NEW statproject(1).Rmd

---
title: "Untitled"
output: html_document
---

```{r}
library(readr)
mydata <- read_csv("Nutrition__Physical_Activity__and_Obesity_-_Behavioral_Risk_Factor_Surveillance_System.csv")
#import dataset

```
```{r}
head(mydata)
names(mydata)

```
```{r}
mydata2=mydata[,c(1,4,8,11,20,30)] # select the data that I want
mydata2
names(mydata2)
```
```{r}
mydata3=which(mydata2[,6]=="Education") #we are interested in Education,just select education from stratificaitonCategory1
data1=mydata2[mydata3,]#combine the data into our data set 
head(data1)
names(data1)
```
```{r}
data2=which(data1[,3]=="Percent of adults aged 18 years and older who have obesity")#choose our question 

question=data1[data2,] 
head(question) # select our question 

```

```{r}
table(question$LocationDesc)
data3=which(question[,2]!="National") #remove national
data_location=question[data3,]
data_location
data4=which(data_location[,2]!="Guma")#remove Guma 
data_location2=data_location[data4,]
data_location2
data5=which(data_location2[,2]!="Puerto Rico") #remove Puerto Rico 
finaldata=data_location2[data5,]

finaldata
```

```{r}
#datavisulaization 

data11=which(finaldata[,1]=="2011") #choose year 2011 data 
data2011=finaldata[data11,]
data2011=data2011[,c(2,4,5)] # choose relevant data 
data2011

data_edu=which(data2011[,3]=="College graduate") # choose college graduate
data_college=data2011[data_edu,]
data_college
```
# 2012 year 
```{r}
data12=which(finaldata[,1]=="2012")
data2012=finaldata[data12,]
data2012=data2012[,c(2,4,5)]
data_edu12012=which(data2012[,3]=="College graduate")
data_college2012=data2012[data_edu12012,]
data_college2012
```
```{r}
#2013
data13=which(finaldata[,1]=="2013")
data2013=finaldata[data13,]
data2013=data2013[,c(2,4,5)]
data2013
data_edu2013=which(data2013[,3]=="College graduate")
data_college2013=data2013[data_edu2013,]
data_college2013


```
```{r}
#2014
data14=which(finaldata[,1]=="2014")
data2014=finaldata[data14,]
data2014=data2014[,c(2,4,5)]
data2014
data_edu2014=which(data2014[,3]=="College graduate")
data_college2014=data2014[data_edu2014,]
data_college2014
```
```{r}
#2015
data15=which(finaldata[,1]=="2015")
data2015=finaldata[data15,]
data2015=data2015[,c(2,4,5)]
data2015
data_edu2015=which(data2015[,3]=="College graduate")
data_college2015=data2015[data_edu2015,]
data_college2015
```
```{r}
#2016
data16=which(finaldata[,1]=="2016")
data2016=finaldata[data16,]
data2016=data2016[,c(2,4,5)]
data2016
data_edu2016=which(data2016[,3]=="College graduate")
data_college2016=data2016[data_edu2016,]
data_college2016
```
```{r}
#college graduate & BMI
library(ggplot2)
F_2011=ggplot(data_college,mapping=aes(x=Data_Value,y=LocationDesc))+geom_bar(stat = "identity",color="black") #data visualization 
F_2011+ggtitle("2011")+xlab("College$BMI")

F_2012=ggplot(data_college2012,mapping = aes(x=Data_Value,y=LocationDesc))+geom_bar(stat="identity",color="black")
F_2012+ggtitle("2012")+xlab("College$BMI")

F_2013=ggplot(data_college2013,mapping=aes(x=Data_Value,y=LocationDesc))+geom_bar(stat = "identity",color="black")
F_2013+ggtitle("2013")+xlab("College$BMI")

F_2014=ggplot(data_college2014,mapping=aes(x=Data_Value,y=LocationDesc))+geom_bar(stat = "identity",color="black")
F_2014+ggtitle("2014")+xlab("College$BMI")

F_2015=ggplot(data_college2015,mapping=aes(x=Data_Value,y=LocationDesc))+geom_bar(stat = "identity",color="black")
F_2015+ggtitle("2015")+xlab("College$BMI")

F_2016=ggplot(data_college2016,mapping=aes(x=Data_Value,y=LocationDesc))+geom_bar(stat = "identity",color="black")
F_2016+ggtitle("2016")+xlab("College$BMI")





```



```{r}
install.packages("ggpubr")
library("ggpubr")
```

```{r}
par(mfrow=c(1,2))
ggdensity(finaldata$Data_Value,xlab = "Obesity rate",main="Density plot of obesity rate") #test normality 
ggqqplot(finaldata$Data_Value)
shapiro.test(finaldata$Data_Value) # our data is normally distributed 
```

```{r}
# high school or lower high school degree=0, otherwise =1

table(finaldata$Education)

Less_High=which(finaldata$Education=='Less than high school') #choose less than high school graduate and equal 0
finaldata$Education[Less_High]=0

High=which(finaldata$Education=='High school graduate')
finaldata$Education[High]=0


College_Tech=which(finaldata$Education=='Some college or technical school')
finaldata$Education[College_Tech]=1

Collge_graduate=which(finaldata$Education=='College graduate')
finaldata$Education[Collge_graduate]=1

finaldata
finaldata$Education=as.numeric(finaldata$Education) #convert string to numeric 
typeof(finaldata$Education) 
```

```{r}

par(mfrow=c(1,2))
hist(finaldata$Data_Value,xlab = "BMI",main="Histogram of BMI")
boxplot(finaldata$Data_Value~finaldata$Education,xlab = "Education Level",ylab = "BMI")
boxplot(finaldata$Data_Value~finaldata$YearStart,xlab="Years",ylab="BMI")

```
```{r}
# BMI <25 =0, BMI>25=1
finaldata$Data_Value[finaldata$Data_Value<=25]<-0
finaldata$Data_Value[finaldata$Data_Value>25]<-1
finaldata
```

```{r}
#Rregression 

fit_lm=glm((finaldata$Data_Value)~finaldata$YearStart+finaldata$Education,family=binomial()) # Logistic model 
summary(fit_lm)
anova(fit_lm)

coef(fit_lm)

```

```{r}
#test VIF
install.packages("MPV")
install.packages("car")
library("car")
library("MPV")
vif(fit_lm)
```
```{r}





```









Note that the `echo = FALSE` parameter was added to the code chunk to prevent printing of the R code that generated the plot.