Exercise 1: filtering data

The goal of this exercise is to pare down a dataset using the filter function with as many conditions as you need in order to filter the data down to a single penguin. This game is a throw back to those days on long road trips when you were down to one last game… the game of Guess Who.

You were BORED because you had already played all the games, done all the puzzles, no one would stop for snacks, and you were still miles from home. So you began, Am I bigger than a bread box? Am I the color green? In this exercise you are writing code to winnow down the list of possibilities. FUN STUFF!

Using the penguins dataset in the palmerpenguins package, use the filter function (with multiple conditions if necessary) to find the penguin that meets one of the following sets of criteria. What is its secret ID number?

1. I am a female penguin living on Biscoe Island. I was observed in the latest year and my friends tell me I have a very handsome bill depth of 17.7 mm. Who am I?
2. I am a male Chinstrap penguin weighing exactly 4000 g or maybe more. Probably more. My bill is a bit over 50 mm long, but I am most proud of my short flippers, which are less than 200 mm long. Who am I?
3. Hi! I am a female Gentoo penguin that likes to wear colorful hats. My bill length is a smidgeon under 45 mm long. Last time I checked, I weighed over 5000 g but that was quite a while ago. Who am I?

Load the penguin data

library(tidyverse)
library(palmerpenguins)

penguins <- penguins

Add a top secret ID column with mutate

library(tidyverse)
library(palmerpenguins)

penguins <- penguins

penguins <- mutate(penguins, 
                   secret_id = 1:n()) 

Ack! Typing is hard. To help avoid typing errors it can help to copy column names. You can use names(penguins) or glimpse(penguins) to print the columns to the console and copy the column name that you need.

p1 <- penguins %>% 
      filter(sex == "female",
             island == _______,
             _______    _______ ) 
  

p1 <- penguins %>%
      filter(sex == "female", 
             island == "Biscoe", 
             year == 2009, 
             bill_depth_mm == 17.7)
  

p2 <- penguins %>% 
      filter(sex == _______ ,
             species == _________ ,
             body_mass_g >= ________ ,
             _______    _______ ,
             _______    _______ ) 
  

p2 <- penguins %>% 
      filter(sex == "male" ,
             species == "Chinstrap" ,
             body_mass_g > 4000, 
             bill_length_mm > 50, 
             flipper_length_mm < 200)
  

p3 <- penguins %>% 
      filter(sex == _______ ,
             species == _________ ,
             _______    _______ ,
             _______    _______ ) 
  

p3 <- penguins %>% 
      filter(sex == "female" ,
             species == "Gentoo" ,
             bill_length_mm < 45,  
             body_mass_g > 5000)
  

For a more advanced challenge, try to isolate these penguins.

4. I forgot my name and I really do not know many details about myself. I know I am an Adelie penguin that lives on Torgersen, but all my physical measurements are missing.
   
5. I have the lowest weight of the male penguins. Who am I?
6. I am the heaviest female Adelie penguin. Who am I?
7. I have the longest flippers of the Adelie penguins on Dream Island. Who am I?

# is.na() checks if a columns value is missing
p4 <- penguins %>% 
      filter(species == _______,
             island == _______ ,
             is.na( __________ ))
      

p4 <- penguins %>% 
      filter(species == "Adelie",
             island == "Torgersen" ,
             is.na(bill_length_mm))
  

# Filter by sex first
p5 <- penguins %>% 
      filter(sex == _______ )
      
# And then pipe it to filter by body_mass_g
# Remember to use `na.rm = TRUE` to account for NA values
p5 <- penguins %>% 
      filter(sex == ________ ) %>%
      filter(body_mass_g == _________ ) 
  

p5 <- penguins %>% 
      filter(sex == "male" ) %>%
      filter(body_mass_g == min(body_mass_g, na.rm = TRUE)) 
  

Exercise 2: Sorting with case_when( )

I love grouping things, which was part of the reason I loved being a baking bot. I could spend all day sorting measuring spoons and measuring cups into groups like “tiny”, “super tiny”, and “the tiny-est of the tiny”. Maybe I can put my grouping skills to use on the penguin data?

Let’s use case_when() to add some sorting columns to the data. Pick one of the options below to split the penguins into various groups based on their measurements.

1. Assign the penguins a value for the new column flipper_group based on the following criteria:
   • First check if flipper length is greater than 210 mm, if it is they are "big flips"
   • Otherwise they are "small flips"
2. Assign the penguins a value for weight_class based on the following criteria:
   • First check if body mass is less than or equal to 3500 g, if so they are "littles"
   • Then check if body mass is less than 4300 g, then they are "middles"
   • Otherwise they are "biggles"
3. Assign the penguins a value for bill_class based on the following criteria:
   • First check if bill depth is missing, then they are "unknown"
   • Then check if bill depth is more than or equal to 18, then they are "deep"
   • Then check if bill depth is more than 16, then they are "middling"
   • Otherwise they are "shallow"

penguins %>% 
  mutate(flipper_group = case_when(flipper_length_mm  > _______ ~ ________,
                                   TRUE ~ _________ ))
  

penguins %>% 
    mutate(flipper_group = case_when(flipper_length_mm  > 210 ~ "big flips",
                                     TRUE ~ "small flips" ))
  

BON VOYAGE!