md-07a-pivoting-complete.qmd

---
title: "Data Visualisation"
author: ""
format:
  html:
    embed-resources: true
editor: visual
execute:
  warning: false
editor_options: 
  chunk_output_type: console
---

# Part 1: Data preparation

## Task 1: Load packages

The required packages for this homework exercise have already been added.

1.  Run the code chunk below to load the required packages. Tipp: Click on the green play button in the top right corner of the code chunk.
2.  What's the `tidyverse` Package? Describe in maximum two sentences below.

```{r}
library(tidyverse)
```

## Task 2: Import data

1.  Use the `read_csv()` (**Note**: Watch out for the `_` and don't use the `.` as in `read.csv()`) function to import the "msw-generation-and-composition-by-income.csv" data from the `data` directory and assign it to an object with the name `waste_data`.

```{r}
waste_data <- read_csv("data/msw-generation-and-composition-by-income.csv")
```

## Task 3: Vector coersion

1.  Use `waste_data` and `count()` to create a frequency table for the `income_cat` variable.

2.  Use then `c()` function to create a vector with a sensible order for the the values in `income_cat`. Use the assignment operator `<-` to assign the vector to an object with the name `levels_income_cat`.

3.  Starting with the `waste_data` object, use the pipe operator and the `mutate()` function to convert the `income_cat` variable from a variable of type character to a variable of type factor. Use the levels you defined in the previous step.ories using the following code to identify the correct spelling of the categories in the variable `income_cat`.

4.  Assign the created data frame to an object with the name `waste_data_fct`.

5.  Render and fix any errors

```{r}

waste_data |> 
  count(income_cat)

levels_income_cat <- c("high income", 
                       "upper-middle income",
                       "lower-middle income",
                       "low income")

waste_data_fct <- waste_data |> 
  mutate(income_cat = factor(income_cat, levels = levels_income_cat))

```

## Task 4: From wide to long

1.  Starting with the `waste_data_fct` object, use the `pivot_longer()` function to convert the data frame from a wide to a long format. Apply the following:

-   bring all columns from `food_organic_waste` to `yard_garden_green_waste` into a long format
-   send the variable names to a column named "waste_category"
-   send the values of the variables to a column named "percent"

2.  Remove all `NA`s from the `percent` variable

3.  Assign the created dataframe to an object with the name `waste_data_long`

4.  Render and fix any errors

```{r}

waste_data_long <- waste_data_fct |> 
  pivot_longer(cols = food_organic_waste:yard_garden_green_waste,
               names_to = "waste_category",
               values_to = "percent") |> 
  filter(!is.na(percent))
```

# Part 2: Data summary

## Task 1: Import data

I have stored the data that I would have expected at the end of the previous task and import it here.

1.  Run the code in the code chunk below.

```{r}
waste_data_long <- read_rds("data/msw-generation-and-composition-by-income-long.rds")
```

## Task 2: Summarise data

1.  Starting with `waste_data_long`, group the data by`income_cat` and `waste_category`, then create a summary table containing the mean of percentages (call this mean_percent) for each group.

2.  Assign the created data frame to an object with the name `waste_data_long_mean`.

```{r}

waste_data_long_mean <- waste_data_long |> 
  group_by(income_cat, waste_category) |> 
  summarise(mean_percent = mean(percent)) 

```

## Task 3: Table display

1.  Starting with the `waste_data_long_mean` object, execute the code and observe the output in the Console. Would you publish this table in this format in a report?

```{r}
waste_data_long_mean
```

## Task 4: From long to wide

1.  Starting with the `waste_data_long_mean` object, use the pipe operator to add another line of code which uses the `pivot_wider()` function to bring the data from a long format into a wide format using names for variables from `waste_category` and corresponding values from `mean_percent`

2.  Execute the code and observe the output in the Console. Would you publish this table in a report in this format?

3.  Render and fix any errors

```{r}
waste_data_long_mean |> 
  pivot_wider(names_from = waste_category,
              values_from = mean_percent) 


```

# Part 3: Data visualization (stretch goal)

## Task 1: Import data

I have stored the data that I would have expected at the end of the previous task and import it here.

1.  Run the code in the code chunk below.

```{r}
waste_data_long_mean <- read_rds("data/msw-generation-and-composition-by-income-long-mean.rds")
```

## Task 2: Reproduce a plot

1.  Render and fix any errors.

2.  Reproduce the plot that you see as an image below when you render the report and view the output in your Viewer tab in the bottom right window.

**Hint:** To get those bars displayed next to each other, use the `geom_col()` function and apply the `position = position_dodge()` argument and value. The colors don't have to be exactly the same colors, just not the default color scale.

**Note:** The size of the plot will be different. That is alright and does not need to match.

```{r}
ggplot(data = waste_data_long_mean, 
       mapping = aes(x = mean_percent, 
                     y = waste_category, 
                     fill = income_cat)) + 
  geom_col(position = position_dodge()) + 
  labs(title = "Waste Composition", 
       subtitle = "Mean percentages of nine waste categories displayed by income categories",
       x = "mean (percent)", 
       y = "waste category", 
       fill = "Income category", 
       caption = "Data from: https://datacatalog.worldbank.org/search/dataset/0039597") + 
  scale_x_continuous(breaks = seq(0, 50, 5)) + 
  scale_fill_brewer(type = "qual", palette = 3) +
  theme_minimal() + 
  theme(panel.grid.minor = element_blank(), 
        panel.grid.major.y = element_blank())

  
```

![](figs/fig-01-mean-waste-composition-by-income.png)

# Part 4: Complete the assignment

## Task 1: Data communication

**In the YAML header (between the three dashes at the top of the document)**

1.  Add your name as the author of this document
2.  Render the document and fix any errors

## Task 2: Stage, Commit & Push to GitHub

1.  Open the Git pane in RStudio. It's in the top right corner in a separate tab.

2.  **Stage** your changes by checking appropriate box next to all files (if you select one file with your mouse, you can then highlight them all with Ctrl + A on your keyboard and check all boxes)

3.  Write a meaningful commit message (e.g. "Completed homework assignment 07.) in the **Commit message** box

4.  Click **Commit**. Note that every commit needs to have a commit message associated with it.

### Open an issue on GitHub

Once you have ensured that the Quarto document renders without errors and you have pushed all your changes to GitHub, you can complete the assignment by opening an issue on

1.  Open [github.com](https://github.com/) in your browser.

2.  Navigate to the GitHub organisation for the course.

3.  Find the repository md-07-assignments that ends with your GitHub username.

4.  Click on the "Issues" tab.

5.  Click on the green "New issue" button.

6.  In the "Title" field write: "Completed module 7 assignments".

7.  In the "Leave a comment" field, tag the course instructors @larnsce @mianzg @sskorik01 and ask some questions, if you like.