This is easily my favorite story of 2020, and probably one of my favorite stories ever. Secretary of State Boockvar told me it was one of the most in-depth analyses of Judicial Watch’s claims she had read when I interviewed her for a profile before the Nov. 3 election. It also popped up on a few Twitter threads when Judicial Watch made similar claims during the Iowa caucus counting debacle. I think this story is also just a good example of good data journalism practices. I collected as much data as I could in order to replicate the data Judicial Watch’s entire analysis was based on. Until the group confirmed their census data table, I used multiple possible tables to test their results against. I read through U.S. Census data manuals to better understand what I should and shouldn t use it for (direct comparisons to other surveys is not advised). When I found inconsistencies in Bucks, I expanded to the state. When the state didn’t make sense, I tested the data nationally.