On November 8th, I drove to Cambridge, MA from Hartford to attend the Civic Analytics Network’s Summit on Data-Smart Government. Conveniently this event was within driving distance for me, otherwise, given the tight budget in the State of Connecticut, I might not have been able to attend. I’ve been casually following the Civic Analytics Network since its inception; periodically browsing the website and through Twitter. When the group published An Open Letter to the Open Data Community, I saw a group that shared many of the same thoughts as I do. In fact I “liked” and re-tweeted it; which in this day and age represents a fairly hearty endorsement (despite what we all say in our bios). What resonated more with me, was that the group was able to develop some consensus about where “open data portals” should be headed, share that out into the community; and that the private sector responded.
It turns out that if you don’t include hidden common causes in your model, you’ll estimate causal effects incorrectly. This raises a question: can we possibly hope to include all of the hidden common causes? What other alternative is there, if this approach fails?