Alexander Coppock

Alexander Coppock studies political persuasion, meta-analysis, and design visualization. He is the author of two books: Persuasion in Parallel (University of Chicago Press) and Research Design: Declaration, Diagnosis, Redesign (Princeton University Press).

What do you do with data science?

My data scientific work falls in two main categories. First, I write and maintain (with collaborators) statistical software to aid in the design and analysis of empirical research studies. These packages for R include DeclareDesign, estimatr, randomizr, and ri2. Second, I research and teach “design visualization,” which is related to the more familiar field of data visualization. Research designs produce data, so data visualizations should respect and reflect the details of the designs that produced them. We can also visualize research designs themselves in order to better understand their properties and to select stronger designs over weaker ones.