Instructor Information
Correlation and/vs. Causation
Overview
This reading builds on the discussions of basic descriptive statistics and the analysis of measures of central tendency and variation by introducing basic ideas about correlation. Correlation encompasses circumstances in which two or more data measures vary together in some way. The reading can provide the basis of a classroom activity focused on assessing variation and then reflecting on what correlation between two quantitative variables means.
Learning Goals
Understand the concept of correlation. Distinguish between correlation and causation. Introduce the idea of a confounding variable.
Readings
Matthew J.C. Crump, Answering Questions With Data, “Correlation”: 3.0-3.4 and “Interpreting Correlations”: 3.6.1-3.6.1.2
Key Concept Review Activity (15 minutes)
Divide students into four groups and assign one of the questions below to each group. Have students record their answers on a blank google slide. Before moving on, ask each group to share their slide to the class as a whole.
- What is covariation?
- What is the difference between correlation and causation?
- Draw a graph depicting positive, negative and no correlation in a scatterplot.
- What is Pearson’s r and how is it calculated?
Note: This reading can be paired with “Hans Rosling’s 200 Countries, 200 Years, 4 Minutes” (YouTube video, also available on the Gapminder website), the Gapminder World Health Chart, as well as the World Happiness Report dataset both of which are included in the Data Advocacy Toolkit.