Lecture 19 (April 4): Design a Controlled Experiment – A Hands-on Activity

Using TELLab, you will design and implement an online experiment to test some hypothesis of your own design related to visual search.Quantitative experiments (e.g., A/B tests) are great, but they are actually very hard to design such that the data can be trusted. In the process of designing your own experiment, you will learn how to be paranoid about accidentally introducing confounds into your experimental design.  You will work in teams of 3-4 (these can be, but do not have to be, your project 2 teams).

Bring your laptop!

Attendance today is mandatory

Required prep (post your written response before class):

Materials:

If you’ve missed L18:

  • Find some way of catching up on the concepts of visual search and visual pop-out
  • Create an account on TELLab:
    • Register for an account at http://lab.tellab.org/signup
    • Once you’ve registered and logged in, go to My Home -> Groups -> Enroll in Group and enter 56f7fc30138fd061762b5d51 as the group code (this will enroll you in the CS 179 group on TELLab).

Lecture Slides

About the author: Krzysztof Gajos

You must be logged in to post a comment.