Week 9: Apply it to your data 8

Import your data

colony <- read_excel("../00_data/myData.xlsx")

set.seed(123)

colony_small <- colony %>%
    sample_n(10) %>%
    select(year, state, colony_lost, colony_added)

colony_small

## # A tibble: 10 × 4
##     year state      colony_lost colony_added
##    <dbl> <chr>            <dbl> <chr>       
##  1  2017 Utah              2700 2900        
##  2  2017 Vermont            170 390         
##  3  2015 Texas            25000 13000       
##  4  2017 Hawaii             130 970         
##  5  2016 Florida          45000 36000       
##  6  2019 Wyoming           3300 100         
##  7  2021 Kansas            1400 2300        
##  8  2020 California       69000 61000       
##  9  2018 Florida          30000 53000       
## 10  2018 Texas            22000 118000

Chapter 13

What are primary keys in your data?

colony_small %>% count(state, colony_lost)

## # A tibble: 10 × 3
##    state      colony_lost     n
##    <chr>            <dbl> <int>
##  1 California       69000     1
##  2 Florida          30000     1
##  3 Florida          45000     1
##  4 Hawaii             130     1
##  5 Kansas            1400     1
##  6 Texas            22000     1
##  7 Texas            25000     1
##  8 Utah              2700     1
##  9 Vermont            170     1
## 10 Wyoming           3300     1

colony_small %>% count(state, colony_lost) %>% filter(n > 1)

## # A tibble: 0 × 3
## # … with 3 variables: state <chr>, colony_lost <dbl>, n <int>

Can you divide your data into two?

Divide it using dplyr::select in a way the two have a common variable, which you could use to join the two.

colony_year.state <- colony_small %>% select(year:colony_added)
colony_year.state

## # A tibble: 10 × 4
##     year state      colony_lost colony_added
##    <dbl> <chr>            <dbl> <chr>       
##  1  2017 Utah              2700 2900        
##  2  2017 Vermont            170 390         
##  3  2015 Texas            25000 13000       
##  4  2017 Hawaii             130 970         
##  5  2016 Florida          45000 36000       
##  6  2019 Wyoming           3300 100         
##  7  2021 Kansas            1400 2300        
##  8  2020 California       69000 61000       
##  9  2018 Florida          30000 53000       
## 10  2018 Texas            22000 118000

colony_state.lost <- colony_small %>% select(colony_added:state:colony_lost)

## Warning in x:y: numerical expression has 3 elements: only the first used

colony_state.lost

## # A tibble: 10 × 2
##    colony_added colony_lost
##    <chr>              <dbl>
##  1 2900                2700
##  2 390                  170
##  3 13000              25000
##  4 970                  130
##  5 36000              45000
##  6 100                 3300
##  7 2300                1400
##  8 61000              69000
##  9 53000              30000
## 10 118000             22000

Can you join the two together?

Use tidyr::left_join or other joining functions.

left_join(colony_year.state, colony_state.lost)

## Joining with `by = join_by(colony_lost, colony_added)`

## # A tibble: 10 × 4
##     year state      colony_lost colony_added
##    <dbl> <chr>            <dbl> <chr>       
##  1  2017 Utah              2700 2900        
##  2  2017 Vermont            170 390         
##  3  2015 Texas            25000 13000       
##  4  2017 Hawaii             130 970         
##  5  2016 Florida          45000 36000       
##  6  2019 Wyoming           3300 100         
##  7  2021 Kansas            1400 2300        
##  8  2020 California       69000 61000       
##  9  2018 Florida          30000 53000       
## 10  2018 Texas            22000 118000

Chapter 14

Tools

Detect matches (str_detect)

x <- c("state", "months", "colony_lost")
str_detect(x, "s")

## [1] TRUE TRUE TRUE

Extract matches (str_extract)

str_extract(x, "state")

## [1] "state" NA      NA

Replacing matches (str_replace)

str_replace(x, "months", "NA")

## [1] "state"       "NA"          "colony_lost"