Brushing up on my statistics I was reminded of the difference between a sample survey and an experiment. When the laboratory, population and subject of your efforts is yourself it is difficult to have a control group. You can’t have a double blind experiment when you are the experimenter and the target of the experiment. Even gathering information can be difficult as self reporting is entirely subjective. Here are three things I have found:
Surprise yourself – Design data collection in a way that a silent alarm triggers you to enter some information. Your conscious mind will forget that you are studying yourself and you will get a truer measure. For example, when I entered information about my mood based on when my mood was bad I would enter it inconsistently. When I set a silent alarm and entered my mood every time it alerted me I got a much nicer survey result.
Use Devices – Devices on their own can be a distraction if they are acquired for their own sake. When quietly observing your activities and collecting on an ongoing basis as part of a study you are conducting they are excellent. The data collected can be surprising and useful. And use your devices to support your studies. Don’t buy them for their own sake.
Keep Study Period Short – A fourteen day study can give you a good sample set. I have let my studies go for a very long time and though I had a mountain of data at the end, the insights could have come in shorter periods. Shorter periods keep you moving through topics quickly and can prevent exhaustion.
Quantified Self studies are in early days. We have a lot to learn, and a lot we need to discover. Keep surprise and time on your side.