02-Transform.Rmd

---
title: "Transform Data"
output: html_notebook
---

<!-- This file by Charlotte Wickham is licensed under a Creative Commons Attribution 4.0 International License, adapted from the orignal work at https://github.com/rstudio/master-the-tidyverse by RStudio. -->

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

# Toy dataset to use
pollution <- tribble(
       ~city,   ~size, ~amount, 
  "New York", "large",      23,
  "New York", "small",      14,
    "London", "large",      22,
    "London", "small",      16,
   "Beijing", "large",      121,
   "Beijing", "small",      56
)
```

## gapminder

```{r}
gapminder
```

## Your Turn 1

See if you can use the logical operators to manipulate our code below to show:

The data for United States
```{r}
filter(gapminder, country == "New Zealand")
```

All data for countries in Oceania
```{r}
filter(gapminder, country == "New Zealand")
```

Rows where the life expectancy is greater than 82
```{r}
filter(gapminder, country == "New Zealand")
```


## Your Turn 2

Use Boolean operators to alter the code below to return only the rows that contain:

* United States before 1970

```{r}
filter(gapminder, country == "New Zealand", year > 2000)
```

*  Countries where life expectancy in 2007 is below 50

```{r}
filter(gapminder, country == "New Zealand", year > 2000)
```

* Records for any of "New Zealand", "Canada" or "United States"

```{r}
filter(gapminder, country == "New Zealand", year > 2000)
```

## Your Turn 3

Use `filter()` to get the records for the US, then plot the life expectancy over time.

```{r}
gapminder
```

## Your Turn 4

Find the records with the smallest population.
```{r}

```

Find the records with the largest GDP per capita.
```{r}

```

## Quiz 

A function that returns a vector the same length as the input is called **vectorized**.

Which of the following functions are vectorized?

  * `ifelse()`
  * `diff()`
  * `sum()`

You might try these:
```{r}
gapminder %>% 
  mutate(size = ifelse(pop < 10e06, "small", "large"))
```

```{r, error = TRUE}
gapminder %>% 
  mutate(diff_pop = diff(pop))
```

```{r}
gapminder %>% 
  mutate(total_pop = sum(as.numeric(pop)))
```

## Your Turn 5

Alter the code to add a `prev_lifeExp` column that contains the life expectancy from the previous record.

(Hint: use cheatsheet, you want to offset elements by one)

Extra challenge: Why isn't this quite the 'life expectency five years ago'?

```{r}
gapminder %>%
  mutate()
```

## Your Turn 6

Use summarise() to compute three statistics about the data:

* The first (minimum) year in the dataset
* The last (maximum) year in the dataset
* The number of countries represented in the data (Hint: use cheatsheet)

```{r}
gapminder 
```

## Your Turn 7

Extract the rows where continent == "Africa" and year == 2007. 

Then use summarise() and summary functions to find:

1. The number of unique countries
2. The median life expectancy

```{r}
gapminder 
```

## Your Turn 8

Find the median life expectancy by continent in 2007.

```{r}
gapminder %>%
  filter(year == 2007)
```

## Your Turn 9

Brainstorm with your neighbor the sequence of operations to find:  the country with biggest jump in life expectancy  (between any two consecutive records) for each continent.

## Your Turn 10

Find the country with biggest jump in life expectancy (between any two consecutive records) for each continent.

```{r}


```

## Your Turn 11

Use `left_join()` to add the country codes in `country_codes` to the gapminder data.

```{r}
country_codes
```

**Challenge**: Which codes in country_codes have no matches in gapminder?

```{r}

```


***

# Take aways

* Extract cases with `filter()`  
* Make new variables, with `mutate()`  
* Make tables of summaries with `summarise()`  
* Do groupwise operations with `group_by()`
* Connect operations with `%>%`  
* Joins are two table verbs