The first major unit in AP Statistics is all about summarizing data. To begin our summarizing one variable data, we’ll begin with pictures…using graphs to help us summarize data. The first day of graphical displays of data focuses on the difference between how to create graphs for categorical and quantitative data. We’ll explore bar graphs, pie charts, dotplots and stemplots. We’re leaving histograms for the next day.

The class will begin with a warm up focused on the question:

There will be directions of what the only possible choices are (set color categories) and I’ll ask them to write their response on a sticky note. My students will then place their response on the white board. I’ll ask them to summarize the data we just collected. Hopefully, the conversation will end up at showing frequencies for each category.

Next we’ll want to create a graphical representation:

We’ll use excel to create a pie chart and this will be the only time that we ever create a pie chart together. The only expectations they have involving creating a graph for categorical data is: 1) They can create a bar graph, 2) they can interpret any graph they’re given.

Then we’ll check for understanding:

And I’ll try to be entertaining (or I’ll at least humor myself):

Then the focus will change to quantitative data. I’ll pull out a funsized bag of M&Ms and ask them to tell me how many are in the package. They’ll say they don’t know…I’ll ask them how we could figure it out…eventually, they’ll tell me to open and count…so we will. Then, I’ll take out another funsized bag of M&Ms and ask the same question. They’ll say the same as the last bag…I’ll ask if they’ll bet money on it…they’ll say no…we’ll discuss variability. Then we’ll explore the variability. Each student will get their own package and count the M&Ms and they’ll organize the data:

I have to be careful not to lead too much. I want them to decide that a dotplot makes sense to organize their data. After they collect their data, I’ll take out one more funsized package of M&Ms and ask how many are in it. They’ll eventually decide that the center of their distribution is their best guess, but the number could be in the range that’s shown on the dotplot. We’ll lay the foundation to make the values for the mean and standard deviation important!!

I’ll give them another “practice” set of data:

The idea behind this set of data is that a dotplot won’t help. The data is too spread out. This will create a need for a new type of graph: a stemplot! We’ll discuss stemplots and they’ll practice.

This is a lot to get done in a 50 minute class period. I’m going to need to see if and where I can streamline some of this. Any thoughts?