EMR20: Statistical Analysis, Part 3. Correlation vs. Causation
2018
In the final of three videos on basic statistical analysis, Professor David Cutler, moves beyond asking if the relationship between two variables, health and education, is "statistically significant" to the question about the nature of their relationship. Specifically, he asks what is the difference between correlation and causation, and how is this determined? He explains the importance of identifying other factors that could be influential on the relationship, and demonstrates how one can control for those other factors. In other words, after taking these other variables into account (e.g., race), is the original relationship identified between health and education still robust and statistically significant?
The first two videos in this collection, intended to precede this video, are shown below:
- EMR20: Part 1. Relationship between Two Variables (~15 min)
- EMR20: Part 2. Is the Relationship Important? (~28 min)
- EMR20: Part 3. Correlation vs. Causation (~21 min)
Videos prototyped and produced in collaboration with the Global Health Education and Learning Incubator at Harvard University (GHELI) as part of a residency in pedagogical innovation and multimodal learning.
Related Files:
Source:
Cutler D. EMR20: Statistical Analysis, Part 3. Correlation vs. Causation. Video. Global Health Education and Learning Incubator at Harvard University 2018; Nov. https://vimeo.com/299560492