06-EDA_Graphs.Rmd

# Exploratory data analysis


## Prepare folder and data

### Set the working directory

## Prepare folder and data

## Set the working directory

This can be done in 2 ways:

1. Using codes
2. Using point and click


To use point and click, use the down arrow button next to *More* . Then click 'Set as working directory'

### List the files inside the working directory

All files will be displayed when you click 'Files'.

Or you can use this code,

```{r}
list.files()
```

### Reading dataset from SPSS file (.sav) 

Dataset in SPSS format will end with .sav. To read SPSS data into R we use 'foreign' library. 

Create a object to represent the SPSS data that we will read into R.  

```{r}
library(foreign)
dataSPSS<-read.spss('qol.sav', to.data.frame = TRUE)
```


## Describing data

Let us examine the data

```{r}
str(dataSPSS)
```

Now, let us summarize our data

```{r}
summary(dataSPSS)
```

## Graphing or Plotting data

You must ask yourselves these:
1.  Which variable do you want to plot?
2.  What is the type of that variable? Factor? Numerical?
3.  Are you going to plot another variable together?

### One variable: A categorical or factor variable

We can create a simple barchart

```{r}
dist.sex<-table(dataSPSS$sex)
barplot(dist.sex,
        main='Sex distribution',
        xlab='Sex')
```


### One variable: A numerical variable

histogram

```{r}
hist(dataSPSS$age, main = 'Age',
     xlab='Age in years',
     ylab='Count')
```

### Two variables : A numerical with another numerical variable

We will use *scatterplot* to plot 

```{r}
plot(dataSPSS$tahundx, dataSPSS$age,
     main = 'Duration having DM VS age',
     xlab = 'Duration of DM', ylab = 'Age',
     pch = 19)
```

Let us make a fit line

```{r}
plot(dataSPSS$tahundx, dataSPSS$age,
     main = 'Duration having DM VS age',
     xlab = 'Duration of DM', ylab = 'Age',
     pch = 19)
abline(lm(dataSPSS$age~dataSPSS$tahundx), col = 'red')
```

and a lowess 

```{r}
plot(dataSPSS$tahundx, dataSPSS$age,
     main = 'Duration having DM VS age',
     xlab = 'Duration of DM', ylab = 'Age',
     pch = 19)
lines(lowess(dataSPSS$tahundx,dataSPSS$age), col = 'blue')
```

### Two variables : A categorical variable with a categorical variable

Now, we will plot 2 categorical variables simultenously.

First, we will use stacked barchart

```{r}
compl.sex<-table(dataSPSS$complica,dataSPSS$sex)
compl.sex
barplot(compl.sex,
        main='Complications by sex',
        xlab='Sex',
        col=c('blue','red'),
        legend=c('No','Yes'))
```

Next, we will use grouped barchart

```{r}
compl.sex
barplot(compl.sex,
        main = 'Complications according to sex',
        xlab = 'Sex',
        col = c('blue','red'),
        legend = c('no','yes'),
        beside = TRUE)
```