# Lessons from Brown’s pre-college course in Applied Statistics

After three-weeks of teaching and two weeks of tying up some loose ends, I have some pretty neat stuff to share from my pre-college course “Applied Statistics,” which I taught earlier this summer at Brown.

22 students from seven countries came to Brown, all for three weeks, and mostly all entirely came to take my one class in statistics. It was fairly humbling to have that much intelligence and excitement in the room!

Here are a few of the coolest things to come out of the class.

1- How to read statistics in the news.

I tried to stress the importance of how important it was to become capable statisticians as far as interpreting news stories which attempt to use statistics (either properly or improperly) to make a point. One of my least favorite articles ever on statistics is here. As a class, I had each student write a letter to the editor of the Washington Post, and I shared the best two with the Post when class ended! Sadly, we are yet to hear back!

2- When are high school students drinking soda?

One of my favorite projects looked at the frequency with which student’s drank soda in the Brown cafeteria. I loved this idea for a few reasons, including the method with which the group gathered the data, the hypothesis itself (are students more likely to drink soda at dinner or lunch? and how does this compare to other beverages?) and because this particular group used R to come up with some pretty neat plots, including the one you see below. The students used chi-square tests to show significant differences between these counts, and while these results were interesting, I thought the graph told the whole story.

In the graph above, the BLUE line is soda, and the RED line is other drinks.

I was amazed that the consumption of other drinks stayed fairly constant while soda consumption was much higher at lunch. It’s pretty clear to me: if we want to lower soda consumption, we need to do so during lunch time, even for high schoolers.

3- Is cell phone use associated with grade point average?

One group sent out a poll to their friends on facebook, asking for, among other information, hours per day spent on the phone, and grade point average. While it’s not necessarily a causal effect just yet, the negative association was stronger than any of us anticipated, suggesting that as time spent on the phone goes up, GPA goes down.

Interestingly enough, future groups were smart enough to try and include phone type (i.e., smart phone or non-smart phone), and the results suggested that phone type was not associated with GPA.

4- Last, but certainly not least, I was fortunate to have three outstanding guest speakers – Hopkins’ and Harvard’s Sarah Peskoe, UAB’s Stella Aslibekyan, and Stonehill’s David Hurley, who each shared how she or he used statistics in research. On my course evaluations, nearly every student went out of their way to say how useful those three were as far as connecting course material to how statistics is shaping the world! So, thanks guys!  Virtual high-5.

All in all, it was a great experience, and, as usually happens in a statistics course, I learned more from them than they did from me. I look forward to teaching the same course again next year – if enough kids sign up!