Upload
reynold-hood
View
213
Download
1
Embed Size (px)
Citation preview
Turnout in multi-level systems
André Blais
• Why is there a turnout gap between regional, national and European election?
• Usual assumption : first order and second order elections.
The Approach
• 1. Comparing turnout across levels• 2. Comparing importance across levels• 3. Determining whether differences in
importance explain differences in turnout• 4. Examining the sources of importance: policy impact vs attachment 5. Importance does not explain everything:
Adding visibility
Turnout level in the various elections
France Germany Spain
Iles de France Provence Bavaria Lower Saxony
Catalonia Madrid
TurnoutEurope 43,1% 43,3% 40,9% 49,1% 46,2% 46,5%
National 54,2% 56,1% 70% 73,4% 65,9% 73,3%
Regional - - 63,9% 59,4% 69,8% N-A
Municipal 54,9% 50,3% - - - -
Mean Importance Ratings of Regional, National, and European Elections
France Germany Spain
Iles de France Provence Bavaria Lower Saxony
Catalonia Madrid
Importance election
Europe 5,9(3010)
5,8(2648)
4,1(5651)
4,8(2788)
5,2 (2857)
5,4(1913)
National 7,3(3021)
7,3(2657)
7(5705)
7,2(2804)
6,1(2859)
6,9 (1914)
Regional - - 6,8(5703)
6,6(2804)
7 (2864)
6,9(1917)
Municipal 7,4(3027)
7,7(2665)
- - - -
The Relative Importance of European and Regional Elections in % (Compared to National Elections)
National elections are less important
National electionsare as important
National electionsare more important
Paris European 9 35 56
Municipal 26 47 28
Provence European 9 35 56
Municipal 30 48 21
Catalonia European 14 40 46
Regional 37 49 14
Madrid European 6 42 53
Regional 15 64. 21
Lower Saxony European 3 29 68
Regional 11 52 36
Bavaria European 2 20 78
Regional 18 52 31
The Impact of Perceived Importance on the Propensity to Vote in Paris, France
Model 1 Model 2 Model 3 Model 4
Perceived Importance 3.9** (.34)+ 3.89**
Information 0.77*
Political Interest 3.11** 1.89** 1.78**
Education 0.22 0.0 -0.08
Gender 0.06 0.01 0.04
Age 0.03** 0.02** 0.02**
Municipal 0.03 0.09 0.15 0.15
Europe -0.45** -0.47** 0.0 .031
_Intercept 0.16** -3.2** -4.72** -4.93**
Predictive Margins
Municipal 0 1
0.51**0.52**
0.51**0.53**
0.52**0.55**
0.52**0.55**
Europe 0 1 0.55**
0.43**0.54**.45**
0.53**0.53**
0.52**0.57**
N 3149 3125 3017 2582
+ : variation in the predicted probability of voting when the importance score is 8 rather than 4.
The Impact of Perceived Importance on the Propensity to Vote in Provence, France
Model 1 Model 2 Model 3
Perceived Importance 3.86** (.34+) 3.74**
Information 0.71*
Political Interest 3.02** 1.55** 1.72**
Education 0.38 0.26 -0.08
Gender -0.02 -0.08 0.05
Age 0.03** 0.02** 0.02**
Municipal -0.23** -0.03 -0.01 0.07
Europe -0.515** -0.54** -0.01 0.27
_Intercept 0.25** -3.46** -4.8** -5.11**
Predictive Margins
Municipal 0 1
0.51**0.46**
0.5**0.5**
0.51**0.51**
0.51**0.52**
Europe 0 1
0.55**0.42**
0.54**0.43**
0.51**0.51**
0.49**0.54**
N 2747 2728 2647 2154
+ : variation in the predicted probability of voting when the importance score is 8 rather than 4.
The Impact of Perceived Importance on the Propensity to Vote in Catalonia, Spain
Model 1 Model 2 Model 3 Model 4
Perceived Importance 2.74** (.24+) 3.05**
Information 1.15**
Political Interest 2.44** 1.34** 1.04**
Education 0.48 0.04 -0.29
Gender -0.19 -0.27** -0.17
Age 0.02** 0.01** 0.01
Regional 0.21* 0.18 -0.04 -0.19
Europe -0.78** -0.89** -0.95** -0.29
_Intercept 0.63** -1.5** -2.13** -2.48**
Predictive Margins
Regional 01
0.58**0.64*
0.59**0.63**
0.62**0.61**
0.630.59
Europe 0 1
0.67**0.48**
0.67**0.48**
0.68**0.5**
0.64**0.59**
N 2929 2923 2858 2376
+ : variation in the predicted probability of voting when the importance score is 8 rather than 4.
The Impact of Perceived Importance on the Propensity to Vote in Madrid, Spain
Model 1 Model 2 Model 3 Model 4
Perceived Importance 3.2** (.27+) 3.09**
Information 1.23**
Political Interest 2.87** 1.62** 1.52**
Education 0.39 0.15 -0.1
Gender 0.08 0.03 0.06
Age 0.01** 0.0 -0.0
Europe -1.15** -1.25** -0.76** -0.2
_Intercept 1.01** -1.28** -2.14** -2.78**
Predictive Margins
Europe 0 1
0.73**0.47**
0.72**0.48**
0.68**0.54**
0.62**0.59**
N 1950 1945 1910 1696
+ : variation in the predicted probability of voting when the importance score is 8 rather than 4.
The Impact of Perceived Importance on the Propensity to Vote in Lower Saxony, Germany
Model 1 Model 2 Model 3 Model 4
Perceived Importance 3.39** (.28+) 3.23**
Information 1.2**
Political Interest 3.62** 2.1** 2.04**
Education 0.59 0.38 -0.27
Gender -0.03 -0.05 0.01
Age 0.01* 0.01* 0.01
Regional -0.63** -0.74** -0.41* 0.02
Europe -1.05** -1.28** -0.53** -0.07
_Intercept 1.02** -1.8** -3.06** -3.6**
Predictive Margins
Regional 01 0.65**
0.51**0.65**0.52**
0.65**0.58**
0.63**0.64**
Europe 0 1 0.69**
0.44**0.69**0.44**
0.66**0.57**
0.64**0.63**
N 2936 2913 2793 1794
+ : variation in the predicted probability of voting when the importance score is 8 rather than 4.
The Impact of Perceived Importance on the Propensity to Vote in Bavaria, Germany
Model 1 Model 2 Model 3 Model 4
Perceived Importance 3.68** (.31+) 3.41**
Political Interest 3.1** 1.71** 1.68**
Information 1.44**
Education 1.5** 1.34** 0.97**
Gender 0.02 0.09 -0.17*
Age 0.01** 0.01 0.01**
Regional -0.28** -0.28** -0.03 0.13
Europe -1.22** -1.52** -0.38** 0.02
_Intercept 0.85** -2.05** -3.79** -4.22**
Predictive Margins
Regional 01
0.64**0.58**
0.64**0.58**
0.63**0.63**
0.63**0.65**
Europe 0 1
0.67**0.38**
0.68**0.37**
0.64**0.58**
0.64**0.64**
N 13492 13469 12969 8903
+ : variation in the predicted probability of voting when the importance score is 8 rather than 4.
• Perceived importance is the most important explanatory variable.
• However, we must understand what shapes perceived importance.
The Effect of Impact and Attachment on the Perceived Importance of Local and National Elections
Policy Impact Attachment
Paris 0.13** 0.01
Provence 0.10** 0.05**
Catalonia 0.04* 0.26**
Madrid 0.07** 0.02+
Lower Saxony 0.24** 0.08**
Bavaria 0.22** 0.07**
The Effect of Impact and Attachment on the Perceived Importance of European and National Elections
Policy Impact Attachment
Paris 0.23** 0.06**
Provence 0.21** 0.1**
Catalonia 0.13** 0.1**
Madrid 0.25** 0.07**
Lower Saxony 0.4** 0.12**
Bavaria 0.35** 0.17**
And visibility
• The measure: knowledge of the top candidates and campaign slogans
The mean level of information in various elections
Paris National .639Paris Municipal .494Paris Europe .114Provence National .649Provence Municipal .490Provence Europe .245Barcelona National .657Barcelona Regional .720Barcelona Europe .217Madrid National .762Madrid Regional -Madrid Europe .249Lower Saxony National .630Lower Saxony Regional .317Lower Saxony Europe .200Bavaria National .616Bavaria Regional .524Bavaria Europe .200
• The turnout gap between elections can be explained by a few variables, especially the perceived importance of election.
• The perceived importance of elections can be explained by attachment and policy impact.
• Also relevant is voters’ knowledge of the candidates and slogans. This possibly reflects party mobilization.