This week we talked about the t-distribution and how to use it to do inference about means. Below are one-page summaries of the different methods.
Today we did the following workshop to practice two sample t-distribution techniques:
Then we introduced the idea of a one-sample matched pairs t-distribution techniques. The idea is that whenever you have two correlated quantitative variables and you want to say something about the average difference, you should focus on the differences not the individual numbers. Here is the example we did in class:
And finally, here is a video I made last year that explains this example.
Today we introduced correlation coefficients and least square regression lines for scatterplots. For the most part, we followed Section 8.1 and Section 8.2 from the book pretty closely, so read those if you didn’t get the notes. One important thing that the book doesn’t cover is how to use a spreadsheet to make a scatterplot with a linear regression trend line. Here is a quick video that explains how:
Today we talked more about correlation and regression. We explained the concept of regression to the mean, and we talked about what \(R^2\) tells us about the data (it’s the percent of the variability in the y-values that is accounted for by the trend). We did this workshop in class: