Commit b5491161 authored by Arndt Leininger's avatar Arndt Leininger

substantial update: new models and streamlined text

parent c295d752
This diff is collapsed.
......@@ -56,6 +56,8 @@ df <-
'Non saprei' = as.character(NA)),
approval = as.numeric(str_extract(govapproval_pre, '\\d+')),
# govapproval_pre = ordered(govapproval_pre, levels = c('')),
opinion = as.numeric(str_extract(S26, '\\d+')),
undecided = opinion == 5 | S26 == 'Non saprei',
voteintention = recode(S28, 'Sì a favore della riforma' = 'Yes, for the reform',
'No, contro la riforma' = 'No, against the reform',
'Non ho ancora deciso' = 'I have not decided yet',
......@@ -120,16 +122,21 @@ df <-
'ho votato scheda bianca' = 'blank',
'non ho votato' = 'did not vote'),
vote = ifelse(votechoice == 'Yes', 1, ifelse(votechoice == 'No', 0, NA)),
pid = D11_W9,
pid = as.character(D11_W9),
pid = ifelse(pid != 'preferisco non rispondere', pid, NA),
pid_gov = pid %in% c('Partito Democratico',
'Area Popolare/Nuovo Centrodestra-UDC',
"Scelta Civica/Cittadini per l'Italia (include ALA)")
"Scelta Civica/Cittadini per l'Italia (include ALA)"),
pid_none = pid == 'nessun partito',
pid_3 = ifelse(pid_gov == T, 'government', ifelse(pid_none == T, 'none',
'opposition')),
pid_opp = pid_3 == 'opposition'
)
df <- df %>% select(gender, female, yearbirth, contains('age'), education, uni,
region,
contains('eco'), contains('approval'), contains('vote'),
contains('know'), pid, pid_gov)
contains('eco'), contains('approval'), opinion, undecided,
contains('vote'), contains('know'), contains('pid'))
# save data --------------------------------------------------------------------
......
# Economic Voting in Referendums
# created: 2017-06-05
# updated: 2017-06-05
# updated: 2018-12-13
# rm(list = ls())
......@@ -10,6 +10,7 @@ library(dplyr)
library(ggplot2)
library(scales)
library(extrafont)
loadfonts()
df <- read.csv('private/data/data.csv', stringsAsFactors = F) %>% tbl_df()
......@@ -24,7 +25,7 @@ ggplot(df, aes(x = eco_soc)) +
pdf('figures/f_eco_soc.pdf')
f_eco_soc + theme(text = element_text(family = 'CMU Serif',
size = 20))
size = 30))
dev.off()
# Sociotropic economic evaluations (post-wave)
......@@ -37,7 +38,7 @@ f_eco_soc_post <-
pdf('figures/f_eco_soc_post.pdf')
f_eco_soc + theme(text = element_text(family = 'CMU Serif',
size = 20))
size = 30))
dev.off()
......@@ -50,5 +51,6 @@ f_eco_ego <-
theme_bw()
pdf('figures/f_eco_ego.pdf')
f_eco_ego + theme(text = element_text(family = 'CMU Serif'))
f_eco_ego + theme(text = element_text(family = 'CMU Serif',
size = 30))
dev.off()
......@@ -29,46 +29,97 @@ dfig$se_fit <- predict(m1, newdata = dfig, type = 'response', se.fit = T)$se.fit
dfig$lwr <- dfig$prediction - 1.96 * dfig$se_fit
dfig$upr <- dfig$prediction + 1.96 * dfig$se_fit
dfig_m1 <- dfig
f_m1 <-
ggplot(dfig, aes(x = eco_soc, y = prediction, ymin = lwr, ymax = upr)) +
geom_errorbar(width = 0) +
geom_point(size = 2) +
scale_y_continuous(breaks = seq(.1, .9, .2)) +
ylab('Probability to vote yes') + xlab('Sociotropic economic evaluation') +
theme_bw()
pdf('figures/f_m1.pdf', width = 3, height = 3)
f_m1 + theme(text = element_text(family ='CMU Serif', size = 12))
pdf('figures/f_m1.pdf', width = 4, height = 4)
f_m1 + theme(text = element_text(family ='CMU Serif', size = 11))
dev.off()
# m2 (Interaction model (main text))
# # m2 (Interaction model (main text))
#
# f_m2 <-
# interplot(m2, 'eco_soc', 'know', hist = T, point = T) +
# xlab('Referendum knowledge') + ylab('Marginal effect') +
# scale_y_continuous(breaks = seq(0,1.5, .5)) +
# theme_bw()
#
# pdf('figures/f_m2.pdf', width = 3, height = 3)
# f_m2 + theme(text = element_text(family ='CMU Serif', size = 12))
# dev.off()
# # Model 3 in table in manuscript
#
# dfig <- data.frame(eco_soc = 1:5, know = 2, female = TRUE,
# age = median(df$age, na.rm = T),
# uni = T)
#
# dfig$prediction <- predict(m3, newdata = dfig, type = 'response')
# dfig$se_fit <- predict(m3, newdata = dfig, type = 'response', se.fit = T)$se.fit
# dfig$lwr <- dfig$prediction - 1.96 * dfig$se_fit
# dfig$upr <- dfig$prediction + 1.96 * dfig$se_fit
#
# f_m3 <-
# ggplot(dfig, aes(x = eco_soc, y = prediction, ymin = lwr, ymax = upr)) +
# geom_errorbar(width = 0) +
# geom_point(size = 2) +
# scale_y_continuous(breaks = seq(2, 8, 2)) +
# ylab('Government approval') + xlab('Sociotropic economic evaluation') +
# theme_bw()
#
# pdf('figures/f_m3.pdf', width = 3, height = 3)
# f_m3 + theme(text = element_text(family ='CMU Serif', size = 12))
# dev.off()
# Model 2 in table in manuscript
dfig <- data.frame(eco_soc = 1:5, know = 2, female = TRUE,
age = median(df$age, na.rm = T),
uni = T, pid_gov = F)
dfig$prediction <- predict(m2, newdata = dfig, type = 'response')
dfig$se_fit <- predict(m2, newdata = dfig, type = 'response', se.fit = T)$se.fit
dfig$lwr <- dfig$prediction - 1.96 * dfig$se_fit
dfig$upr <- dfig$prediction + 1.96 * dfig$se_fit
dfig_m2 <- dfig
f_m2 <-
interplot(m2, 'eco_soc', 'know', hist = T, point = T) +
xlab('Referendum knowledge') + ylab('Marginal effect') +
scale_y_continuous(breaks = seq(0,1.5, .5)) +
theme_bw()
ggplot(dfig, aes(x = eco_soc, y = prediction, ymin = lwr, ymax = upr)) +
geom_errorbar(width = 0) +
geom_point(size = 2) +
scale_y_continuous(breaks = seq(.1,.9,.2)) +
ylab('Probability to vote yes') + xlab('Sociotropic economic evaluation') +
theme_bw()
pdf('figures/f_m2.pdf', width = 3, height = 3)
f_m2 + theme(text = element_text(family ='CMU Serif', size = 12))
dev.off()
# Model 3 in table in manuscript
dfig <- data.frame(eco_soc = 1:5, know = 2, female = TRUE,
age = median(df$age, na.rm = T),
uni = T)
uni = T, vote_gov = 0)
dfig$prediction <- predict(m3, newdata = dfig, type = 'response')
dfig$se_fit <- predict(m3, newdata = dfig, type = 'response', se.fit = T)$se.fit
dfig$lwr <- dfig$prediction - 1.96 * dfig$se_fit
dfig$upr <- dfig$prediction + 1.96 * dfig$se_fit
dfig_m3 <- dfig
f_m3 <-
ggplot(dfig, aes(x = eco_soc, y = prediction, ymin = lwr, ymax = upr)) +
geom_errorbar(width = 0) +
geom_point(size = 2) +
scale_y_continuous(breaks = seq(2, 8, 2)) +
ylab('Government approval') + xlab('Sociotropic economic evaluation') +
scale_y_continuous(breaks = seq(.1,.9,.2)) +
ylab('Probability to vote yes') + xlab('Sociotropic economic evaluation') +
theme_bw()
pdf('figures/f_m3.pdf', width = 3, height = 3)
......@@ -76,7 +127,6 @@ f_m3 + theme(text = element_text(family ='CMU Serif', size = 12))
dev.off()
# Model 4 in table in manuscript
dfig <- data.frame(eco_soc = 1:5, know = 2, female = TRUE,
age = median(df$age, na.rm = T),
uni = T, approval = 5)
......@@ -86,6 +136,8 @@ dfig$se_fit <- predict(m4, newdata = dfig, type = 'response', se.fit = T)$se.fit
dfig$lwr <- dfig$prediction - 1.96 * dfig$se_fit
dfig$upr <- dfig$prediction + 1.96 * dfig$se_fit
dfig_m4 <- dfig
f_m4 <-
ggplot(dfig, aes(x = eco_soc, y = prediction, ymin = lwr, ymax = upr)) +
geom_errorbar(width = 0) +
......@@ -97,6 +149,91 @@ pdf('figures/f_m4.pdf', width = 3, height = 3)
f_m4 + theme(text = element_text(family ='CMU Serif', size = 12))
dev.off()
# Model 5 in table in manuscript
dfig <- data.frame(eco_soc = 1:5, know = 2, female = TRUE,
age = median(df$age, na.rm = T),
uni = T, approval_renzi = 5)
dfig$prediction <- predict(m5, newdata = dfig, type = 'response')
dfig$se_fit <- predict(m5, newdata = dfig, type = 'response', se.fit = T)$se.fit
dfig$lwr <- dfig$prediction - 1.96 * dfig$se_fit
dfig$upr <- dfig$prediction + 1.96 * dfig$se_fit
dfig_m5 <- dfig
f_m5 <-
ggplot(dfig, aes(x = eco_soc, y = prediction, ymin = lwr, ymax = upr)) +
geom_errorbar(width = 0) +
geom_point(size = 2) +
ylab('Probability to vote yes') + xlab('Sociotropic economic evaluation') +
theme_bw()
pdf('figures/f_m5.pdf', width = 3, height = 3)
f_m5 + theme(text = element_text(family ='CMU Serif', size = 12))
dev.off()
dfig_m1_m4 <- bind_rows(dfig_m1, dfig_m4) %>%
mutate(model = ifelse(is.na(approval), '(1)', '(2)'))
f_m1_m4 <-
ggplot(dfig_m1_m4, aes(x = eco_soc, y = prediction, ymin = lwr, ymax = upr)) +
geom_errorbar(width = 0) +
geom_point(size = 2) + facet_wrap(~model) +
scale_y_continuous(breaks = seq(.1, .9, .2)) +
ylab('Probability to vote yes') + xlab('Sociotropic economic evaluation') +
theme_bw() + theme(strip.background = element_rect(fill = 'white'))
pdf('figures/f_m1_m4.pdf', width = 6, height = 3)
f_m1_m4 + theme(text = element_text(family ='CMU Serif', size = 12))
dev.off()
# Figure for PID groups
dfig <- data.frame(eco_soc = 1:5, know = 2, female = TRUE,
age = median(df$age, na.rm = T),
uni = T)
dfig$prediction <- predict(m1_pidg, newdata = dfig, type = 'response')
dfig$se_fit <- predict(m1, newdata = dfig, type = 'response', se.fit = T)$se.fit
dfig$lwr <- dfig$prediction - 1.96 * dfig$se_fit
dfig$upr <- dfig$prediction + 1.96 * dfig$se_fit
dfig_m1_pidg <- dfig %>% mutate(model = 'PID (Government)')
dfig$prediction <- predict(m1_pidn, newdata = dfig, type = 'response')
dfig$se_fit <- predict(m1, newdata = dfig, type = 'response', se.fit = T)$se.fit
dfig$lwr <- dfig$prediction - 1.96 * dfig$se_fit
dfig$upr <- dfig$prediction + 1.96 * dfig$se_fit
dfig_m1_pidn <- dfig %>% mutate(model = 'PID (None)')
dfig$prediction <- predict(m1_pido, newdata = dfig, type = 'response')
dfig$se_fit <- predict(m1, newdata = dfig, type = 'response', se.fit = T)$se.fit
dfig$lwr <- dfig$prediction - 1.96 * dfig$se_fit
dfig$upr <- dfig$prediction + 1.96 * dfig$se_fit
dfig_m1_pido <- dfig %>% mutate(model = 'PID (Opposition)')
dfig <- bind_rows(dfig_m1_pidg, dfig_m1_pidn, dfig_m1_pido)
f_models1 <-
ggplot(dfig, aes(x = eco_soc, y = prediction, ymin = lwr, ymax = upr)) +
geom_errorbar(width = 0) +
geom_point(size = 2) +
scale_y_continuous(breaks = seq(.1, .9, .2)) +
facet_wrap(~model) +
ylab('Probability to vote yes') + xlab('Sociotropic economic evaluation') +
theme_bw() + theme(strip.background = element_rect(fill = 'white'))
pdf('figures/f_models1.pdf', width = 6, height = 3)
f_models1 + theme(text = element_text(family ='CMU Serif', size = 12))
dev.off()
# appendix ---------------------------------------------------------------------
# Vote intentions --
......
......@@ -49,3 +49,21 @@ rownames(t) <- c('Mode', 'Reduced model', 'Full model')
colnames(t) <- c('False', 'Correct')
t %>% xtable %>% print(floating = F, file = 'tables/t_predicted.tex')
# Regression table
stargazer(list(m1_ne, m1), type = 'latex', float = F,
out = 'tables/t_m_reduced_full.tex',
covariate.labels =
c('Sociotropic economy', 'Referendum knowledge',
'Government approval', 'Female', 'Age',
'University education',
NA),
dep.var.caption = 'Dependent variable: Vote choice (Yes)',
dep.var.labels.include = F,
digits = 2,
align = T,
column.sep.width = '1pt',
initial.zero = F,
keep.stat = c('n', 'rsq', 'll'))
# Economic Voting in Referendums
# created: 2018-03-21
# updated: 2018-03-21
# updated: 2018-12-14
# rm(list = ls())
......@@ -13,7 +13,8 @@ library(xtable)
df <- read.csv('private/data/data.csv', stringsAsFactors = F) %>% as_tibble()
dt <-
df %>% select(vote, eco_soc, know, approval,
df %>% dplyr::select(vote, eco_soc, know, pid_gov, vote_gov, approval,
approval_renzi,
female, age, uni) %>%
gather() %>%
group_by(key) %>%
......@@ -28,10 +29,13 @@ dt <-
`female` = 'Female',
`know` = 'Referendum knowledge',
`uni` = 'University education',
`vote` = 'Vote choice')) %>%
`vote` = 'Vote choice',
`pid_gov` = 'PID (Government)',
`vote_gov` = 'Vote (Government)',
`approval_renzi` = 'Approval (PM)')) %>%
rename(Variable = key)
dt <- dt[c(7, 3, 5, 2, 1, 4, 6),]
dt <- dt[c(9, 4, 7, 10, 2, 3, 6, 1, 5, 8),]
xtab <- xtable(dt)
......
No preview for this file type
No preview for this file type
No preview for this file type
No preview for this file type
No preview for this file type
No preview for this file type
No preview for this file type
\subsection*{Pre-referendum wave}
%\subsubsection{S15 (Vote choice in the last national election)}
%
%\emph{Original}
%
%Mi può dire per quale partito ha votato alla Camera?
%
%\begin{enumerate}
% \item Rivoluzione Civile
% \item Sinistra Ecologia Libertà
% \item Partito Democratico (Pd)
% \item Movimento 5 Stelle di Grillo
% \item Scelta Civica Con Monti Per L''italia
% \item Unione Di Centro
% \item Futuro E Libertà
% \item Fare Per Fermare Il Declino
% \item Il Popolo Della Libertà (PdL)
% \item Lega Nord
% \item La Destra
% \item Fratelli D'Italia
% \item Altro partito
% \item Ho votato scheda Bianca/Nulla
% \item Preferisco non rispondere
%\end{enumerate}
%
%\emph{English}
\subsubsection*{S18 (sociotropic economic evaluation)}
\emph{Original}
......
% Table created by stargazer v.5.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Wed, Jun 20, 2018 - 11:17:52 AM
% Date and time: Fri, Dec 14, 2018 - 04:43:41 PM
% Requires LaTeX packages: dcolumn
\begin{tabular}{@{\extracolsep{0pt}}lD{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} }
\\[-1.8ex]\hline
......@@ -13,16 +13,16 @@
Sociotropic economy & 1.15^{***} & .59^{***} & & \\
& (.07) & (.13) & & \\
& & & & \\
Political Knowledge & -.44^{***} & -1.29^{***} & .03 & -1.09^{***} \\
Knowledge & -.44^{***} & -1.29^{***} & .03 & -1.09^{***} \\
& (.06) & (.19) & (.05) & (.18) \\
& & & & \\
\multilinebox{Sociotropic economy $\times$\\ Political knowledge} & & .33^{***} & & \\
\multilinebox{Sociotropic economy $\times$\\ Knowledge} & & .33^{***} & & \\
& & (.07) & & \\
& & & & \\
Sociotropic economy (after referendum) & & & 1.29^{***} & .69^{***} \\
& & & (.06) & (.10) \\
& & & & \\
\multilinebox{Sociotropic economy (after referendum) $\times$\\ Political knowledge} & & & & .39^{***} \\
\multilinebox{Sociotropic economy (after referendum) $\times$\\ Knowledge} & & & & .39^{***} \\
& & & & (.06) \\
& & & & \\
Female & .23^{**} & .27^{**} & -.02 & -.001 \\
......
% Table created by stargazer v.5.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Wed, Jun 20, 2018 - 11:17:53 AM
% Date and time: Fri, Dec 14, 2018 - 06:12:49 PM
% Requires LaTeX packages: dcolumn
\begin{tabular}{@{\extracolsep{1pt}}lD{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} }
\begin{tabular}{@{\extracolsep{1pt}}lD{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} }
\\[-1.8ex]\hline
\hline \\[-1.8ex]
& \multicolumn{4}{c}{\textit{Dependent variable:}} \\
\cline{2-5}
\\[-1.8ex] & \multicolumn{2}{c}{Vote Choice (Yes)} & \multicolumn{1}{c}{Government approval} & \multicolumn{1}{c}{Vote Choice (Yes)} \\
\\[-1.8ex] & \multicolumn{2}{c}{\textit{logistic}} & \multicolumn{1}{c}{\textit{OLS}} & \multicolumn{1}{c}{\textit{logistic}} \\
\\[-1.8ex] & \multicolumn{1}{c}{(1)} & \multicolumn{1}{c}{(2)} & \multicolumn{1}{c}{(3)} & \multicolumn{1}{c}{(4)}\\
& \multicolumn{5}{c}{\textit{Dependent variable:}} \\
\cline{2-6}
\\[-1.8ex] & \multicolumn{5}{c}{Vote Choice (Yes)} \\
\\[-1.8ex] & \multicolumn{1}{c}{(1)} & \multicolumn{1}{c}{(2)} & \multicolumn{1}{c}{(3)} & \multicolumn{1}{c}{(4)} & \multicolumn{1}{c}{(5)}\\
\hline \\[-1.8ex]
Egotropic economy & .95^{***} & .52^{***} & 1.49^{***} & .35^{***} \\
& (.07) & (.11) & (.06) & (.08) \\
& & & & \\
Political knowledge & .06 & -.82^{***} & -.06 & .13^{***} \\
& (.04) & (.19) & (.05) & (.05) \\
& & & & \\
\multilinebox{Egotropic economy $\times$\\ Political knowledge} & & .31^{***} & & \\
& & (.07) & & \\
& & & & \\
Government approval & & & & .58^{***} \\
& & & & (.02) \\
& & & & \\
Female & -.05 & -.04 & -.04 & -.06 \\
& (.09) & (.09) & (.10) & (.10) \\
& & & & \\
Age & .01^{***} & .01^{***} & .01^{***} & .01^{***} \\
& (.003) & (.003) & (.003) & (.003) \\
& & & & \\
University education & .28^{***} & .27^{***} & .49^{***} & .07 \\
& (.09) & (.09) & (.10) & (.11) \\
& & & & \\
Constant & -3.94^{***} & -2.74^{***} & -.60^{**} & -4.74^{***} \\
& (.25) & (.34) & (.23) & (.30) \\
& & & & \\
Egotropic economy & .95^{***} & .70^{***} & .91^{***} & .35^{***} & .50^{***} \\
& (.07) & (.08) & (.07) & (.08) & (.08) \\
& & & & & \\
Referendum Knowledge & .06 & -.04 & .05 & .13^{***} & .14^{***} \\
& (.04) & (.05) & (.04) & (.05) & (.05) \\
& & & & & \\
PID (Government) & -.05 & -.08 & -.06 & -.06 & -.11 \\
& (.09) & (.10) & (.09) & (.10) & (.11) \\
& & & & & \\
Vote (Government) & & 2.80^{***} & & & \\
& & (.13) & & & \\
& & & & & \\
Government approval & & & 1.20^{***} & & \\
& & & (.13) & & \\
& & & & & \\
Approval (PM) & & & & .58^{***} & \\
& & & & (.02) & \\
& & & & & \\
Female & & & & & .45^{***} \\
& & & & & (.02) \\
& & & & & \\
Age & .01^{***} & .01^{**} & .01^{***} & .01^{***} & .01^{***} \\
& (.003) & (.003) & (.003) & (.003) & (.003) \\
& & & & & \\
University education & .28^{***} & .13 & .24^{***} & .07 & .11 \\
& (.09) & (.10) & (.09) & (.11) & (.11) \\
& & & & & \\
Constant & -3.94^{***} & -3.27^{***} & -3.79^{***} & -4.74^{***} & -4.33^{***} \\
& (.25) & (.28) & (.26) & (.30) & (.30) \\
& & & & & \\
\hline \\[-1.8ex]
Observations & \multicolumn{1}{c}{2,689} & \multicolumn{1}{c}{2,689} & \multicolumn{1}{c}{2,946} & \multicolumn{1}{c}{2,658} \\
R$^{2}$ & & & \multicolumn{1}{c}{.17} & \\
Log Likelihood & \multicolumn{1}{c}{-1,647.76} & \multicolumn{1}{c}{-1,636.33} & & \multicolumn{1}{c}{-1,217.66} \\
Observations & \multicolumn{1}{c}{2,689} & \multicolumn{1}{c}{2,689} & \multicolumn{1}{c}{2,579} & \multicolumn{1}{c}{2,658} & \multicolumn{1}{c}{2,572} \\
Log Likelihood & \multicolumn{1}{c}{-1,647.76} & \multicolumn{1}{c}{-1,327.45} & \multicolumn{1}{c}{-1,541.75} & \multicolumn{1}{c}{-1,217.66} & \multicolumn{1}{c}{-1,215.44} \\
\hline
\hline \\[-1.8ex]
\textit{Note:} & \multicolumn{4}{r}{$^{*}$p$<$0.1; $^{**}$p$<$0.05; $^{***}$p$<$0.01} \\
\textit{Note:} & \multicolumn{5}{r}{$^{*}$p$<$0.1; $^{**}$p$<$0.05; $^{***}$p$<$0.01} \\
\end{tabular}
% Table created by stargazer v.5.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Wed, Jun 20, 2018 - 11:17:54 AM
% Date and time: Fri, Dec 14, 2018 - 06:21:09 PM
% Requires LaTeX packages: dcolumn
\begin{tabular}{@{\extracolsep{1pt}}lD{.}{.}{-2} D{.}{.}{-2} }
\begin{tabular}{@{\extracolsep{1pt}}lD{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} }
\\[-1.8ex]\hline
\hline \\[-1.8ex]
& \multicolumn{2}{c}{\textit{Dependent variable:}} \\
\cline{2-3}
\\[-1.8ex] & \multicolumn{2}{c}{Vote Choice (Yes)} \\
\\[-1.8ex] & \multicolumn{1}{c}{(1)} & \multicolumn{1}{c}{(2)}\\
& \multicolumn{5}{c}{\textit{Dependent variable:}} \\
\cline{2-6}
\\[-1.8ex] & \multicolumn{5}{c}{Vote Choice (Yes)} \\
\\[-1.8ex] & \multicolumn{1}{c}{(1)} & \multicolumn{1}{c}{(2)} & \multicolumn{1}{c}{(3)} & \multicolumn{1}{c}{(4)} & \multicolumn{1}{c}{(5)}\\
\hline \\[-1.8ex]
Sociotropic economy & .82^{***} & .40^{***} \\
& (.06) & (.11) \\
& & \\
Egotropic economy & .52^{***} & .36^{***} \\
& (.08) & (.12) \\
& & \\
Political knowledge & .05 & -1.05^{***} \\
& (.05) & (.22) \\
& & \\
\multilinebox{Sociotropic economy $\times$\\ Political knowledge} & & .28^{***} \\
& & (.06) \\
& & \\
\multilinebox{Egotropic economy $\times$\\ Political knowledge} & & .12^{*} \\
& & (.07) \\
& & \\
Female & -.01 & .01 \\
& (.09) & (.09) \\
& & \\
Age & .02^{***} & .02^{***} \\
& (.003) & (.003) \\
& & \\
University education & .19^{**} & .18^{**} \\
& (.09) & (.09) \\
& & \\
Constant & -5.02^{***} & -3.43^{***} \\
& (.28) & (.38) \\
& & \\
Sociotropic economy & .82^{***} & .59^{***} & .82^{***} & .04 & .22^{***} \\
& (.06) & (.07) & (.07) & (.08) & (.08) \\
& & & & & \\
Egotropic economy & .52^{***} & .41^{***} & .49^{***} & .33^{***} & .41^{***} \\
& (.08) & (.08) & (.08) & (.09) & (.09) \\
& & & & & \\
Referendum Knowledge & .05 & -.04 & .04 & .13^{**} & .13^{**} \\
& (.05) & (.05) & (.05) & (.05) & (.05) \\
& & & & & \\
PID (Government) & & 2.63^{***} & & & \\
& & (.13) & & & \\
& & & & & \\
Vote (Government) & & & 1.11^{***} & & \\
& & & (.13) & & \\
& & & & & \\
Government approval & & & & .57^{***} & \\
& & & & (.03) & \\
& & & & & \\
Approval (PM) & & & & & .43^{***} \\
& & & & & (.02) \\
& & & & & \\
Female & -.01 & -.05 & -.01 & -.05 & -.09 \\
& (.09) & (.10) & (.09) & (.11) & (.11) \\
& & & & & \\
Age & .02^{***} & .01^{***} & .01^{***} & .01^{***} & .01^{***} \\
& (.003) & (.003) & (.003) & (.003) & (.003) \\
& & & & & \\
University education & .19^{**} & .06 & .16^{*} & .07 & .09 \\
& (.09) & (.10) & (.09) & (.11) & (.11) \\
& & & & & \\
Constant & -5.02^{***} & -4.08^{***} & -4.86^{***} & -4.79^{***} & -4.59^{***} \\
& (.28) & (.30) & (.29) & (.31) & (.31) \\
& & & & & \\
\hline \\[-1.8ex]
Observations & \multicolumn{1}{c}{2,675} & \multicolumn{1}{c}{2,675} \\
Log Likelihood & \multicolumn{1}{c}{-1,543.73} & \multicolumn{1}{c}{-1,526.02} \\
Observations & \multicolumn{1}{c}{2,675} & \multicolumn{1}{c}{2,675} & \multicolumn{1}{c}{2,565} & \multicolumn{1}{c}{2,649} & \multicolumn{1}{c}{2,563} \\
Log Likelihood & \multicolumn{1}{c}{-1,543.73} & \multicolumn{1}{c}{-1,282.53} & \multicolumn{1}{c}{-1,447.39} & \multicolumn{1}{c}{-1,211.59} & \multicolumn{1}{c}{-1,206.07} \\
\hline
\hline \\[-1.8ex]
\textit{Note:} & \multicolumn{2}{r}{$^{*}$p$<$0.1; $^{**}$p$<$0.05; $^{***}$p$<$0.01} \\
\textit{Note:} & \multicolumn{5}{r}{$^{*}$p$<$0.1; $^{**}$p$<$0.05; $^{***}$p$<$0.01} \\
\end{tabular}
% Table created by stargazer v.5.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Fri, Dec 14, 2018 - 04:43:41 PM
% Requires LaTeX packages: dcolumn
\begin{tabular}{@{\extracolsep{0pt}}lD{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} }
\\[-1.8ex]\hline
\hline \\[-1.8ex]
& \multicolumn{3}{c}{\textit{Dependent variable:}} \\
\cline{2-4}
\\[-1.8ex] & \multicolumn{3}{c}{Vote choice (Yes)} \\
\\[-1.8ex] & \multicolumn{1}{c}{(1)} & \multicolumn{1}{c}{(2)} & \multicolumn{1}{c}{(3)}\\
\hline \\[-1.8ex]
Sociotropic economy & .36^{***} & .99^{***} & .73^{***} \\
& (.07) & (.06) & (.06) \\
& & & \\
Approval of PM & .44^{***} & & \\
& (.02) & & \\
& & & \\
Voted for governing party & & 1.14^{***} & \\
& & (.13) & \\
& & & \\
PID with governing party & & & 2.66^{***} \\
& & & (.13) \\
& & & \\
Knowledge & .15^{***} & .05 & -.02 \\
& (.05) & (.05) & (.05) \\
& & & \\
Female & -.07 & -.01 & -.04 \\
& (.11) & (.09) & (.10) \\
& & & \\
Age & .01^{***} & .01^{***} & .01^{**} \\
& (.003) & (.003) & (.003) \\
& & & \\
University education & .10 & .17^{*} & .07 \\
& (.11) & (.09) & (.10) \\
& & & \\
Constant & -3.84^{***} & -3.98^{***} & -3.35^{***} \\
& (.26) & (.24) & (.25) \\
& & & \\
\hline \\[-1.8ex]
Observations & \multicolumn{1}{c}{2,567} & \multicolumn{1}{c}{2,572} & \multicolumn{1}{c}{2,682} \\
Log Likelihood & \multicolumn{1}{c}{-1,221.48} & \multicolumn{1}{c}{-1,472.11} & \multicolumn{1}{c}{-1,299.71} \\
\hline
\hline \\[-1.8ex]
\textit{Note:} & \multicolumn{3}{r}{$^{*}$p$<$0.1; $^{**}$p$<$0.05; $^{***}$p$<$0.01} \\
\end{tabular}
% Table created by stargazer v.5.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu
% Date and time: Thu, Dec 13, 2018 - 10:20:12 PM
% Requires LaTeX packages: dcolumn
\begin{tabular}{@{\extracolsep{0pt}}lD{.}{.}{-2} D{.}{.}{-2} D{.}{.}{-2} }
\\[-1.8ex]\hline
\hline \\[-1.8ex]
& \multicolumn{3}{c}{Subgroups} \\
\cline{2-4}
\\[-1.8ex] & \multicolumn{1}{c}{PID (Government)} & \multicolumn{1}{c}{PID (None)} & \multicolumn{1}{c}{PID (Opposition)} \\
\\[-1.8ex] & \multicolumn{1}{c}{(1)} & \multicolumn{1}{c}{(2)} & \multicolumn{1}{c}{(3)}\\
\hline \\[-1.8ex]
Sociotropic economy & .54^{***} & .83^{***} & .72^{***} \\
& (.15) & (.13) & (.09) \\
& & & \\
Knowledge & -.05 & -.05 & -.08 \\
& (.13) & (.10) & (.07) \\
& & & \\
Female & -.47^{*} & -.21 & -.10 \\
& (.25) & (.20) & (.15) \\
& & & \\
Age & .03^{***} & .01^{*} & .002 \\