Project Proposal
OMER OZEREN
Data Preparation
The data comes from the American National Election Studies (ANES), I uploaded file to my github : https://raw.githubusercontent.com/omerozeren/DATA606/master/Project/anes_timeseries_2016_rawdata.txt
I created sample data with selected columns that i believed have effect on presidential elections. The sample data survey questions are respondent’s gender , whether respondent voted for previous presidential election, health insurance, economic review , high school graduation status , marriage status , whether respondent believe in built a wall ,allow refugues to come USA, belive in global warming, how often goes Church, how frequently watch media,tempature to Democrat party.
I also uploaded data variable user guidance to understand each variable’s characteristics in my github :https://github.com/omerozeren/DATA606/blob/master/Project/anes_timeseries_2016_userguidecodebook.pdf .
# Load data via a pipe-delimited file, using character type for all columns
url <-"https://raw.githubusercontent.com/omerozeren/DATA606/master/Project/anes_timeseries_2016_rawdata.txt"
rawData <- read_delim(url,"|",
col_types= paste(rep("c",1290),sep="",collapse=""))
# Empty data frame
sample <- data.frame(ID = as.character(rawData$V160001))
sample$sex <- fct_recode(rawData$V161002,
"Male" = " 1", "Female" = " 2", NULL = "-1")
sample$votePre <- fct_recode(rawData$V161005,
"Yes" = " 1", "No" = " 2",
NULL = "-8", NULL = "-9")
sample$health_ins <- fct_recode(rawData$V161112,
"Yes" = " 1", "No" = " 2",
NULL = "-8", NULL = "-9")
sample$heat_to_democ <- rawData$V162078
sample$econ_review <- fct_recode(rawData$V161139,
"Very good" = " 1",
"Good" = " 2",
"Neither good nor bad" = " 3",
"Bad" = " 4",
"Very bad" = " 5",
"Don't know" = "-8",
"Refused" = "-9")
sample$high_school <- fct_recode(rawData$V161271,
"Yes" = " 1",
"No" = " 2",
"Inapplicable" = "-1",
"Refused" = "-9")
sample$marriage <- fct_recode(rawData$V161268,
"Married" = " 1",
"Married" = " 2",
"Widowed" = " 3",
"Divorced" = " 4",
"Separated" = " 5",
"Never married" = " 6",
"Refused" = "-9")
sample$built_wall <- fct_recode(rawData$V161196x,
"Favor a great deal" = " 1",
"Favor a moderate amount" = " 2",
"Favor a little" = " 3",
"Neither favor nor oppose" = " 4",
"Oppose a little" = " 5",
"Oppose a moderate amount" = " 6",
"Oppose a great deal" = " 7",
"Don't know" = "-8",
"Refused" = "-9")
sample$allow_refugees <- fct_recode(rawData$V161214x,
"Favor a great deal" = " 1",
"Favor a moderate amount" = " 2",
"Favor a little" = " 3",
"Neither favor nor oppose" = " 4",
"Oppose a little" = " 5",
"Oppose a moderate amount" = " 6",
"Oppose a great deal" = " 7",
"Don't know" = "-8",
"Refused" = "-9")
sample$global_warming <- fct_recode(rawData$V161221,
"Yes" = " 1",
"No" = " 2",
"Don't know" = "-8",
"Refused" = "-9")
sample$church <- fct_recode(rawData$V161245,
"Every week" = " 1",
"Almost every week" = " 2",
"Once or twice a month" = " 3",
"A few times a year" = " 4",
"Never" = " 5",
"Inapplicable" = "-1",
"Refused" = "-9")
sample$media <- fct_recode(rawData$V161009,
"A great deal" = " 1",
"A lot" = " 2",
"A moderate amount" = " 3",
"A little" = " 4",
"None at all" = " 5",
"Inapplicable" = "-1",
"Refused" = "-9")
sample$clinton <- as.numeric(rawData$V161086)
sample$clinton[sample$clinton < 0] <- NA
sample$trump <- as.numeric(rawData$V161087)
sample$trump[sample$trump < 0] <- NAResearch Question
Does a correlation exist between a voter’s characteristics/reviews to the presidential candidates in the 2016 election?
Cases
Each case is a person surveyed. There are 4270 observations and 1290 variables in the data set.
Data Collection
Data was collected by survey.
Type of Study
This is an observational study.
Data Source
Data was retreived from the ANES website
Response
The response variable is rating of candidates “Trump ,Clinton” that we will be using numeric(Quantitative).
Explanatory
The explanatory variables are the respondent’s frequency of use of various media (web, print, TV), repondent’s gender status, repondent’s previous election vote status,health insurance status, economic review , high school graduation status , marriage status , belive in build a wall , global warming, and allowing refegues , frequency of going church,tempature to democrat party(Quantitative).
Relevant summary statistics
sample %>% filter(!is.na(clinton), !is.na(trump)) %>% summary(.)## ID sex votePre health_ins heat_to_democ
## 1 : 1 Male : 558 Yes :3082 Yes :3835 Length:4208
## 2 : 1 Female: 616 No :1110 No : 367 Class :character
## 3 : 1 NA's :3034 NA's: 16 NA's: 6 Mode :character
## 4 : 1
## 5 : 1
## 6 : 1
## (Other):4202
## econ_review high_school marriage
## Don't know : 3 Inapplicable:3421 Refused : 25
## Refused : 4 Refused : 0 Married :2116
## Very good : 67 Yes : 651 Widowed : 295
## Good : 953 No : 136 Divorced : 656
## Neither good nor bad:1354 Separated : 93
## Bad :1356 Never married:1023
## Very bad : 471
## built_wall allow_refugees
## Oppose a great deal :1494 Oppose a great deal :1355
## Neither favor nor oppose: 911 Neither favor nor oppose:1171
## Favor a great deal : 871 Oppose a moderate amount: 519
## Favor a moderate amount : 405 Favor a moderate amount : 431
## Oppose a moderate amount: 345 Favor a great deal : 386
## Oppose a little : 84 Favor a little : 158
## (Other) : 98 (Other) : 188
## global_warming church media
## Don't know: 31 Inapplicable :1694 Inapplicable : 74
## Refused : 19 Refused : 3 Refused : 1
## Yes :3410 Every week : 779 A great deal : 903
## No : 748 Almost every week : 533 A lot :1057
## Once or twice a month: 463 A moderate amount:1376
## A few times a year : 688 A little : 727
## Never : 48 None at all : 70
## clinton trump
## Min. : 0.00 Min. : 0.00
## 1st Qu.: 2.00 1st Qu.: 0.00
## Median : 40.00 Median : 30.00
## Mean : 42.07 Mean : 36.98
## 3rd Qu.: 70.00 3rd Qu.: 70.00
## Max. :100.00 Max. :100.00
## High Scholl graduated and global warming belivers
# summary statistics
ratings <- sample %>% gather(key = "candidate", value = "rating",
clinton, trump) %>%
filter(high_school == "Yes", global_warming == "Yes") %>%
select(media, candidate, rating)
# Box plot of individual candidate ratings
ratings %>% filter(!is.na(rating)) %>%
ggplot(aes(x=candidate, y=rating, fill=candidate)) +
geom_boxplot(fill=c("blue","red")) +
ggtitle("Respondent Ratings of Candidates") +
xlab("Candidate") +
ylab("Rating") +
scale_x_discrete(labels=c("Hillary Clinton","Donald Trump"))
# Box plot of Trump ratings
ratings %>% filter(candidate=="trump", !is.na(rating)) %>%
ggplot(aes(x=media, y=rating)) +
geom_boxplot(fill="grey") +
ggtitle("Respondent Ratings of Donald Trump") +
xlab("How frequently watch Media") +
ylab("Rating") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
# Box plot of Clinton ratings
ratings %>% filter(candidate=="clinton", !is.na(rating)) %>%
ggplot(aes(x=media, y=rating)) +
geom_boxplot(fill="grey") +
ggtitle("Respondent Ratings of Hillary Clinton") +
xlab("How frequently watch Media") +
ylab("Rating") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))