AP Stat Week 1

I love teaching AP Statistics.  I love the content, but it’s also the course I’ve taught the longest.  I feel like it’s the class I teach the best.  In AP Statistics I often impress myself, in other classes that is less often the case :).

This year in AP Statistics I’m trying to move towards using simulation more throughout the course to support the concepts of inference, rather than just tacking it on as it’s own section during probability.  The simulation section has always felt like a weird not connected topic to me, which I don’t like, and it has so much potential I’m not making use of right now.  I think this will be a positive change.

We have had three different bell schedules at school this week.  Tuesday and Wednesday students were dismissed early after we ran each class for 30 minutes with no lunch.  Even with two planning periods these days were hectic!  I always try to start my courses off by spending the first few days doing what I feel like is the major work of the course, rather than going over the syllabus or reviewing material from previous classes.  I like the tone this sets and the message it sends about what is important in this class.  Then after a week or so we back up and kind of start from the beginning.

In statistics this year (and last year) I started off gathering some basic data on heights and looking at distributions.  I originally got this activity from a PDF on engaging activities for AP Statistics.  It does a great job of getting students engaged immediately, there’s a little bit of a mystery to it, and it gathers data that students are interested in (because it’s about them!) without it being too cumbersome or time consuming to collect the data.  After they make their post its I ask them to make some predictions about what set of guesses they think will be “best,” which usually leads to a good discussion about what best means in this context.  Usually someone says best is closest to the true average height (so already we’re getting that idea of a fixed but unknown true value), but then we discuss how that makes sense for a single observation but not a group of observations.

Then we make “dot plots” out of them on the board.  I have them do this themselves and it gets them interacting with each other.  They almost always run into some problems (how to scale it, what to do about the person who wrote an average height of 580″, how to deal with decimal guesses, etc) that we can discuss as a class.  Then I lead a discussion using the dot plots on things like distributions, shape, center, spread, precision vs. accuracy, bias, etc.

Usually the data is very cooperative about becoming more precise (less variable) as the guesses go on, and generally the guesses after the first group are fairly accurate.

Last we put their actual heights into the lists on the calculator and I show them how to do 1-var stats and discuss what the different values it produces mean, which ones we care about, etc.

This year this took the two 30 minute periods.

On the third day we moved into the Westvaco case study.  This is originally from Statistics in Action but I got it from Statsmonkey.  This activity does several things I really like.  It has some controversy that students get to have an opinion about (there either was or was not age discrimination) with no clear cut answer.  It has a largish but not too large data set for them to work with, and then it follows up by using some simulation to introduce some of the logic of inference.

The first day I gave them the case study and let them go to town.  They were making a lot of dot plots and finding various ways to process/color code the data.  The second day I gave them the data in a google sheet and showed them how to do some basic sorting and calculating in spreadsheets.  I have found my students generally have no idea how to use a spreadsheet so I try to introduce the basics early and often.  I also showed them how to use statkey to make dot plots.

Eventually I will have them take the work they’ve done on this case study, pick a side on the age discrimination issue, and make a statistical argument (1-2 pages) to support their argument, and I want them to use graphs and other appropriate tools to support their argument.  With statkey and google sheets they can hopefully just copy and paste these graphs in.  I will mainly be looking for them to have an argument that uses statistical evidence and clear writing when I grade their assignments, since they are still novices even at basic statistical methods.

Next week we will spend a day or two doing the simulation with Westvaco.  We will use index cards and keep our simulations physical, for now.  I find this helps slow things down and lets them focus on the main idea (the simulation) vs getting bogged down and distracted by simulating with technology.  Later in the course we will do simulations with google sheets and XLMiner.

After that comes Stats Modeling the World chapters 1-3!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s