Life generates a lot of data – the number of hours you spend awake and asleep and at your desk and in your car; the foods you eat; the pace of your morning run. I just read a great article in the New York Times Magazine, The Data-Driven Life, about geeky guys who are obsessed with self-tracking – gathering data about their daily lives and distilling from these data information about how to improve their lives. How to work more efficiently, run faster, feel happier. This got me thinking about how self-tracking might be a great hook to get people interested in science – and in one of my favorite little corners of science, data analysis.
The guys in the article were for the most part tech-savvy nerds. They recorded copious data, like what they ate, how much money they spent, and… every single idea they had since 1984. This allowed them to figure out things, like exactly how much time one man spent cleaning up after a messy roommate. Or how many tablespoons of flaxseed oil another man needs to take to maximize his concentration.
I think people could really get into this. Everyone likes to know more about himself. And there could be some tangible benefits: in today's bad economic times, people could really benefit from knowing exactly how they spend their money. And in this image-obsessed society, understanding how diet and exercise affect your own weight – not just the weight of Hollywood types – could be very very appealing. Is that $10 skin cream just as good as the $100 skin cream? Run your own test (controlling for potentially influencing co-variables!) and find out for yourself.
Collecting data is easy. There are all kinds of ways to automate data collection (companies like Google collect tons of data about you already). Depending on the kind of data you want to collect, cheap sensors might already exist, such as accelerometers for your running shoes, or tiny temperature loggers (I tend to sleep better with the window open – is that because the room is cooler?). And smartphones make it easy to collect data on the fly – and can remind you to record your mood or what you ate.
The next step, data analysis, is the best part. This is where people can really learn about how science works. Using their own life as an example, people can understand concepts like correlation versus causation, signal versus noise, controlling for confounding variables, and the importance of replication. It would be super to have a website to help you track your data – and easily analyze and graph it. The website Gapminder is an elegant model.
Last week, I started keeping track of how much time I spend working on each of my many work projects each day. Myriad web-based programs are out there to help you track your time. For me, self-tracking is simultaneously a bit tedious and super interesting. I can see how, once I get into the habit of recording what I'm doing, it can become an obsession. There are so many questions I can ask about my own work habits and efficiency! Now I need a spreadsheet to track more variables – only then will I figure out if my ability to concentrate correlates with the number of cups of coffee I consume!