Letters & Science 39G:
Health, Human Behavior, and Data

Fall 2015
Visiting Prof. Ryan D. Edwards

L&S 39G is one of five new 2-unit connector courses at Cal that supports Stat 94, Foundations of Data Science, the new Data Sciences core course for undergraduates at UC Berkeley.

Stat 94 is designed to instruct first-year students in statistical methods and theory and in the mechanics of manipulating large data sets through programming in Python.

In L&S 39G, students study how and why to apply these analytical and programming skills to real-world problems in health economics and population studies. Health inequalities in the U.S. and worldwide are large and remain the focus of great political and research interest. But the sources of health inequalities remain unclear, especially when the inferential challenges of observational data are binding constraints.

Like Stat 94, L&S 39G is a new course that is evolving to best meet the needs of students in a world that is increasingly data-rich but where cogent analysis sometimes may be lacking. The didactic emphasis on L&S 39G begins with understanding the open questions in health economics and the statistical methods available to answer them, and it continues with hands-on examination of real data designed to elucidate key messages.

---

Syllabus: available as a PDF

Slides from class:

1. Background and motivation
2. Health metrics, social indicators, inequality, and policy
3. Mortality rates and life expectancy
4. How to think like a health economist: The Grossman model of health capital
5. How to think like a health economist: Causal influences, variance
6. Randomized controlled trials: Lind's Scurvy Experiments
7. Randomized controlled trials: The RAND Health Insurance Experiment revisited
8. Observational studies: In utero influences
9. Strong exogeneity: Weather, beer and Student's t, wine quality models
10. More strong exogeneity: Are recessions good for your health?
11. Natural experiments and instrumental variables: War and wages
12. Case studies: Season of birth and later outcomes
13. Case studies: The minimum legal drinking age
14. Perception and behavior: Risk over the life bicycle

Datasets using in class:

8. Data: c08_r89smokeweight.csv
   Doc: c08_r89smokeweight.html
9. Data: c09_ashenfelter.csv
   Doc: c09_ashenfelter.html
10. Data: c10_sweden.csv
    Doc: c10_sweden.html
11. Data: c11_wwii.csv
    Doc: c11_wwii.html
12. Data: c12_b1960.csv | c12_b1970.csv | c12_b1980.csv
    Doc: c12_data.html
13. Data: c13_drinkage.csv | c13_drinkage_u21.csv | c13_drinkage_o21.csv
    Doc: c13_drinkage.html
14. Data: c14_cycling.csv
    Doc: c14_cycling.html