library(haven)
TEDS_2016<-read_stata("https://github.com/datageneration/home/blob/master/DataProgramming/data/TEDS_2016.dta?raw=true")
TEDS_2016$Tondu<-as.numeric(TEDS_2016$Tondu,labels=c("Unificationnow","Statusquo,unif.infuture","Statusquo,decidelater",
"Statusquoforever","Statusquo,indep.infuture","Independencenow","Noresponse"))
# Generate Frequency Table
table(TEDS_2016$Tondu)
##
## 1 2 3 4 5 6 9
## 27 180 546 328 380 108 121
# Barplot of Tondu Variable
barplot(TEDS_2016$Tondu, main="Frequency of Responses",
xlab="Response Frequency",
ylab="Response")
# Multiple Linear Regression to Explore Relationship between Tondu and other variables
explore <- lm(Tondu ~ age + female + DPP, data=TEDS_2016)
summary(explore) # show results
##
## Call:
## lm(formula = Tondu ~ age + female + DPP, data = TEDS_2016)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.5414 -1.1640 -0.3789 0.8656 5.4216
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.253427 0.140359 23.179 < 2e-16 ***
## age 0.010484 0.002528 4.148 3.53e-05 ***
## female 0.491240 0.084980 5.781 8.85e-09 ***
## DPP 0.341330 0.088999 3.835 0.00013 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.743 on 1686 degrees of freedom
## Multiple R-squared: 0.03739, Adjusted R-squared: 0.03568
## F-statistic: 21.83 on 3 and 1686 DF, p-value: 7.182e-14
# Multiple Linear Regression with income instead of DPP
explore2 <- lm(Tondu ~ age + female + income, data=TEDS_2016)
summary(explore2) # show results
##
## Call:
## lm(formula = Tondu ~ age + female + income, data = TEDS_2016)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.8385 -1.1767 -0.4111 0.9705 5.4310
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.806978 0.175860 21.648 < 2e-16 ***
## age 0.008199 0.002581 3.176 0.001518 **
## female 0.466677 0.085058 5.487 4.72e-08 ***
## income -0.058233 0.015855 -3.673 0.000247 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.743 on 1686 degrees of freedom
## Multiple R-squared: 0.0367, Adjusted R-squared: 0.03499
## F-statistic: 21.41 on 3 and 1686 DF, p-value: 1.302e-13
# Multiple Linear Regression with edu instead of income
explore3 <- lm(Tondu ~ age + female + edu, data=TEDS_2016)
summary(explore3) # show results
##
## Call:
## lm(formula = Tondu ~ age + female + edu, data = TEDS_2016)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.8795 -1.0725 -0.4116 0.9992 5.5664
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.893961 0.257545 19.002 < 2e-16 ***
## age -0.003801 0.003173 -1.198 0.231
## female 0.441796 0.084293 5.241 1.80e-07 ***
## edu -0.243427 0.036035 -6.755 1.96e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.713 on 1676 degrees of freedom
## (10 observations deleted due to missingness)
## Multiple R-squared: 0.05617, Adjusted R-squared: 0.05448
## F-statistic: 33.25 on 3 and 1676 DF, p-value: < 2.2e-16
# Multiple Linear Regression
explore4 <- lm(Tondu ~ Taiwanese + female + edu, data=TEDS_2016)
summary(explore4) # show results
##
## Call:
## lm(formula = Tondu ~ Taiwanese + female + edu, data = TEDS_2016)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.0742 -1.0963 -0.2877 0.7432 5.9037
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.11799 0.12410 33.184 < 2e-16 ***
## Taiwanese 0.78268 0.08484 9.225 < 2e-16 ***
## female 0.37782 0.08255 4.577 5.07e-06 ***
## edu -0.20433 0.02761 -7.400 2.15e-13 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.671 on 1676 degrees of freedom
## (10 observations deleted due to missingness)
## Multiple R-squared: 0.101, Adjusted R-squared: 0.0994
## F-statistic: 62.77 on 3 and 1676 DF, p-value: < 2.2e-16
# Multiple Linear Regression
explore5 <- lm(Tondu ~ Taiwanese + female + Econ_worse, data=TEDS_2016)
summary(explore5) # show results
##
## Call:
## lm(formula = Tondu ~ Taiwanese + female + Econ_worse, data = TEDS_2016)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.7556 -1.1450 -0.3203 0.6797 5.6797
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.51557 0.08831 39.809 < 2e-16 ***
## Taiwanese 0.82469 0.08704 9.475 < 2e-16 ***
## female 0.41538 0.08386 4.953 8.03e-07 ***
## Econ_worse -0.19522 0.08439 -2.313 0.0208 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.713 on 1686 degrees of freedom
## Multiple R-squared: 0.07027, Adjusted R-squared: 0.06861
## F-statistic: 42.47 on 3 and 1686 DF, p-value: < 2.2e-16
# Correlation between Tondu and votetsai
tsai_cor <- lm(Tondu ~ votetsai, data=TEDS_2016)
summary(tsai_cor) # show results
##
## Call:
## lm(formula = Tondu ~ votetsai, data = TEDS_2016)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.3785 -1.3588 -0.3588 0.6215 5.6412
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.35881 0.06918 48.55 <2e-16 ***
## votetsai 1.01967 0.08741 11.67 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.501 on 1259 degrees of freedom
## (429 observations deleted due to missingness)
## Multiple R-squared: 0.09755, Adjusted R-squared: 0.09683
## F-statistic: 136.1 on 1 and 1259 DF, p-value: < 2.2e-16