Exploratory data I: penguins
STA 101
June 28, 2022
Bulletin
- first lab today, due Friday at 11:59pm on Gradescope
- be sure to complete prepare material (on the schedule) before each class
Today
By the end of today you will…
- gain familiarity with the ‘big seven’ functions of
dplyr - define and compute various statistics
- begin to gain familiarity with
ggplot
Getting started
Download this application exercise by pasting the code below into your console (bottom left of screen)
download.file("https://sta101.github.io/static/appex/ae2.Rmd",
destfile = "ae2.rmd")- A note on YAML. Create a new
.rmd
Load packages and data
library(tidyverse)
library(palmerpenguins)Type ?palmerpenguins to learn more about this package.
Or better yet, check it out here.
data(penguins)Exercise 1:
Look at the data, how many observations are there? How many variables?
A package within a package…
When we load the tidyverse library, dplyr
is packaged with it.
dplyr, a grammar of data manipulation offers intuitive ‘verb’ functions that describe actions we commonly want to perform with data. The big 7 we’ll cover today are:
mutate()adds new variables that are functions of existing variablesselect()picks variables based on their names.filter()picks cases based on their values.group_by()sets us up to summarize across groupssummarize()reduces multiple values down to a single summary.arrange()changes the ordering of the rows.slice()select, remove and duplicate rows based on their index
(as described at https://dplyr.tidyverse.org/)
- Check out documentation with
?
Mutate
Approximate bill area (in \(mm^2\)) as bill length * bill depth:
penguins %>%
mutate(bill_area_mm2 = bill_length_mm * bill_depth_mm)## # A tibble: 344 × 9
## species island bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
## <fct> <fct> <dbl> <dbl> <int> <int>
## 1 Adelie Torgersen 39.1 18.7 181 3750
## 2 Adelie Torgersen 39.5 17.4 186 3800
## 3 Adelie Torgersen 40.3 18 195 3250
## 4 Adelie Torgersen NA NA NA NA
## 5 Adelie Torgersen 36.7 19.3 193 3450
## 6 Adelie Torgersen 39.3 20.6 190 3650
## 7 Adelie Torgersen 38.9 17.8 181 3625
## 8 Adelie Torgersen 39.2 19.6 195 4675
## 9 Adelie Torgersen 34.1 18.1 193 3475
## 10 Adelie Torgersen 42 20.2 190 4250
## # … with 334 more rows, and 3 more variables: sex <fct>, year <int>,
## # bill_area_mm2 <dbl>Select
It’s hard to see bill length, depth and area in the same output, let’s select a smaller subset of the variables to look at.
# Example 1
penguins %>%
mutate(bill_area_mm2 = bill_length_mm * bill_depth_mm) %>%
select(-year)## # A tibble: 344 × 8
## species island bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
## <fct> <fct> <dbl> <dbl> <int> <int>
## 1 Adelie Torgersen 39.1 18.7 181 3750
## 2 Adelie Torgersen 39.5 17.4 186 3800
## 3 Adelie Torgersen 40.3 18 195 3250
## 4 Adelie Torgersen NA NA NA NA
## 5 Adelie Torgersen 36.7 19.3 193 3450
## 6 Adelie Torgersen 39.3 20.6 190 3650
## 7 Adelie Torgersen 38.9 17.8 181 3625
## 8 Adelie Torgersen 39.2 19.6 195 4675
## 9 Adelie Torgersen 34.1 18.1 193 3475
## 10 Adelie Torgersen 42 20.2 190 4250
## # … with 334 more rows, and 2 more variables: sex <fct>, bill_area_mm2 <dbl># Example 2
penguins %>%
mutate(bill_area_mm2 = bill_length_mm * bill_depth_mm) %>%
select(species, island, bill_length_mm, bill_depth_mm, flipper_length_mm, body_mass_g, sex, bill_area_mm2)## # A tibble: 344 × 8
## species island bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
## <fct> <fct> <dbl> <dbl> <int> <int>
## 1 Adelie Torgersen 39.1 18.7 181 3750
## 2 Adelie Torgersen 39.5 17.4 186 3800
## 3 Adelie Torgersen 40.3 18 195 3250
## 4 Adelie Torgersen NA NA NA NA
## 5 Adelie Torgersen 36.7 19.3 193 3450
## 6 Adelie Torgersen 39.3 20.6 190 3650
## 7 Adelie Torgersen 38.9 17.8 181 3625
## 8 Adelie Torgersen 39.2 19.6 195 4675
## 9 Adelie Torgersen 34.1 18.1 193 3475
## 10 Adelie Torgersen 42 20.2 190 4250
## # … with 334 more rows, and 2 more variables: sex <fct>, bill_area_mm2 <dbl>- A note on pipes
%>%and a note on style.
Filter
Let’s just examine penguins on Dream island
penguins %>%
mutate(bill_area_mm2 = bill_length_mm * bill_depth_mm) %>%
select(-year)
# code hereGroup by + Summarize
Exercise 2:
Find mean bill area across sex. Fill in the blanks
penguins %>%
mutate(bill_area_mm2 = bill_length_mm * bill_depth_mm) %>%
select(-year) %>%
# filter for Dream
group_by(___) %>%
summarize(mean_bill_area_mm2 = ___)Arrange + Slice
Let’s use arrange() and slice() to report
the five penguins with the greatest bill area.
penguins %>%
mutate(bill_area_mm2 = bill_length_mm * bill_depth_mm) %>%
select(bill_area_mm2, bill_length_mm) %>%
arrange(desc(bill_area_mm2))## # A tibble: 344 × 2
## bill_area_mm2 bill_length_mm
## <dbl> <dbl>
## 1 1127. 54.2
## 2 1105. 55.8
## 3 1076. 52
## 4 1065. 53.5
## 5 1056 52.8
## 6 1050. 51.7
## 7 1043. 52.7
## 8 1032. 58
## 9 1021. 51.3
## 10 1013. 59.6
## # … with 334 more rowsExercise 3:
Are these the same five penguins with the longest bills?
Optional hint: if you want to be exactly precise about which penguins are which, you could add an ID column, e.g.
penguins %>%
mutate(id = seq(1:nrow(penguins)))## # A tibble: 344 × 9
## species island bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
## <fct> <fct> <dbl> <dbl> <int> <int>
## 1 Adelie Torgersen 39.1 18.7 181 3750
## 2 Adelie Torgersen 39.5 17.4 186 3800
## 3 Adelie Torgersen 40.3 18 195 3250
## 4 Adelie Torgersen NA NA NA NA
## 5 Adelie Torgersen 36.7 19.3 193 3450
## 6 Adelie Torgersen 39.3 20.6 190 3650
## 7 Adelie Torgersen 38.9 17.8 181 3625
## 8 Adelie Torgersen 39.2 19.6 195 4675
## 9 Adelie Torgersen 34.1 18.1 193 3475
## 10 Adelie Torgersen 42 20.2 190 4250
## # … with 334 more rows, and 3 more variables: sex <fct>, year <int>, id <int>This takes advantage of the nrow() function. Can you
guess what it returns?
Exercise 4:
Compute the average bill length, bill depth, flipper length and body mass across all islands.
Exercise 5:
Is every species on every island?
Summary statistics
What is a statistic? It’s any mathematical function of the data. Sometimes, a statistic is referred to as “sample statistic” since you compute it from a finite sample (the data) and not the entire population.
For example, penguins is a sample of penguins in
Antarctica, not an exhaustive list of the entire population.
Examples of statistics:
measure of central tendency:
- mean
- median
- mode
measures of spread:
- standard deviation
- variance
- range
- quartiles
- inter-quartile range (IQR)
order statistics:
- quantiles
- minimum (0 percentile)
- median (50th percentile)
- maximum (100 percentile)
… and any other arbitrary function of the data you can come up with!
Exercise 6:
Come up with your own statistic and write it in the narrative below.
To access a column of the data, we’ll use
data$column.
Let’s write down the R function for each of the above.
Exercise 7:
Try to compute the above statistics for the penguin bill length column.
Do you receive an error? Why?
# code hereLet’s take a look at the distribution of bill length among all penguins.
penguins %>% # data
ggplot(aes(x = bill_length_mm)) + # columns we want to look at
geom_histogram() # geometry of the visualization## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 2 rows containing non-finite values (stat_bin).
Let’s visualize some of our summary statistics on the plot.
Exercise 8:
Uncomment the code below and fill in the blank with the mean.
penguins %>%
ggplot(aes(x = bill_length_mm)) +
geom_histogram() #+
#geom_vline(xintercept = __, color = 'red')Exercise 9
Add another geom_vline with the median and mode. Use
separate colors for each.
Exercise 10
Finally, let’s try out another geometry geom_density and
clean up our graph with some axes labels.
# live coding here