my own frequent and intimate experiences that data usually just doesn’t make any difference has led me to a few conclusions: 1. most interventions to change a system don’t do anything, so bringing data is not unusually ineffective as a tactic. 1/?

I really liked this thread. I’ve long pushed back in the urbanism circles when people say “if only we could show X the data” (where X is often a city councilor)

When I worked in data analysis one of the biggest issues we had to deal with was people's expectations of what we could do Vs the reality of it. People really want to think it's magic.

2. it is worth thinking about the conditions under which data does have an impact, and evaluate how the current conditions compare. historians of intelligence agencies have useful things to say here. 2/?

Best illustration of this is when I had a principal tell the staff he was a “data guy”. Problem was we were a continuation high school, where “data” was the absolutely last thing these kids needed - most of whom came from extremely shitty circumstances. Kind of summed up Ed reform for me.