Monday, January 02, 2023

It's not just about data


The title of this post would shock the people at the company I used to work for. After all, our job was to make hardware and software that let you collect, store and make sense of data in the real world. It was all about the data. But just collecting data doesn't do any good unless you know how to use it. And it doesn't do other people any good if you haven't collected the correct data. If you are measuring the temperature of the ocean but never put your temperature sensor in the water, the data you have and write about and collect money for is useless and could lead to real harm.

This book, like any good book, is a mixture of a lot of disciplines from mathematics to economics and from psychology to philosophy. And it is very easy to understand. It gives you tools to use to not only make sense of data but to make sense of every day information. It will help you look through lies and sloppy thinking. It will help you uncover "fake news" and outright lies.

I find the rules set down by Mr. Hartford useful when I'm reading the paper or when I'm studying the Bible. When you are scanning YouTube trying to learn something new or looking to expand your knowledge about some subject, it is too easy to fall into the trap of believing the well edited and well organized presentations.

The first of ten rules is "Search Your Feelings". This chapter starts with a story about an art critic being fooled by a forgery. And because this critic was so respected, it was accepted by the larger art world as well. How had an expert been fooled? Because he wanted so badly to find a lost painting by the master Vermeer. And, as Mr. Holford points out, this is why we buy things we don't need and why we do things only to impress our friends. So, one of the first rules in making sense of statistics is really about human psychology. You have to not let your emotions cloud your judgement.

The second rule is "Ponder Your Personal Experience". When we see some statistic, sometimes it helps if we have some personal experience that makes us question the data. Not that we should just throw out any information that we can't verify ourselves but just let it be an impetus to look further. As someone once said, it's not the education or experience that makes a good scientist. Great scientific discoveries come from people who notice something and say, "Well, that's strange. How could that happen?" 

The third rule is "Avoid Premature Enumeration". He starts this chapter with a discussion of a study that looked into infant mortality in the United Kingdom. The rate was much different in different sections of the country. Could they look at what was being done inn the area where mortality was high and change their procedures to that of the area where the mortality rate was low? It turned out that most of the difference in mortality rates was the naming of various stages of gestation. In some areas, babies deaths were labeled as miscarriages instead of an infant dying. And if we want to gather statistics about "children", say what their interests are or what they like to read, don't we first need to define what we mean as a child? This happens a lot. We often think that to solve a problem, we need to jump right in and start measuring things and counting things. Our first step needs to be defining what we are looking at and what we are searching for.

I won't list all ten rules here. I hope I've gotten you interested enough to explore this book yourself. You will not be disappointed no matter what your job or hobby or life is like. Everyone should be given a copy of this book when they enter high school and then be taught to use these rules for the rest of their life.

No comments: