I wonder how the knowledge that you are tracking yourself affects your results. Studies suggest it would. People paid for coffee more often on the honor system, for instance, when researchers placed a picture of eyes above the collection basket (PDF). But it would seem hard to test—how would you track yourself without knowing it?
In some experiments, you can blind yourself. Often you can have a friend help blind you, too. Ideally you have the kind of friend who will track data about your without you ever mentioning it so that you won’t know that it’s being tracked, just because the friend knows you would want the data. But perhaps that’s not a realistic scenario.
If we build complex enough software that tracks all sorts of stuff the software will probably track some variables that you didn’t know about.
If you have a iPhone that iPhone simply stores all location data for you without telling you and it only fairly recent that the behavior became public knowledge.
After you discover the data tracking you can use the feature for data analysis.
At this year’s QS Conference Kevin Kelly devoted a part of his closing keynote to talk about his notion of the “Triple Blind” study. I could summarize it, but Ethan Zuckerman has already done a fantastic job:
You can find more of Ethan’s thoughts about the QS Conference here. Probably the most well-written account of that wonderful weekend.
Thanks for this Ernesto.
The effects of observation in self-tracking: are these a feature or a bug? As a feature we might call it: “self-awareness.” As a bug we might call it: “self-consciousness.” (In the social sense of excessive self awareness, embarrassment, or shame.) Ryan’s paper on the cultural/legal questions raised by changing the observation environment through non-human observation has been really useful for me:
People Can Be So Fake: A New Dimension to Privacy and Technology Scholarship
M. Ryan Calo, Stanford Law School
Penn State Law Review, Vol. 114, No. 3, 2010
The effects of observation on behaviour can be used to a great advantage. If you blind the experiments (the medical model) you are effectively choosing a probability function, which means every possible type of behaviour will, over time be displayed. However if you can use observation to gain optimum results by using conscious intent (quantum physics) to select the behaviour you desire to see. Thus you have in a sense “the collapse of the Schrodinger equation” so only that behaviour that you select will be seen.
To optimize observation of behaviour, to get the best result, use what I call “a mental prescription”, in which you state the behaviour that you desire to see right at the start and state it as if it has already been observed. You will be surprised how much you can affect /enhance your performance.