ae-11b-data-visualisation.qmd

---
title: "Programming Exercise - Data Visualisation"
format: html
editor: visual
bibliography: references.bib
execute:
  echo: true
  message: false
  warning: false
editor_options: 
  chunk_output_type: console
---

# Task 0: Load R packages

1.  Add a new code-chunk below the three outlined steps for Task 0

2.  Load the `gapminder`, `ggplot2` and `dplyr` R Packages

3.  Run the code contained in the code-chunk and fix any errors



# Task 1: Data import

We are not importing any external data, we will use data from the `gapminder` R Package. The step of "Data import" is listed to remind us of this step in the data science lifecycle.

# Task 2: Data tidying

The data is tidy and we can directly work with it. The step is here as reminder.

# Task 3: Data transformation

## Data for 2007

**Fill in the gaps**

1.  A code-chunk has already been created

2.  Start with the `gapminder` object

3.  Add the pipe operator (`%>%`) and on a new line use the `filter()` function to narrow down the data for observation of the year 2007

4.  Use the assignment operator (`<-`) to assign the data to an object named `gapminder_yr_2007`

5.  Run the code contained in the code-chunk and fix any errors

6.  Next to the code-chunk option `#| eval:` change the value from `false` to `true`

7.  Render

8.  Add, Commit, Push

```{r}
#| eval: false

# A subset only for 2007

___ <- ___ %>% 
  ___(___ == ___)
```

## Median life expectancy and country frequency by continent

**Fill in the gaps**

1.  A code-chunk has already been created

2.  Start with the `gapminder_yr_2007` object

3.  Add the pipe operator (`%>%`) and on a new line use the `group_by()` function to group the following operations by the variable continent.

4.  Add the pipe operator (`%>%`) and on a new line use the `summarise()` function to summarise the data for the count and the median life expectancy.

5.  Use the assignment operator (`<-`) to assign the data to an object named `gapminder_summary_continent_2007`

6.  Run the code contained in the code-chunk and fix any errors

7.  Next to the code-chunk option `#| eval:` change the value from `false` to `true`

8.  Render

9.  Add, Commit, Push

```{r}
#| eval: false

# A subset for median life expectancy in 2007 by continent

gapminder_summary_continent_2007 <- gapminder_yr_2007 %>% 
  ___(___) %>% 
  ___(
    count = n(),
    lifeExp = median(___)
  )

```

## Median life expectancy by continent and year

**Fill in the gaps**

1.  A code-chunk has already been created

2.  Start with the `gapminder` object

3.  Add the pipe operator (`%>%`) and on a new line use the `group_by()` function to group the following operations by the variables continent and year.

4.  Add the pipe operator (`%>%`) and on a new line use the `summarise()` function to summarise the data for the median life expectancy and name the new colum 'lifeExp'.

5.  Use the assignment operator (`<-`) to assign the data to an object named `gapminder_summary_continent_year`

6.  Run the code contained in the code-chunk and fix any errors

7.  Next to the code-chunk option `#| eval:` change the value from `false` to `true`

8.  Render

9.  Add, Commit, Push

```{r}
#| eval: false

# A subest for median life expectancy by continent and year

gapminder_summary_continent_year <- gapminder %>%
  ___(continent, ___) %>% 
  summarise(___ = ___(___))

```

# Task 4: Data visualisation

## Boxplot

1.  Add a new code-chunk below

2.  Use the `ggplot()` function and the `gapminder_yr_2007` object to create a boxplot with the following aesthetic mappings:

-   continent to the x-axis;
-   life expactancy to the y-axis;
-   continent to color using the `fill = continent` argument inside `aes()`

3.  Do not display (ignore) the outliers in the plot.

4.  Run the code contained in the code-chunk and fix any errors



## Timeseries plot

1.  Add a new code-chunk below

2.  Use the `ggplot()` function and the `gapminder_summary_continent_year` object to create a connected scatterplot (also timeseries plot) using the `geom_line()` and `geom_point()`functions with the following aesthetic mappings:

-   year to the x-axis;
-   life expactancy to the y-axis;
-   continent to color using the `color = continent` argument inside `aes()`

3.  Run the code contained in the code-chunk and fix any errors



## Barplot

### with `geom_col()`

1.  Add a new code-chunk below

2.  Use the `ggplot()` function and the `gapminder_summary_continent_2007` object to create a barplot using the `geom_col()` function with the following aesthetic mappings:

-   continent to the x-axis;
-   count to the y-axis;

3.  Run the code contained in the code-chunk and fix any errors



### with `geom_bar()`

1.  Add a new code-chunk below

2.  Use the `ggplot()` function and the `gapminder_yr_2007` object to create a barplot using the `geom_bar()` function with the following aesthetic mappings:

-   continent to the x-axis;

3.  Run the code contained in the code-chunk and fix any errors

4.  The plot is identical to the plot created with `geom_col()`. Why? What does the `geom_bar()` function do? Write your text here:




## Histogram

1.  Add a new code-chunk below

2.  Use the `ggplot()` function and the `gapminder_yr_2007` object to create a histogram using the `geom_histogram()` function with the following aesthetic mappings:

-   life expactancy to the x-axis;
-   continent to color using the `fill = continent` argument inside `aes()`

3.  Run the code contained in the code-chunk and fix any errors.

4.  Inside the `geom_histogram()` function, add the following arguments and values:

-   `col = "grey30"`
-   `breaks = seq(40, 85, 2.5)`

5.  Run the code contained in the code-chunk and fix any errors.

6.  Describe how the `geom_histogram()` function is similar to the `geom_bar()` function.

7.  What happens by adding the 'breaks' argument? Play around with the numbers inside of `seq()` to see what changes.




## Scatterplot and faceting

1.  Use the `ggplot()` function and assign `gapminder_yr_2007` to the data argument and create a scatterplot using the `geom_point()`function with the following aesthetic mappings:

-   gdpPercap the x-axis;
-   lifeExp to the y-axis;
-   population to the size argument;
-   country to color using the `color = continent` argument inside `aes()`

2.  Run the code contained in the code-chunk and fix any errors.

3.  Use the variable continent to facet the plot by adding: `facet_wrap(~continent)`

4.  Run the code contained in the code-chunk and fix any errors

5.  Describe in your own words what the `facet_wrap()`function does

6.  Adapt the scaling of the size by adding `scale_size(range = c(2, 12))` on the next line

7.  Run the code contained in the code-chunk and fix any errors.

8.  Adapt the scaling of the x-axis to log10 by adding `scale_x_log10(labels = scales::label_log(), limits = c(100, 100000))` on the next line

9.  Run the code contained in the code-chunk and fix any errors




## Lineplot and faceting

1.  A code-chunk with complete code has already been prepared

2.  Run the code contained in the code-chunk and fix any errors

3.  Remove \# sign at the line that starts with the `scale_color_manual()` function

4.  What is stored in the `country_colors` object? Find out by executing the object in the Console (type it to the Console and hit enter). Do the same again, but with a question mark `?country`.

5.  Next to the code-chunk option `#| eval:` change the value from `false` to `true`

6.  Render

```{r}
#| eval: false

ggplot(data = gapminder,
       mapping = aes(x = year, 
                     y = lifeExp, 
                     group = country, 
                     color = country)) +
  geom_line(lwd = 1, show.legend = FALSE) + 
  facet_wrap(~continent) +
  # scale_color_manual(values = country_colors) +
  theme_minimal() 

```

## Choropleth Maps

You can also prepare maps with `ggplot2`. It's beyond the scope of the class to teach you the foundations of spatial data in R, but a popular package to work with spatial data is the `sf` (Simple Feautures) R Package. The `rnaturalearth` R Package facilitates world mapping by making [Natural Earth](https://www.naturalearthdata.com/) map data more easily available to R users.

The code-chunk below contains code that for a world map that shows countries by income group. If you want to view that map, do the following:

1.  Install the `sf` and the `rnaturalearth` R Packages.

2.  Load the R Packages at the beginning of this file alongside the other loaded R Packages.

3.  Run the code contained in the code-chunk and fix any errors

4.  Next to the code-chunk option `#| eval:` change the value from `false` to `true`

5.  Render

6.  Add, Commit, Push

```{r}
#| eval: false

world <- ne_countries(scale = "small", returnclass = "sf")

world %>% 
  mutate(income_grp = factor(income_grp, ordered = T)) %>% 
  ggplot() + 
  geom_sf(aes(fill = income_grp)) +
  theme_void() +
  theme(legend.position = "top") +
  labs(fill = "Income Group:") +
  guides(fill = guide_legend(nrow = 2, byrow = TRUE))


```

The code for the code-chunk is taken from here: More here: https://bookdown.org/alhdzsz/data_viz_ir/maps.html

## Working with spatial data in R

If you are interested in working with spatial data in R, then I recommend the following resources:

-   Geocompuation with R - Book: https://geocompr.robinlovelace.net/
-   Simple Features for R - Article: https://r-spatial.github.io/sf/articles/sf1.html
-   tmap: thematic maps in R - R Package: https://r-tmap.github.io/tmap/

# Task 5: Data communication

1.  Use the Quarto Documentation Guide for HTML format to figure out how to add a theme to this Quarto document.

https://quarto.org/docs/guide/

2.  Apply a theme to this Quarto document.

# Task 6: Complete assignment

Open an issue on the repo for this exercise to let us know you completed it. Use the @larnsce mention.