![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Meat meal [teaching]
Oct. 13th, 2021 03:28 pmLife is kind of boring right now, so a short story from teaching yesterday:
We're in the midst of the R and Statistics portion of the course, which a lot of students struggle with. For this particular class, the earliest stages of working with R and learning statistical concepts were pretty rough, but by this point most students are starting to get the hang of things. We're now at the stage where they're starting to learn about things like two-way Analysis of Variance and how to conduct post-hoc comparisons.
For as much as I might think we do a great job of teaching these topics in the lecture videos we created last fall, there are still a couple of sticking points. So there's some value in taking a couple extra minutes during our class period to address those sticking points in particular.
One that I anticipated students struggling with is the application of the appropriate letters above each bar to indicate which treatment groups are statistically similar vs. different from each other. I have spent a lot of time during office hours past going over this with students. So, why not go over it once, with everyone, during lecture instead?
The only thing I was missing was a fake dataset of some sort or another.
If you're an R user, you know that R comes stocked with a set of datasets. One of the more classic datasets, "mtcars" is in there, for instance (Motor Trends data on automobiles - basically the only dataset we ever used for everything back when I took SAS-based Regression and Analysis of Variance courses in grad school).
But we needed a dataset that had at least some Biology connection.
Also, I decided all this about 10 minutes before it was time to go and teach.
So I quickly popped open R and ran:
data()
Scanning the list the first simple dataset that caught my eye was one called "chickwts" - chicken weights by feed type. There were 6 different feed types, so I was good to go.
It wasn't until I got to class that I discovered that one of the feed types was called "meatmeal". Yum!
Still, it worked for my purposes, and the students were into it.
To make up for that, I made up a second example based on ice cream flavors and their deliciousness.
And that is all for now. Other than that I'm mostly grading papers and reading articles right now, so I don't have a ton to show for myself.
We're in the midst of the R and Statistics portion of the course, which a lot of students struggle with. For this particular class, the earliest stages of working with R and learning statistical concepts were pretty rough, but by this point most students are starting to get the hang of things. We're now at the stage where they're starting to learn about things like two-way Analysis of Variance and how to conduct post-hoc comparisons.
For as much as I might think we do a great job of teaching these topics in the lecture videos we created last fall, there are still a couple of sticking points. So there's some value in taking a couple extra minutes during our class period to address those sticking points in particular.
One that I anticipated students struggling with is the application of the appropriate letters above each bar to indicate which treatment groups are statistically similar vs. different from each other. I have spent a lot of time during office hours past going over this with students. So, why not go over it once, with everyone, during lecture instead?
The only thing I was missing was a fake dataset of some sort or another.
If you're an R user, you know that R comes stocked with a set of datasets. One of the more classic datasets, "mtcars" is in there, for instance (Motor Trends data on automobiles - basically the only dataset we ever used for everything back when I took SAS-based Regression and Analysis of Variance courses in grad school).
But we needed a dataset that had at least some Biology connection.
Also, I decided all this about 10 minutes before it was time to go and teach.
So I quickly popped open R and ran:
data()
Scanning the list the first simple dataset that caught my eye was one called "chickwts" - chicken weights by feed type. There were 6 different feed types, so I was good to go.
It wasn't until I got to class that I discovered that one of the feed types was called "meatmeal". Yum!
Still, it worked for my purposes, and the students were into it.
To make up for that, I made up a second example based on ice cream flavors and their deliciousness.
And that is all for now. Other than that I'm mostly grading papers and reading articles right now, so I don't have a ton to show for myself.
