Today’s lecture presented the idea of correlation in data sets: observing correlations through scatter plots; measuring them with the correlation coefficient; and using hypothesis testing to see whether that gives evidence to distinguish them from chance coincidence. In this way we get increasingly more precise and sensitive measures for detecting correlation.
Although, remember: correlation does not imply causation. More on that next time.
The recording made at this lecture has no sound. As a replacement, I recommend the first 30 minutes or so of the recording from March 2015. That does not have the TopHat questions, nor the review of earlier content from Lecture 17, but does include the key material on correlations and hypothesis testing. The second half of that recording has more content on statistical testing, causality, and where this all goes wrong — all of this will appear next Tuesday.
1. Read This
- The way you’re revising may let you down in exams — and here’s why
Tom Stafford, The Guardian, 7 May 2016
2. Do This
Find your own statistically significant results. Analyse 60 years of data on the US economy to see the effect of having Republicans or Democrats in power.
- Hack Your Way To Scientific Glory