Additional Resources

There is no reading list for this course as such: your lecturer provides slides and videos, and everything else you’ll need (quizzes, problem sets, etc.) is on this site. With that said, there are many outside resources that are helpful for understanding the course material or engaging more deeply with it. Here are some of my favorites.

Summer reading

My suggested summer reading is Naked Statistics by Charles Wheelan. This book provides a light and friendly introduction to some of the main concepts from the course. I will post some reading quiz questions on this material on Canvas.

Further Reading

• Data Analysis for Social Science by Llaudet and Imai is a hands-on introduction to data analysis in R. Our course doesn’t involve real data analysis, but the Q-Step component of your politics course does, so this is a good bridge. There’s a copy in the LMH library.

• Thinking Clearly with Data by Bueno de Mesquita and Fowler provides a critical evaluation of the kinds of quantitative arguments you’ll frequently encounter in research and the media. While it’s not a traditional textbook, reading it will teach you a lot about probability and statistics along the way.

• An Introduction to Probability and Inductive Logic by Ian Hacking is a more philosophical introduction to probability and inductive reasoning. It connects nicely to the induction topic in your first-year General Philosophy. It also helps to draw a connection between probability theory (which concerns sets) and the standard deductive logic that you’ll encounter in The Logic Manual from your Philosophy course.

Discrete Random Variables

My notes and videos on discrete random variables are lightly edited versions of my lectures for Econ 103, the introductory statistics course I taught at the University of Pennsylvania.

• Discrete Random Variables I covers random variables, the pmf and CDF, the Bernoulli and binomial random variables, and joint, marginal, and conditional distributions. Its sections on joint, marginal, and conditional distributions are required preparation for Class 2; the rest is optional.
• Discrete Random Variables II covers expected value, variance, conditional expectation, covariance, and sums of random variables. It is required preparation for Class 3.

From my blog

These posts from my blog go beyond the material in the lecture videos. The syllabus links each one from the class it relates to. The two posts on the confidence interval for a proportion explain the weaknesses of the standard interval from lecture video S5b. That standard interval is the one used in this course and its exams.

• Is It Better to Improve Sensitivity or Specificity? works through Bayes’ rule for a medical test. It uses odds to show how the answer changes with the base rate, the sensitivity, and the specificity.
• The Independence Zoo distinguishes independence, mean independence, and zero correlation. It includes examples of random variables that are dependent but uncorrelated.
• Random Variables Cheatsheet points to my reference sheet of the main rules for expected value, variance, and covariance.
• Lessons from the Oxford Vaccination Survey shows how a survey to which only about half of those asked responded can give a misleading estimate for the whole population.
• Thirty Isn’t the Magic Number shows that how quickly the Central Limit Theorem takes effect depends on the distribution of the data, not on a fixed sample size such as 30.
• Don’t Use the Textbook CI for a Proportion shows that the standard interval can cover the true proportion much less often than its stated confidence level.
• The Wilson Confidence Interval for a Proportion derives a better interval by inverting a different test.
• Overlapping Confidence Intervals shows that two confidence intervals can overlap even when the difference between the two estimates is statistically significant.
• Overlapping Confidence Intervals: Part II extends this to correlated estimates, such as two treatments compared with the same control group.
• Untangling Cause and Effect Without Experiments is a written version of my webinar on causal inference for a general audience, prepared for LMH News.