Sexism & Election Cross-Country Study
Alex Flores
2026-09-09
Prep
Data & libraries
library(tidyverse)
library(psych)
library(ggplot2)
library(lmerTest)
library(lme4)
library(knitr)
library(interactions)
library(patchwork)
library(sjPlot)
library(cowplot)
library(hrbrthemes)
library(ordinal)
library(marginaleffects)
d1 <- read.csv("Sexism_PostElection_Study1comb.csv", header = T)
d2 <- read.csv("Sexism_PostElection_Study2comb.csv", header = T)
d3 <- read.csv("Sexism_PostElection_Study3comb.csv", header = T)Study 1
VotedFor: If you could have voted in the 2024 U.S. Presidential Election/were an American citizen this week, who would you have voted for?
Var Construction
d1$ASI_3.r <- 5 - d1$ASI_3
d1$ASI_6.r <- 5 - d1$ASI_6
d1$ASI_7.r <- 5 - d1$ASI_7
d1$ASI_13.r <- 5 - d1$ASI_13
d1$ASI_18.r <- 5 - d1$ASI_18
d1$ASI_21.r <- 5 - d1$ASI_21
d1$ASI <- rowMeans(d1[,c("ASI_1", "ASI_2", "ASI_3.r", "ASI_4", "ASI_5", "ASI_6.r", "ASI_7.r", "ASI_8", "ASI_9", "ASI_10", "ASI_11", "ASI_12", "ASI_13.r", "ASI_14", "ASI_15", "ASI_16", "ASI_17", "ASI_18.r", "ASI_19", "ASI_20", "ASI_21.r", "ASI_22")], na.rm = T) # ambivalent sexism inventory
d1$BS <- rowMeans(d1[,c("ASI_1", "ASI_3.r", "ASI_6.r", "ASI_8", "ASI_9", "ASI_12", "ASI_13.r", "ASI_17", "ASI_19", "ASI_20", "ASI_22")], na.rm = T) # benevolent sexism
d1$HS <- rowMeans(d1[,c("ASI_2", "ASI_4", "ASI_5", "ASI_7.r", "ASI_10", "ASI_11", "ASI_14", "ASI_15", "ASI_16", "ASI_18.r", "ASI_21.r")], na.rm = T) # hostile sexism
d1$SRQ_1.r <- 100 - d1$SRQ_1
d1$SRQ_2.r <- 100 - d1$SRQ_2
d1$SRQ_3.r <- 100 - d1$SRQ_3
d1$SRQ_4.r <- 100 - d1$SRQ_4
d1$SRQ_5.r <- 100 - d1$SRQ_5
d1$SRQ <- rowMeans(d1[,c("SRQ_1.r", "SRQ_2.r", "SRQ_3.r", "SRQ_4.r", "SRQ_5.r", "SRQ_6", "SRQ_7", "SRQ_8", "SRQ_9", "SRQ_10", "SRQ_11", "SRQ_12", "SRQ_13")], na.rm = T) # Social roles questionnaire
d1$SDO7_3.r <- 8 - d1$SDO7_3
d1$SDO7_4.r <- 8 - d1$SDO7_4
d1$SDO7_7.r <- 8 - d1$SDO7_7
d1$SDO7_8.r <- 8 - d1$SDO7_8
d1$SDO <- rowMeans(d1[,c("SDO7_1", "SDO7_2", "SDO7_3.r", "SDO7_4.r", "SDO7_5", "SDO7_6", "SDO7_7.r", "SDO7_8.r")], na.rm = T) # social dominance orientation
d1$VotedHarris_num <- ifelse(d1$VotedFor == 1, 1, 0)
d1$VotedHarris <- as.factor(d1$VotedHarris_num)
psych::alpha(d1[,c("Mood", "MentalHealth", "StressAnxiety", "Safety")])##
## Reliability analysis
## Call: psych::alpha(x = d1[, c("Mood", "MentalHealth", "StressAnxiety",
## "Safety")])
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.92 0.92 0.9 0.74 11 0.0084 4.4 1.8 0.74
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.9 0.92 0.93
## Duhachek 0.9 0.92 0.94
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## Mood 0.91 0.91 0.87 0.76 9.7 0.011 0.00251 0.78
## MentalHealth 0.87 0.87 0.82 0.69 6.8 0.014 0.00022 0.70
## StressAnxiety 0.90 0.90 0.86 0.75 9.1 0.011 0.00231 0.78
## Safety 0.90 0.90 0.87 0.75 9.2 0.011 0.00447 0.78
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## Mood 244 0.87 0.88 0.82 0.78 4.9 1.8
## MentalHealth 244 0.94 0.94 0.92 0.88 4.1 2.0
## StressAnxiety 244 0.90 0.89 0.84 0.80 4.1 2.1
## Safety 244 0.89 0.89 0.83 0.80 4.3 2.0
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## Mood 0.07 0.06 0.09 0.16 0.23 0.15 0.25 0
## MentalHealth 0.16 0.11 0.09 0.16 0.21 0.12 0.16 0
## StressAnxiety 0.18 0.11 0.11 0.12 0.16 0.13 0.19 0
## Safety 0.14 0.08 0.12 0.16 0.17 0.11 0.21 0Setting Classes & centering
# classes
d1$Demo_Age <- as.numeric(d1$Demo_Age) #Participant age
d1$Demo_Gender_num <- as.factor(d1$Demo_Gender) #0 = male, 1 = female
d1$Demo_Gender <- recode_factor(d1$Demo_Gender_num,
`0` = "Man",
`1` = "Woman") #0 = male, 1 = female
d1$Demo_Ethnicity <- as.factor(d1$Demo_Ethnicity) #Participant ethnicity (0 = non-Hispanic, 1 = Hispanic)
d1$Demo_Race <- as.factor(d1$Demo_Race) #Participant race
d1$Demo_Education <- as.numeric(d1$Demo_Education, na.rm = TRUE)
d1$BS <- as.numeric(d1$BS) #Benevolent sexism
d1$HS <- as.numeric(d1$HS) #Hostile sexism
d1$SRQ <- as.numeric(d1$SRQ) #Traditional gender role beliefs (higher = more traditional)
d1$SDO <- as.numeric(d1$SDO)
d1$CanadianCitizen <- as.factor(d1$CanadianCitizen)
d1$OftenThought <- as.numeric(d1$OftenThought)
d1$InfluenceCanada <- as.numeric(d1$InfluenceCanada)
d1$VotedFor <- as.factor(d1$VotedFor)
# centering
d1$age.c <- d1$Demo_Age - mean(d1$Demo_Age, na.rm = T)
d1$edu.c <- d1$Demo_Education - mean(d1$Demo_Education, na.rm = T)
d1$BS.c <- d1$BS - mean(d1$BS, na.rm = T)
d1$HS.c <- d1$HS - mean(d1$HS, na.rm = T)
d1$SRQ.c <- d1$SRQ - mean(d1$SRQ, na.rm = T)
d1$SDO.c <- d1$SDO - mean(d1$SDO, na.rm = T)
d1$Impact.c <- d1$Impact_index - mean(d1$Impact_index, na.rm = T)
d1$InfluenceCanada <- d1$InfluenceCanada - mean(d1$InfluenceCanada, na.rm = T)Assessing between country metrics, normality, etc.
ASI & factors
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 1.62 0.86 1.5 1.59 1.01 0 3.68 3.68 0.26 -0.9 0.08
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 1.83 0.88 1.91 1.84 1.08 0.05 3.68 3.64 -0.12 -0.94 0.08d1 %>%
ggplot( aes(x=ASI, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .25) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Ambivalent Sexism Inventory") +
facet_grid(Country~.)
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 1.94 0.93 1.91 1.9 0.94 0 4.45 4.45 0.31 -0.32 0.08
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 2.18 1.03 2.18 2.2 1.21 0 4.09 4.09 -0.17 -0.84 0.09d1 %>%
ggplot( aes(x=BS, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .25) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Benevolent Sexism")+
facet_grid(Country~.)
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 1.3 1.09 1.09 1.17 1.08 0 4.36 4.36 0.85 -0.19 0.1
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 1.48 1.13 1.27 1.42 1.35 0 4.73 4.73 0.46 -0.77 0.1d1 %>%
ggplot( aes(x=HS, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .25) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Hostile Sexism")+
facet_grid(Country~.)
SRQ
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 26.15 15.81 23.08 25.28 17.11 0 64.62 64.62 0.43 -0.67 1.44
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 31.34 17.47 30 31.22 21.67 0 72.31 72.31 0.11 -1.01 1.58d1 %>%
ggplot( aes(x=SRQ, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 5) +
# coord_cartesian(xlim = c(0,100)) +
scale_x_continuous(breaks = seq(0,70,5)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Social Roles Questionnaire")+
facet_grid(Country~.)
SDO
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 2.34 1.17 2 2.2 1.11 1 5.75 4.75 1 0.43 0.11
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 2.43 1.26 2.12 2.3 1.48 1 6.62 5.62 0.79 -0.02 0.11d1 %>%
ggplot( aes(x=SDO, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .5) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Social Dominance Orientation")+
facet_grid(Country~.)
Personal Impact Measures
Mood = “How much has the U.S. election result impacted your overall mood this week?”
OftenThought = “How often have you thought about the U.S. election results in the past few days?”
StressAnxiety = “How much stress or anxiety do you feel regarding the outcome of the U.S. election?”
Safety = “Do you feel the U.S. election results have affected your sense of safety or well-being?”
MentalHealth = “How have the election results affected your overall mental health this week?”
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 4.64 1.71 5 4.76 1.48 1 7 6 -0.53 -0.46 0.16
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 5.11 1.87 5 5.33 2.97 1 7 6 -0.68 -0.64 0.17d1 %>%
ggplot( aes(x=Mood, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 1) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Mood")+
facet_grid(Country~.)
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 3.99 2.06 4 3.99 2.97 1 7 6 -0.11 -1.31 0.19
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 4.26 2.23 4 4.32 2.97 1 7 6 -0.17 -1.44 0.2d1 %>%
ggplot( aes(x=StressAnxiety, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 1) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Stress & Anxiety")+
facet_grid(Country~.)
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 4.16 1.89 4 4.2 1.48 1 7 6 -0.23 -0.98 0.17
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 4.5 2.16 5 4.63 2.97 1 7 6 -0.27 -1.32 0.19d1 %>%
ggplot( aes(x=Safety, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 1) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Stress & Anxiety")+
facet_grid(Country~.)
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 3.9 1.93 4 3.88 1.48 1 7 6 -0.15 -1.1 0.18
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 4.39 2.07 5 4.48 2.97 1 7 6 -0.3 -1.25 0.19d1 %>%
ggplot( aes(x=MentalHealth, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 1) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Mental Health")+
facet_grid(Country~.)
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 5.48 1.4 6 5.66 1.48 2 7 5 -0.83 -0.04 0.13
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 5.6 1.61 6 5.83 1.48 1 7 6 -0.93 -0.33 0.15d1 %>%
ggplot( aes(x=OftenThought, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 1) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "How often have you thought about the U.S. election this week?")+
facet_grid(Country~.)
##
## Descriptive statistics by group
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 121 3.9 1.93 4 3.88 1.48 1 7 6 -0.15 -1.1 0.18
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 123 4.39 2.07 5 4.48 2.97 1 7 6 -0.3 -1.25 0.19d1 %>%
ggplot( aes(x=MentalHealth, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 1) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Mental Health")+
facet_grid(Country~.)
Study 2
VotedFor: If you could have voted in the 2024 U.S. Presidential Election/were an American citizen this week, who would you have voted for?
Var Construction
d2$ASI_3.r <- 5 - d2$ASI_3
d2$ASI_6.r <- 5 - d2$ASI_6
d2$ASI_7.r <- 5 - d2$ASI_7
d2$ASI_13.r <- 5 - d2$ASI_13
d2$ASI_18.r <- 5 - d2$ASI_18
d2$ASI_21.r <- 5 - d2$ASI_21
d2$ASI <- rowMeans(d2[,c("ASI_1", "ASI_2", "ASI_3.r", "ASI_4", "ASI_5", "ASI_6.r", "ASI_7.r", "ASI_8", "ASI_9", "ASI_10", "ASI_11", "ASI_12", "ASI_13.r", "ASI_14", "ASI_15", "ASI_16", "ASI_17", "ASI_18.r", "ASI_19", "ASI_20", "ASI_21.r", "ASI_22")], na.rm = T) # ambivalent sexism inventory
d2$BS <- rowMeans(d2[,c("ASI_1", "ASI_3.r", "ASI_6.r", "ASI_8", "ASI_9", "ASI_12", "ASI_13.r", "ASI_17", "ASI_19", "ASI_20", "ASI_22")], na.rm = T) # benevolent sexism
d2$HS <- rowMeans(d2[,c("ASI_2", "ASI_4", "ASI_5", "ASI_7.r", "ASI_10", "ASI_11", "ASI_14", "ASI_15", "ASI_16", "ASI_18.r", "ASI_21.r")], na.rm = T) # hostile sexism
d2$SRQ_1.r <- 100 - d2$SRQ_1
d2$SRQ_2.r <- 100 - d2$SRQ_2
d2$SRQ_3.r <- 100 - d2$SRQ_3
d2$SRQ_4.r <- 100 - d2$SRQ_4
d2$SRQ_5.r <- 100 - d2$SRQ_5
d2$SRQ <- rowMeans(d2[,c("SRQ_1.r", "SRQ_2.r", "SRQ_3.r", "SRQ_4.r", "SRQ_5.r", "SRQ_6", "SRQ_7", "SRQ_8", "SRQ_9", "SRQ_10", "SRQ_11", "SRQ_12", "SRQ_13")], na.rm = T) # Social roles questionnaire
d2$SDO7_3.r <- 8 - d2$SDO7_3
d2$SDO7_4.r <- 8 - d2$SDO7_4
d2$SDO7_7.r <- 8 - d2$SDO7_7
d2$SDO7_8.r <- 8 - d2$SDO7_8
d2$SDO <- rowMeans(d2[,c("SDO7_1", "SDO7_2", "SDO7_3.r", "SDO7_4.r", "SDO7_5", "SDO7_6", "SDO7_7.r", "SDO7_8.r")], na.rm = T) # social dominance orientation
d2$VotedHarris_num <- ifelse(d2$VotedFor == 1, 1, 0)
d2$VotedHarris <- as.factor(d2$VotedHarris_num)
psych::alpha(d2[,c("Mood", "StressAnxiety", "Safety")])##
## Reliability analysis
## Call: psych::alpha(x = d2[, c("Mood", "StressAnxiety", "Safety")])
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.84 0.84 0.79 0.63 5.1 0.012 3.7 1.7 0.57
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.81 0.84 0.86
## Duhachek 0.81 0.84 0.86
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## Mood 0.86 0.86 0.76 0.76 6.3 0.011 NA 0.76
## StressAnxiety 0.72 0.72 0.56 0.56 2.5 0.023 NA 0.56
## Safety 0.73 0.73 0.57 0.57 2.7 0.022 NA 0.57
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## Mood 593 0.81 0.82 0.65 0.60 3.9 1.9
## StressAnxiety 593 0.90 0.89 0.84 0.76 3.6 2.0
## Safety 593 0.89 0.89 0.83 0.75 3.7 2.0
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## Mood 0.16 0.13 0.11 0.18 0.23 0.09 0.10 0
## StressAnxiety 0.22 0.14 0.13 0.16 0.16 0.11 0.09 0
## Safety 0.20 0.15 0.08 0.15 0.20 0.12 0.09 0d2$Impact_index <- rowMeans(d2[,c("Mood", "StressAnxiety", "Safety")], na.rm = T)
d2$ideology <- rowMeans(d2[,grep("Ideology", colnames(d2))], na.rm = T)
d2$Ideology <- 8 - d2$ideology
describe(d2[,c("Country","Ideology","Ideology_1","Ideology_2","Ideology_3","Ideology_4")])## vars n mean sd median trimmed mad min max range skew kurtosis
## Country* 1 593 2.50 1.12 3.00 2.50 1.48 1 4 3 0.00 -1.36
## Ideology 2 593 3.36 1.46 3.25 3.28 1.11 1 7 6 0.44 -0.16
## Ideology_1 3 593 4.51 1.64 5.00 4.57 1.48 1 7 6 -0.33 -0.62
## Ideology_2 4 593 4.89 1.60 5.00 5.02 1.48 1 7 6 -0.58 -0.40
## Ideology_3 5 593 4.45 1.58 4.00 4.49 1.48 1 7 6 -0.31 -0.45
## Ideology_4 6 593 4.74 1.50 5.00 4.83 1.48 1 7 6 -0.48 -0.16
## se
## Country* 0.05
## Ideology 0.06
## Ideology_1 0.07
## Ideology_2 0.07
## Ideology_3 0.06
## Ideology_4 0.06Setting Classes
# classes
d2$Demo_Age <- as.numeric(d2$Demo_Age) #Participant age
d2$Demo_Gender_num <- as.factor(d2$Demo_Gender) #0 = male, 1 = female
d2$Demo_Gender <- recode_factor(d2$Demo_Gender_num,
`0` = "Man",
`1` = "Woman") #0 = male, 1 = female
d2$Demo_Ethnicity <- as.factor(d2$Demo_Ethnicity) #Participant ethnicity (0 = non-Hispanic, 1 = Hispanic)
d2$Demo_Race <- as.factor(d2$Demo_Race) #Participant race
d2$Demo_Education <- as.numeric(d2$Demo_Education, na.rm = TRUE)
d2$BS <- as.numeric(d2$BS) #Benevolent sexism
d2$HS <- as.numeric(d2$HS) #Hostile sexism
d2$SRQ <- as.numeric(d2$SRQ) #Traditional gender role beliefs (higher = more traditional)
d2$SDO <- as.numeric(d2$SDO)
d2$CanadianCitizen <- as.factor(d2$CanadianCitizen)
d2$OftenThought <- as.numeric(d2$OftenThought)
d2$InfluenceCanada <- as.numeric(d2$InfluenceCanada)
d2$VotedFor <- as.factor(d2$VotedFor)
d2$Country <- as.factor(d2$Country)
# centering
d2$age.c <- d2$Demo_Age - mean(d2$Demo_Age, na.rm = T)
d2$edu.c <- d2$Demo_Education - mean(d2$Demo_Education, na.rm = T)
d2$BS.c <- d2$BS - mean(d2$BS, na.rm = T)
d2$HS.c <- d2$HS - mean(d2$HS, na.rm = T)
d2$SRQ.c <- d2$SRQ - mean(d2$SRQ, na.rm = T)
d2$SDO.c <- d2$SDO - mean(d2$SDO, na.rm = T)
d2$Impact.c <- d2$Impact_index - mean(d2$Impact_index, na.rm = T)
d2$InfluenceCanada.c <- d2$InfluenceCanada - mean(d2$InfluenceCanada, na.rm = T)
d2$Ideology.c <- d2$Ideology - 4 # meaningful midpoint--midpoint centeringAssessing between country metrics, normality, etc.
ASI & factors
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 1.77 0.94 1.82 1.75 1.01 0.05 4.05 4 0.19 -0.72 0.08
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 149 1.73 0.98 1.86 1.73 1.21 0 3.73 3.73 -0.07 -1.11 0.08
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 150 2.01 0.95 2.07 2.03 1.11 0 4.09 4.09 -0.15 -0.86 0.08
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 1.89 0.93 1.86 1.89 1.01 0 4.45 4.45 0.01 -0.54 0.08d2 %>%
ggplot( aes(x=ASI, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .25) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Ambivalent Sexism Inventory") +
facet_wrap(~Country)
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 2.03 0.99 2.18 2.02 0.94 0 4.73 4.73 0.04 -0.55 0.08
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 149 1.92 1.07 2 1.91 1.21 0 4.73 4.73 0.05 -0.73 0.09
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 150 2.25 0.97 2.32 2.27 0.88 0 4.45 4.45 -0.14 -0.24 0.08
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 2.19 1.09 2.27 2.19 1.08 0 5 5 0.05 -0.26 0.09d2 %>%
ggplot( aes(x=BS, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .25) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Benevolent Sexism")+
facet_wrap(~Country)
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 1.51 1.13 1.45 1.45 1.35 0 4.36 4.36 0.39 -0.95 0.09
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 149 1.53 1.16 1.45 1.47 1.48 0 4.36 4.36 0.29 -1.03 0.09
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 150 1.76 1.17 1.64 1.71 1.35 0 4.64 4.64 0.31 -0.92 0.1
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 1.58 1.17 1.45 1.5 1.48 0 4.18 4.18 0.42 -0.87 0.1d2 %>%
ggplot( aes(x=HS, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .25) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Hostile Sexism") +
facet_wrap(~Country)
SRQ
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 27.26 16.33 26.15 26.48 17.11 0 70 70 0.39 -0.65 1.35
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 149 25.74 17.82 25.38 25.2 23.95 0 66.92 66.92 0.22 -1.14 1.46
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 150 31.13 16.05 30 31.04 19.39 0 66.92 66.92 0.09 -1.03 1.31
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 31.21 18.62 30.77 31.03 23.95 0 76.15 76.15 0.07 -1.05 1.54d2 %>%
ggplot( aes(x=SRQ, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 5) +
# coord_cartesian(xlim = c(0,100)) +
scale_x_continuous(breaks = seq(0,70,5)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Social Roles Questionnaire")+
facet_wrap(~Country)
SDO
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 2.6 1.12 2.5 2.53 1.11 1 6.25 5.25 0.59 0.06 0.09
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 149 2.51 1.25 2.38 2.41 1.48 1 7 6 0.81 0.59 0.1
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 150 2.77 1.17 2.88 2.74 1.3 1 5.75 4.75 0.16 -0.79 0.1
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 2.44 1.39 2.25 2.3 1.85 1 5.88 4.88 0.57 -0.85 0.11d2 %>%
ggplot( aes(x=SDO, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .5) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Social Dominance Orientation")+
facet_wrap(~Country)
Personal Impact Measures
Mood = “How much has the U.S. election result impacted your overall mood this week?”
StressAnxiety = “How much stress or anxiety do you feel regarding the outcome of the U.S. election?”
Safety = “Do you feel the U.S. election results have affected your sense of safety or well-being?”
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 3.91 1.78 4 3.89 1.48 1 7 6 -0.09 -0.85 0.15
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 149 3.79 1.82 4 3.77 1.48 1 7 6 -0.06 -1.12 0.15
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 150 3.53 1.82 4 3.46 1.48 1 7 6 0.1 -1.05 0.15
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 4.28 2.07 5 4.34 2.97 1 7 6 -0.29 -1.21 0.17d2 %>%
ggplot( aes(x=Mood, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 1) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Mood") +
facet_wrap(~Country)
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 3.52 1.79 4 3.46 1.48 1 7 6 0.13 -1.02 0.15
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 149 3.66 1.92 4 3.61 2.97 1 7 6 0.06 -1.25 0.16
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 150 3.43 1.85 3 3.34 2.97 1 7 6 0.21 -1.1 0.15
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 3.73 2.28 4 3.66 2.97 1 7 6 0.14 -1.51 0.19d2 %>%
ggplot( aes(x=StressAnxiety, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 1) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Stress & Anxiety")+
facet_wrap(~Country)
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 3.73 1.86 4 3.71 2.97 1 7 6 -0.04 -1.19 0.15
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 149 3.62 1.95 4 3.57 2.97 1 7 6 0.05 -1.31 0.16
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 150 3.53 1.86 4 3.46 2.97 1 7 6 0.15 -1.18 0.15
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 4.06 2.25 4 4.08 2.97 1 7 6 -0.15 -1.47 0.19d2 %>%
ggplot( aes(x=Safety, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 1) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Stress & Anxiety")+
facet_wrap(~Country)
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 4.5 1.69 5 4.56 1.48 1 7 6 -0.33 -0.8 0.14
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 149 4.34 1.59 4 4.32 1.48 1 7 6 -0.06 -0.92 0.13
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 150 4.03 1.8 4 4.03 1.48 1 7 6 -0.09 -0.96 0.15
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 4.59 1.86 5 4.72 1.48 1 7 6 -0.45 -0.81 0.15d2 %>%
ggplot( aes(x=OftenThought, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 1) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "How often have you thought about the U.S. election this week?")+
facet_wrap(~Country)
Ideology
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 3.12 1.17 3 3.08 1.48 1 7 6 0.41 0.48 0.1
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 149 3.36 1.49 3 3.29 1.48 1 7 6 0.48 -0.29 0.12
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 150 3.41 1.26 3.5 3.35 0.93 1 7 6 0.39 0.27 0.1
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 147 3.54 1.81 3.75 3.46 2.22 1 7 6 0.23 -1 0.15d2 %>%
ggplot( aes(x=Ideology, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .5) +
scale_x_continuous(breaks = seq(1,7,1)) +
coord_cartesian(xlim = c(1,7)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Ideology (Liberal to Conservative)")+
facet_wrap(~Country)
Study 3
VotedFor: If you could have voted in the 2024 U.S. Presidential Election/were an American citizen this week, who would you have voted for?
Var Construction
d3$ASI_3.r <- 5 - d3$ASI_3
d3$ASI_6.r <- 5 - d3$ASI_6
d3$ASI_7.r <- 5 - d3$ASI_7
d3$ASI_13.r <- 5 - d3$ASI_13
d3$ASI_18.r <- 5 - d3$ASI_18
d3$ASI_21.r <- 5 - d3$ASI_21
d3$ASI <- rowMeans(d3[,c("ASI_1", "ASI_2", "ASI_3.r", "ASI_4", "ASI_5", "ASI_6.r", "ASI_7.r", "ASI_8", "ASI_9", "ASI_10", "ASI_11", "ASI_12", "ASI_13.r", "ASI_14", "ASI_15", "ASI_16", "ASI_17", "ASI_18.r", "ASI_19", "ASI_20", "ASI_21.r", "ASI_22")], na.rm = T) # ambivalent sexism inventory
d3$BS <- rowMeans(d3[,c("ASI_1", "ASI_3.r", "ASI_6.r", "ASI_8", "ASI_9", "ASI_12", "ASI_13.r", "ASI_17", "ASI_19", "ASI_20", "ASI_22")], na.rm = T) # benevolent sexism
d3$HS <- rowMeans(d3[,c("ASI_2", "ASI_4", "ASI_5", "ASI_7.r", "ASI_10", "ASI_11", "ASI_14", "ASI_15", "ASI_16", "ASI_18.r", "ASI_21.r")], na.rm = T) # hostile sexism
d3$SRQ_1.r <- 100 - d3$SRQ_1
d3$SRQ_2.r <- 100 - d3$SRQ_2
d3$SRQ_3.r <- 100 - d3$SRQ_3
d3$SRQ_4.r <- 100 - d3$SRQ_4
d3$SRQ_5.r <- 100 - d3$SRQ_5
d3$SRQ <- rowMeans(d3[,c("SRQ_1.r", "SRQ_2.r", "SRQ_3.r", "SRQ_4.r", "SRQ_5.r", "SRQ_6", "SRQ_7", "SRQ_8", "SRQ_9", "SRQ_10", "SRQ_11", "SRQ_12", "SRQ_13")], na.rm = T) # Social roles questionnaire
d3$SDO7_3.r <- 8 - d3$SDO7_3
d3$SDO7_4.r <- 8 - d3$SDO7_4
d3$SDO7_7.r <- 8 - d3$SDO7_7
d3$SDO7_8.r <- 8 - d3$SDO7_8
d3$SDO <- rowMeans(d3[,c("SDO7_1", "SDO7_2", "SDO7_3.r", "SDO7_4.r", "SDO7_5", "SDO7_6", "SDO7_7.r", "SDO7_8.r")], na.rm = T) # social dominance orientation
d3$VotedHarris_num <- ifelse(d3$VotedFor == 1, 1, 0)
d3$VotedHarris <- as.factor(d3$VotedHarris_num)
d3$ideology <- rowMeans(d3[,grep("Ideology", colnames(d3))], na.rm = T)
d3$Ideology <- 8 - d3$ideology
describe(d3[,c("Country","Ideology","Ideology_1","Ideology_2","Ideology_3","Ideology_4")])## vars n mean sd median trimmed mad min max range skew kurtosis
## Country* 1 995 2.50 1.12 3.00 2.50 1.48 1 4 3 0.00 -1.36
## Ideology 2 995 3.49 1.55 3.25 3.43 1.85 1 7 6 0.32 -0.59
## Ideology_1 3 995 4.36 1.71 4.00 4.39 1.48 1 7 6 -0.23 -0.88
## Ideology_2 4 995 4.79 1.72 5.00 4.92 1.48 1 7 6 -0.54 -0.63
## Ideology_3 5 995 4.32 1.66 4.00 4.35 1.48 1 7 6 -0.19 -0.77
## Ideology_4 6 995 4.59 1.59 5.00 4.66 1.48 1 7 6 -0.38 -0.63
## se
## Country* 0.04
## Ideology 0.05
## Ideology_1 0.05
## Ideology_2 0.05
## Ideology_3 0.05
## Ideology_4 0.05psych::alpha(d3[,c("WomanTraits_1", "WomanTraits_2", "WomanTraits_3", "WomanTraits_4", "WomanTraits_5")])##
## Reliability analysis
## Call: psych::alpha(x = d3[, c("WomanTraits_1", "WomanTraits_2", "WomanTraits_3",
## "WomanTraits_4", "WomanTraits_5")])
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.77 0.78 0.75 0.41 3.5 0.011 6.2 0.67 0.4
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.75 0.77 0.8
## Duhachek 0.75 0.77 0.8
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## WomanTraits_1 0.78 0.78 0.74 0.48 3.6 0.012 0.0074 0.46
## WomanTraits_2 0.75 0.75 0.71 0.43 3.0 0.012 0.0195 0.42
## WomanTraits_3 0.70 0.71 0.65 0.38 2.4 0.015 0.0056 0.36
## WomanTraits_4 0.71 0.72 0.68 0.39 2.6 0.014 0.0122 0.36
## WomanTraits_5 0.71 0.71 0.66 0.38 2.5 0.014 0.0099 0.37
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## WomanTraits_1 994 0.56 0.62 0.45 0.40 6.7 0.67
## WomanTraits_2 994 0.73 0.70 0.57 0.51 6.0 1.08
## WomanTraits_3 994 0.79 0.78 0.73 0.63 6.1 0.98
## WomanTraits_4 994 0.78 0.76 0.68 0.60 5.9 1.03
## WomanTraits_5 994 0.76 0.78 0.71 0.63 6.4 0.80
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## WomanTraits_1 0 0.00 0.00 0.01 0.04 0.19 0.75 0
## WomanTraits_2 0 0.00 0.02 0.07 0.16 0.35 0.40 0
## WomanTraits_3 0 0.01 0.01 0.04 0.18 0.37 0.39 0
## WomanTraits_4 0 0.00 0.01 0.08 0.21 0.34 0.36 0
## WomanTraits_5 0 0.00 0.00 0.02 0.11 0.34 0.53 0psych::alpha(d3[,c("WomanTraits_6", "WomanTraits_7", "WomanTraits_8", "WomanTraits_9", "WomanTraits_10")])##
## Reliability analysis
## Call: psych::alpha(x = d3[, c("WomanTraits_6", "WomanTraits_7", "WomanTraits_8",
## "WomanTraits_9", "WomanTraits_10")])
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.91 0.91 0.9 0.67 10 0.0045 5.4 1.1 0.67
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.9 0.91 0.92
## Duhachek 0.9 0.91 0.92
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## WomanTraits_6 0.89 0.89 0.87 0.67 8.1 0.0058 0.0086 0.68
## WomanTraits_7 0.89 0.89 0.87 0.67 8.2 0.0057 0.0046 0.67
## WomanTraits_8 0.91 0.91 0.88 0.71 9.7 0.0049 0.0026 0.71
## WomanTraits_9 0.88 0.88 0.87 0.65 7.6 0.0061 0.0056 0.65
## WomanTraits_10 0.89 0.89 0.88 0.67 8.2 0.0057 0.0062 0.67
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## WomanTraits_6 994 0.87 0.87 0.83 0.79 5.2 1.3
## WomanTraits_7 994 0.86 0.86 0.83 0.78 5.6 1.3
## WomanTraits_8 994 0.82 0.81 0.75 0.71 5.4 1.3
## WomanTraits_9 994 0.89 0.89 0.86 0.82 5.5 1.3
## WomanTraits_10 994 0.86 0.86 0.82 0.78 5.4 1.3
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## WomanTraits_6 0.01 0.03 0.06 0.18 0.31 0.23 0.19 0
## WomanTraits_7 0.01 0.02 0.03 0.12 0.25 0.28 0.29 0
## WomanTraits_8 0.01 0.02 0.04 0.17 0.27 0.28 0.22 0
## WomanTraits_9 0.01 0.02 0.04 0.13 0.25 0.30 0.26 0
## WomanTraits_10 0.01 0.03 0.03 0.15 0.27 0.28 0.25 0Setting Classes & Centering
# classes
d3$Demo_Age <- as.numeric(d3$Demo_Age) #Participant age
d3$Demo_Gender_num <- as.factor(d3$Demo_Gender) #0 = male, 1 = female
d3$Demo_Gender <- recode_factor(d3$Demo_Gender_num,
`0` = "Man",
`1` = "Woman") #0 = male, 1 = female
d3$Demo_Ethnicity <- as.factor(d3$Demo_Ethnicity) #Participant ethnicity (0 = non-Hispanic, 1 = Hispanic)
d3$Demo_Race <- as.factor(d3$Demo_Race) #Participant race
d3$Demo_Education <- as.numeric(d3$Demo_Education, na.rm = TRUE)
d3$BS <- as.numeric(d3$BS) #Benevolent sexism
d3$HS <- as.numeric(d3$HS) #Hostile sexism
d3$SRQ <- as.numeric(d3$SRQ) #Traditional gender role beliefs (higher = more traditional)
d3$SDO <- as.numeric(d3$SDO)
d3$Citizen <- as.factor(d3$Citizen)
d3$VotedFor <- as.factor(d3$VotedFor)
d3$Country <- as.factor(d3$Country)
# centering
d3$age.c <- d3$Demo_Age - mean(d3$Demo_Age, na.rm = T)
d3$edu.c <- d3$Demo_Education - mean(d3$Demo_Education, na.rm = T)
d3$BS.c <- d3$BS - mean(d3$BS, na.rm = T)
d3$HS.c <- d3$HS - mean(d3$HS, na.rm = T)
d3$SRQ.c <- d3$SRQ - mean(d3$SRQ, na.rm = T)
d3$SDO.c <- d3$SDO - mean(d3$SDO, na.rm = T)
d3$Ideology.c <- d3$Ideology - 4 # meaningful midpoint--midpoint centering
d3$warmth.c <- d3$WomanTraits_warmth - mean(d3$WomanTraits_warmth, na.rm = T)
d3$competence.c <- d3$WomanTraits_competence - mean(d3$WomanTraits_competence, na.rm = T)Assessing between country metrics, normality, etc.
ASI & factors
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 247 1.7 0.9 1.68 1.67 1.01 0 3.95 3.95 0.27 -0.56 0.06
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 250 1.85 0.86 1.86 1.86 0.98 0 4.05 4.05 -0.01 -0.71 0.05
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 1.87 0.91 2 1.88 1.01 0 4 4 -0.15 -0.73 0.06
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 1.71 1.05 1.64 1.68 1.21 0 4.45 4.45 0.25 -0.84 0.07d3 %>%
ggplot( aes(x=ASI, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .25) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Ambivalent Sexism Inventory") +
facet_wrap(~Country)
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 247 2 0.94 2 1.99 0.94 0 4.45 4.45 0.09 -0.45 0.06
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 250 2.09 0.92 2.09 2.1 0.94 0 4.64 4.64 -0.08 -0.3 0.06
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 2.06 1 2.09 2.08 1.08 0 4.55 4.55 -0.11 -0.57 0.06
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 1.94 1.13 2 1.93 1.35 0 5 5 0.08 -0.76 0.07d3 %>%
ggplot( aes(x=BS, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .25) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Benevolent Sexism")+
facet_wrap(~Country)
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 247 1.4 1.12 1.27 1.3 1.21 0 5 5 0.71 -0.05 0.07
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 250 1.62 1.14 1.55 1.54 1.21 0 5 5 0.5 -0.29 0.07
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 1.67 1.12 1.64 1.63 1.35 0 4.64 4.64 0.27 -0.72 0.07
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 1.48 1.28 1.18 1.35 1.35 0 5 5 0.69 -0.47 0.08d3 %>%
ggplot( aes(x=HS, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .25) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Hostile Sexism") +
facet_wrap(~Country)
SRQ
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 247 27.93 16.32 26.15 27.53 20.53 0 66.15 66.15 0.21 -0.95 1.04
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 250 29.94 16.38 29.23 29.51 18.25 0 76.15 76.15 0.23 -0.6 1.04
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 29.9 17.51 29.23 29.23 19.86 0 80 80 0.34 -0.52 1.11
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 29.38 20.12 27.69 28.56 26.23 0 90 90 0.34 -0.77 1.28d3 %>%
ggplot( aes(x=SRQ, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = 5) +
coord_cartesian(xlim = c(0,75)) +
scale_x_continuous(breaks = seq(0,75,5)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Social Roles Questionnaire")+
facet_wrap(~Country)
SDO
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 247 2.43 1.17 2.25 2.33 1.3 1 6.5 5.5 0.66 -0.05 0.07
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 250 2.55 1.26 2.5 2.45 1.48 1 7 6 0.62 -0.19 0.08
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 2.74 1.37 2.62 2.66 1.85 1 7 6 0.44 -0.57 0.09
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 2.4 1.58 1.75 2.17 1.11 1 7 6 0.98 -0.05 0.1d3 %>%
ggplot( aes(x=SDO, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .5) +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Social Dominance Orientation")+
facet_wrap(~Country)
Ideology
##
## Descriptive statistics by group
## group: Aus
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 247 3.22 1.3 3 3.15 1.48 1 7 6 0.45 -0.11 0.08
## ------------------------------------------------------------
## group: Canada
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 250 3.59 1.52 3.75 3.57 1.85 1 7 6 0.16 -0.64 0.1
## ------------------------------------------------------------
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 3.68 1.51 3.5 3.64 1.85 1 7 6 0.23 -0.66 0.1
## ------------------------------------------------------------
## group: US
## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 249 3.45 1.8 3.25 3.35 1.85 1 7 6 0.36 -0.92 0.11d3 %>%
ggplot( aes(x=Ideology, fill=Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.6, position = 'identity',
binwidth = .5) +
scale_x_continuous(breaks = seq(1,7,1)) +
coord_cartesian(xlim = c(1,7)) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
labs(fill="Gender",
x = "Ideology (Liberal to Conservative)")+
facet_wrap(~Country)
Analyses: Voting for Harris
Study 1
Codes
d1$US_Can <- NA
d1$US_Can[d1$Country == "US"] <- -1/2
d1$US_Can[d1$Country == "Canada"] <- 1/2
d1$Country <- factor(d1$Country, levels = c("US","Canada"))
d1$man_woman <- NA
d1$man_woman[d1$Demo_Gender == "Woman"] <- 1/2
d1$man_woman[d1$Demo_Gender == "Man"] <- -1/2
contrasts(d1$Demo_Race) <- contr.poly(nlevels(d1$Demo_Race))Original Models
Models
Dropping interaction terms with country because of model instability
Sexist Attitudes x Gender
s1_glm1 <- glm(VotedHarris ~ (HS.c + BS.c + SRQ.c + SDO.c) * man_woman + Country
+ (edu.c + age.c + Demo_Race + Demo_Ethnicity),
data = d1,
family = binomial)
# summary(s1_glm1)
tab_model(s1_glm1,
show.stat = T,
show.se = T)| Voted Harris | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| (Intercept) | 0.58 | 0.29 | 0.21 – 1.53 | -1.11 | 0.266 |
| HS c | 0.30 | 0.09 | 0.16 – 0.51 | -4.14 | <0.001 |
| BS c | 0.92 | 0.23 | 0.56 – 1.52 | -0.31 | 0.756 |
| SRQ c | 0.99 | 0.02 | 0.95 – 1.02 | -0.78 | 0.433 |
| SDO c | 0.63 | 0.14 | 0.40 – 0.96 | -2.13 | 0.034 |
| man woman | 0.79 | 0.34 | 0.33 – 1.80 | -0.56 | 0.579 |
| Country [Canada] | 1.63 | 0.65 | 0.75 – 3.61 | 1.22 | 0.222 |
| edu c | 0.88 | 0.07 | 0.75 – 1.03 | -1.60 | 0.109 |
| age c | 1.02 | 0.02 | 0.98 – 1.05 | 0.93 | 0.353 |
| Demo Race [.L] | 0.64 | 0.78 | 0.06 – 8.81 | -0.36 | 0.716 |
| Demo Race [.Q] | 0.18 | 0.20 | 0.02 – 1.73 | -1.54 | 0.124 |
| Demo Race [.C] | 2.91 | 2.61 | 0.52 – 18.84 | 1.19 | 0.234 |
| Demo Race [^4] | 0.46 | 0.32 | 0.12 – 1.76 | -1.12 | 0.263 |
| Demo Race [^5] | 2.49 | 1.31 | 0.89 – 7.06 | 1.74 | 0.081 |
| Demo Ethnicity [1] | 5.30 | 4.26 | 1.12 – 26.69 | 2.08 | 0.038 |
| HS c × man woman | 0.12 | 0.07 | 0.03 – 0.37 | -3.51 | <0.001 |
| BS c × man woman | 1.90 | 0.97 | 0.71 – 5.25 | 1.27 | 0.206 |
| SRQ c × man woman | 0.97 | 0.03 | 0.90 – 1.04 | -0.91 | 0.364 |
| SDO c × man woman | 3.47 | 1.65 | 1.40 – 9.15 | 2.62 | 0.009 |
| Observations | 242 | ||||
| R2 Tjur | 0.446 | ||||
Reported Personal Impact x Gender
s1_glm2 <- glm(VotedHarris ~ man_woman * Impact.c
+ (age.c + edu.c + Demo_Race + Demo_Ethnicity + Country),
data = d1,
family = binomial)
# summary(s1_glm2)
tab_model(s1_glm2,
show.stat = T,
show.se = T)| Voted Harris | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| (Intercept) | 0.91 | 0.41 | 0.37 – 2.23 | -0.21 | 0.834 |
| man woman | 1.99 | 0.69 | 1.01 – 3.99 | 1.98 | 0.048 |
| Impact c | 2.21 | 0.25 | 1.79 – 2.81 | 6.89 | <0.001 |
| age c | 0.99 | 0.01 | 0.96 – 1.01 | -1.03 | 0.304 |
| edu c | 0.96 | 0.07 | 0.84 – 1.10 | -0.55 | 0.583 |
| Demo Race [.L] | 3.33 | 4.09 | 0.30 – 46.25 | 0.98 | 0.328 |
| Demo Race [.Q] | 0.84 | 0.94 | 0.08 – 8.22 | -0.16 | 0.876 |
| Demo Race [.C] | 2.72 | 2.39 | 0.49 – 17.05 | 1.13 | 0.257 |
| Demo Race [^4] | 1.33 | 0.86 | 0.37 – 4.78 | 0.44 | 0.657 |
| Demo Race [^5] | 1.45 | 0.68 | 0.58 – 3.64 | 0.80 | 0.425 |
| Demo Ethnicity [1] | 2.53 | 1.78 | 0.65 – 10.34 | 1.32 | 0.186 |
| Country [Canada] | 3.24 | 1.22 | 1.57 – 6.92 | 3.12 | 0.002 |
| man woman × Impact c | 1.15 | 0.25 | 0.75 – 1.78 | 0.65 | 0.516 |
| Observations | 242 | ||||
| R2 Tjur | 0.335 | ||||
Full Model: Impact Index & Sexist Attitudes x Gender
s1_glm3 <- glm(VotedHarris ~ man_woman * (Impact.c + HS.c + BS.c + SRQ.c + SDO.c )
+ (age.c + edu.c + Demo_Race + Country),
data = d1,
family = binomial)
summary(s1_glm3)##
## Call:
## glm(formula = VotedHarris ~ man_woman * (Impact.c + HS.c + BS.c +
## SRQ.c + SDO.c) + (age.c + edu.c + Demo_Race + Country), family = binomial,
## data = d1)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.257592 0.476828 0.540 0.58905
## man_woman -0.398945 0.503405 -0.792 0.42807
## Impact.c 0.774115 0.144871 5.343 9.12e-08 ***
## HS.c -0.842090 0.336671 -2.501 0.01238 *
## BS.c -0.316316 0.285617 -1.107 0.26809
## SRQ.c -0.008923 0.019933 -0.448 0.65439
## SDO.c -0.587368 0.252581 -2.325 0.02005 *
## age.c -0.002275 0.017471 -0.130 0.89641
## edu.c -0.135079 0.085349 -1.583 0.11350
## Demo_Race.L -0.084813 1.325243 -0.064 0.94897
## Demo_Race.Q -0.651606 1.195202 -0.545 0.58563
## Demo_Race.C 0.821435 0.992173 0.828 0.40772
## Demo_Race^4 -0.452393 0.761557 -0.594 0.55249
## Demo_Race^5 0.307842 0.574083 0.536 0.59180
## CountryCanada 0.785682 0.441183 1.781 0.07494 .
## man_woman:Impact.c -0.089828 0.282170 -0.318 0.75022
## man_woman:HS.c -1.823945 0.693774 -2.629 0.00856 **
## man_woman:BS.c 1.154907 0.564780 2.045 0.04087 *
## man_woman:SRQ.c -0.021067 0.038970 -0.541 0.58879
## man_woman:SDO.c 0.850898 0.527576 1.613 0.10678
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 319.27 on 242 degrees of freedom
## Residual deviance: 163.07 on 223 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 203.07
##
## Number of Fisher Scoring iterations: 6| Voted Harris | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| (Intercept) | 1.29 | 0.62 | 0.51 – 3.37 | 0.54 | 0.589 |
| man woman | 0.67 | 0.34 | 0.24 – 1.76 | -0.79 | 0.428 |
| Impact c | 2.17 | 0.31 | 1.66 – 2.94 | 5.34 | <0.001 |
| HS c | 0.43 | 0.15 | 0.21 – 0.80 | -2.50 | 0.012 |
| BS c | 0.73 | 0.21 | 0.41 – 1.28 | -1.11 | 0.268 |
| SRQ c | 0.99 | 0.02 | 0.95 – 1.03 | -0.45 | 0.654 |
| SDO c | 0.56 | 0.14 | 0.33 – 0.91 | -2.33 | 0.020 |
| age c | 1.00 | 0.02 | 0.96 – 1.03 | -0.13 | 0.896 |
| edu c | 0.87 | 0.07 | 0.74 – 1.03 | -1.58 | 0.113 |
| Demo Race [.L] | 0.92 | 1.22 | 0.07 – 15.15 | -0.06 | 0.949 |
| Demo Race [.Q] | 0.52 | 0.62 | 0.05 – 5.99 | -0.55 | 0.586 |
| Demo Race [.C] | 2.27 | 2.26 | 0.34 – 17.46 | 0.83 | 0.408 |
| Demo Race [^4] | 0.64 | 0.48 | 0.14 – 2.78 | -0.59 | 0.552 |
| Demo Race [^5] | 1.36 | 0.78 | 0.43 – 4.17 | 0.54 | 0.592 |
| Country [Canada] | 2.19 | 0.97 | 0.93 – 5.32 | 1.78 | 0.075 |
| man woman × Impact c | 0.91 | 0.26 | 0.52 – 1.59 | -0.32 | 0.750 |
| man woman × HS c | 0.16 | 0.11 | 0.04 – 0.58 | -2.63 | 0.009 |
| man woman × BS c | 3.17 | 1.79 | 1.08 – 9.99 | 2.04 | 0.041 |
| man woman × SRQ c | 0.98 | 0.04 | 0.91 – 1.06 | -0.54 | 0.589 |
| man woman × SDO c | 2.34 | 1.24 | 0.85 – 6.82 | 1.61 | 0.107 |
| Observations | 243 | ||||
| R2 Tjur | 0.563 | ||||
Data Visualization
d1 %>% ggplot(aes(x = Country,
y = VotedHarris_num,
fill = Demo_Gender)) +
geom_bar(stat = "summary",
fun = "mean",
position = position_dodge(.9),
alpha = .8) +
stat_summary(geom = "errorbar",
color = "black",
width = .1,
show.legend = F,
position = position_dodge(.9)) +
theme_bw() +
scale_y_continuous(breaks = seq(0,1,.2)) +
coord_cartesian(ylim = c(0,1)) +
scale_fill_manual("Gender",
values = c("#69b3a2", "#404080")) +
labs(y = "Proportion Voted for Harris",
title = "Vote Choice by Gender and Country")
d1 %>% ggplot() +
geom_jitter(aes(x = Impact_index,
y = VotedHarris,
color = Demo_Gender,
group = Demo_Gender),
alpha = .7) +
# geom_smooth(aes(x = Impact_index,
# y = VotedHarris,
# color = Demo_Gender,
# group = Demo_Gender),
# method = "loess", se = F) +
geom_smooth(aes(x = Impact_index,
y = VotedHarris,
color = Demo_Gender,
group = Demo_Gender),
method = "lm", se = F, fullrange = T) +
labs(x = "Election Impact on Wellbeing",
y = "Voted for Harris",
title = "Likelihood of Voting Harris by Election Impact, Gender, and Country") +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_y_discrete(labels = c("No","Yes")) +
theme_bw() +
scale_color_manual("Gender",
values = c("#69b3a2", "#404080")) +
facet_wrap(~Country)
d1 %>% ggplot() +
geom_jitter(aes(x = HS,
y = VotedHarris,
color = Demo_Gender,
group = Demo_Gender),
alpha = .7) +
# geom_smooth(aes(x = HS,
# y = VotedHarris,
# color = Demo_Gender,
# group = Demo_Gender),
# method = "loess", se = F) +
geom_smooth(aes(x = HS,
y = VotedHarris,
color = Demo_Gender,
group = Demo_Gender),
method = "lm", se = F, fullrange = T) +
labs(x = "Hostile Sexism",
y = "Voted for Harris",
title = "Likelihood of Voting Harris by Hostile Sexism and Gender") +
scale_x_continuous(breaks = seq(0,7,1)) +
scale_y_discrete(labels = c("No","Yes")) +
coord_cartesian(xlim = c(0,5)) +
theme_bw() +
scale_color_manual("Gender",
values = c("#69b3a2", "#404080"))
d1 %>% ggplot() +
geom_jitter(aes(x = SDO,
y = VotedHarris,
color = Demo_Gender,
group = Demo_Gender),
alpha = .7) +
# geom_smooth(aes(x = SDO,
# y = VotedHarris,
# color = Demo_Gender,
# group = Demo_Gender),
# method = "loess", se = F) +
geom_smooth(aes(x = SDO,
y = VotedHarris,
color = Demo_Gender,
group = Demo_Gender),
method = "lm", se = F, fullrange = T) +
labs(x = "Social Dominance Orientation",
y = "Voted for Harris",
title = "Likelihood of Voting Harris by SDO, Gender, and Country") +
scale_x_continuous(breaks = seq(0,7,1)) +
scale_y_discrete(labels = c("No","Yes")) +
coord_cartesian(xlim = c(1,7)) +
theme_bw() +
scale_color_manual("Gender",
values = c("#69b3a2", "#404080")) 
Study 2
Codes
d2$US_oth <- NA
d2$US_oth[d2$Country == "US"] <- 3
d2$US_oth[d2$Country == "Canada"] <- -1
d2$US_oth[d2$Country == "Aus"] <- -1
d2$US_oth[d2$Country == "UK"] <- -1
d2$Can_AusUK <- NA
d2$Can_AusUK[d2$Country == "US"] <- 0
d2$Can_AusUK[d2$Country == "Canada"] <- 2
d2$Can_AusUK[d2$Country == "Aus"] <- -1
d2$Can_AusUK[d2$Country == "UK"] <- -1
d2$Aus_UK <- NA
d2$Aus_UK[d2$Country == "US"] <- 0
d2$Aus_UK[d2$Country == "Canada"] <- 0
d2$Aus_UK[d2$Country == "Aus"] <- -1
d2$Aus_UK[d2$Country == "UK"] <- 1
d2$Country <- factor(d2$Country, levels = c("US","Canada", "Aus", "UK"))
d2$man_woman <- ifelse(d2$Demo_Gender == "Woman", 1/2, -1/2)
contrasts(d2$Demo_Race) <- contr.poly(nlevels(d2$Demo_Race))Models
Insufficient power in race, ethnicity variables; omitting for these
Sexist Attitudes x Gender x Country
s2_glm1 <- glm(VotedHarris ~ (HS.c + BS.c + SRQ.c + SDO.c) * (US_oth + Can_AusUK + Aus_UK) * man_woman
+ (age.c + edu.c),
data = d2,
family = binomial)
# summary(s2_glm1)
tab_model(s2_glm1,
show.stat = T,
show.se = T)| Voted Harris | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| (Intercept) | 1.65 | 0.21 | 1.30 – 2.13 | 3.99 | <0.001 |
| HS c | 0.50 | 0.08 | 0.36 – 0.68 | -4.41 | <0.001 |
| BS c | 1.11 | 0.18 | 0.80 – 1.54 | 0.60 | 0.545 |
| SRQ c | 0.96 | 0.01 | 0.94 – 0.98 | -3.70 | <0.001 |
| SDO c | 0.83 | 0.10 | 0.65 – 1.04 | -1.61 | 0.107 |
| US oth | 0.81 | 0.05 | 0.70 – 0.92 | -3.19 | 0.001 |
| Can AusUK | 0.84 | 0.08 | 0.69 – 1.02 | -1.73 | 0.084 |
| Aus UK | 0.71 | 0.14 | 0.48 – 1.04 | -1.74 | 0.081 |
| man woman | 1.17 | 0.29 | 0.71 – 1.91 | 0.61 | 0.540 |
| age c | 1.02 | 0.01 | 1.00 – 1.04 | 2.47 | 0.014 |
| edu c | 1.01 | 0.04 | 0.94 – 1.09 | 0.27 | 0.788 |
| HS c × US oth | 0.97 | 0.08 | 0.82 – 1.14 | -0.33 | 0.745 |
| HS c × Can AusUK | 1.18 | 0.15 | 0.91 – 1.53 | 1.28 | 0.200 |
| HS c × Aus UK | 1.30 | 0.30 | 0.82 – 2.08 | 1.11 | 0.267 |
| BS c × US oth | 0.93 | 0.08 | 0.79 – 1.10 | -0.81 | 0.418 |
| BS c × Can AusUK | 0.92 | 0.13 | 0.70 – 1.21 | -0.63 | 0.528 |
| BS c × Aus UK | 0.99 | 0.25 | 0.60 – 1.61 | -0.05 | 0.962 |
| SRQ c × US oth | 1.01 | 0.01 | 1.00 – 1.02 | 1.62 | 0.106 |
| SRQ c × Can AusUK | 1.01 | 0.01 | 0.99 – 1.02 | 0.81 | 0.418 |
| SRQ c × Aus UK | 1.00 | 0.02 | 0.97 – 1.04 | 0.24 | 0.807 |
| SDO c × US oth | 0.97 | 0.06 | 0.85 – 1.09 | -0.55 | 0.580 |
| SDO c × Can AusUK | 0.85 | 0.08 | 0.70 – 1.03 | -1.67 | 0.096 |
| SDO c × Aus UK | 1.15 | 0.21 | 0.81 – 1.65 | 0.79 | 0.428 |
| HS c × man woman | 0.96 | 0.30 | 0.52 – 1.79 | -0.12 | 0.905 |
| BS c × man woman | 1.38 | 0.45 | 0.72 – 2.65 | 0.96 | 0.336 |
| SRQ c × man woman | 1.01 | 0.02 | 0.97 – 1.05 | 0.31 | 0.757 |
| SDO c × man woman | 0.93 | 0.22 | 0.58 – 1.48 | -0.31 | 0.758 |
| US oth × man woman | 1.00 | 0.13 | 0.77 – 1.30 | 0.01 | 0.989 |
| Can AusUK × man woman | 1.04 | 0.21 | 0.70 – 1.55 | 0.19 | 0.852 |
| Aus UK × man woman | 0.91 | 0.35 | 0.42 – 1.97 | -0.24 | 0.808 |
|
(HS c × US oth) × man woman |
1.17 | 0.20 | 0.85 – 1.65 | 0.95 | 0.340 |
|
(HS c × Can AusUK) × man woman |
0.68 | 0.18 | 0.41 – 1.13 | -1.47 | 0.142 |
|
(HS c × Aus UK) × man woman |
2.00 | 0.93 | 0.80 – 5.10 | 1.48 | 0.140 |
|
(BS c × US oth) × man woman |
0.97 | 0.16 | 0.70 – 1.34 | -0.20 | 0.841 |
|
(BS c × Can AusUK) × man woman |
1.72 | 0.48 | 1.00 – 3.02 | 1.93 | 0.053 |
|
(BS c × Aus UK) × man woman |
0.32 | 0.16 | 0.12 – 0.83 | -2.30 | 0.022 |
|
(SRQ c × US oth) × man woman |
0.99 | 0.01 | 0.96 – 1.01 | -1.21 | 0.227 |
|
(SRQ c × Can AusUK) × man woman |
0.98 | 0.02 | 0.94 – 1.01 | -1.28 | 0.200 |
|
(SRQ c × Aus UK) × man woman |
0.98 | 0.03 | 0.92 – 1.05 | -0.48 | 0.632 |
|
(SDO c × US oth) × man woman |
1.14 | 0.14 | 0.90 – 1.47 | 1.09 | 0.276 |
|
(SDO c × Can AusUK) × man woman |
1.21 | 0.24 | 0.83 – 1.79 | 0.98 | 0.328 |
|
(SDO c × Aus UK) × man woman |
1.02 | 0.37 | 0.50 – 2.07 | 0.06 | 0.949 |
| Observations | 581 | ||||
| R2 Tjur | 0.357 | ||||
Personal Impact Index x Gender x Country
s2_glm2 <- glm(VotedHarris ~ (US_oth + Can_AusUK + Aus_UK) * man_woman * Impact.c
+ (age.c + edu.c),
data = d2,
family = binomial)
# summary(s2_glm2)
tab_model(s2_glm2,
show.stat = T,
show.se = T)| Voted Harris | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| (Intercept) | 1.60 | 0.18 | 1.29 – 2.00 | 4.27 | <0.001 |
| US oth | 0.74 | 0.05 | 0.65 – 0.84 | -4.57 | <0.001 |
| Can AusUK | 0.99 | 0.09 | 0.83 – 1.20 | -0.08 | 0.939 |
| Aus UK | 0.75 | 0.11 | 0.55 – 1.00 | -1.91 | 0.056 |
| man woman | 2.07 | 0.46 | 1.35 – 3.22 | 3.30 | 0.001 |
| Impact c | 2.06 | 0.15 | 1.79 – 2.39 | 9.84 | <0.001 |
| age c | 1.02 | 0.01 | 1.01 – 1.04 | 3.04 | 0.002 |
| edu c | 0.99 | 0.04 | 0.92 – 1.06 | -0.27 | 0.784 |
| US oth × man woman | 0.80 | 0.10 | 0.62 – 1.03 | -1.72 | 0.085 |
| Can AusUK × man woman | 1.00 | 0.18 | 0.70 – 1.46 | 0.01 | 0.989 |
| Aus UK × man woman | 0.71 | 0.22 | 0.38 – 1.28 | -1.14 | 0.255 |
| US oth × Impact c | 1.06 | 0.04 | 0.98 – 1.16 | 1.49 | 0.136 |
| Can AusUK × Impact c | 1.04 | 0.06 | 0.93 – 1.18 | 0.66 | 0.508 |
| Aus UK × Impact c | 1.00 | 0.10 | 0.82 – 1.22 | 0.03 | 0.978 |
| man woman × Impact c | 1.26 | 0.18 | 0.95 – 1.69 | 1.60 | 0.110 |
|
(US oth × man woman) × Impact c |
0.92 | 0.08 | 0.78 – 1.09 | -0.96 | 0.339 |
|
(Can AusUK × man woman) × Impact c |
1.11 | 0.14 | 0.88 – 1.43 | 0.88 | 0.379 |
|
(Aus UK × man woman) × Impact c |
1.03 | 0.21 | 0.69 – 1.53 | 0.14 | 0.891 |
| Observations | 581 | ||||
| R2 Tjur | 0.285 | ||||
Ideology x Gender x Country
s2_glm3 <- glm(VotedHarris ~ man_woman * Ideology.c * (US_oth + Can_AusUK + Aus_UK)
+ (age.c + edu.c),
data = d2,
family = binomial)
# summary(s2_glm3)
tab_model(s2_glm3,
show.stat = T,
show.se = T)| Voted Harris | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| (Intercept) | 0.78 | 0.10 | 0.61 – 0.99 | -1.99 | 0.047 |
| man woman | 1.84 | 0.45 | 1.14 – 2.98 | 2.48 | 0.013 |
| Ideology c | 0.35 | 0.04 | 0.28 – 0.43 | -9.93 | <0.001 |
| US oth | 0.80 | 0.06 | 0.68 – 0.92 | -2.89 | 0.004 |
| Can AusUK | 1.04 | 0.10 | 0.86 – 1.24 | 0.38 | 0.702 |
| Aus UK | 0.89 | 0.15 | 0.64 – 1.25 | -0.67 | 0.506 |
| age c | 1.03 | 0.01 | 1.02 – 1.05 | 3.80 | <0.001 |
| edu c | 0.99 | 0.03 | 0.92 – 1.06 | -0.31 | 0.759 |
| man woman × Ideology c | 0.79 | 0.16 | 0.52 – 1.18 | -1.13 | 0.258 |
| man woman × US oth | 0.89 | 0.14 | 0.66 – 1.22 | -0.73 | 0.464 |
| man woman × Can AusUK | 0.93 | 0.17 | 0.64 – 1.34 | -0.40 | 0.687 |
| man woman × Aus UK | 0.78 | 0.26 | 0.40 – 1.51 | -0.73 | 0.463 |
| Ideology c × US oth | 0.89 | 0.06 | 0.78 – 1.00 | -1.89 | 0.058 |
| Ideology c × Can AusUK | 1.03 | 0.08 | 0.89 – 1.20 | 0.44 | 0.661 |
| Ideology c × Aus UK | 1.16 | 0.18 | 0.86 – 1.59 | 0.97 | 0.333 |
|
(man woman × Ideology c) × US oth |
0.88 | 0.11 | 0.68 – 1.12 | -1.03 | 0.302 |
|
(man woman × Ideology c) × Can AusUK |
0.98 | 0.15 | 0.72 – 1.33 | -0.11 | 0.914 |
|
(man woman × Ideology c) × Aus UK |
1.02 | 0.31 | 0.56 – 1.90 | 0.06 | 0.952 |
| Observations | 581 | ||||
| R2 Tjur | 0.351 | ||||
Impact index, sexist attitudes, ideology x Gender x Country
s2_glm4 <- glm(VotedHarris ~ man_woman * (US_oth + Can_AusUK + Aus_UK) * (Impact.c + HS.c + BS.c + SRQ.c + SDO.c + Ideology.c)
+ age.c + edu.c,
data = d2,
family = binomial)
# summary(s2_glm4)
tab_model(s2_glm4,
show.stat = T,
show.se = T)| Voted Harris | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| (Intercept) | 1.54 | 0.28 | 1.09 – 2.20 | 2.41 | 0.016 |
| man woman | 1.12 | 0.40 | 0.55 – 2.26 | 0.31 | 0.757 |
| US oth | 0.72 | 0.07 | 0.58 – 0.87 | -3.36 | 0.001 |
| Can AusUK | 0.81 | 0.11 | 0.62 – 1.07 | -1.48 | 0.138 |
| Aus UK | 0.84 | 0.23 | 0.48 – 1.45 | -0.61 | 0.541 |
| Impact c | 1.99 | 0.20 | 1.64 – 2.44 | 6.76 | <0.001 |
| HS c | 0.66 | 0.13 | 0.45 – 0.97 | -2.12 | 0.034 |
| BS c | 0.93 | 0.19 | 0.62 – 1.39 | -0.37 | 0.715 |
| SRQ c | 0.96 | 0.01 | 0.94 – 0.99 | -2.97 | 0.003 |
| SDO c | 1.05 | 0.16 | 0.78 – 1.41 | 0.30 | 0.763 |
| Ideology c | 0.57 | 0.07 | 0.44 – 0.73 | -4.39 | <0.001 |
| age c | 1.03 | 0.01 | 1.01 – 1.05 | 2.73 | 0.006 |
| edu c | 0.96 | 0.05 | 0.88 – 1.05 | -0.84 | 0.404 |
| man woman × US oth | 0.93 | 0.18 | 0.63 – 1.37 | -0.35 | 0.727 |
| man woman × Can AusUK | 1.06 | 0.29 | 0.61 – 1.83 | 0.20 | 0.842 |
| man woman × Aus UK | 0.97 | 0.54 | 0.32 – 2.86 | -0.06 | 0.949 |
| man woman × Impact c | 1.06 | 0.21 | 0.71 – 1.57 | 0.27 | 0.790 |
| man woman × HS c | 1.47 | 0.58 | 0.68 – 3.24 | 0.98 | 0.326 |
| man woman × BS c | 0.90 | 0.37 | 0.40 – 2.01 | -0.25 | 0.804 |
| man woman × SRQ c | 1.02 | 0.03 | 0.97 – 1.07 | 0.70 | 0.484 |
| man woman × SDO c | 0.98 | 0.29 | 0.55 – 1.77 | -0.06 | 0.952 |
| man woman × Ideology c | 0.70 | 0.18 | 0.42 – 1.16 | -1.39 | 0.164 |
| US oth × Impact c | 1.07 | 0.07 | 0.95 – 1.21 | 1.03 | 0.303 |
| US oth × HS c | 1.01 | 0.12 | 0.79 – 1.28 | 0.06 | 0.954 |
| US oth × BS c | 0.79 | 0.10 | 0.61 – 1.00 | -1.93 | 0.053 |
| US oth × SRQ c | 1.02 | 0.01 | 1.00 – 1.04 | 2.44 | 0.015 |
| US oth × SDO c | 0.96 | 0.08 | 0.81 – 1.14 | -0.47 | 0.641 |
| US oth × Ideology c | 0.89 | 0.06 | 0.76 – 1.02 | -1.67 | 0.096 |
| Can AusUK × Impact c | 0.99 | 0.07 | 0.85 – 1.15 | -0.20 | 0.844 |
| Can AusUK × HS c | 1.17 | 0.18 | 0.87 – 1.58 | 1.02 | 0.309 |
| Can AusUK × BS c | 0.87 | 0.14 | 0.63 – 1.19 | -0.90 | 0.368 |
| Can AusUK × SRQ c | 1.02 | 0.01 | 1.00 – 1.04 | 1.57 | 0.116 |
| Can AusUK × SDO c | 0.89 | 0.11 | 0.70 – 1.13 | -0.96 | 0.337 |
| Can AusUK × Ideology c | 0.99 | 0.10 | 0.81 – 1.20 | -0.10 | 0.919 |
| Aus UK × Impact c | 1.21 | 0.18 | 0.91 – 1.63 | 1.31 | 0.189 |
| Aus UK × HS c | 1.25 | 0.34 | 0.73 – 2.17 | 0.81 | 0.416 |
| Aus UK × BS c | 0.98 | 0.28 | 0.56 – 1.72 | -0.06 | 0.950 |
| Aus UK × SRQ c | 1.00 | 0.02 | 0.96 – 1.03 | -0.25 | 0.806 |
| Aus UK × SDO c | 1.24 | 0.26 | 0.82 – 1.90 | 1.02 | 0.306 |
| Aus UK × Ideology c | 0.97 | 0.19 | 0.66 – 1.43 | -0.14 | 0.885 |
|
(man woman × US oth) × Impact c |
0.95 | 0.12 | 0.74 – 1.21 | -0.40 | 0.686 |
|
(man woman × US oth) × HS c |
1.11 | 0.27 | 0.70 – 1.81 | 0.42 | 0.676 |
|
(man woman × US oth) × BS c |
0.83 | 0.20 | 0.50 – 1.32 | -0.76 | 0.447 |
|
(man woman × US oth) × SRQ c |
1.00 | 0.02 | 0.97 – 1.03 | -0.07 | 0.942 |
|
(man woman × US oth) × SDO c |
1.24 | 0.21 | 0.89 – 1.76 | 1.25 | 0.210 |
|
(man woman × US oth) × Ideology c |
0.76 | 0.11 | 0.57 – 1.01 | -1.88 | 0.060 |
|
(man woman × Can AusUK) × Impact c |
1.16 | 0.18 | 0.86 – 1.57 | 0.97 | 0.331 |
|
(man woman × Can AusUK) × HS c |
0.60 | 0.18 | 0.33 – 1.09 | -1.69 | 0.091 |
|
(man woman × Can AusUK) × BS c |
1.78 | 0.57 | 0.96 – 3.39 | 1.79 | 0.073 |
|
(man woman × Can AusUK) × SRQ c |
0.97 | 0.02 | 0.93 – 1.01 | -1.29 | 0.197 |
|
(man woman × Can AusUK) × SDO c |
1.68 | 0.41 | 1.05 – 2.75 | 2.11 | 0.035 |
|
(man woman × Can AusUK) × Ideology c |
0.96 | 0.19 | 0.65 – 1.43 | -0.18 | 0.856 |
|
(man woman × Aus UK) × Impact c |
1.23 | 0.36 | 0.69 – 2.20 | 0.70 | 0.486 |
|
(man woman × Aus UK) × HS c |
3.04 | 1.67 | 1.04 – 9.18 | 2.02 | 0.044 |
|
(man woman × Aus UK) × BS c |
0.41 | 0.23 | 0.13 – 1.24 | -1.57 | 0.117 |
|
(man woman × Aus UK) × SRQ c |
0.96 | 0.04 | 0.89 – 1.03 | -1.16 | 0.248 |
|
(man woman × Aus UK) × SDO c |
0.86 | 0.36 | 0.36 – 1.94 | -0.37 | 0.713 |
|
(man woman × Aus UK) × Ideology c |
0.80 | 0.31 | 0.37 – 1.72 | -0.59 | 0.558 |
| Observations | 581 | ||||
| R2 Tjur | 0.541 | ||||
Data Visualization
d2 %>% ggplot(aes(x = Country,
y = VotedHarris_num,
fill = Demo_Gender)) +
geom_bar(stat = "summary",
fun = "mean",
position = position_dodge(.9),
alpha = .8) +
stat_summary(geom = "errorbar",
color = "black",
width = .1,
show.legend = F,
position = position_dodge(.9)) +
theme_bw() +
scale_y_continuous(breaks = seq(0,1,.2)) +
coord_cartesian(ylim = c(0,1)) +
scale_fill_manual("Gender",
values = c("#69b3a2", "#404080")) +
labs(y = "Proportion Voted for Harris",
title = "Vote Choice by Gender and Country")
d2 %>% ggplot(aes(x = Ideology, y = VotedHarris_num, fill = Demo_Gender, color = Demo_Gender)) +
geom_jitter(size = 1, alpha = .7) +
# geom_smooth(method = "loess", se = F, fullrange = T) +
geom_smooth(method = "lm", se = F, fullrange = T, fullrange = T) +
theme_bw() +
# facet_wrap(~Country) +
scale_fill_manual(values = c("#69b3a2", "#404080"))+
scale_color_manual(values = c("#69b3a2", "#404080")) +
labs(fill = "Gender", color = "Gender",
y = "Voted for Harris",
x = "Ideology (Liberal to Conseravtive)",
title = "Likelihood of Voting Harris by Ideology and Gender",
caption= "Interestingly, ideology has narrower range for 'yes' voters than 'no' voters") +
scale_y_continuous(breaks = c(0,1),
labels = c("No",
"Yes")) +
scale_x_continuous(breaks = seq(1,7,1))
d2 %>% ggplot() +
geom_jitter(aes(x = Impact_index,
y = VotedHarris_num),
color = "#69b3a2",
size = 1) +
# geom_smooth(aes(x = Impact_index,
# y = VotedHarris_num),
# method = "loess", se = F, color = "#404080") +
geom_smooth(aes(x = Impact_index,
y = VotedHarris_num),
method = "lm", se = F, color = "#404080", fullrange = T) +
labs(x = "Election Impact on Wellbeing",
y = "Voted for Harris",
title = "Likelihood of Voting Harris by Impact Index") +
scale_x_continuous(breaks = seq(1,7,1)) +
scale_y_continuous(labels = c("No","Yes"),
breaks = c(0,1)) +
theme_bw()
d2 %>% ggplot() +
geom_jitter(aes(x = SRQ,
y = VotedHarris,
color = Demo_Gender,
group = Demo_Gender),
size = 1, alpha = .7) +
# geom_smooth(aes(x = SRQ,
# y = VotedHarris,
# color = Demo_Gender,
# group = Demo_Gender),
# method = "loess", se = F, fullrange = T) +
geom_smooth(aes(x = SRQ,
y = VotedHarris,
color = Demo_Gender,
group = Demo_Gender),
method = "lm", se = F, fullrange = T, fullrange = T) +
labs(x = "Social Roles Questionnaire",
y = "Voted for Harris",
title = "Likelihood of Voting Harris by SDO, Gender, and Country") +
scale_x_continuous(breaks = seq(0,80,10)) +
scale_y_discrete(labels = c("No",
"Yes")) +
coord_cartesian(xlim = c(0,80)) +
theme_bw() +
scale_color_manual("Gender",
values = c("#69b3a2", "#404080")) +
facet_wrap(~Country)
Study 3
Codes
d3$US_oth <- NA
d3$US_oth[d3$Country == "US"] <- 3
d3$US_oth[d3$Country == "Canada"] <- -1
d3$US_oth[d3$Country == "Aus"] <- -1
d3$US_oth[d3$Country == "UK"] <- -1
d3$Can_AusUK <- NA
d3$Can_AusUK[d3$Country == "US"] <- 0
d3$Can_AusUK[d3$Country == "Canada"] <- 2
d3$Can_AusUK[d3$Country == "Aus"] <- -1
d3$Can_AusUK[d3$Country == "UK"] <- -1
d3$Aus_UK <- NA
d3$Aus_UK[d3$Country == "US"] <- 0
d3$Aus_UK[d3$Country == "Canada"] <- 0
d3$Aus_UK[d3$Country == "Aus"] <- -1
d3$Aus_UK[d3$Country == "UK"] <- 1
d3$Country <- factor(d3$Country, levels = c("US","Canada", "Aus", "UK"))
d3$man_woman <- ifelse(d3$Demo_Gender == "Woman", 1/2, -1/2)Models
Sexist attitudes x Country x Gender
s3_glm1 <- glm(VotedHarris ~ (HS.c + BS.c + SRQ.c + SDO.c) * (US_oth + Can_AusUK + Aus_UK) * man_woman
+ (age.c + edu.c),
data = d2,
family = binomial)
# summary(s3_glm1)
tab_model(s3_glm1,
show.stat = T,
show.se = T)| Voted Harris | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| (Intercept) | 1.65 | 0.21 | 1.30 – 2.13 | 3.99 | <0.001 |
| HS c | 0.50 | 0.08 | 0.36 – 0.68 | -4.41 | <0.001 |
| BS c | 1.11 | 0.18 | 0.80 – 1.54 | 0.60 | 0.545 |
| SRQ c | 0.96 | 0.01 | 0.94 – 0.98 | -3.70 | <0.001 |
| SDO c | 0.83 | 0.10 | 0.65 – 1.04 | -1.61 | 0.107 |
| US oth | 0.81 | 0.05 | 0.70 – 0.92 | -3.19 | 0.001 |
| Can AusUK | 0.84 | 0.08 | 0.69 – 1.02 | -1.73 | 0.084 |
| Aus UK | 0.71 | 0.14 | 0.48 – 1.04 | -1.74 | 0.081 |
| man woman | 1.17 | 0.29 | 0.71 – 1.91 | 0.61 | 0.540 |
| age c | 1.02 | 0.01 | 1.00 – 1.04 | 2.47 | 0.014 |
| edu c | 1.01 | 0.04 | 0.94 – 1.09 | 0.27 | 0.788 |
| HS c × US oth | 0.97 | 0.08 | 0.82 – 1.14 | -0.33 | 0.745 |
| HS c × Can AusUK | 1.18 | 0.15 | 0.91 – 1.53 | 1.28 | 0.200 |
| HS c × Aus UK | 1.30 | 0.30 | 0.82 – 2.08 | 1.11 | 0.267 |
| BS c × US oth | 0.93 | 0.08 | 0.79 – 1.10 | -0.81 | 0.418 |
| BS c × Can AusUK | 0.92 | 0.13 | 0.70 – 1.21 | -0.63 | 0.528 |
| BS c × Aus UK | 0.99 | 0.25 | 0.60 – 1.61 | -0.05 | 0.962 |
| SRQ c × US oth | 1.01 | 0.01 | 1.00 – 1.02 | 1.62 | 0.106 |
| SRQ c × Can AusUK | 1.01 | 0.01 | 0.99 – 1.02 | 0.81 | 0.418 |
| SRQ c × Aus UK | 1.00 | 0.02 | 0.97 – 1.04 | 0.24 | 0.807 |
| SDO c × US oth | 0.97 | 0.06 | 0.85 – 1.09 | -0.55 | 0.580 |
| SDO c × Can AusUK | 0.85 | 0.08 | 0.70 – 1.03 | -1.67 | 0.096 |
| SDO c × Aus UK | 1.15 | 0.21 | 0.81 – 1.65 | 0.79 | 0.428 |
| HS c × man woman | 0.96 | 0.30 | 0.52 – 1.79 | -0.12 | 0.905 |
| BS c × man woman | 1.38 | 0.45 | 0.72 – 2.65 | 0.96 | 0.336 |
| SRQ c × man woman | 1.01 | 0.02 | 0.97 – 1.05 | 0.31 | 0.757 |
| SDO c × man woman | 0.93 | 0.22 | 0.58 – 1.48 | -0.31 | 0.758 |
| US oth × man woman | 1.00 | 0.13 | 0.77 – 1.30 | 0.01 | 0.989 |
| Can AusUK × man woman | 1.04 | 0.21 | 0.70 – 1.55 | 0.19 | 0.852 |
| Aus UK × man woman | 0.91 | 0.35 | 0.42 – 1.97 | -0.24 | 0.808 |
|
(HS c × US oth) × man woman |
1.17 | 0.20 | 0.85 – 1.65 | 0.95 | 0.340 |
|
(HS c × Can AusUK) × man woman |
0.68 | 0.18 | 0.41 – 1.13 | -1.47 | 0.142 |
|
(HS c × Aus UK) × man woman |
2.00 | 0.93 | 0.80 – 5.10 | 1.48 | 0.140 |
|
(BS c × US oth) × man woman |
0.97 | 0.16 | 0.70 – 1.34 | -0.20 | 0.841 |
|
(BS c × Can AusUK) × man woman |
1.72 | 0.48 | 1.00 – 3.02 | 1.93 | 0.053 |
|
(BS c × Aus UK) × man woman |
0.32 | 0.16 | 0.12 – 0.83 | -2.30 | 0.022 |
|
(SRQ c × US oth) × man woman |
0.99 | 0.01 | 0.96 – 1.01 | -1.21 | 0.227 |
|
(SRQ c × Can AusUK) × man woman |
0.98 | 0.02 | 0.94 – 1.01 | -1.28 | 0.200 |
|
(SRQ c × Aus UK) × man woman |
0.98 | 0.03 | 0.92 – 1.05 | -0.48 | 0.632 |
|
(SDO c × US oth) × man woman |
1.14 | 0.14 | 0.90 – 1.47 | 1.09 | 0.276 |
|
(SDO c × Can AusUK) × man woman |
1.21 | 0.24 | 0.83 – 1.79 | 0.98 | 0.328 |
|
(SDO c × Aus UK) × man woman |
1.02 | 0.37 | 0.50 – 2.07 | 0.06 | 0.949 |
| Observations | 581 | ||||
| R2 Tjur | 0.357 | ||||
Ideology x Country x Gender
s3_glm2 <- glm(VotedHarris ~ Ideology.c * (US_oth + Can_AusUK + Aus_UK) * man_woman
+ (age.c + edu.c),
data = d2,
family = binomial)
# summary(s3_glm2)
tab_model(s3_glm2,
show.stat = T,
show.se = T)| Voted Harris | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| (Intercept) | 0.78 | 0.10 | 0.61 – 0.99 | -1.99 | 0.047 |
| Ideology c | 0.35 | 0.04 | 0.28 – 0.43 | -9.93 | <0.001 |
| US oth | 0.80 | 0.06 | 0.68 – 0.92 | -2.89 | 0.004 |
| Can AusUK | 1.04 | 0.10 | 0.86 – 1.24 | 0.38 | 0.702 |
| Aus UK | 0.89 | 0.15 | 0.64 – 1.25 | -0.67 | 0.506 |
| man woman | 1.84 | 0.45 | 1.14 – 2.98 | 2.48 | 0.013 |
| age c | 1.03 | 0.01 | 1.02 – 1.05 | 3.80 | <0.001 |
| edu c | 0.99 | 0.03 | 0.92 – 1.06 | -0.31 | 0.759 |
| Ideology c × US oth | 0.89 | 0.06 | 0.78 – 1.00 | -1.89 | 0.058 |
| Ideology c × Can AusUK | 1.03 | 0.08 | 0.89 – 1.20 | 0.44 | 0.661 |
| Ideology c × Aus UK | 1.16 | 0.18 | 0.86 – 1.59 | 0.97 | 0.333 |
| Ideology c × man woman | 0.79 | 0.16 | 0.52 – 1.18 | -1.13 | 0.258 |
| US oth × man woman | 0.89 | 0.14 | 0.66 – 1.22 | -0.73 | 0.464 |
| Can AusUK × man woman | 0.93 | 0.17 | 0.64 – 1.34 | -0.40 | 0.687 |
| Aus UK × man woman | 0.78 | 0.26 | 0.40 – 1.51 | -0.73 | 0.463 |
|
(Ideology c × US oth) × man woman |
0.88 | 0.11 | 0.68 – 1.12 | -1.03 | 0.302 |
|
(Ideology c × Can AusUK) × man woman |
0.98 | 0.15 | 0.72 – 1.33 | -0.11 | 0.914 |
|
(Ideology c × Aus UK) × man woman |
1.02 | 0.31 | 0.56 – 1.90 | 0.06 | 0.952 |
| Observations | 581 | ||||
| R2 Tjur | 0.351 | ||||
Sexist Attitudes x Gender x Country + Ideology (can’t handle fully interactive model)
s3_glm3 <- glm(VotedHarris ~ (HS.c + BS.c + SRQ.c + SDO.c + Ideology.c) * man_woman * (US_oth + Can_AusUK + Aus_UK)
+ (age.c + edu.c),
data = d2,
family = binomial)
# summary(s3_glm3)
tab_model(s3_glm3,
show.stat = T,
show.se = T)| Voted Harris | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| (Intercept) | 1.09 | 0.17 | 0.80 – 1.47 | 0.53 | 0.598 |
| HS c | 0.60 | 0.10 | 0.42 – 0.83 | -3.02 | 0.003 |
| BS c | 1.11 | 0.20 | 0.78 – 1.59 | 0.56 | 0.574 |
| SRQ c | 0.96 | 0.01 | 0.94 – 0.99 | -3.10 | 0.002 |
| SDO c | 0.99 | 0.13 | 0.77 – 1.29 | -0.04 | 0.965 |
| Ideology c | 0.47 | 0.06 | 0.37 – 0.60 | -6.09 | <0.001 |
| man woman | 1.15 | 0.36 | 0.63 – 2.11 | 0.45 | 0.652 |
| US oth | 0.76 | 0.07 | 0.63 – 0.90 | -3.11 | 0.002 |
| Can AusUK | 0.86 | 0.10 | 0.68 – 1.08 | -1.27 | 0.205 |
| Aus UK | 0.70 | 0.17 | 0.44 – 1.11 | -1.49 | 0.136 |
| age c | 1.03 | 0.01 | 1.01 – 1.05 | 3.04 | 0.002 |
| edu c | 0.98 | 0.04 | 0.91 – 1.07 | -0.39 | 0.699 |
| HS c × man woman | 0.97 | 0.33 | 0.49 – 1.91 | -0.10 | 0.923 |
| BS c × man woman | 1.11 | 0.40 | 0.55 – 2.25 | 0.28 | 0.780 |
| SRQ c × man woman | 1.02 | 0.02 | 0.97 – 1.06 | 0.65 | 0.515 |
| SDO c × man woman | 1.01 | 0.27 | 0.60 – 1.69 | 0.03 | 0.980 |
| Ideology c × man woman | 0.70 | 0.17 | 0.43 – 1.12 | -1.47 | 0.141 |
| HS c × US oth | 1.08 | 0.11 | 0.88 – 1.32 | 0.75 | 0.451 |
| HS c × Can AusUK | 1.17 | 0.16 | 0.89 – 1.53 | 1.15 | 0.250 |
| HS c × Aus UK | 1.29 | 0.31 | 0.80 – 2.10 | 1.05 | 0.295 |
| BS c × US oth | 0.90 | 0.09 | 0.74 – 1.10 | -1.02 | 0.308 |
| BS c × Can AusUK | 0.92 | 0.13 | 0.69 – 1.23 | -0.57 | 0.566 |
| BS c × Aus UK | 0.98 | 0.25 | 0.58 – 1.62 | -0.09 | 0.929 |
| SRQ c × US oth | 1.01 | 0.01 | 1.00 – 1.02 | 1.55 | 0.120 |
| SRQ c × Can AusUK | 1.01 | 0.01 | 0.99 – 1.03 | 0.98 | 0.329 |
| SRQ c × Aus UK | 1.00 | 0.02 | 0.97 – 1.03 | 0.04 | 0.968 |
| SDO c × US oth | 0.96 | 0.07 | 0.84 – 1.11 | -0.50 | 0.616 |
| SDO c × Can AusUK | 0.90 | 0.10 | 0.73 – 1.12 | -0.92 | 0.357 |
| SDO c × Aus UK | 1.22 | 0.24 | 0.84 – 1.80 | 1.04 | 0.299 |
| Ideology c × US oth | 0.85 | 0.06 | 0.74 – 0.97 | -2.28 | 0.023 |
| Ideology c × Can AusUK | 0.98 | 0.09 | 0.82 – 1.18 | -0.18 | 0.859 |
| Ideology c × Aus UK | 0.95 | 0.18 | 0.66 – 1.37 | -0.28 | 0.781 |
| man woman × US oth | 0.98 | 0.17 | 0.70 – 1.39 | -0.10 | 0.916 |
| man woman × Can AusUK | 1.01 | 0.24 | 0.63 – 1.61 | 0.04 | 0.966 |
| man woman × Aus UK | 0.85 | 0.40 | 0.33 – 2.15 | -0.35 | 0.729 |
|
(HS c × man woman) × US oth |
1.24 | 0.25 | 0.84 – 1.87 | 1.06 | 0.288 |
|
(HS c × man woman) × Can AusUK |
0.67 | 0.18 | 0.39 – 1.14 | -1.49 | 0.137 |
|
(HS c × man woman) × Aus UK |
1.89 | 0.92 | 0.73 – 5.02 | 1.31 | 0.191 |
|
(BS c × man woman) × US oth |
0.84 | 0.17 | 0.56 – 1.24 | -0.84 | 0.399 |
|
(BS c × man woman) × Can AusUK |
1.73 | 0.51 | 0.98 – 3.13 | 1.86 | 0.063 |
|
(BS c × man woman) × Aus UK |
0.35 | 0.18 | 0.12 – 0.96 | -2.02 | 0.044 |
|
(SRQ c × man woman) × US oth |
0.99 | 0.01 | 0.97 – 1.02 | -0.61 | 0.539 |
|
(SRQ c × man woman) × Can AusUK |
0.98 | 0.02 | 0.94 – 1.01 | -1.26 | 0.209 |
|
(SRQ c × man woman) × Aus UK |
0.98 | 0.03 | 0.92 – 1.05 | -0.51 | 0.611 |
|
(SDO c × man woman) × US oth |
1.15 | 0.17 | 0.87 – 1.53 | 0.97 | 0.331 |
|
(SDO c × man woman) × Can AusUK |
1.34 | 0.29 | 0.88 – 2.07 | 1.35 | 0.178 |
|
(SDO c × man woman) × Aus UK |
1.01 | 0.39 | 0.46 – 2.15 | 0.02 | 0.985 |
|
(Ideology c × man woman) × US oth |
0.81 | 0.11 | 0.61 – 1.06 | -1.49 | 0.137 |
|
(Ideology c × man woman) × Can AusUK |
0.99 | 0.18 | 0.68 – 1.43 | -0.05 | 0.957 |
|
(Ideology c × man woman) × Aus UK |
0.81 | 0.30 | 0.39 – 1.68 | -0.58 | 0.564 |
| Observations | 581 | ||||
| R2 Tjur | 0.462 | ||||
Data Visualization
d3 %>% ggplot(aes(x = Country,
y = VotedHarris_num,
fill = Demo_Gender)) +
geom_bar(stat = "summary",
fun = "mean",
position = position_dodge(.9),
alpha = .8) +
stat_summary(geom = "errorbar",
color = "black",
width = .1,
show.legend = F,
position = position_dodge(.9)) +
theme_bw() +
scale_y_continuous(breaks = seq(0,1,.2)) +
coord_cartesian(ylim = c(0,1)) +
scale_fill_manual("Gender",
values = c("#69b3a2", "#404080")) +
labs(y = "Proportion Voted for Harris",
title = "Vote Choice by Gender and Country")
d3 %>% ggplot(aes(x = HS, y = VotedHarris_num, group = Demo_Gender, color = Demo_Gender)) +
geom_jitter(size = 1, alpha = .7) +
# geom_smooth(method = "loess", se = F, fullrange = T) +
geom_smooth(method = "lm", se = F, fullrange = T, fullrange = T) +
theme_bw() +
# facet_wrap(~Country) +
scale_fill_manual(values = c("#69b3a2", "#404080"))+
scale_color_manual(values = c("#69b3a2", "#404080")) +
labs(fill = "Gender", color = "Gender",
y = "Voted for Harris",
x = "Hostile Sexism",
title = "Likelihood of Voting Harris by Hostile Sexism, Gender, and Country") +
scale_y_continuous(breaks = c(0,1),
labels = c("No",
"Yes")) +
scale_x_continuous(breaks = seq(0,7,1)) +
coord_cartesian(xlim = c(0,5))
d3 %>% ggplot(aes(x = SRQ, y = VotedHarris_num, group = Demo_Gender, color = Demo_Gender)) +
geom_jitter(size = 1, alpha = .7) +
# geom_smooth(method = "loess", se = F, fullrange = T) +
geom_smooth(method = "lm", se = F, fullrange = T, fullrange = T) +
theme_bw() +
# facet_wrap(~Country) +
scale_fill_manual(values = c("#69b3a2", "#404080"))+
scale_color_manual(values = c("#69b3a2", "#404080")) +
labs(fill = "Gender", color = "Gender",
y = "Voted for Harris",
x = "Social Roles Questionnaire",
title = "Likelihood of Voting Harris by SRQ, Gender, and Country") +
scale_y_continuous(breaks = c(0,1),
labels = c("No",
"Yes")) +
scale_x_continuous(breaks = seq(0,90,10)) +
coord_cartesian(xlim = c(0,90))
d3 %>% ggplot(aes(x = Ideology, y = VotedHarris_num, group = Demo_Gender, color = Demo_Gender)) +
geom_jitter(size = 1, alpha = .7) +
# geom_smooth(method = "loess", se = F, fullrange = T) +
geom_smooth(method = "lm", se = F, fullrange = T, fullrange = T) +
theme_bw() +
facet_wrap(~Country) +
scale_fill_manual(values = c("#69b3a2", "#404080"))+
scale_color_manual(values = c("#69b3a2", "#404080")) +
labs(fill = "Gender", color = "Gender",
y = "Voted for Harris",
x = "Ideology (Liberal to Conseravtive)",
title = "Likelihood of Voting Harris by Ideology, Gender, and Country") +
scale_y_continuous(breaks = c(0,1),
labels = c("No",
"Yes")) +
scale_x_continuous(breaks = seq(1,7,1)) +
coord_cartesian(xlim = c(1,7))
Analyses: Candidate Experiment
Only uses study 3 data.
Codes
d3$US_oth <- NA
d3$US_oth[d3$Country == "US"] <- 3
d3$US_oth[d3$Country == "Canada"] <- -1
d3$US_oth[d3$Country == "Aus"] <- -1
d3$US_oth[d3$Country == "UK"] <- -1
d3$Can_AusUK <- NA
d3$Can_AusUK[d3$Country == "US"] <- 0
d3$Can_AusUK[d3$Country == "Canada"] <- 2
d3$Can_AusUK[d3$Country == "Aus"] <- -1
d3$Can_AusUK[d3$Country == "UK"] <- -1
d3$Aus_UK <- NA
d3$Aus_UK[d3$Country == "US"] <- 0
d3$Aus_UK[d3$Country == "Canada"] <- 0
d3$Aus_UK[d3$Country == "Aus"] <- -1
d3$Aus_UK[d3$Country == "UK"] <- 1
d3$Country <- factor(d3$Country, levels = c("US","Canada", "Aus", "UK"))
contrasts(d3$Country) <- contr.helmert(nlevels(d3$Country))
d3$man_woman <- ifelse(d3$Demo_Gender == "Woman", 1/2, -1/2)
contrasts(d3$Demo_Race) <- contr.poly(nlevels(d3$Demo_Race))
d3$LikelihoodofVoting <- as.factor(d3$LikelihoodofVoting)Models
Interactive models
s3_olr1 <- clm(LikelihoodofVoting ~ man_woman * (US_oth + Can_AusUK + Aus_UK)
+ age.c + edu.c,
data = d3,
link = "logit"
)
# summary(s3_olr1)
tab_model(s3_olr1,
show.se = T,
show.stat = T)| Likelihoodof Voting | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| 1|2 | 0.23 | 0.02 | 0.19 – 0.27 | -17.87 | <0.001 |
| 2|3 | 0.58 | 0.04 | 0.51 – 0.66 | -8.11 | <0.001 |
| 3|4 | 1.75 | 0.12 | 1.54 – 2.00 | 8.26 | <0.001 |
| man woman | 1.81 | 0.21 | 1.44 – 2.29 | 5.02 | <0.001 |
| US oth | 0.99 | 0.03 | 0.93 – 1.06 | -0.18 | 0.858 |
| Can AusUK | 0.87 | 0.04 | 0.79 – 0.95 | -3.04 | 0.002 |
| Aus UK | 0.80 | 0.07 | 0.68 – 0.95 | -2.56 | 0.010 |
| age c | 1.00 | 0.00 | 1.00 – 1.00 | -1.17 | 0.242 |
| edu c | 1.09 | 0.02 | 1.05 – 1.14 | 4.06 | <0.001 |
| man woman × US oth | 0.91 | 0.06 | 0.80 – 1.04 | -1.35 | 0.177 |
| man woman × Can AusUK | 0.99 | 0.09 | 0.82 – 1.20 | -0.06 | 0.949 |
| man woman × Aus UK | 1.09 | 0.19 | 0.78 – 1.52 | 0.50 | 0.617 |
| Observations | 980 | ||||
| R2 Nagelkerke | 0.067 | ||||
s3_olr2 <- clm(LikelihoodofVoting ~ man_woman * (US_oth + Can_AusUK + Aus_UK) * (HS.c + BS.c + SDO.c + SRQ.c)
+ age.c + edu.c,
data = d3,
link = "logit"
)
# summary(s3_olr2)
tab_model(s3_olr2,
show.stat = T,
show.se=T)| Likelihoodof Voting | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| 1|2 | 0.17 | 0.02 | 0.14 – 0.21 | -18.44 | <0.001 |
| 2|3 | 0.53 | 0.04 | 0.46 – 0.62 | -8.21 | <0.001 |
| 3|4 | 2.02 | 0.16 | 1.74 – 2.36 | 9.15 | <0.001 |
| man woman | 1.21 | 0.16 | 0.94 – 1.55 | 1.44 | 0.149 |
| US oth | 0.95 | 0.04 | 0.88 – 1.02 | -1.38 | 0.167 |
| Can AusUK | 0.87 | 0.05 | 0.78 – 0.96 | -2.72 | 0.007 |
| Aus UK | 0.88 | 0.08 | 0.73 – 1.05 | -1.40 | 0.161 |
| HS c | 0.62 | 0.05 | 0.53 – 0.73 | -5.83 | <0.001 |
| BS c | 1.14 | 0.10 | 0.96 – 1.35 | 1.47 | 0.140 |
| SDO c | 0.73 | 0.05 | 0.64 – 0.82 | -5.00 | <0.001 |
| SRQ c | 0.98 | 0.01 | 0.97 – 1.00 | -2.63 | 0.009 |
| age c | 1.00 | 0.00 | 1.00 – 1.00 | 0.00 | 0.999 |
| edu c | 1.07 | 0.03 | 1.02 – 1.12 | 2.84 | 0.004 |
| man woman × US oth | 0.89 | 0.07 | 0.77 – 1.03 | -1.57 | 0.117 |
| man woman × Can AusUK | 1.05 | 0.11 | 0.85 – 1.29 | 0.43 | 0.665 |
| man woman × Aus UK | 1.16 | 0.21 | 0.81 – 1.67 | 0.80 | 0.426 |
| man woman × HS c | 0.73 | 0.12 | 0.53 – 1.01 | -1.91 | 0.056 |
| man woman × BS c | 0.87 | 0.15 | 0.62 – 1.22 | -0.81 | 0.420 |
| man woman × SDO c | 1.13 | 0.14 | 0.88 – 1.44 | 0.95 | 0.342 |
| man woman × SRQ c | 1.03 | 0.01 | 1.00 – 1.05 | 2.25 | 0.024 |
| US oth × HS c | 1.01 | 0.05 | 0.92 – 1.11 | 0.24 | 0.813 |
| US oth × BS c | 0.98 | 0.05 | 0.90 – 1.08 | -0.37 | 0.715 |
| US oth × SDO c | 0.99 | 0.03 | 0.93 – 1.06 | -0.21 | 0.837 |
| US oth × SRQ c | 1.00 | 0.00 | 1.00 – 1.01 | 0.88 | 0.381 |
| Can AusUK × HS c | 1.03 | 0.07 | 0.90 – 1.16 | 0.40 | 0.688 |
| Can AusUK × BS c | 0.98 | 0.07 | 0.85 – 1.13 | -0.25 | 0.804 |
| Can AusUK × SDO c | 0.96 | 0.05 | 0.87 – 1.06 | -0.87 | 0.384 |
| Can AusUK × SRQ c | 1.01 | 0.00 | 1.00 – 1.02 | 1.59 | 0.111 |
| Aus UK × HS c | 1.03 | 0.12 | 0.82 – 1.29 | 0.25 | 0.804 |
| Aus UK × BS c | 0.82 | 0.11 | 0.64 – 1.06 | -1.48 | 0.138 |
| Aus UK × SDO c | 0.98 | 0.09 | 0.81 – 1.18 | -0.21 | 0.831 |
| Aus UK × SRQ c | 1.01 | 0.01 | 0.99 – 1.03 | 0.90 | 0.370 |
|
(man woman × US oth) × HS c |
0.89 | 0.08 | 0.74 – 1.06 | -1.30 | 0.194 |
|
(man woman × US oth) × BS c |
0.98 | 0.09 | 0.81 – 1.18 | -0.22 | 0.827 |
|
(man woman × US oth) × SDO c |
1.00 | 0.07 | 0.87 – 1.14 | -0.03 | 0.974 |
|
(man woman × US oth) × SRQ c |
1.01 | 0.01 | 1.00 – 1.02 | 1.25 | 0.211 |
|
(man woman × Can AusUK) × HS c |
1.03 | 0.13 | 0.80 – 1.33 | 0.26 | 0.798 |
|
(man woman × Can AusUK) × BS c |
0.82 | 0.12 | 0.62 – 1.09 | -1.37 | 0.170 |
|
(man woman × Can AusUK) × SDO c |
1.15 | 0.12 | 0.94 – 1.40 | 1.35 | 0.176 |
|
(man woman × Can AusUK) × SRQ c |
1.01 | 0.01 | 0.99 – 1.03 | 0.75 | 0.455 |
|
(man woman × Aus UK) × HS c |
1.00 | 0.23 | 0.63 – 1.57 | -0.01 | 0.996 |
|
(man woman × Aus UK) × BS c |
0.93 | 0.24 | 0.56 – 1.55 | -0.28 | 0.779 |
|
(man woman × Aus UK) × SDO c |
1.09 | 0.20 | 0.75 – 1.57 | 0.44 | 0.662 |
|
(man woman × Aus UK) × SRQ c |
1.00 | 0.02 | 0.96 – 1.04 | -0.10 | 0.917 |
| Observations | 980 | ||||
| R2 Nagelkerke | 0.305 | ||||
s3_olr3 <- clm(LikelihoodofVoting ~ (US_oth + Can_AusUK + Aus_UK) * man_woman + (warmth.c + competence.c)
+ age.c + edu.c,
data = d3,
link = "logit"
)
# summary(s3_olr3)
tab_model(s3_olr3,
show.stat = T,
show.se = T)| Likelihoodof Voting | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| 1|2 | 0.21 | 0.02 | 0.18 – 0.25 | -18.26 | <0.001 |
| 2|3 | 0.54 | 0.04 | 0.47 – 0.62 | -8.68 | <0.001 |
| 3|4 | 1.77 | 0.12 | 1.54 – 2.03 | 8.14 | <0.001 |
| US oth | 1.00 | 0.03 | 0.93 – 1.07 | -0.03 | 0.977 |
| Can AusUK | 0.87 | 0.04 | 0.79 – 0.96 | -2.87 | 0.004 |
| Aus UK | 0.86 | 0.07 | 0.72 – 1.02 | -1.75 | 0.081 |
| man woman | 1.86 | 0.23 | 1.46 – 2.37 | 5.01 | <0.001 |
| warmth c | 1.68 | 0.10 | 1.50 – 1.88 | 8.98 | <0.001 |
| competence c | 0.60 | 0.06 | 0.49 – 0.72 | -5.28 | <0.001 |
| age c | 1.00 | 0.00 | 1.00 – 1.00 | -1.57 | 0.117 |
| edu c | 1.10 | 0.02 | 1.06 – 1.15 | 4.40 | <0.001 |
| US oth × man woman | 0.93 | 0.06 | 0.81 – 1.06 | -1.10 | 0.273 |
| Can AusUK × man woman | 1.01 | 0.10 | 0.83 – 1.22 | 0.09 | 0.931 |
| Aus UK × man woman | 1.10 | 0.19 | 0.78 – 1.55 | 0.56 | 0.577 |
| Observations | 979 | ||||
| R2 Nagelkerke | 0.165 | ||||
s3_olr4 <- clm(LikelihoodofVoting ~ man_woman * (US_oth + Can_AusUK + Aus_UK) * Ideology.c
+ age.c + edu.c,
data = d3,
link = "logit"
)
# summary(s3_olr4)
tab_model(s3_olr4,
show.stat = T,
show.se = T)| Likelihoodof Voting | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| 1|2 | 0.25 | 0.02 | 0.21 – 0.30 | -15.84 | <0.001 |
| 2|3 | 0.70 | 0.05 | 0.61 – 0.81 | -4.82 | <0.001 |
| 3|4 | 2.39 | 0.18 | 2.05 – 2.78 | 11.28 | <0.001 |
| man woman | 1.47 | 0.19 | 1.14 – 1.89 | 3.00 | 0.003 |
| US oth | 0.99 | 0.04 | 0.92 – 1.06 | -0.28 | 0.777 |
| Can AusUK | 0.89 | 0.05 | 0.81 – 0.99 | -2.22 | 0.026 |
| Aus UK | 0.88 | 0.08 | 0.73 – 1.06 | -1.32 | 0.187 |
| Ideology c | 0.62 | 0.03 | 0.57 – 0.68 | -10.91 | <0.001 |
| age c | 1.00 | 0.00 | 1.00 – 1.00 | -0.33 | 0.742 |
| edu c | 1.09 | 0.02 | 1.04 – 1.14 | 3.67 | <0.001 |
| man woman × US oth | 0.92 | 0.07 | 0.80 – 1.06 | -1.17 | 0.240 |
| man woman × Can AusUK | 1.08 | 0.11 | 0.88 – 1.32 | 0.76 | 0.448 |
| man woman × Aus UK | 1.23 | 0.23 | 0.85 – 1.78 | 1.11 | 0.268 |
| man woman × Ideology c | 0.99 | 0.08 | 0.84 – 1.17 | -0.09 | 0.925 |
| US oth × Ideology c | 1.02 | 0.02 | 0.98 – 1.06 | 0.99 | 0.320 |
| Can AusUK × Ideology c | 1.06 | 0.04 | 0.99 – 1.14 | 1.75 | 0.081 |
| Aus UK × Ideology c | 0.94 | 0.06 | 0.82 – 1.07 | -1.00 | 0.320 |
|
(man woman × US oth) × Ideology c |
0.98 | 0.04 | 0.90 – 1.07 | -0.46 | 0.644 |
|
(man woman × Can AusUK) × Ideology c |
1.03 | 0.07 | 0.90 – 1.18 | 0.39 | 0.698 |
|
(man woman × Aus UK) × Ideology c |
1.24 | 0.17 | 0.96 – 1.62 | 1.63 | 0.103 |
| Observations | 980 | ||||
| R2 Nagelkerke | 0.202 | ||||
s3_olr5 <- clm(LikelihoodofVoting ~ man_woman * (US_oth + Can_AusUK + Aus_UK) * (Ideology.c + warmth.c + competence.c)
+ age.c + edu.c,
data = d3,
link = "logit"
)
# summary(s3_olr5)
tab_model(s3_olr5,
show.stat = T,
show.se = T)| Likelihoodof Voting | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| 1|2 | 0.22 | 0.02 | 0.19 – 0.27 | -16.22 | <0.001 |
| 2|3 | 0.65 | 0.05 | 0.56 – 0.76 | -5.58 | <0.001 |
| 3|4 | 2.37 | 0.19 | 2.02 – 2.77 | 10.65 | <0.001 |
| man woman | 1.54 | 0.21 | 1.18 – 2.01 | 3.21 | 0.001 |
| US oth | 1.00 | 0.04 | 0.93 – 1.08 | 0.13 | 0.899 |
| Can AusUK | 0.92 | 0.05 | 0.83 – 1.02 | -1.57 | 0.117 |
| Aus UK | 0.90 | 0.09 | 0.74 – 1.09 | -1.09 | 0.277 |
| Ideology c | 0.64 | 0.03 | 0.58 – 0.69 | -9.93 | <0.001 |
| warmth c | 1.61 | 0.10 | 1.43 – 1.82 | 7.74 | <0.001 |
| competence c | 0.62 | 0.06 | 0.51 – 0.76 | -4.57 | <0.001 |
| age c | 1.00 | 0.00 | 1.00 – 1.00 | -0.85 | 0.396 |
| edu c | 1.09 | 0.03 | 1.04 – 1.15 | 3.75 | <0.001 |
| man woman × US oth | 0.94 | 0.07 | 0.81 – 1.09 | -0.85 | 0.397 |
| man woman × Can AusUK | 1.11 | 0.12 | 0.90 – 1.37 | 0.96 | 0.338 |
| man woman × Aus UK | 1.32 | 0.26 | 0.90 – 1.96 | 1.41 | 0.157 |
| man woman × Ideology c | 0.99 | 0.09 | 0.83 – 1.18 | -0.07 | 0.944 |
| man woman × warmth c | 0.87 | 0.11 | 0.68 – 1.10 | -1.15 | 0.250 |
| man woman × competence c | 0.76 | 0.16 | 0.50 – 1.13 | -1.36 | 0.175 |
| US oth × Ideology c | 1.02 | 0.02 | 0.98 – 1.07 | 0.94 | 0.349 |
| US oth × warmth c | 0.97 | 0.03 | 0.91 – 1.03 | -0.97 | 0.332 |
| US oth × competence c | 0.96 | 0.06 | 0.85 – 1.07 | -0.77 | 0.441 |
| Can AusUK × Ideology c | 1.07 | 0.04 | 1.00 – 1.15 | 1.98 | 0.048 |
| Can AusUK × warmth c | 0.95 | 0.05 | 0.87 – 1.05 | -0.95 | 0.343 |
| Can AusUK × competence c | 1.03 | 0.08 | 0.87 – 1.21 | 0.33 | 0.739 |
| Aus UK × Ideology c | 0.96 | 0.07 | 0.84 – 1.11 | -0.51 | 0.608 |
| Aus UK × warmth c | 0.89 | 0.08 | 0.74 – 1.06 | -1.31 | 0.192 |
| Aus UK × competence c | 0.92 | 0.14 | 0.68 – 1.23 | -0.57 | 0.568 |
|
(man woman × US oth) × Ideology c |
0.97 | 0.04 | 0.89 – 1.06 | -0.71 | 0.475 |
|
(man woman × US oth) × warmth c |
0.89 | 0.06 | 0.78 – 1.01 | -1.80 | 0.072 |
|
(man woman × US oth) × competence c |
0.95 | 0.11 | 0.76 – 1.20 | -0.42 | 0.675 |
|
(man woman × Can AusUK) × Ideology c |
1.03 | 0.07 | 0.89 – 1.18 | 0.35 | 0.728 |
|
(man woman × Can AusUK) × warmth c |
0.99 | 0.10 | 0.82 – 1.21 | -0.06 | 0.954 |
|
(man woman × Can AusUK) × competence c |
0.82 | 0.14 | 0.59 – 1.13 | -1.20 | 0.231 |
|
(man woman × Aus UK) × Ideology c |
1.16 | 0.16 | 0.88 – 1.53 | 1.05 | 0.295 |
|
(man woman × Aus UK) × warmth c |
1.33 | 0.25 | 0.92 – 1.91 | 1.52 | 0.127 |
|
(man woman × Aus UK) × competence c |
1.65 | 0.50 | 0.92 – 2.97 | 1.67 | 0.094 |
| Observations | 979 | ||||
| R2 Nagelkerke | 0.282 | ||||
Full model
gender main effects are pretty unstable across these models
s3_olr6 <- clm(LikelihoodofVoting ~ man_woman
* (warmth.c + competence.c + Ideology.c + HS.c + BS.c + SDO.c + SRQ.c)
* (US_oth + Can_AusUK + Aus_UK)
+ age.c + edu.c,
data = d3,
link = "logit"
)
# summary(s3_olr6)
tab_model(s3_olr6,
show.stat = T,
show.se = T)| Likelihoodof Voting | |||||
|---|---|---|---|---|---|
| Predictors | Odds Ratios | std. Error | CI | Statistic | p |
| 1|2 | 0.16 | 0.02 | 0.13 – 0.20 | -17.18 | <0.001 |
| 2|3 | 0.54 | 0.05 | 0.45 – 0.64 | -7.12 | <0.001 |
| 3|4 | 2.27 | 0.20 | 1.91 – 2.70 | 9.28 | <0.001 |
| man woman | 1.21 | 0.18 | 0.90 – 1.63 | 1.28 | 0.200 |
| warmth c | 1.52 | 0.10 | 1.34 – 1.73 | 6.30 | <0.001 |
| competence c | 0.52 | 0.06 | 0.42 – 0.64 | -6.01 | <0.001 |
| Ideology c | 0.85 | 0.05 | 0.76 – 0.95 | -2.80 | 0.005 |
| HS c | 0.64 | 0.06 | 0.54 – 0.77 | -4.96 | <0.001 |
| BS c | 1.12 | 0.10 | 0.93 – 1.34 | 1.20 | 0.231 |
| SDO c | 0.85 | 0.06 | 0.74 – 0.97 | -2.35 | 0.019 |
| SRQ c | 0.98 | 0.01 | 0.97 – 0.99 | -3.30 | 0.001 |
| US oth | 0.98 | 0.04 | 0.91 – 1.07 | -0.41 | 0.681 |
| Can AusUK | 0.92 | 0.06 | 0.81 – 1.04 | -1.38 | 0.167 |
| Aus UK | 0.86 | 0.10 | 0.70 – 1.07 | -1.31 | 0.189 |
| age c | 1.00 | 0.00 | 1.00 – 1.00 | -0.57 | 0.569 |
| edu c | 1.07 | 0.03 | 1.02 – 1.13 | 2.74 | 0.006 |
| man woman × warmth c | 0.95 | 0.13 | 0.73 – 1.23 | -0.41 | 0.684 |
| man woman × competence c | 0.76 | 0.17 | 0.50 – 1.17 | -1.24 | 0.215 |
| man woman × Ideology c | 0.87 | 0.10 | 0.70 – 1.09 | -1.21 | 0.225 |
| man woman × HS c | 0.77 | 0.14 | 0.54 – 1.08 | -1.50 | 0.134 |
| man woman × BS c | 0.85 | 0.16 | 0.59 – 1.23 | -0.87 | 0.385 |
| man woman × SDO c | 1.09 | 0.16 | 0.83 – 1.44 | 0.63 | 0.526 |
| man woman × SRQ c | 1.03 | 0.01 | 1.00 – 1.05 | 2.23 | 0.026 |
| man woman × US oth | 0.89 | 0.07 | 0.76 – 1.05 | -1.34 | 0.179 |
| man woman × Can AusUK | 1.04 | 0.13 | 0.82 – 1.32 | 0.30 | 0.765 |
| man woman × Aus UK | 1.48 | 0.33 | 0.96 – 2.28 | 1.76 | 0.078 |
| warmth c × US oth | 0.95 | 0.03 | 0.89 – 1.02 | -1.38 | 0.167 |
| warmth c × Can AusUK | 0.94 | 0.05 | 0.85 – 1.05 | -1.04 | 0.298 |
| warmth c × Aus UK | 0.96 | 0.10 | 0.79 – 1.17 | -0.40 | 0.691 |
| competence c × US oth | 0.95 | 0.06 | 0.84 – 1.08 | -0.75 | 0.453 |
| competence c × Can AusUK | 1.09 | 0.09 | 0.92 – 1.29 | 0.95 | 0.342 |
| competence c × Aus UK | 0.91 | 0.14 | 0.67 – 1.24 | -0.60 | 0.546 |
| Ideology c × US oth | 1.03 | 0.03 | 0.97 – 1.09 | 1.05 | 0.293 |
| Ideology c × Can AusUK | 1.05 | 0.05 | 0.96 – 1.14 | 0.97 | 0.333 |
| Ideology c × Aus UK | 0.90 | 0.08 | 0.76 – 1.07 | -1.18 | 0.238 |
| HS c × US oth | 0.99 | 0.05 | 0.90 – 1.09 | -0.24 | 0.807 |
| HS c × Can AusUK | 1.04 | 0.07 | 0.90 – 1.19 | 0.55 | 0.584 |
| HS c × Aus UK | 1.01 | 0.13 | 0.79 – 1.30 | 0.10 | 0.918 |
| BS c × US oth | 1.01 | 0.05 | 0.91 – 1.11 | 0.18 | 0.857 |
| BS c × Can AusUK | 0.97 | 0.07 | 0.83 – 1.12 | -0.46 | 0.648 |
| BS c × Aus UK | 0.88 | 0.12 | 0.67 – 1.16 | -0.88 | 0.380 |
| SDO c × US oth | 0.97 | 0.04 | 0.90 – 1.04 | -0.93 | 0.352 |
| SDO c × Can AusUK | 0.95 | 0.06 | 0.85 – 1.07 | -0.88 | 0.379 |
| SDO c × Aus UK | 1.03 | 0.11 | 0.84 – 1.26 | 0.26 | 0.793 |
| SRQ c × US oth | 1.00 | 0.00 | 1.00 – 1.01 | 0.82 | 0.410 |
| SRQ c × Can AusUK | 1.01 | 0.01 | 1.00 – 1.02 | 1.60 | 0.109 |
| SRQ c × Aus UK | 1.01 | 0.01 | 0.99 – 1.03 | 0.77 | 0.442 |
|
(man woman × warmth c) × US oth |
0.85 | 0.06 | 0.74 – 0.97 | -2.36 | 0.018 |
|
(man woman × warmth c) × Can AusUK |
1.04 | 0.11 | 0.84 – 1.29 | 0.36 | 0.721 |
|
(man woman × warmth c) × Aus UK |
1.35 | 0.27 | 0.91 – 1.99 | 1.50 | 0.135 |
|
(man woman × competence c) × US oth |
0.93 | 0.12 | 0.73 – 1.18 | -0.61 | 0.543 |
|
(man woman × competence c) × Can AusUK |
0.91 | 0.16 | 0.65 – 1.28 | -0.54 | 0.592 |
|
(man woman × competence c) × Aus UK |
1.39 | 0.44 | 0.75 – 2.59 | 1.05 | 0.293 |
|
(man woman × Ideology c) × US oth |
0.96 | 0.06 | 0.85 – 1.07 | -0.78 | 0.433 |
|
(man woman × Ideology c) × Can AusUK |
0.91 | 0.08 | 0.76 – 1.09 | -0.98 | 0.327 |
|
(man woman × Ideology c) × Aus UK |
1.33 | 0.24 | 0.94 – 1.89 | 1.60 | 0.110 |
|
(man woman × HS c) × US oth |
0.92 | 0.09 | 0.76 – 1.12 | -0.82 | 0.414 |
|
(man woman × HS c) × Can AusUK |
1.18 | 0.17 | 0.89 – 1.56 | 1.16 | 0.246 |
|
(man woman × HS c) × Aus UK |
0.79 | 0.20 | 0.48 – 1.29 | -0.93 | 0.350 |
|
(man woman × BS c) × US oth |
1.01 | 0.10 | 0.83 – 1.23 | 0.13 | 0.896 |
|
(man woman × BS c) × Can AusUK |
0.80 | 0.12 | 0.60 – 1.08 | -1.44 | 0.149 |
|
(man woman × BS c) × Aus UK |
0.94 | 0.26 | 0.54 – 1.62 | -0.23 | 0.818 |
|
(man woman × SDO c) × US oth |
1.00 | 0.08 | 0.86 – 1.16 | 0.04 | 0.971 |
|
(man woman × SDO c) × Can AusUK |
1.26 | 0.15 | 1.00 – 1.59 | 1.96 | 0.049 |
|
(man woman × SDO c) × Aus UK |
1.06 | 0.22 | 0.70 – 1.60 | 0.28 | 0.780 |
|
(man woman × SRQ c) × US oth |
1.01 | 0.01 | 0.99 – 1.02 | 0.77 | 0.440 |
|
(man woman × SRQ c) × Can AusUK |
1.01 | 0.01 | 0.99 – 1.03 | 0.53 | 0.599 |
|
(man woman × SRQ c) × Aus UK |
1.00 | 0.02 | 0.96 – 1.03 | -0.23 | 0.817 |
| Observations | 979 | ||||
| R2 Nagelkerke | 0.391 | ||||
Assumption
H0: Parallel Regression Assumption holds
Data Visualization
Experiment Vote Choice
d3 %>% ggplot(aes(x = as.numeric(LikelihoodofVoting), fill = Demo_Gender)) +
geom_histogram( color="#e9ecef", alpha=0.8, position = 'dodge',
binwidth = .5) +
scale_fill_manual(values=c("#69b3a2", "#404080")) +
theme_bw() +
scale_x_continuous( labels = c("Masc Man",
"Femme Man",
"Masc Woman",
"Femme Woman"),
breaks = seq(1,4,1)) +
scale_y_continuous(breaks = seq(0,80,10)) +
labs(x = "Candidate Selected",
y = "Frequency",
fill = "Participant
Gender",
title = "Number Voting for Each Candidate by Gender and Country") +
facet_wrap(~Country)
Predicting experimental vote choice
s3_olr.plot <- clm(LikelihoodofVoting ~ Demo_Gender * (Ideology + WomanTraits_warmth + WomanTraits_competence + HS + BS + SDO + SRQ) * Country,
data = d3,
link = "logit"
)
## Plot 1: Female candidate traits: Warmth
preds_olr.plot <- predictions(
s3_olr.plot,
newdata = datagrid(
# Plotting variables
WomanTraits_warmth = seq(
min(d3$WomanTraits_warmth, na.rm = TRUE),
max(d3$WomanTraits_warmth, na.rm = TRUE),
length.out = 100
),
Demo_Gender = unique(na.omit(d3$Demo_Gender)),
Country = unique(na.omit(d3$Country)),
# Variables to hold constant
Ideology = mean(d3$Ideology, na.rm = TRUE),
WomanTraits_competence = mean(d3$WomanTraits_competence, na.rm = TRUE),
HS = mean(d3$HS, na.rm = TRUE),
BS = mean(d3$BS, na.rm = TRUE),
SDO = mean(d3$SDO, na.rm = TRUE),
SRQ = mean(d3$SRQ, na.rm = TRUE)
)
)
# Ensure prediction data is sorted cleanly for smooth lines
preds <- preds_olr.plot[order(preds_olr.plot$Demo_Gender, preds_olr.plot$group, preds_olr.plot$WomanTraits_warmth), ]
preds$group_labeled <- factor(
preds$group,
labels = c("Masc Man",
"Femme Man",
"Masc Woman",
"Femme Woman")
)
# Plot
preds %>% ggplot(aes(x = WomanTraits_warmth, y = estimate, color = factor(group_labeled), group = factor(group_labeled))) +
geom_line(linewidth = 1) +
geom_ribbon(aes(ymin = conf.low, ymax = conf.high, fill = factor(group_labeled)), alpha = 0.15, color = NA) +
facet_grid(Demo_Gender ~ Country) +
labs(
x = "Woman Traits Warmth",
y = "Predicted Probability of Voting",
color = "Candidate",
fill = "Candidate",
caption = "Note: Model-predicted values. Other continuous predictors held at mean.",
title = "Likelihood of voting for each candidate by warmth importance, shown by gender and country"
) +
scale_x_continuous(breaks = seq(1,7,1)) +
theme_minimal()
## Plot 2: SRQ
preds_olr.plot2 <- predictions(
s3_olr.plot,
newdata = datagrid(
# Plotting variables
SRQ = seq(
min(d3$SRQ, na.rm = TRUE),
max(d3$SRQ, na.rm = TRUE),
length.out = 100
),
Demo_Gender = unique(na.omit(d3$Demo_Gender)),
Country = unique(na.omit(d3$Country)),
# Variables to hold constant
Ideology = mean(d3$Ideology, na.rm = TRUE),
WomanTraits_competence = mean(d3$WomanTraits_competence, na.rm = TRUE),
WomanTraits_warmth = mean(d3$WomanTraits_warmth, na.rm = TRUE),
HS = mean(d3$HS, na.rm = TRUE),
BS = mean(d3$BS, na.rm = TRUE),
SDO = mean(d3$SDO, na.rm = TRUE)
)
)
# Ensure prediction data is sorted cleanly for smooth lines
preds2 <- preds_olr.plot2[order(preds_olr.plot2$Demo_Gender, preds_olr.plot2$group, preds_olr.plot2$WomanTraits_warmth), ]
preds2$group_labeled <- factor(
preds2$group,
labels = c("Masc Man",
"Femme Man",
"Masc Woman",
"Femme Woman")
)
# Plot
preds2 %>% ggplot(aes(x = SRQ, y = estimate, color = factor(group_labeled), group = factor(group_labeled))) +
geom_line(linewidth = 1) +
geom_ribbon(aes(ymin = conf.low, ymax = conf.high, fill = factor(group_labeled)), alpha = 0.15, color = NA) +
facet_grid(Demo_Gender ~ Country) +
labs(
x = "Social Roles Questionnaire (0 to 100)",
y = "Predicted Probability of Voting",
color = "Candidate",
fill = "Candidate",
caption = "Note: Model-predicted values. Other continuous predictors held at mean.",
title = "Likelihood of voting for each candidate by SRQ, shown by gender and country"
) +
scale_x_continuous(breaks = seq(0,100,25)) +
theme_minimal()
Summary of Findings
DV: Voting for Harris
Conditional Effects
Across all three studies, there is no effect of gender in likelihood of voting for Harris (compared with any other choice) once controlling for other variables.
Hostile sexism consistently negatively predicts likelihood of voting Harris across all three studies, and social roles questionnaire negatively predicts it in studies 2 and 3. Impact index positively predicts and ideology (increasing conservatism) negatively predicts likelihood of voting Harris when measured (studies 1 and 2 for impact index, studies 2 and 3 for ideology).
Interaction Effects
Across all three studies, there were only two cross-country higher-order interactions (SRQ in Study 2, Ideology in Study 3). Higher-order gender interactions were only observed in study 1 (with hostile sexism and benevolent sexism).
DV: Experimental Candidate Selection
Conditional Effects
Participants’ rated importance of a female candidate’s warmth positively predicted voting for the feminine woman candidate, whereas rated importance of a female candidate’s competence negatively predicted voting for the feminine woman. Increasing conservatism, hostile sexism, social dominance orientation, and social roles questionnaire also negatively predicted voting for the feminine woman candidate. Education positively predicted voting for the feminine woman.
Interaction Effects
There was an interaction between participant gender (man vs. woman) and social roles questionnaire on voting for the feminine woman candidate. There was also a three-way interaction between gender, country, and rated importance of warmth in likelihood of voting for the feminine woman candidate.