{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/b866d2ef092c48458bf2dae08e5b8632\" frameborder=\"0\" width=\"1920\" height=\"1440\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1440,"width":1920,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1440,"thumbnail_width":1920,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/b866d2ef092c48458bf2dae08e5b8632-86dfc1d106a48cc3.gif","duration":2901.5,"title":"Data Analysis and Connecting Make.com to a Local Model","description":"I scoped EDA for a high dimensional civil rights discipline dataset, loaded layout, school, and counts, and highlighted what is actually doable before heavy compute. The key finding was discipline is highly concentrated, with the top 1 percent of schools producing about 40 percent of discipline events, and some measures like expulsions can’t be trusted as baselines because of missing or ghost reporting. I also covered practical next steps like ratio disparity, suspension ratios, and within district access and equality variants. No viewer action was required, just pick a question and poke at the data for five minutes if you want to try it."}