Fred Truman
Design & Code Creative
QS Review – Tracking the Subjective against the Quantitative
September 24, 2011
As part of this week’s assignment for DIY Health, we were asked to take a look at the archive of Quantified Self talks and review a couple by considering some question, like:
Are objective / quantified metrics always better? Are subjective/qualitative measures always suspect? Is QS a way of discreetly organizing our body? Are we beginning to set up a shared language about the body and the self? What the hell might this lead to? Will this look more like Gattaca? Or something else completely? Can we ever really know our bodies? Can we ever really know our selves?
I watched about ten of these, mostly on the first couple of pages of the vimeo group and found some more interesting than others. I wondered why some people were investing all this time to disrupt their normal lives to do a particular experiment if they had no clear understanding of they were either trying to prove or disprove. The couple talks I did find interesting had very explicit commentary on what they believed we should be trying to achieve through self-tracking. For example, Quinn Norton gives a really quick but useful talk on self-tracking for insight:
What I found to be valuable here is that she stresses that the results of our experiments will only ever be as good as our questions – something possibly more obvious to a scientist but less so to us techno-tinkerers. We should set out to either prove or disprove something and that if we do find our results to be negative, we should just accept it and move on. If looking at the data we’ve collected we find something useful, the most important step is to act on what is the data is telling us. The data is in fact the means, not the end. So – she stresses – we should put down our notebooks and our data and apply its insights to improve our subjective experiences of the world which is in fact what we should be optimizing (not data). In order to best do this, we should try to track our subjective experience alongside quantifiable data – something I believe would improve a project like this:
This is no doubt a beautiful and very nice project, however I believe that it’s emphasis on the subjective without a corresponding emphasis on some quantifiable data doesn’t allow any real insights to shine through (but yes – she did mention that using skype did improve her mood).
In addition to tracking a subjective experience against quantifiable data, I found that Seth Robert’s suggestion that we can use the QS methodology to prove or disprove assertions made by those more established and brainy thinkers that otherwise lack the tools to test their ideas through experimentation. He suggests specifically that QS methods could be used to test out the benefits of the Paleo Diet. Although I will admit to knowing relatively nothing about this diet, I do think that this builds on what Quinn Norton was saying. We should really have a good idea of what we’re asking of our experiment, and while we’re at it why not try and take a crack at some of these ideas already floating around out there and work towards some kind of resolution.
The last one I want to mention is this talk by Kai Chang as to the benefits of CrossFit:
I think he has a nice insight here, which gets totally lost, and that is the idea that joining a gym is not a sufficient motivating factor for personal fitness. This is where BJ Fogg’s model might come in handy. Knowing that the factor he has isolated is motivation, he might better be able to communicate the value of his experiment. Without doing this by the end of the talk you start to believe that this guy is saying he exercised and got in better shape. Great. We all know this. Looking carefully, though we can see that what he’s saying is not that CrossFit got him in shape, it’s really that CrossFit acted as great “spark trigger” by giving him the repeating metrics to motivate his participation.
In summary, my takeaways are this:
- Focus on ideas to prove or disprove
- Track the subjective and the quantifiable at the same time
- Communicate insights by building (on existing or new) models
[image credit: Fellowship of the Rich]