In Algebra II tomorrow, I’m going to finish up talking about linear regressions. And in honor of that, I’ve created their Thanksgiving homework: a worksheet.

I’m having students

1. pick 10 words
2. check to see how many wikipedia hits these words got in October 2008
3. check to see how many google hits there are for those words

(Example: “monkey” was looked up 122,694 in October 2008 and has about 140,000,000 google hits.)

Then they’re going to see if the data looks linear, and calculate a line of best fit, and use it to predict.

Even though the worksheet needs a lot of work in terms of the questioning and phrasing (I am so tired that I couldn’t think of great questions… I just felt the need to pound this out), I still think this is one of my better activities.

If you want: 2008-11-25-worksheet-on-linear-regressions

Perhaps I’ll compile all the data from all the students and we’ll have a larger data set. We’ll get to talk about outliers (e.g. if you look the word “water” up, things are crazy). I personally am curious what will happen.

(FYI, for the 10 words I chose, I got an r value of about 0.7.)