Part 3: Errors, `ggplot2`, factors, boxplots, `dplyr`: subsetting using `filter()`/`select()`

Function of the Week Assignment

Please make sure to sign up for a slot for the function of the week assignment if you haven’t already.

Please refer to the function_assignment project in the RStudio Cloud workspace. We will go over this in class.

Checking your Data Types

We will discuss assignment 2.

As an aside, it’s important to deal with data type issues as soon as you can. That’s why it’s important to check your data and use the na argument in read_excel() or read_csv().

Why here()?

here() and why you’d use that if everything is already in the project

The here function is still really muddy for me. I do not quite understand the purpose of it.

We’ll talk more about this with a demonstration.

Questions about {ggplot2}

I still struggle with understanding how to read ggplot2 code.

If you are still struggling, please contact Eric or Me for a review session.

How to deal with lots of missing data when creating ggplots

This is something to be aware of - ggplot2 does give you a warning when a variable that you’ve mapped to an aesthetic has a lot of missing values - it will tell you the number of rows dropped. The default ggplot2 behavior is to drop rows that have missing values.

In general, missing values should be dealt with {tidyr} or {dplyr} before you plot. I will talk about one approach next time when we talk about reshaping data.

I think it would be interesting to learn more about ggplot (can we do more with facet_wrap() for example or is there any other useful tricks we should know)

I’ve added a short section on more ggplot2 fundamentals (scales and colors).

Questions about {dplyr}

Still wrapping my head around how exactly to use select(), filter() and arrange(), especially in sequence.

Still trying to understand the tidyverse data-wrangling functions, but I think practice will help!

Some thoughts. I have found it helpful to compare them to operations in Excel.

• filter() is much like a Data Filter in Excel.
• select() is much like selecting columns in Excel (such as A1-G1).
• arrange() is the equivalent of Sorting your data.

This would have been helpful

What %>% is and its function were very clear. That info would have been so helpful for BSTA!

Learning about the %>% function was soooo clarifying

Glad to help. Again, I feel your frustration.