PCA
pca_data <- pretest_data[,3:8] #pull just the gender characteristic columns
data_normalized <- scale(pca_data)
head(data_normalized) #normalize the data
Decisive.Indecisive Strong.Weak Inspiring.Uninspiring Dishonest.Honest
[1,] -1.0385484 -0.6000738 -0.3349686 0.9702373
[2,] 0.6589557 0.8926097 0.5251443 -1.3012595
[3,] 0.6589557 -0.6000738 -0.3349686 0.9702373
[4,] 0.6589557 0.8926097 0.5251443 0.2130717
[5,] 1.5077078 1.6389515 1.3852573 -1.3012595
[6,] -0.1897963 0.1462680 -0.3349686 0.2130717
Harsh.Compassionate Against.equality.For.equality
[1,] 0.5208243 0.3557871
[2,] -1.2433294 -1.2442498
[3,] 1.4029011 1.1558055
[4,] 0.5208243 0.3557871
[5,] -1.2433294 -1.2442498
[6,] -0.3612526 0.3557871
data.pca <- princomp(data_normalized)
summary(data.pca)
Importance of components:
Comp.1 Comp.2 Comp.3 Comp.4 Comp.5
Standard deviation 1.8922151 1.0125529 0.66090518 0.60921557 0.55774044
Proportion of Variance 0.5982495 0.1713077 0.07298265 0.06201308 0.05197633
Cumulative Proportion 0.5982495 0.7695571 0.84253980 0.90455288 0.95652920
Comp.6
Standard deviation 0.5100681
Proportion of Variance 0.0434708
Cumulative Proportion 1.0000000
Contribution to PCA
Comp.1 Comp.2 Comp.3
Decisive.Indecisive 0.2374737 0.8086153 0.51932318
Strong.Weak 0.3886839 0.4106341 -0.70760931
Inspiring.Uninspiring 0.4538161 -0.1156917 -0.13194661
Dishonest.Honest -0.4575839 0.1753043 -0.06414359
Harsh.Compassionate -0.4168051 0.3004526 -0.45582129
Against.equality.For.equality -0.4510792 0.2076849 0.01717992
Scree Plot
fviz_eig(data.pca, addlabels = TRUE)
Biplot of the attributes
fviz_pca_var(data.pca, col.var = "black")
Mapping onto PartyId
Decisive / Indecisive mapped onto respondent party id
very_filtered_data <- pretest_data |>
filter(partyid != 0) |>
mutate(partyid_numeric = as.numeric(partyid))
correlation_spearman <- cor(very_filtered_data$partyid_numeric, very_filtered_data$Decisive.Indecisive, method = "spearman")
print(correlation_spearman)
Strong / Weak mapped onto party id
very_filtered_data <- pretest_data |>
filter(partyid != 0) |>
mutate(partyid_numeric = as.numeric(partyid))
correlation_spearman <- cor(very_filtered_data$partyid_numeric, very_filtered_data$Strong.Weak, method = "spearman")
print(correlation_spearman)
Decisive/Indecisive mapped onto partyid by condition
Condition 1
filtered_data_condition_1<- pretest_data |>
filter(Condition.number == 1) |>
filter(partyid != 0) |>
mutate(partyid_numeric = as.numeric(partyid))
correlation_spearman <- cor(filtered_data_condition_1$partyid_numeric, filtered_data_condition_1$Decisive.Indecisive, method = "spearman")
print(correlation_spearman)
Condition 2
filtered_data_condition_2<- pretest_data |>
filter(Condition.number == 2) |>
filter(partyid != 0) |>
mutate(partyid_numeric = as.numeric(partyid))
correlation_spearman <- cor(filtered_data_condition_2$partyid_numeric, filtered_data_condition_2$Decisive.Indecisive, method = "spearman")
print(correlation_spearman)
Condition 3
filtered_data_condition_3<- pretest_data |>
filter(Condition.number == 3) |>
filter(partyid != 0) |>
mutate(partyid_numeric = as.numeric(partyid))
correlation_spearman <- cor(filtered_data_condition_3$partyid_numeric, filtered_data_condition_3$Decisive.Indecisive, method = "spearman")
print(correlation_spearman)
Condition 4
filtered_data_condition_4<- pretest_data |>
filter(Condition.number == 4) |>
filter(partyid != 0) |>
mutate(partyid_numeric = as.numeric(partyid))
correlation_spearman <- cor(filtered_data_condition_4$partyid_numeric, filtered_data_condition_4$Decisive.Indecisive, method = "spearman")
print(correlation_spearman)
Condition 5
)
print(kruskal_test)
filtered_data_condition_5<- pretest_data |>
filter(Condition.number == 5) |>
filter(partyid != 0) |>
mutate(partyid_numeric = as.numeric(partyid))
correlation_spearman <- cor(filtered_data_condition_5$partyid_numeric, filtered_data_condition_5$Decisive.Indecisive, method = "spearman")
print(correlation_spearman)
Condition 6
filtered_data_condition_6<- pretest_data |>
filter(Condition.number == 6) |>
filter(partyid != 0) |>
mutate(partyid_numeric = as.numeric(partyid))
correlation_spearman <- cor(filtered_data_condition_6$partyid_numeric, filtered_data_condition_6$Decisive.Indecisive, method = "spearman")
print(correlation_spearman)
filtered_data_condition_6<- pretest_data |>
filter(Condition.number == 6) |>
filter(partyid != 0) |>
mutate(partyid_numeric = as.numeric(partyid))
correlation_spearman <- cor(filtered_data_condition_6$partyid_numeric, filtered_data_condition_6$Decisive.Indecisive, method = "spearman")
print(correlation_spearman)
Characteristics mapped onto respondent perception of party for statement
Decisive/Indecisive by perceived party
correlation_spearman <- cor(pretest_data$perceived.party, pretest_data$Decisive.Indecisive, method = "spearman")
print(correlation_spearman)
Strong / Weak by perceived party
correlation_spearman <- cor(pretest_data$perceived.party, pretest_data$Strong.Weak, method = "spearman")
print(correlation_spearman)
Mapping characteristic onto perceived threat to democracy
Decisive/Indecisive mapped onto threat to democracy
correlation_spearman <- cor(pretest_data$Good.for.democracy...1...disagree..5...agree., pretest_data$Decisive.Indecisive, method = "spearman")
print(correlation_spearman)
Strong/Weak mapped onto threat to democracy
correlation_spearman <- cor(pretest_data$Good.for.democracy...1...disagree..5...agree., pretest_data$Strong.Weak, method = "spearman")
print(correlation_spearman)
Harsh / Compassionate as threat to democracy
correlation_spearman <- cor(pretest_data$Good.for.democracy...1...disagree..5...agree., pretest_data$Harsh.Compassionate, method = "spearman")
print(correlation_spearman)
Against equality/ For equality as threat to democracy
correlation_spearman <- cor(pretest_data$Good.for.democracy...1...disagree..5...agree., pretest_data$Against.equality.For.equality, method = "spearman")
print(correlation_spearman)
Dishonest / Honest as threat to democracy
correlation_spearman <- cor(pretest_data$Good.for.democracy...1...disagree..5...agree., pretest_data$Dishonest.Honest, method = "spearman")
print(correlation_spearman)
Inspiring / Uninspiring as threat to democracy
correlation_spearman <- cor(pretest_data$Good.for.democracy...1...disagree..5...agree., pretest_data$Inspiring.Uninspiring, method = "spearman")
print(correlation_spearman)
