ποΈ Week 03 - Exploratory Data Analysis
We start with numerical summaries of one, two, and many variables (summary statistics, correlation, and summaries by group), followed by categorical data and the chi-square test, using the Palmer penguins.
The second half asks a bigger question: where does the hypothesis go? Following Donohoβs 50 Years of Data Science, we contrast the statistical (generative) culture, where the hypothesis comes first, with the data-mining (predictive) culture, where the pattern is found in the data and credibility is bought back with a held-out test set. We then do Exploratory Data Analysis the modern way, following Yu & Barterβs Veridical Data Science: the data science life cycle, exploratory vs. explanatory analysis, the question-and-answer workflow, and the PCS framework (predictability, computability, stability) for deciding whether a finding is real.
Finally we move on to Graphs in R, covering both base R functions and ggplot2.
π Readings for this week
π Homework
- HW2 is due Monday, Oct 12, 11:59 pm.
- HW3 (exploratory data analysis and visualization) is released Tuesday, Oct 13, and is due Tuesday, Oct 20, 11:59 pm.
π¨βπ« Lecture Slides
Either click on the slide area below or click here to view it in fullscreen. Use your keypad to navigate the slides.