NCI Voting

This is an R Markdown Notebook. When you execute code within the notebook, the results appear beneath the code.

Try executing this chunk by clicking the Run button within the chunk or by placing your cursor inside it and pressing Ctrl+Shift+Enter.

library(haven)
library(tidyverse)
library(dplyr)

library(readxl)
NCI_17_18 <- read_excel("NCI 17-18.xlsx")
View(NCI_17_18)

library(readxl)
state_disprov <- read_excel("state disprov.xlsx")
View(state_disprov)

NCI_merge <- left_join(NCI_17_18, state_disprov, by = "STATENUM")
glimpse(NCI_merge)

NCI.select <- NCI_merge %>%
  select(.,
         STATENUM,
         STATENAM,
         DIS_PROV,
         VOTE15,
         AGE,
         RACE15_AMERIN,
         RACE15_ASIAN,
         RACE15_BLACK,
         RACE15_DK,
         RACE15_HAWAII,
         RACE15_WHITE,
         RACE15_LATINO,
         RACE15_OTHER,
         LGLSTAT15,
         LEVELMR08,
         EXPRESS15,
         MOBILITY08,
         GENDER)

NCI.clean <- NCI.select %>%
  filter(!(LEVELMR08 %in%
             c("5", "6", "98")))%>%
  filter(!(VOTE15 %in%
             c("99")))%>%
  filter(!(EXPRESS15 %in%
             c("99")))%>%
  filter(!(MOBILITY08 %in%
             c("99")))%>%
  filter(!(LGLSTAT15 %in%
             c("99")))%>%
  mutate(., 
         vote.re = case_when(
           VOTE15 == 1 ~ 0,
           VOTE15 == 2 ~ 1,
           VOTE15 == 3 ~ 1
         )) %>%
  mutate(.,
         mobility.re = case_when(
           MOBILITY08 == 1 ~ 0,
           MOBILITY08 == 2 ~ 1,
           MOBILITY08 == 3 ~ 1
         ))  %>%
  mutate(.,
         communication = case_when(
           EXPRESS15 == 1 ~ 0,
           EXPRESS15 == 2 ~ 1,
           EXPRESS15 == 3 ~ 1,
           EXPRESS15 == 4 ~ 1,
           EXPRESS15 == 5 ~ 1
         )) %>%
  mutate (.,
          guardian = case_when(
            LGLSTAT15 == 1 ~ 0,
            LGLSTAT15 == 2 ~ 1,
            LGLSTAT15 == 3 ~ 1,
            LGLSTAT15 == 4 ~ 1
          )) %>%
  mutate(.,
         white = case_when(
           RACE15_WHITE == 1 ~ 0,
           RACE15_WHITE == 2 ~ 1
         ))  %>%
  mutate(.,
         black = case_when(
           RACE15_BLACK == 1 ~ 0,
           RACE15_BLACK == 2 ~ 1
         ))  %>%
 mutate(.,
         latino = case_when(
           RACE15_LATINO == 1 ~ 0,
           RACE15_LATINO == 2 ~ 1
         )) %>%
   mutate(.,
         other = case_when(
           RACE15_AMERIN == 1 ~ 0,
           RACE15_AMERIN == 2 ~ 1,
           RACE15_ASIAN == 1 ~ 0,
           RACE15_ASIAN == 2 ~ 1,
           RACE15_HAWAII == 1 ~ 0,
           RACE15_HAWAII == 2 ~ 1,
           RACE15_OTHER == 1 ~ 0,
           RACE15_OTHER == 2 ~ 1
         )) %>%
  mutate(.,
         gender.re = case_when(
           GENDER == 1 ~ 0,
           GENDER == 2 ~ 1,
           GENDER == 3 ~ 2
         ))%>%
  mutate(.,
         levelid = as.factor(LEVELMR08))
           
age.cat <- rep(NA,19862)
age.cat[NCI.clean$AGE <= 29] <- 1
age.cat[NCI.clean$AGE >= 30 & NCI.clean$AGE <=44] <- 2
age.cat[NCI.clean$AGE >=45 & NCI.clean$AGE <=59] <- 3
age.cat[NCI.clean$AGE >=60] <- 4
NCI.clean$age.cat <- as.factor(age.cat)

glimpse(NCI.clean)

``

table(NCI.clean$white)

table(NCI.clean$STATENAM)

table(NCI.clean$gender.re)

table(NCI.clean$age.cat)

table(NCI.clean$guardian)

table(NCI.clean$levelid)

table(NCI.clean$communication)

table(NCI.clean$mobility.re)

table(NCI.clean$DIS_PROV)

table(NCI.clean$black)

table(NCI.clean$latino)

table(NCI.clean$other)

table(NCI.clean$vote.re)

library(lme4)
model.null <- glmer(vote.re ~ (1|STATENUM), family = binomial, data = NCI.clean)
summary(model.null)

model.2 <- glmer(vote.re ~ age.cat + levelid + communication + mobility.re + gender.re + other + black +  white + latino + guardian+ (1|STATENUM), family = binomial, data = NCI.clean)
summary(model.2)

model.3 <- glmer(vote.re ~ age.cat + levelid + communication + mobility.re + gender.re + other + black +  white + latino + guardian + DIS_PROV + (1|STATENUM), family = binomial, data = NCI.clean)
summary(model.3)

ICC.null <- 0.3244 / (0.3244 + (pi^2/3))
ICC.null

library(janitor)

tabyl(NCI.clean$vote.re, sort=FALSE)

tabyl(NCI.clean$white, sort=FALSE)

tabyl(NCI.clean$black, sort=FALSE)

tabyl(NCI.clean$latino, sort=FALSE)

tabyl(NCI.clean$other, sort=FALSE)

tabyl(NCI.clean$age.cat, sort=FALSE)

tabyl(NCI.clean$gender.re, sort=FALSE)

tabyl(NCI.clean$mobility.re, sort=FALSE)

tabyl(NCI.clean$communication, sort=FALSE)

tabyl(NCI.clean$LEVELMR08, sort=FALSE)

tabyl(NCI.clean$guardian, sort=FALSE)

tabyl(NCI.clean$DIS_PROV, sort=FALSE)

tabyl(NCI.clean$STATENAM, sort=FALSE)

library(modelsummary)
library(broom.mixed)

models <- list(model.null, model.2, model.3)
modelsummary(models, output = 'default')

model.4 <- glmer(vote.re ~ age.cat + levelid + communication + mobility.re + gender.re + other + black +  white + latino + guardian + DIS_PROV + (guardian|STATENUM), family = binomial, data = NCI.clean)
summary(model.4)

model.5 <- glmer(vote.re ~ age.cat + levelid + communication + mobility.re + gender.re + other + black +  white + latino + guardian + DIS_PROV + (levelid|STATENUM), family = binomial, data = NCI.clean)
summary(model.5)

model.6 <- glmer(vote.re ~ age.cat + levelid + communication + mobility.re + gender.re + other + black +  white + latino + guardian + DIS_PROV + guardian:DIS_PROV + (1|STATENUM), family = binomial, data = NCI.clean)
summary(model.6)

model.7 <- glmer(vote.re ~ age.cat + levelid + communication + mobility.re + gender.re + other + black +  white + latino + guardian + DIS_PROV + levelid:DIS_PROV + (1|STATENUM), family = binomial, data = NCI.clean)
summary(model.7)

models2 <- list(model.7)
modelsummary(models2, output = 'default')

library(ggplot2)

library(jtools)

help("cat_plot")
library(interactions)

interact_plot(model.7, pred = DIS_PROV, modx = levelid)

table(NCI.clean$STATENAM)

#odds ratios and probabilities for model 2--probably not useful#
age2.odds <- exp(.374891)
age2.odds

age2.prob <- (1.454833)/(1+1.454833)
age2.prob

age3.odds <- exp(.30341)
age3.prob <- exp(.30341)/(1+exp(.30341))
age3.odds
age3.prob

age4.odds <-exp(.19816)
age4.prob <- exp(.19816)+(1+exp(.19816))
age4.odds
age4.prob

id2.odds <- exp(-.83680)
id2.prob <- exp(-.8368)/(1+exp(-.8368))
id2.odds
id2.prob

id3.odds <- exp(-1.56064)
id3.prob <- exp(-1.56064)/(1+exp(-1.56064))
id3.odds
id3.prob

id4.odds <- exp(-2.04651)
id4.prob <- exp(-2.04651)/(1+exp(-2.04651))
id4.odds
id4.prob

black.odds <- exp(.63635)
black.prob <- exp(.63635)/(1+exp(.63635))
black.odds
black.prob

white.odds <- exp(.46405)
white.prob <- exp(.46405)/(1+exp(.46405))
white.odds
white.prob

guardian.odds <- exp(-.45986)
guardian.prob <- exp(-.45986)/(1+exp(-.45986))
guardian.odds
guardian.prob

#model 7 disenfranchisement provision odds and probability#
dis.odds <- exp(-0.30423)
dis.prob <- exp(-0.30423)/(1+exp(-0.30423))
dis.odds
dis.prob

#model 7 interaction id moderate (2)#
id2.dis.odds <-exp(-.7912+-.65939+-.30423+-.25213)
id2.dis.odds

id2.dis.prob <- 0.134398/(1+0.134398)
id2.dis.prob

#model 7 interaction id severe (3)#
id3.dis.odds <- exp(-.7912+-1.25549 + -.30423 + -.45732)
id3.dis.odds

id3.dis.prob <- 0.06031105/(1+0.06031105)
id3.dis.prob

#model 7 interaction id profound (4)#
id4.dis.odds <- exp(-.7912+-1.4816 +-.30423 +-.96945)
id4.dis.odds

id4.dis.prob <- 0.02882593/(1+0.02882593)
id4.dis.prob

#model 7 main effect id moderate (2)#
model7id2.odds <-exp(-.7912+-.65393)
model7id2.odds

model7id2.prob <- 0.2357154/(1+0.2357154)
model7id2.prob

#model 7 main effect id severe (3)#
model7id3.odds <-exp(-.7912+-1.25549)
model7id3.odds

model2id3.prob <- 0.1317525/(1+0.1317525)
model2id3.prob

#model 7 main effect id profound (4)#
model7id4.odds <- exp(-.7912 + -1.48416)
model7id4.odds

model7id4.prob <- 0.1027599/(1+0.1027599)

#model 7 main effect ID mild (1)
model7id1.odds <- exp(-.7912)
model7id1.odds

model7id1.prob <- 0.4533005/(1+0.4533005)
model7id1.prob

#model 7 id mild (1) and disenfranchisement#
model7id1.dis.odds <-exp(-.7912+-.30423)
model7id1.dis.odds

model7id1.dis.prob <- 0.3343958/(1+0.3343958)
model7id1.dis.prob

```

Add a new chunk by clicking the Insert Chunk button on the toolbar or by pressing Ctrl+Alt+I.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the Preview button or press Ctrl+Shift+K to preview the HTML file).

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike Knit, Preview does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.