Data Visualization_summary.Rmd

---
title: "Data Visualization"
author: "Alex"
date: "2023-09-28"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(warning=FALSE, message=FALSE)
knitr::opts_chunk$set(echo = TRUE)
```

## 3. Data Visualization

The data is the life expectation.
* Year: the year of birth
* Race: the races represented in the measured population
* Sex: the sex of the measured population
* Avg_Life_Expec: the average life expectancy, in years, at birth of people born in a given year
* Age_Adj_Death_Rate: the age adjusted death rate of people born in a given year — the age adjusted death rate is a metric that adjusts death rate for populations' age distributions to make population comparisons equitable

Before data visualization, data has to be cleaned. 


### 3.1 Read data and check the data

```{r}
library(readr)
data <- read_csv('life_expec.csv')

# Class of each column
library(purrr)
data %>% map_df(class)

# Check missing values
colSums(is.na(data))

# Check the unique categorical values
library(dplyr)
data %>% distinct(Race)
data %>% distinct(Sex)

```


### 3.2 Data Visualization

#### 3.2.1 Data distribution (one colume)

```{r}
library(ggplot2)

# Histogram
data %>% ggplot(aes(x=Avg_Life_Expec)) + geom_histogram(bins=20)
data %>% ggplot(aes(x=Avg_Life_Expec, fill=Race)) + geom_histogram(bins=20)


# Density
data %>% ggplot(aes(x=Avg_Life_Expec)) + geom_density()
data %>% ggplot(aes(x=Avg_Life_Expec, fill=Race)) + geom_density(alpha=0.3) 
data %>% ggplot(aes(x=Avg_Life_Expec, color=Race)) + geom_density()

# Barplot, good for discrete data, count different categories

data %>% ggplot(aes(x=Race, fill='red')) + geom_bar() +  theme_bw()


```
#### 3.2.2 Data trends
```{r}
# Line plot
library(ggplot2)
data %>% ggplot(aes(x=Year, y=Avg_Life_Expec)) + geom_line()

data %>% filter(Race=='All Races' & Sex != 'Both Sexes') %>% ggplot(aes(x=Year, y=Avg_Life_Expec, color=Sex)) + geom_line() 

data %>% filter(Race=='All Races' & Sex != 'Both Sexes') %>% ggplot(aes(x=Year, y=Avg_Life_Expec, lty=Sex)) + geom_line()


data %>% filter(Race!='All Races' & Sex != 'Both Sexes') %>% ggplot(aes(x=Year, y=Avg_Life_Expec, color= Race, lty=Sex)) + geom_line()

# Scatter plot
data %>% ggplot(aes(x=Year, y=Avg_Life_Expec)) + geom_point()

# Smooth
data %>% ggplot(aes(x=Year, y=Avg_Life_Expec)) + geom_point() + geom_smooth(method = 'lm', se=FALSE)

# boxplot
data %>% ggplot(aes(x=Avg_Life_Expec)) + geom_density() 


# boxplot
data %>% ggplot(aes(x=Sex, y=Avg_Life_Expec)) + geom_boxplot() 
data %>% ggplot(aes(x=Race, y=Avg_Life_Expec)) + geom_boxplot()


# Multiple figures
data  %>% ggplot(aes(x=Year, y=Avg_Life_Expec)) + geom_line() + facet_wrap(vars(Sex), ncol=2)

data  %>% ggplot(aes(x=Year, y=Avg_Life_Expec)) + geom_line() + facet_grid(vars(Sex))

data  %>% ggplot(aes(x=Year, y=Avg_Life_Expec)) + geom_line() + facet_grid(rows=vars(Sex), cols=vars(Race))

```

#### 3.2.3 Aesthetics

```{r}

# Add title, xlabel, ylabel and adjust the position, fontsize, and color

data %>% ggplot(aes(x=Year, y=Avg_Life_Expec)) + geom_line(color='red', linewidth=0.2, linetype='dashed')+ labs(title='Average life expectation vs Year', x= 'Year', y='Average life expectation') + theme_bw() + theme(plot.title=element_text(hjus=0.5, size=18, color='orange'), axis.title=element_text(size=16, color='blue'), axis.text = element_text(size=14, color='red')) 



# Add legend
library(tidyr)
df <- data %>% mutate(Age_Adj_Death_Rate=Age_Adj_Death_Rate/10) %>% pivot_longer(cols=c(Avg_Life_Expec, Age_Adj_Death_Rate), names_to='column', values_to='value')

df %>% filter(Race == 'All Races' & Sex == 'Both Sexes') %>% ggplot(aes(x=Year, y=value, lty=column)) + geom_line()+scale_color_manual(
    name='Trend', 
    values= c('forestgreen', 'orangered'),
    labels=c('Avg. Life Expectancy', 'Death Rate (Age Adjusted)')
) +
  theme(legend.position = 'top')


```