A/B Testing, Design and Analysis
This course looks at how A/B testing helps measure causal effects across different industries and fields of public policy studies. We aim at answering questions such as: how does the demand for a product change when the price does or the ratings do? how can we anticipate how sales and profits change if the firm changes its business strategy? how can we measure whether introducing technology in schools, universities and/or classrooms affects the performance of students? Many companies and Governmental agencies ask and try to answer questions of this type every day. Moreover, they are now also actively seeking tools that can take advantage of big datasets to answer these questions as well as skillful individuals that know how to use them confidently.
This course introduces fundamental concepts to correctly ask this type of question. We study frameworks to measure causal effects and we discuss their pros and cons. Every tool is discussed in the context of a specific example that students work on using real world datasets. Significant effort is placed on understanding how to design randomized experiments (aka A/B tests) to measure causal effects. We also discuss the most common challenges that arise when trying to design such experiments in the wild. The concepts and tools discussed in this course are general in nature and can be applied in different settings. The examples discussed in class will be mostly drawn from our own work at the Heinz College on the media industry and education policy. Lectures are 3 hours long. In the first half of each lecture we go over concepts related to A/B testing and what to do when A/B tests are unavailable. The discussion is based on the ideas and intuition behind these concepts. In the second half of each lecture we go over specific examples -- we study several real-world datasets and the code used to analyze them properly. Student evaluation is based on five weekly homeworks and a brief term project to be developed in teams.
Students will learn the fundamentals of randomized control trials (aka A/B tests), namely what they achieve, how to design, implement and analyze their outcomes as well as their shortcomings and work arounds. Students will also learn tools that can be used to analyze data from observational studies where randomization was not implemented.