Topic outline

  • General

    THIS COURSE HAS BEEN POSTPONED TO SPRING 2022. New dates will be announced with Tuutti at a later time.

    S-E400202 Special Course in Computer Science D: Seminar on Experimental Design

    This course is targeted mainly for doctoral students. Participation is limited. The course will aim to answer the difficult and exciting question: how does experimental design affect how are we able to answer causal questions from data? We will be covering selected parts of the book Shadish et al. "Experimental and Quasi-Experimental Designs for Generalized Causal Inference, 2nd Edition". Students are expected to attend all classes in Period 2 and participate in discussions. There will be weekly essay assignments based on the topics and discussion.

    The chapters to be covered during Fall 2021 include:

    1. EXPERIMENTS AND GENERALIZED CAUSAL INFERENCE

    2. STATISTICAL CONCLUSION VALIDITY AND INTERNAL VALIDITY

    3. CONSTRUCT VALIDITY AND EXTERNAL VALIDITY

    4. QUASI-EXPERIMENTAL DESIGNS THAT EITHER LACK A CONTROL GROUP OR LACK PRETEST OBSERVATIONS ON THE OUTCOME

    5. QUASI-EXPERIMENTAL DESIGNS THAT USE BOTH CONTROL GROUPS AND PRETESTS

    8. RANDOMIZED EXPERIMENTS: RATIONALE, DESIGNS, AND CONDITIONS CONDUCIVE TO DOING THEM.

    9. PRACTICAL PROBLEMS 1: ETHICS, PARTICIPANT RECRUiTMENT, AND RANDOM ASSIGNMENT

    The course will meet on Zoom. The link will be announced on the day of the first class. Details for passing will be covered during the first class.