Section outline

  • Live Lecture: DAGS

    We will review how to draw and use Directed Acyclic Graphs.  This will cover applying the d-separation rule, identifying sufficient and minimally sufficient sets, and DAGs to represent common biases in epidemiology.  We will consider representations of alternative study designs and how these representations help identify potential design problems.  Finally, we will discuss limitations of DAGs and controversies about the usefulness of DAGs.

    Faculty:  Maria Glymour

    Location:      Mission Hall 1406

    Just a reminder to please RSVP for next week's 02/07/2018 session: https://www.eventbrite.com/e/epi-tools-workshop-hailey-banack-tickets-41222983052  so we can have a headcount and send out materials!

    • Session Slides:

    • epi207 mmg DAGsJan30 v7 File
    • DAG slides final version post-class Jan 31 File

    • Session Audio/Video Recording (Access restricted to registered students):

    • Watch URL
      Not available unless: You belong to a group in Registered Students Only

    • Required Reading:

      Modern Epidemiology 3rd edition. Chapter 12. Causal Diagrams.

      Thorpe Jr, R.J., Szanton, S.L., Bell, C.N. and Whitfield, K.E., 2016. Education, income and disability in African Americans. Ethnicity & disease, 23(1), pp.12-17.

    • Optional Reading:

      "Using Causal Diagrams to Understand Common Problems in Social Epidemiology", M Glymour.  chapter 18 in Methods in Social Epidemiology, 2nd edn. (2017) Oakes and Kaufman eds. Wiley  (This is an updated and somewhat friendlier version of the chapter in Modern Epi).

      Glymour MM, Weuve J, Berkman LF, Kawachi I, Robins JM. When is baseline adjustment useful in analyses of change? An example with education and cognitive change. American journal of epidemiology. 2005 Aug 1;162(3):267-78. (This is an illustration of a particular problem that could helpfully be represented with DAGs)

      Hernán MA, Hernández-Díaz S, Robins JM. A structural approach to selection bias. Epidemiology. 2004 Sep 1;15(5):615-25. (This was a very influential paper that reconceptualized how we think about selection bias).

      VanderWeele TJ, Hernán MA. Results on differential and dependent measurement error of the exposure and the outcome using signed directed acyclic graphs. American journal of epidemiology. 2012 May 8;175(12):1303-10.



    • Assignment: HW4 has now been posted.  It is due before class on 2/14/18 (INSTEAD OF THE ORIGINAL DUE DATE OF 2/7/14)

    • HW4 DAGs Problem Set DUE Feb 14, 2018 File
    • data file for DAG hw: American Community Survey

    • Assignment Due Date: February 14, 2018 at the beginning of Small Group Section

    • Week #4 Homework Submission Assignment

    • Assignment Answer Key (access restricted to registered students):

    • HW4 DAGs KEY File
      Not available unless: You belong to Registered Students Only