230 A Appendix. Electoral vp1 No Yes In 1979 drop out coded as. Variable in dataset Categories years Remarks. Independent variables

Transcription

1 A Appendix Springer International Publishing 2017 A.C. Goldberg, The Impact of Cleavages on Swiss Voting Behaviour, Contributions to Political Science, DOI /

2 230 A Appendix Table A.1 Operationalisation of variables for longitudinal analyses Original variable Missing Variable in dataset Categories years Remarks Dependent variables Party choice vdn1b FDP (LPS) CVP (CSP) SP SVP GP Electoral vp1 No Yes In 1979 drop out coded as participation no Independent variables Denomination Religion Protestant Catholic Other/no denomination Church Churchg Often Sometimes Rarely Never 1987, 1991 Respondents with no attendance denomination coded as never Combined religion Oesch 8class scheme Catholic often Catholic sometimes Catholic rarely Catholic never Other/no Protestant often Protestant sometimes Protestant rarely Protestant never denomination clas8_x Socio-cultural specialists Managers and administrators Service workers Technical specialists Production workers 1979, 1987, 1991 Clerks Liberal professions and large employers Small business owners 1987, 1991 Combination of religion and churchg Added values for 1971 and 1975 from separate data files, recoding based on ISCO values

3 A Appendix 231 Reduced3class scheme Rural-urban (dummy) Rural-urban (5 cat.) Language region clas8_x Service and production workers Socio-cultural and technical specialists, managers and administrators, clerks Liberal professions, large employers and small business owners Collapsed version of clas8_x, recoded professions for 1979, 1987 and 1991 sg9 Urban Rural 1979, 1987, 1991 Coding for 1971, 1975 and UrbRur, AggGross Central city in big agglomeration Central city in small agglomeration Other municipality in big agglomeration Other municipality in small agglomeration Rural municipality isolated city 1995 based on community number (BfS) UrbRur and AggGross for recoded based on Bfs community codes sg1 German French Italian 1979 Additional recoding based on interview language when possible/adequate Control variables Age Age Continuous (18 98) Sex Sex Male Female Education Educ Low Middle High

4 232 A Appendix Table A.2 Operationalisation of variables for contextual analyses (Chaps. 4 and 5) Original variable in Variable dataset 07/11 Categories Remarks Dependent variables Party choice f11801 FDP (LPS) CVP (CSP) SP SVP GP Voting propensity f14400a-e FDP CVP SP SVP GP Independent variables Denomination f20800 Protestant Catholic Other/no denomination Church attendance f20900 Several times a week/once a week Social class (Oesch) clas8_x Socio-cultural specialists EU/immigration scale f15430(r), f15440(r) Tax/social spending scale f15480(r), f15420(r) Rural-urban UrbRur, AggGross Managers and administrators Join EU/equal chances Tax and spending increase Central city in big agglomeration Once or twice per month Several times a year Service workers Technical specialists Clerks Liberal professions and large employers Neutral Stay out of EU/better chances Neutral Tax and spending cut Central city in small Other agglomeration municipality in big agglomeration Once a year Only on special occasions Never Respondents with no denomination coded as never (f20800) Production workers Reduced five category scheme for random slope models (see Sect. 5.4) Small business owners Other municipality in small agglomeration Rural municipality isolated city UrbRur and AggGross for 2007 recoded based on Bfs community codes; reduced three categories for random slope models (see Sect. 5.4) Context variables Catholic-Protestant ratio Natural logarithm of ratio between Catholic and Protestant population per canton in 2010 Data from BfS Wealth inequality (gini) Gini coefficient measuring inequality in wealth in 2003 Data from Jeitzinger and Peters (2007) Urbanisation Percentage of residents living in urban areas 2010 Data from BfS

5 A Appendix 233 Control variables Age Age Continuous (18 96) Sex Sex Male Female Education educ_r Low Middle High Language region lregion German French Italian Additional recoding based on interview language in 2007; 2011 Rhaeto-Romanic = German 0:75 Threshold of exclusion Natural logarithm of proportion of votes a party can receive without winning a seat: 1Cm district magnitude with m as

6 234 A Appendix Table A.3 Multinomial logistic model on party choice including all cleavage variables SVP FDP CVP GP Denomination (ref. Protestant) Catholic 0:14 0:14 1:56 0:07.0:15/.0:15/.0:19/.0:18/ Other/none 0:37 0:40 0:40 0:04.0:18/.0:18/.0:24/.0:18/ Church attendance 0:05 0:09 0:34 0:06.0:05/.0:04/.0:05/.0:05/ Social class (ref. managers/administrators) Socio-cultural specialists 1:06 1:18 0:72 0:32.0:22/.0:20/.0:22/.0:21/ Service workers 0:14 0:32 0:14 0:05.0:25/.0:25/.0:27/.0:32/ Technical specialists 0:88 0:32 0:44 0:47.0:24/.0:22/.0:26/.0:25/ Production workers 0:26 0:73 0:38 0:28.0:23/.0:26/.0:26/.0:30/ Clerks 0:07 0:44 0:15 0:12.0:23/.0:23/.0:26/.0:29/ Liberal professions and large employers 0:12 0:14 0:27 0:79.0:31/.0:28/.0:30/.0:33/ Small business owners 0:21 0:33 0:02 0:09.0:23/.0:23/.0:25/.0:31/ Rural-urban (ref. rural municipality/isolated town) City in big agglomeration 0:99 0:75 1:09 0:26.0:22/.0:21/.0:26/.0:23/ City in small agglomeration 0:47 0:23 0:42 0:08.0:21/.0:21/.0:21/.0:25/ Municipality in big agglomeration 0:06 0:15 0:85 0:50.0:19/.0:19/.0:21/.0:22/ Municipality in small agglomeration 0:08 0:27 0:16 0:12.0:19/.0:19/.0:19/.0:26/ EU/Immigration scale (ref. neutral) Pro 2:41 0:85 0:69 0:05.0:20/.0:14/.0:16/.0:15/ Contra 1:29 0:67 0:66 0:26.0:17/.0:18/.0:19/.0:24/ (continued)

7 A Appendix 235 Table A.3 (continued) SVP FDP CVP GP Tax/Social spending scale (ref. neutral) Increase 1:52 1:56 0:95 0:15.0:14/.0:14/.0:15/.0:15/ Decrease 1:98 2:06 1:95 0:84.0:38/.0:37/.0:39/.0:53/ Woman 0:33 0:01 0:03 0:15.0:14/.0:14/.0:14/.0:15/ Age 0:01 0:01 0:00 0:02.0:00/.0:00/.0:00/.0:00/ Education (ref. middle) Low 0:40 0:18 0:13 0:31.0:23/.0:25/.0:26/.0:35/ High 0:37 0:24 0:03 0:12.0:15/.0:15/.0:16/.0:17/ Language region (ref. German-speaking) French 0:51 0:32 0:21 0:09.0:17/.0:16/.0:17/.0:16/ Italian 1:27 0:16 0:35 0:48.0:21/.0:18/.0:19/.0:25/ Constant 1:36 0:30 0:90 0:19.0:32/.0:34/.0:37/.0:39/ N 4140 Pseudo R Standard errors in parentheses; reference outcome is SP p<0:10; p<0:05; p<0:01 marginal effect catholic ZG AI marginal effect church attendance ZG AI log ratio Catholics Protestants 2010 log ratio Catholics Protestants 2010 Fig. A.1 Cantonal distribution of marginal effects of religion for CVP voting (Refers to Fig. 5.5)

8 236 A Appendix Table A.4 Distribution of contextual variables by canton Catholic-Protestant Canton ratio (log) Wealth inequality (gini) Urbanisation Zürich 0:20 0:84 0:95 Bern 1:27 0:80 0:63 Luzern 1:78 0:83 0:51 Uri 3:20 0:69 0 Schwyz 1:75 0:88 0:80 Obwalden 2:24 0:79 0 Nidwalden 1:96 0:87 0:88 Glarus 0:12 0:77 0 Zug 1:26 0:84 0:96 Fribourg 1:50 0:83 0:56 Solothurn 0:41 0:86 0:77 Basel Stadt 0:07 0:90 1 Basel Land 0:17 0:85 0:92 Schaffhausen 0:55 0:77 0:76 Appenzell Ausserrhoden 0:30 0:77 0:53 Appenzell Innerrhoden 2:10 0:78 0 St. Gallen 0:74 0:77 0:67 Graubünden 0:22 0:72 0:50 Aargau 0:21 0:77 0:66 Thurgau 0:12 0:78 0:50 Ticino 2:73 0:76 0:66 Vaud 0:07 0:91 0:75 Valais 2:46 0:80 0:57 Neuchâtel 0:01 0:83 0:75 Genève 1:14 0:88 0:99 Jura 1:94 0:80 0:31 ; 0:90 0:81 0:60 marginal effect catholic 2.0 ZG AI marginal effect church attendance AI ZG log ratio Catholics Protestants log ratio Catholics Protestants Fig. A.2 Cantonal distribution of marginal effects of religion for SVP voting (Refers to Fig. 5.6)

9 A Appendix 237 Table A.5 Lambda indices on cantonal level Canton Denomination Social class Rural-urban Zürich 41 (09) 84 (13) 72 (13) Bern 93 (20) (21) (24) Luzern 75 (21) 90 (23) 88 (29) Uri 91 (37) 89 (33) NA Schwyz 89 (22) 91 (28) (26) Obwalden 95 (22) (33) NA Nidwalden NA NA NA Glarus NA NA NA Zug 83 (24) 98 (25) 67 (20) Fribourg (29) (50) (24) Solothurn (25) 79 (26) (23) Basel Stadt (31) 90 (30) 55 (24) Basel Land (38) (30) 76 (22) Schaffhausen 94 (29) (39) (45) Appenzell Ausserrhoden 99 (36) 94 (38) 50 (24) Appenzell Innerrhoden 84 (49) NA NA St. Gallen 90 (22) 66 (21) 81 (20) Graubünden (64) 95 (39) 90 (32) Aargau 95 (27) 79 (26) 82 (23) Thurgau (31) (29) 96 (28) Ticino 82 (15) 48 (12) 37 (10) Vaud 80 (22) (35) (22) Valais (26) 77 (28) 87 (28) Neuchâtel 85 (31) (33) (28) Genève 86 (12) 65 (12) 47 (09) Jura (34) (38) (36) ; Standard deviation in parentheses

10 238 A Appendix Table A.6 Contextual coefficients of complete multilevel models excluding once wealth inequality and once urbanisation SVP FDP CVP SP GP Excluding urbanisation LEVEL 2(CONTEXT) Catholic-Protestant ratio 0:265 0:015 0:142 0:160 0:120.0:087/.0:064/.0:079/.0:067/.0:053/ Wealth inequality (gini) 1:104 0:274 1:459 0:416 0:533.1:966/.1:471/.1:767/.1:551/.1:263/ Threshold (log) 0:116 0:144 0:203 0:112 0:005.0:122/.0:088/.0:110/.0:094/.0:073/ Romandie 0:755 0:399 0:301 0:163 0:085.0:207/.0:160/.0:185/.0:169/.0:139/ Constant 4:189 5:402 3:895 7:498 6:552.1:566/.1:176/.1:408/.1:239/.1:013/ Excluding wealth inequality LEVEL 2(CONTEXT) Catholic-Protestant ratio 0:275 0:015 0:116 0:156 0:114.0:084/.0:063/.0:075/.0:067/.0:050/ Urbanisation 0:525 0:052 0:621 0:218 0:300.0:360/.0:278/.0:319/.0:292/.0:230/ Threshold (log) 0:041 0:138 0:084 0:143 0:047.0:130/.0:098/.0:115/.0:103/.0:077/ Romandie 0:775 0:409 0:244 0:172 0:101.0:191/.0:150/.0:170/.0:158/.0:122/ Constant 3:449 5:199 5:169 7:776 6:903.0:359/.0:287/.0:318/.0:300/.0:248/ Standard errors in parentheses; individual coefficients not shown p<0:1; p<0:05; p<0:01

11 A Appendix 239 Table A.7 Multilevel models for religious voting (For Figs. 5.5 and 5.6) Probability to vote CVP LEVEL 1(INDIVIDUAL) Catholic 1:096 0:316.0:139/.0:135/ Other/no denomination 0:004 0:427.0:119/.0:116/ Church attendance 0:346 0:063.0:040/.0:034/ Woman 0:073 0:734.0:063/.0:079/ Age 0:017 0:012.0:002/.0:002/ Low education 0:074 0:069.0:104/.0:129/ High education 0:445 1:335 SVP.0:069/.0:086/ LEVEL 2(CONTEXT) Catholics-Protestants ratio (log) 0:184 0:059.0:170/.0:209/ Catholics-Protestants ratio (log) 2 0:011 0:006.0:074/.0:088/ Threshold (log) 0:154 0:255.0:089/.0:137/ Romandie 0:313 1:389.0:156/.0:216/ INTERACON TERMS Catholic X 0:170 0:449 Catholics-Protestants ratio (log).0:176/.0:175/ Church attendance X 0:019 0:116 Catholics-Protestants ratio (log).0:046/.0:041/ Catholic X 0:150 0:117 Catholics-Protestants ratio (log) 2.0:076/.0:075/ Church attendance X 0:009 0:034 Catholics-Protestants ratio (log) 2.0:020/.0:018/ Constant 4:424 6:419.0:278/.0:388/ Observations (cantons) 8,278 (26) 8,343 (26) Log likelihood 20, , Akaike Inf. Crit. 40, , Bayesian Inf. Crit. 41, , Note: p<0:1; p<0:05; p<0:01; standard errors in parentheses

12 240 A Appendix Table A.8 Multilevel models for class voting (For Figs. 5.7, 5.8 and 5.9) Probability to vote FDP SP SVP LEVEL 1(INDIVIDUAL) Specialists 5:774 0:353 0:394.1:915/.2:747/.1:810/ Workers 4:184 3:552 3:544.1:854/.2:200/.2:341/ Clerks 2:166 0:418 0:179.2:432/.2:988/.3:034/ Small business owners 1:842 0:534 4:588.2:213/.2:988/.2:828/ Woman 0:246 0:653 0:712.0:070/.0:079/.0:083/ Age 0:005 0:020 0:012.0:002/.0:002/.0:002/ Low education 0:170 0:140 0:119.0:119/.0:133/.0:140/ High education 0:415 0:460 0:933.0:081/.0:090/.0:095/ LEVEL 2(CONTEXT) Gini wealth 3:324 1:872 0:756.1:461/.1:680/.2:375/ Threshold (log) 0:190 0:118 0:259.0:060/.0:076/.0:103/ Romandie 0:665 0:205 1:249.0:126/.0:156/.0:195/ INTERACON TERMS WITH GINI WEALTH Specialists 7:916 2:047 1:627.2:346/.3:374/.2:205/ Workers 6:155 4:604 3:701.2:279/.2:705/.2:879/ Clerks 3:153 0:105 0:460.2:972/.3:655/.3:709/ Small business owners 1:698 0:102 4:534.2:717/.3:673/.3:474/ Constant 3:081 2:858 6:610.1:160/.1:322/.1:888/ Observations (cantons) 7,468 (26) 7,520 (26) 7,548 (26) Log likelihood 18, , , Akaike Inf. Crit. 37, , ,4780 Bayesian Inf. Crit. 37, , , Reference category for social class are the two joined categories of liberal professions/large employers with managers/administrators p<0.1; p<5; p<1; standard errors in parentheses

13 A Appendix 241 marginal effect specialists 2.0 AI ZG marginal effect workers 2.0 AI ZG marginal effect clerks 2.0 AI ZG marginal effect small business owners 2.0 AI ZG Fig. A.3 Cantonal distribution of marginal effects of class for FDP voting (Refers to Fig. 5.7) marginal effect specialists ZG marginal effect workers AI ZG AI marginal effect clerks AI ZG marginal effect small business owners AI ZG Fig. A.4 Cantonal distribution of marginal effects of class for SP voting (Refers to Fig. 5.8)

14 242 A Appendix 2.0 marginal effect specialists AI ZG marginal effect workers AI ZG 2.0 marginal effect clerks AI ZG marginal effect small business owners AI ZG Fig. A.5 Cantonal distribution of marginal effects of class for SVP voting (Refers to Fig. 5.9)

15 A Appendix 243 Table A.9 Multilevel models for rural-urban voting (For Figs and 5.11) Probability to vote SP GP SVP FDP CVP LEVEL 1(INDIVIDUAL) City 1:527 0:365 1:025 1:022 2:600.2:471/.1:741/.2:450/.1:734/.1:597/ Municipality 0:660 1:014 0:875 2:603 0:268 (in agglomeration).2:035/.1:781/.2:167/.1:735/.1:798/ Woman 0:671 0:877 0:694 0:263 0:004.0:078/.0:071/.0:081/.0:070/.0:069/ Age 0:024 0:028 0:007 0:003 0:007.0:002/.0:002/.0:002/.0:002/.0:002/ Low education 0:145 0:152 0:158 0:169 0:032.0:131/.0:118/.0:136/.0:117/.0:115/ High education 0:657 0:825 1:264 0:473 0:409.0:086/.0:078/.0:090/.0:077/.0:076/ LEVEL 2(CONTEXT) Urbanisation 3:772 4:138 13:209 9:979 1:865.4:187/.4:700/.6:236/.4:308/.5:920/ Urbanisation 2 3:294 4:407 9:465 7:180 0:024.3:133/.3:490/.4:594/.3:217/.4:292/ Threshold (log) 0:019 0:232 0:195 0:255 0:317.0:135/.0:104/.0:149/.0:097/.0:160/ Romandie 0:515 0:469 1:253 0:564 0:262.0:199/.0:158/.0:212/.0:152/.0:207/ INTERACON TERMS City X 1:364 2:327 5:029 3:098 6:698 Urbanisation.7:328/.5:185/.7:310/.5:171/.4:774/ Municipality (in agg.) X 0:035 0:755 3:469 7:405 1:410 Urbanisation.5:905/.5:211/.6:361/.5:066/.5:240/ City X 0:447 2:589 3:442 1:885 4:100 Urbanisation 2.5:234/.3:742/.5:263/.3:731/.3:454/ Municipality (in agg.) X 0:706 0:727 2:616 5:223 1:612 Urbanisation 2.4:164/.3:710/.4:537/.3:605/.3:708/ Constant 5:783 4:842 1:747 2:113 6:778.1:299/.1:503/.2:004/.1:371/.1:924/ Observations (cantons) 7,663 (26) 7,717 (26) 7,689 (26) 7,595 (26) 7,622 (26) Log likelihood 20, , , , , Akaike Inf. Crit. 40, , , , , Bayesian Inf. Crit. 40, , , , , Note: p<0.1; p<5; p<1; standard errors in parentheses

16 244 A Appendix SP GP marginal effect city 2.0 ZG marginal effect city 2.0 ZG urbanisation SVP urbanisation marginal effect city ZG urbanisation Fig. A.6 Cantonal distribution of marginal effects of city residents for SP, GP and SVP voting (Refers to Fig. 5.10) FDP CVP marginal effect city ZG marginal effect city ZG urbanisation urbanisation Fig. A.7 Cantonal distribution of marginal effects of city residents for FDP and CVP voting (Refers to Fig. 5.11)