getwd()
## [1] "C:/Users/Denis/Documents/RStudio"
setwd("C:/Users/Denis/Documents/RStudio")
italy_data <- ess %>%
filter(cntry == "IT")
italy_data$weight <- italy_data$dweight * italy_data$pweight
survey_design <- svydesign(ids = ~1, data = italy_data, weights = ~weight)
italy_data <- italy_data %>%
mutate(prtvtbit = case_when(
trstprt == 1 ~ 1, # Yes to 1
trstprt == 2 ~ 0, # No to 0
TRUE ~ NA_real_ # Everything else to NA
))
italy_data <- italy_data %>% filter(!is.na(vote))
df <- italy_data
df$year <- NA
replacements <- c(1990, 1992, 1994, 1996, 1998, 2000, 2002, 2004, 2006, 2008, 2010, 2012, 2014, 2016, 2018, 2020)
for(i in 1:10){
df$year[df$essround == i] <- replacements[i]
}
vote_avg <- df %>%
group_by(year) %>%
summarize(vote_avg = mean(vote, na.rm = TRUE))
vote_plot_IT <- ggplot(vote_avg, aes(x = year, y = vote_avg)) +
geom_point(aes(size = vote_avg, color = vote_avg), alpha = 0.7) +
geom_line(aes(group = 1), color = "blue", linetype = "dashed") +
labs(
x = "Survey Year",
y = "Pr(vote)") +
theme_minimal() +
theme(legend.position = "none")
print(vote_plot_IT)
italy_data <- italy_data %>%
# Recoding for feeling about trust in political parties
mutate(trstprl = case_when(
trstprt == 1 ~ "high", # more trust to "high"
trstprt %in% 2:4 ~ "low", # Other categories to "low"
TRUE ~ NA_character_ # Everything else to NA
)) %>%
# Recoding age, adding age group, and cohort with generational cohorts
mutate(
age = ifelse(agea == 999, NA, agea),
adulthood = ifelse(age >= 18 & age <= 30, "Early", ifelse(age >= 31, "Later", NA)),
cohort = ifelse(yrbrn < 1930 | yrbrn > 2000, NA, yrbrn),
gen = case_when(
yrbrn %in% 1900:1945 ~ "Interwar",
yrbrn %in% 1946:1964 ~ "Baby Boomers",
yrbrn %in% 1965:1979 ~ "Gen X",
yrbrn %in% 1980:1996 ~ "Millennials",
yrbrn %in% 1997:2012 ~ "Gen Z",
TRUE ~ as.character(cohort)
),
gen = factor(gen, levels = c("Interwar", "Baby Boomers", "Gen X", "Millennials", "Gen Z"))
) %>%
# Recoding political ID and voter behaviour
mutate(
polID = case_when(
lrscale %in% 0:3 ~ "Left",
lrscale %in% 7:10 ~ "Right",
lrscale %in% 4:6 ~ "Moderate",
lrscale %in% c(77, 88, 99) ~ NA_character_
),
prtvtbit = case_when(
essround < 5 & trstplt == 5 ~ "high trust or more",
essround >= 5 & trstprt > 600 ~ "high trust or more",
TRUE ~ "low trust"
),
trstplt = ifelse(trstplt %in% c(77, 88, 99), NA_integer_, trstplt),
trstprt = ifelse(trstprt %in% c(5555, 7777, 8888), NA_integer_, trstprt),
prtvtbit = factor(prtvtbit, levels = c("Low trust", "high trust or more"))
)
# Setting values greater than 10 to NA
italy_data$trstplt <- ifelse(italy_data$trstplt > 10, NA, italy_data$trstplt)
italy_data$trstplt <- ifelse(italy_data$trstplt > 10, NA, italy_data$trstplt)
italy_data$trstprt <- ifelse(italy_data$trstprt > 10, NA, italy_data$trstprt)
# Creating and rescaling the trust variable
italy_data$trust <- scales::rescale(italy_data$trstplt + italy_data$trstplt + italy_data$trstprt, na.rm = TRUE, to = c(0, 100))
# Here's how you would "flip it" because populist is conceived by N + I as 'anti-trust'
# As Schafer explains there are some issues with that conceptualization, but N + I is also widely applied
italy_data$populist <- scales::rescale(italy_data$trust, na.rm = TRUE, to=c(100,0))
table(italy_data$populist)
##
## 0 3.33333333333333 6.66666666666667 10
## 14 2 4 25
## 13.3333333333333 16.6666666666667 20 23.3333333333333
## 7 13 66 30
## 26.6666666666667 30 33.3333333333333 36.6666666666667
## 38 177 85 89
## 40 43.3333333333333 46.6666666666667 50
## 407 185 147 737
## 53.3333333333333 56.6666666666667 60 63.3333333333333
## 205 174 702 217
## 66.6666666666667 70 73.3333333333333 76.6666666666667
## 178 682 263 170
## 80 83.3333333333333 86.6666666666667 90
## 746 201 160 626
## 93.3333333333333 96.6666666666667 100
## 233 124 2043
italydf <- italy_data %>%
dplyr::select(trstprt, trstprl, gen, populist, polID, prtvtbit, weight)
# Check the first few rows to verify the selection
italydf <- na.omit(italydf)
head(italydf)
## [1] trstprt trstprl gen populist polID prtvtbit weight
## <0 rows> (or 0-length row.names)
# Load italy data
italy_data <- read.fst("italy_data.fst")
# Setting values greater than 10 to NA for trust variables
italy_data$trstplt <- ifelse(italy_data$trstplt > 10, NA, italy_data$trstplt)
italy_data$trstprl <- ifelse(italy_data$trstprl > 10, NA, italy_data$trstprl)
italy_data$trstprt <- ifelse(italy_data$trstprt > 10, NA, italy_data$trstprt)
# Creating and rescaling the trust variable to create a populist scale
italy_data$trust <- scales::rescale(italy_data$trstplt + italy_data$trstprl + italy_data$trstprt, na.rm = TRUE, to = c(0, 100))
# Flip the scale as higher is now anti-trust (populist)
italy_data$populist <- scales::rescale(italy_data$trust, na.rm = TRUE, to = c(100, 0))
# Plot the distribution of the populist scale for italy
ggplot(italy_data, aes(x = populist)) +
geom_histogram(bins = 30, fill = "blue", color = "black") +
theme_minimal() +
labs(title = "Distribution of Populist Scale for italy",
x = "Populist Scale",
y = "Count")
## Warning: Removed 1493 rows containing non-finite values (`stat_bin()`).
# Creating our trust in political party variable
italy_data <- italy_data %>%
mutate(
trstprt = case_when(
essround < 5 & trstprt == 5 ~ "more trust",
essround >= 5 & trstprt > 600 ~ "more trust",
TRUE ~ "low trust"
),
# Handle NAs for trust levels
trstprt = ifelse(trstprt %in% c(77, 88, 99), NA_integer_, trstprt),
trstplt = ifelse(trstplt %in% c(5555, 7777, 8888), NA_integer_, trstplt),
# Explicitly making 'low trust' the reference category
trstprt = factor(trstprt, levels = c("low trust", "more trust"))
)
italy_data <- df %>%
mutate(
# Recoding cohort variable, setting years before 1930 and after 2012 to NA
cohort = ifelse(yrbrn < 1930 | yrbrn > 2012, NA, yrbrn),
# Recoding generational cohorts based on year of birth
gen = case_when(
yrbrn %in% 1900:1945 ~ "1",
yrbrn %in% 1946:1964 ~ "2",
yrbrn %in% 1965:1979 ~ "3",
yrbrn %in% 1980:1996 ~ "4",
yrbrn %in% 1997:2012 ~ "5",
TRUE ~ as.character(cohort) # Keeping other values as character if they do not fit the ranges
),
# Converting cohort to a factor with labels
gen = factor(gen,
levels = c("1", "2", "3", "4", "5"),
labels = c("Interwar", "Baby Boomers", "Gen X", "Millennials", "Gen Z"))
)
table(italy_data$gen)
##
## Interwar Baby Boomers Gen X Millennials Gen Z
## 1470 3179 2636 2031 763
model1 <- lm(trstprt ~ gen, data = italy_data)
modelsummary(model1, fmt = 1,
estimate = c("{estimate} ({std.error}){stars}"),
statistic = NULL,
coef_rename = c("genBaby Boomers" = "Baby Boomers", "genGen X" = "Gen X", "genMillennials" = "Millennials", "genGen Z" = "Gen Z"),
title = 'Table 2. Regression model for authoritarian attitudes')
Table 2. Regression model for authoritarian attitudes
|
|
(1)
|
|
(Intercept)
|
4.2 (0.4)***
|
|
Baby Boomers
|
−0.6 (0.4)
|
|
Gen X
|
0.0 (0.4)
|
|
Millennials
|
0.9 (0.5)*
|
|
Gen Z
|
3.9 (0.6)***
|
|
Num.Obs.
|
8873
|
|
R2
|
0.010
|
|
R2 Adj.
|
0.009
|
|
AIC
|
69888.5
|
|
BIC
|
69931.1
|
|
Log.Lik.
|
−34938.261
|
|
RMSE
|
12.41
|
italy <- df %>%
mutate(ID = case_when(
lrscale >= 0 & lrscale <= 4 ~ "Left",
lrscale >= 6 & lrscale <= 10 ~ "Right",
lrscale > 10 ~ NA_character_, # Set values above 10 as NA
TRUE ~ NA_character_ # Ensure value 5 and any other unexpected values are set as NA
))
table(df$ID)
## < table of extent 0 >
italy_data <- df %>%
mutate(
polID = case_when(
lrscale %in% 0:3 ~ "Left",
lrscale %in% 7:10 ~ "Right",
lrscale %in% 4:6 ~ "Moderate",
lrscale %in% c(77, 88, 99) ~ NA_character_
),
trust = case_when(
trstprt %in% c(77, 88, 99) ~ NA_real_,
TRUE ~ trstprt
)
)
italy_data$weight <- italy_data$dweight * italy_data$pweight
survey_design <- svydesign(ids = ~1, data = italy_data, weights = ~weight)
df <- italy_data
df <- df %>%
mutate(trustplt = trstplt,
trustprt = trstprt,
attitude = lrscale,
prtcl = prtclcit) %>%
mutate(across(c("trustplt", "trustprt", "attitude", "prtcl"),
~ na_if(.x, 7) %>% na_if(8) %>% na_if(9))) %>%
# Apply the reverse coding
mutate(across(c("trustplt", "trustprt", "attitude", "prtcl"), ~ 7 - .x ))
# Now you can calculate 'schwartzauth' after the NA recoding
df$auth <- scales::rescale(df$trustplt +
df$trustprt +
df$attitude +
df$prtcl, to=c(0,100), na.rm=TRUE)
df <- df %>% filter(!is.na(auth))
df <- df %>%
mutate(
polID = case_when(
lrscale %in% 0:3 ~ "Left",
lrscale %in% 7:10 ~ "Right",
lrscale %in% 4:6 ~ "Moderate",
lrscale %in% c(77, 88, 99) ~ NA_character_
),
trust = case_when(
trstprt %in% c(77, 88, 99) ~ NA_real_,
TRUE ~ trstprt
)
)
# MLR without interaction
model2_italy <- lm(auth ~ polID + trust, data = df, weights = weight)
# MLR with interaction
model3_italy <- lm(auth ~ polID + trust + polID*trust, data = df, weights = weight)
# Displaying both models using modelsummary
modelsummary(list(model2_italy, model3_italy),
fmt = 1,
estimate = c( "{estimate} ({std.error}){stars}",
"{estimate} ({std.error}){stars}"),
statistic = NULL)
|
|
(1)
|
(2)
|
|
(Intercept)
|
87.1 (0.6)***
|
86.5 (0.7)***
|
|
polIDModerate
|
−2.8 (0.6)***
|
−2.0 (0.9)*
|
|
polIDRight
|
−0.5 (1.0)
|
0.8 (1.3)
|
|
trust
|
−0.1 (0.1)
|
0.1 (0.2)
|
|
polIDModerate × trust
|
|
−0.3 (0.3)
|
|
polIDRight × trust
|
|
−0.9 (0.6)
|
|
Num.Obs.
|
1221
|
1221
|
|
R2
|
0.022
|
0.024
|
|
R2 Adj.
|
0.020
|
0.020
|
|
AIC
|
8905.0
|
8906.1
|
|
BIC
|
8930.5
|
8941.9
|
|
Log.Lik.
|
−4447.503
|
−4446.068
|
|
RMSE
|
9.23
|
9.22
|
interaction_plot_italy <- effect("polID*trust", model3_italy, na.rm=TRUE)
plot(interaction_plot_italy,
main="Interaction effect",
xlab="Trust in Politicians",
ylab="Authoritarian attitudes scale")
