Ask HN: How do you avoid cherry-picking data? With all the data-powered hot takes on Covid, racism, police violence, government spending, etc., how does someone avoid cherry-picking data in order to support their own or their in-group’s bias? Is this possible? |
2. Play devil's advocate with the information you get.
Personally, i think it's important to remember that the world is very complex and 2 views can be different and at the same time true.
Nickel and Dimed[0] is a book on low-wage jobs and how it is difficult to get by without support from some kind of social program.
Scratch beginnings[1] is the story of a guy that tried to live in a shelter and get by. In ten months he had moved into an apartment, bought a pickup truck, and had saved around $5,300.
Both these stories are opposing, in the logical sense. But they also real. Exploring different perspective helps dispel the notions that there is one "truth".
[0] https://en.wikipedia.org/wiki/Nickel_and_Dimed [1] https://en.wikipedia.org/wiki/Scratch_Beginnings
I do the same. For me it's result of applying "scientific thinking" to everything, and searching a "grand unifying theory".
I agree that conflicting realities should be displayed, as long at there is no effort to try any reconcile them and force some homogeneity.
1. Time. In the moment it's very hard to understand what's going on. Time gives perspective. The down-side to this is that raw data sometimes gets lost with time, and then the fights begin about how to interpret what data there are.
2. Judgment. In the moment, you have to use a combination of historical track record of the data source and your own judgment. New York Governor Cuomo did a good job of explaining the subtleties of this at his daily briefing yesterday (Saturday), which I recommend watching. In summary, there are certain things doctors and the WHO thought were true and recommended several months ago that turned out not to be true, and new information and policy recommendations get disseminated as we learn more.
You try to prove that your hypothesis is wrong. If you can't find evidence that your hypothesis is wrong, you can "accept" that your hypothesis is right.
This forces you to try to find holes in your logic. If you can't find any, probably it is solid.