ra tio nal de ci sion not to spend re - sources on the elec tion cam paign be fore the Eu ro pean elec tions. Ac cord ing to EES data, re spon dents who had all the nec es sary in for ma tion in or der to choose who to vote for in the 2014 Eu ro pean elec tions are much more in clined to vote. Ob vi ously, the num ber of such in formed cit i zens de pends on mo bi li za tion ef forts of po lit i cal par ties and in for ma tion mes sages in pub lic space. In for ma tion si lence and lack of ef forts by par ties to mo bi lize vot ers re sult in sig nif i cantly lower turn - out rates. This fac tor is of the big gest im por tance for the East ern Eu ro pean coun - tries which joined the EU af ter 2004. These coun tries have the low est voter turn - out level among all EU mem ber states (13,05% in Slovakia and 18,2% in Czech Re pub lic in 2014 EP elec tions). Try ing to ex plain this phe nom e non re search ers de fine dif fer ent fac tors in in di vid ual coun tries which caused low turn out in 2014 EP elec tions, but the com mon fea ture of all East ern Eu ro pean coun tries is a low in ten sity of elec tion cam paign. Na tional po lit i cal party give the pri or ity to na - tional po lit i cal arena and con sider EP elec tions as sec ond-or der elec tions. This causes low mo bi li za tion ef forts made by po lit i cal par ties and as a con se quence less in for ma tion about EP elec tions in pub lic space and less pub lic at ten tion [Fislage, 2015]. As the re sult, vot ers also con sider EP elec tions as sec ond-or der elec tions and de cide not to participate. A sense of be long ing to the Eu ro pean com mu nity, a pos i tive as sess ment of re - spon dent’s coun try mem ber ship of the EU, a pos i tive at ti tude to wards deep en ing Eu ro pean in te gra tion also in flu ence cit i zens’ in cli na tion to par tic i pate in the Eu - ro pean Par lia ment elec tions, but these fac tors have been some what less im por - tant, al though their im pact is sta tis ti cally proven. The in di vid ual ori en ta tions shaped by na tional po lit i cal con texts are also im - por tant fac tors of voter be hav ior: trust in na tional au thor i ties, in ter est in pol i tics and a cit i zen’s com mit ment to a par tic u lar party make him or her more in clined to par tic i pate in elec tions. Al though these fac tors do not di rectly de ter mine cit i - Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 155 Individual-level factors influencing voter turnout in the 2014 European Parliament election zens’ at ti tude to wards the Eu ro pean Par lia ment or the EU as a whole, they do form spe cific pat terns of cit i zens’ elec toral be hav ior at na tional level which also ex tend to the be hav ior dur ing the EP elec tions. Sec ond-or der elec tion the ory states that EP elec tion cam paigns are dom i nated by na tional mat ters (top ics rel - e vant to each coun try, po lit i cal sit u a tion in each coun try, etc.) [Reif, Schmitt, 1980; Schmitt, 2005]. Ac cord ingly, if voter be hav ior in the Eu ro pean elec tions is af fected by fac tors shaped in na tional po lit i cal en vi ron ment, such be hav ior can be seen as an ev i dence of the sec ond-or der na ture of the Eu ro pean elec tions. One of the in ter est ing find ings based on the EES data is that trust in the EU in sti tu tions has no sta tis ti cally sig nif i cant ef fect on voter’s de ci sion to par tic i pate in elec - tions. On the con trary, trust in na tional au thor i ties ap peared to be a sig nif i cant vari able. Some re search ers state that trust in na tional au thor i ties is sig nif i cantly lower in post-so cial ist coun tries that low ers par tic i pa tion in na tional pol i tics and con se quently low ers voter turn out in EP elec tions [Rose, 2004]. The sec ond-or der na ture of EP elec tions may also di min ish the ef fect of some macro-level fac tors. The in ten sity of po lit i cal com pe ti tion is as sumed to in crease voter turn out. Close com pe ti tion and ab sence of a clear leader be fore elec tions in - creases the per ceived weight of cit i zen’s in di vid ual vote and per ceived abil ity to in flu ence elec tion re sult that mo ti vates a cit i zen to par tic i pate in elec tion. This logic works in na tional elec tions when elec tions out come is per ceived as im por - tant by both cit i zens and par ties. Out comes of sec ond-or der elec tion are less im - por tant and ab sence of clear win ner does not mo ti vate cit i zens to par tic i pate. Conclusion Ac cord ing to the re sults of anal y sis of EES-2014 post-elec tion sur vey data, the main in di vid ual-level fac tors which make a cit i zen in clined to take part in the Eu ro pean elec tions are the fol low ing: — having the necessary information in order to choose who to vote for in the recent European elections; — importance for a person which particular candidates have been elected as MEPs in his or her country; — feeling close to certain political party; — belief that one’s voice counts at the European level; — being contacted by national political parties regarding one’s vote in the recent European elections; — communication with friends or family about European elections. As we can see, feel ing of po lit i cal ef fi cacy makes vot ing a sen si ble act for a per - son. Feel ing close to cer tain po lit i cal party mo ti vates a per son to sup port his or her party in all types of elec tions. Other fac tors are re lated to the sec ond-or der elec - tions the ory. ‘Hav ing the nec es sary in for ma tion about can di dates’, ‘com mu ni ca - tion of cit i zens with their friends and fam i lies’, ‘be ing con tacted by na tional po lit i - cal par ties’ are the fac tors which are the con se quences of the sec ond-or der na ture of the Eu ro pean elec tions. Pub lic dis cus sions of the EP elec tions topic, avail abil ity of in for ma tion about par ties and elec toral pro cess as a whole, field work with the cit i - zens are the con se quences of par ties’ mo bi li za tion ef forts dur ing the pre-elec tion cam paign. If the par ties do not con sider Eu ro pean elec tions as im por tant, they ra - tio nally de cide not to spend re sources for elec tion cam paign. As the re sult, in for - 156 Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 Serhii Shapovalov ma tion si lence caused by low per ceived im por tance of EP elec tions re in forces the pub lic con sid er ation of the Eu ro pean elec tions as not im por tant. Other in di vid ual mo ti va tions (feel ing at tached to Eu rope, be ing in ter ested in pol i tics, trust in the EU in sti tu tions, pos i tive eval u a tion of one’ coun try mem - ber ship of the EU etc.) do make their con tri bu tion in the pre dic tion of in di vid ual act of vot ing, but the im pact of these vari ables is lower than of the above listed vari ables. The im por tance of fac tors re lated to the sec ond-or der elec tions par a digm is partly con firmed by the voter turn out in 2019 EP elec tions. More mo bi li za tion ef forts by the par ties and more vis i ble elec tion cam paign made the topic of the EP elec tions more pub lic and mo ti vated the voter to turn out. References Blais, A., Carty, R. (1990) Does pro por tional rep re sen ta tion fos ter voter turn out? Eu ro - pean Jour nal of Po lit i cal Re search, 18, 167–181. Blais, A., Aarts, K. (2006) Elec toral Sys tems and Turn out. Acta Politica, 41, 180–196. Blais, A., Dobrzynska, A. (1998) Turn out in elec toral de moc ra cies. Eu ro pean Jour nal of Po lit i cal Re search, 33, 239–261. Blais, A. (2006) What Af fects Voter Turn out? An nual Re view of Po lit i cal Sci ence, 9, 111–125. Blais, A., Massicotte, L., Dobrzynska, A. (2003) Why is Turn out Higher in Some Coun - tries than in Oth ers? Retreived from: https://www.elec tions.ca/res/rec/part/tuh/TurnoutHigher.pdf Clark, N. (2013) Ex plain ing Low Turn out in Eu ro pean Elec tions: The Role of Is sue Sa - lience and In sti tu tional Per cep tions in Elec tions to the Eu ro pean Par lia ment. Jour nal of Eu ro - pean In te gra tion, 36 (4), 339–356. Dyck, J., Lascher, E. (2009) Di rect De moc racy and Po lit i cal Ef fi cacy Re con sid ered. Po lit - i cal Be hav ior, 31 (3), 401–427. Field, A. (2009) Dis cov er ing Sta tis tics Us ing SPSS. Third Edi tion. Lon don: SAGE Pub li ca - tions Ltd. Fiorino, N., Pontarollo, N., Ricciuti, R. (2016) Voter Turn out in Eu ro pean Par lia ment Elec tions: A Spa tial Anal y sis. CESifo Work ing Pa per Se ries, 5910. Fislage, F. (1015) EU Elec tions — Where Are the Vot ers? In: Konrad Adenauer Stiftung. Retreived from: https://www.kas.de/c/doc u ment_li brary/get_file?uuid= d3b13d4c-81ad-5926-f2eb-d41cbc81f95b&groupId=252038. Frank lin, M., Bernhard, W. (2009) Turn ing Out or Turn ing Off: Do Mo bi li za tion and At - ti tudes Ac count for Turn out Dif fer ences be tween New and Es tab lished Mem ber States at the 2004 EP Elec tions? Eu ro pean In te gra tion, 31 (5), 609–626. Frank lin, M. (2004) Voter Turn out and the Dy nam ics of Elec toral Com pe ti tion in Es tab - lished De moc ra cies Since 1945. Cam bridge: Cam bridge Uni ver sity Press. Frank lin, M., Eijk, M. van der, Marsh, M. (1996) What Vot ers Teach Us About Eu - rope-Wide Elec tions: What Eu rope-Wide Elec tions Teach Us About Vot ers. Elec toral Stud ies, 15 (2), 149–166. Gronlund, K., Setala, M. (2007) Po lit i cal Trust, Sat is fac tion and Voter Turn out. Com par - a tive Eu ro pean Pol i tics, 5, 400–422. Hobolt, S., Spoon, J., Til ley, J. (2008) A Vote Against Eu rope? Ex plain ing De fec tion at the 1999 and 2004 Eu ro pean Par lia ment Elec tions. Brit ish Jour nal of Po lit i cal Sci ence, 39 (1), 93–115. Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 157 Individual-level factors influencing voter turnout in the 2014 European Parliament election Jackman, R. (1987) Po lit i cal In sti tu tions and Voter Turn out in the In dus trial De moc ra - cies / Rob ert Jackman. Amer i can Po lit i cal Sci ence Re view, 81 (2). 405–423. Marsh, M. (1998) Test ing the Sec ond-Or der Elec tion Model af ter Four Eu ro pean Elec - tions. Brit ish Jour nal of Po lit i cal Sci ence, 28 (4), 591–607. Mattila, M. (2003) Why bother? De ter mi nants of turn out in the Eu ro pean elec tions. Elec toral Stud ies, 22, 449–468. Oelbermann, K., Pukelsheim, F. (2015) Eu ro pean Elec tions 2014: From Vot ers to Rep re - sen ta tives, in Twenty-eight Ways. Eu ro pean Elec toral Stud ies. 10 (2), 91–124. Powell, B. (2009) Con tem po rary De moc ra cies: Par tic i pa tion, Sta bil ity, and Vi o lence. Har vard: Har vard Uni ver sity Press. Reif, K., Schmitt, H. (1980) Or der Na tional Elec tions–A Con cep tual Frame work for the Anal y sis of Eu ro pean Elec tion Re sults. Eu ro pean Jour nal of Po lit i cal Re search, 8, 3–44. Rose, R. (2004) Eu rope ex pands, turn out falls: The Sig nif i cance of the 2004 Eu ro pean Par - lia ment Elec tion. Glas gow: Cen tre for the Study of Pub lic Pol icy, Uni ver sity of Strathclydes. Schmitt, H., Mannheimer, R. (1991) About vot ing and non-vot ing in the Eu ro pean elec - tions of June 1989. Eu ro pean Jour nal of Po lit i cal Re search, 19, 31–54. Schmitt, H. (2005) The Eu ro pean Par lia ment Elec tions of June 2004: Still Sec ond-Or - der? West Eu ro pean Pol i tics, 28, 650–679. Søderlund, P., Wass, H., Blais, A. (2011) The im pact of mo ti va tional and con tex tual fac - tors on turn out in first- and sec ond-or der elec tions. Elec toral Stud ies, 30, 689–699. Received 12.08.2019 SERHII SHAPOVALOV Individual-level factors influencing voter turnout in the 2014 European Parliament election The par tic i pa tion of EU cit i zens in the Eu ro pean Par lia ment elec tions is much lower than in the na - tional elec tions and dif fers sig nif i cantly across the EU mem ber states. The ar ti cled is aimed at find ing out the fac tors that in flu ence the par tic i pa tion of EU cit i zens in the elec tions to the Eu ro - pean Par lia ment. The study out lines the the o ret i cal ap proaches to un der stand ing the phe nom e non of the Eu ro pean Par lia ment elec tions and the as sump tions of re search ers about the fac tors that may in flu ence the elec toral ac tiv ity of cit i zens. Ac cord ing to sec ond-or der elec tions the ory by Karlheinz Reif and Hermann Schmitt cit i zens per ceive Eu ro pean Par lia ment elec tions as less im - por tant than na tional elec tions that re sults in lower par tic i pa tion. How ever, the be hav ior of vot ers may also be af fected by a num ber of fac tors which may be re garded as in di vid ual-level mo ti va - tions (trust in na tional and Eu ro pean au thor i ties, at ti tudes to wards EU in sti tu tions etc.). With use of bi nary lo gis tic re gres sion method it was de fined which fac tors in flu enced the par tic i pa tion of cit i zens in the Eu ro pean Par lia ment elec tion 2014 and ex plained the dif fer ences in the elec toral ac tiv ity of cit i zens of dif fer ent EU coun tries. The na ture of the iden ti fied fac tors that in flu ence the participation of citizens in the elections to the European Parliament suggests that the sec ond- or der elections theory is still valid. Keywords: elections, voter turnout, EU, factors, attitudes 158 Со ци о ло гия: те о рия, ме то ды, мар ке тинг, 2019, 3 Serhii Shapovalov

Ähnliche Einträge