Lab 7: Functions + Fish

The goal of this lab is learn more about exploring missing data and writing modular code.

The Data

This lab’s data concerns mark-recapture data on four species of trout from the Blackfoot River outside of Helena, Montana. These four species are rainbow trout (RBT), westslope cutthroat trout (WCT), bull trout, and brown trout.

A serene scene of the Blackfoot River in Montana, with a small raft carrying two people navigating the gentle current. The river winds through a landscape of rugged, rocky shores and lush, green pine forests. Rolling hills and distant mountains frame the background under a lightly clouded sky.

Mark-recapture is a common method used by ecologists to estimate a population’s size when it is impossible to conduct a census (count every animal). This method works by tagging animals with a tracking device so that scientists can track their movement and presence.

These images show three examples of animal tracking methods: On this image, a brown bear walks across a road, tagged with a small radio collar around its neck for monitoring.

In this image, a California condor spreads its wings while perched on a rock, displaying a numbered wing tag and a small transmitter for tracking its movements.

In this image, an illustration of a fish, with an inset showing a surgically implanted tracking device that collects data on its location and environment.

Data Exploration

The measurements of each captured fish were taken by a biologist on a raft in the river. The lack of a laboratory setting opens the door to the possibility of measurement errors.

1. Let’s look for missing values in the dataset. Output ONE table that answers BOTH of the following questions:

How many observations have missing values?
What variable(s) have missing values present?

You should use across()!

2. Create ONE thoughtful visualization that explores the frequency of missing values across the different years, sections, and trips.

Rescaling the Data

If I wanted to rescale every quantitative variable in my dataset so that they only have values between 0 and 1, I could use this formula:

\[y_{scaled} = \frac{y_i - min\{y_1, y_2,..., y_n\}}{max\{y_1, y_2,..., y_n\} - min\{y_1, y_2,..., y_n\}}\]

I might write the following R code to carry out the rescaling procedure for the length and weight columns of the BlackfoorFish data:

fish <- fish |> 
  mutate(length = (length - min(length, na.rm = TRUE)) / 
           (max(length, na.rm = TRUE) - min(length, na.rm = TRUE)), 
         weight = (weight - min(weight, na.rm = TRUE)) / 
           (max(weight, na.rm = TRUE) - min(length, na.rm = TRUE)))

This process of duplicating an action multiple times can make it difficult to understand the intent of the process. Additionally, it can make it very difficult to spot mistakes.

3. What is the mistake I made in the above rescaling code?

When you find yourself copy-pasting lines of code, it’s time to write a function, instead!

4. Transform the repeated process above into a rescale_01() function. Your function should…

… take a single vector as input.
… return the rescaled vector.

Efficiency

Are you calling the same function multiple times? You might want to look into the range() function.

5. Let’s incorporate some input validation into your function. Modify your previous code so that the function stops if …

… the input vector is not numeric.
… the length of the input vector is not greater than 1.

Modify Previous Code

Do not create a new code chunk here – simply add these stops to your function above!

Test Your Function

6. Run the code below to test your function. Verify that the maximum of your rescaled vector is 1 and the minimum is 0!

x <- c(1:25, NA)

rescaled <- rescale_01(x)
min(rescaled, na.rm = TRUE)
max(rescaled, na.rm = TRUE)

Next, let’s test the function on the length column of the BlackfootFish data.

7. The code below makes a histogram of the original values of length. Add a plot of the rescaled values of length. Output your plots side-by-side, so the reader can confirm the only aspect that has changed is the scale.

Warning

This will require you to call your rescale_01() function within a mutate() statement in order to create a length_scaled variable.

fish |>  
  ggplot(aes(x = length)) + 
  geom_histogram(binwidth = 45) +
  labs(x = "Original Values of Fish Length (mm)") +
  scale_y_continuous(limits = c(0,4000))

# Code for Q7 plot.

Tip

Set the y-axis limits for both plots to go from 0 to 4000 to allow for direct comparison across plots.
Pay attention to binwidth!
Use a Quarto code chunk option to put the plots side-by-side.

Use Variables within a Dataset

Suppose you would like for your rescale() function to perform operations on a variable within a dataset. Ideally, your function would take in a data frame and a variable name as inputs and return a data frame where the variable has been rescaled.

8. Create a rescale_column() function that accepts two arguments:

a dataframe
the name(s) of the variable(s) to be rescaled

The body of the function should call the original rescale_01() function you wrote previously. Your solution MUST use one of the rlang options from class.

Tip

If you are struggling with this task, I recommend looking back over the data frame functions section of R for Data Science!

9. Use your rescale_column() function to rescale both the length and weight columns.

Warning

I expect that you carry out this process by calling the rescale_column() function only ONE time!

--- title: "Lab 7: Functions + Fish" format: html: code-tools: true toc: true editor: source execute: error: true echo: true eval: false message: false warning: false --- The goal of this lab is learn more about exploring missing data and writing modular code. ```{r} #| label: setup #| echo: false library(tidyverse) fish <- read_csv(here::here("labs", "instructions", "data", "BlackfootFish.csv") ) ``` ## The Data This lab's data concerns mark-recapture data on four species of trout from the Blackfoot River outside of Helena, Montana. These four species are **rainbow trout (RBT)**, **westslope cutthroat trout (WCT)**, **bull trout**, and **brown trout**. ![](https://github.com/earobinson95/stat331-calpoly/blob/master/lab-assignments/lab7/images/blackfoot_river.jpg?raw=true){fig-alt="A serene scene of the Blackfoot River in Montana, with a small raft carrying two people navigating the gentle current. The river winds through a landscape of rugged, rocky shores and lush, green pine forests. Rolling hills and distant mountains frame the background under a lightly clouded sky."} Mark-recapture is a common method used by ecologists to estimate a population's size when it is impossible to conduct a census (count every animal). This method works by *tagging* animals with a tracking device so that scientists can track their movement and presence. <center> ::: columns ::: {.column width="25%"} ![](https://github.com/earobinson95/stat331-calpoly/blob/master/lab-assignments/lab7/images/grizzly.jpg?raw=true){fig-alt="These images show three examples of animal tracking methods: On this image, a brown bear walks across a road, tagged with a small radio collar around its neck for monitoring. "} ::: ::: {.column width="25%"} ![](https://github.com/earobinson95/stat331-calpoly/blob/master/lab-assignments/lab7/images/condor.jpg?raw=true){fig-alt="In this image, a California condor spreads its wings while perched on a rock, displaying a numbered wing tag and a small transmitter for tracking its movements."} ::: ::: {.column width="25%"} ![](https://github.com/earobinson95/stat331-calpoly/blob/master/lab-assignments/lab7/images/fish.PNG?raw=true){fig-alt="In this image, an illustration of a fish, with an inset showing a surgically implanted tracking device that collects data on its location and environment."} ::: ::: </center> ## Data Exploration The measurements of each captured fish were taken by a biologist on a raft in the river. The lack of a laboratory setting opens the door to the possibility of measurement errors. **1. Let's look for missing values in the dataset. Output ONE table that answers BOTH of the following questions:** + **How many observations have missing values?** + **What variable(s) have missing values present?** ::: callout-tip # You should use `across()`! ::: ```{r} #| label: find-missing-values ``` **2. Create ONE thoughtful visualization that explores the frequency of missing values across the different years, sections, and trips.** ```{r} #| label: visual-of-missing-values-over-time ``` ## Rescaling the Data If I wanted to rescale every quantitative variable in my dataset so that they only have values between 0 and 1, I could use this formula: </br> $$y_{scaled} = \frac{y_i - min\{y_1, y_2,..., y_n\}}{max\{y_1, y_2,..., y_n\} - min\{y_1, y_2,..., y_n\}}$$ </br> I might write the following `R` code to carry out the rescaling procedure for the `length` and `weight` columns of the `BlackfoorFish` data: \vspace{0.25cm} ```{r} #| echo: true #| eval: false fish <- fish |> mutate(length = (length - min(length, na.rm = TRUE)) / (max(length, na.rm = TRUE) - min(length, na.rm = TRUE)), weight = (weight - min(weight, na.rm = TRUE)) / (max(weight, na.rm = TRUE) - min(length, na.rm = TRUE))) ``` This process of duplicating an action multiple times can make it difficult to understand the intent of the process. *Additionally, it can make it very difficult to spot mistakes.* **3. What is the mistake I made in the above rescaling code?** When you find yourself copy-pasting lines of code, it's time to write a function, instead! **4. Transform the repeated process above into a `rescale_01()` function. Your function should...** + **... take a single vector as input.** + **... return the rescaled vector.** ::: callout-tip # Efficiency Are you calling the **same** function multiple times? You might want to look into the `range()` function. ::: ```{r} #| label: write-rescale-function ``` **5. Let's incorporate some input validation into your function. Modify your previous code so that the function stops if ...** + **... the input vector is not numeric.** + **... the length of the input vector is not greater than 1.** ::: callout-tip # Modify Previous Code Do not create a new code chunk here -- simply add these stops to your function above! ::: ## Test Your Function **6. Run the code below to test your function. Verify that the maximum of your rescaled vector is 1 and the minimum is 0!** ```{r} x <- c(1:25, NA) rescaled <- rescale_01(x) min(rescaled, na.rm = TRUE) max(rescaled, na.rm = TRUE) ``` Next, let's test the function on the `length` column of the `BlackfootFish` data. **7. The code below makes a histogram of the original values of `length`. Add a plot of the rescaled values of `length`. Output your plots side-by-side, so the reader can confirm the only aspect that has changed is the scale.** ::: callout-warning This will require you to call your `rescale_01()` function within a `mutate()` statement in order to create a `length_scaled` variable. ::: ```{r} fish |> ggplot(aes(x = length)) + geom_histogram(binwidth = 45) + labs(x = "Original Values of Fish Length (mm)") + scale_y_continuous(limits = c(0,4000)) # Code for Q7 plot. ``` ::: callout-tip 1. Set the y-axis limits for both plots to go from 0 to 4000 to allow for direct comparison across plots. 2. Pay attention to `binwidth`! 3. Use a Quarto code chunk option to put the plots side-by-side. ::: ## Use Variables within a Dataset Suppose you would like for your `rescale()` function to perform operations on a **variable within a dataset**. Ideally, your function would take in a data frame and a variable name as inputs and return a data frame where the variable has been rescaled. **8. Create a `rescale_column()` function that accepts two arguments:** + **a dataframe** + **the name(s) of the variable(s) to be rescaled** **The body of the function should call the original `rescale_01()` function you wrote previously. Your solution MUST use one of the `rlang` options from class.** ::: callout-tip If you are struggling with this task, I recommend looking back over the [data frame functions](https://r4ds.hadley.nz/functions.html#data-frame-functions) section of R for Data Science! ::: ```{r} #| label: rescale-data-frame-function ``` **9. Use your `rescale_column()` function to rescale *both* the `length` and `weight` columns.** ::: callout-warning I expect that you carry out this process by calling the `rescale_column()` function only ONE time! ::: ```{r} #| label: rescale-two-columns ```