By Miguel Olvera and Rafael Diaz
(Grades 6–12+) Originally published in February 2017, this lesson focuses on correlation as a way of measuring the strength and direction of a linear association between two numerical variables.
After reviewing concepts of bivariate data sets, their graphical representations through scatterplots, and the meaning of a linear association, students will be asked to collect their own data set by measuring two numerical variables about themselves. Then, students enter their data online into the freely available statistical software SeeIt and produce scatterplots to determine if the association between two variables is linear (and therefore, whether it is appropriate to compute a correlation coefficient for each of these pairs of variables).
Students use the dynamic feature of SeeIt to approximate the strength and direction of a linear association with an ellipse, and thus measure the correlation between two variables. Extensions include comparing strengths of correlation, the Pearson correlation coefficient, and how to fit a median-median line and a least-squares line when modeling a linear relationship between two variables.
Downloads
- PDF Lesson
- Handouts in PDF and Word