Shared Reflection for Personal Informatics

Posted 07 Apr 2017

The paper that Lisa Graham recently presented at GROUP 2016 was the culmination of her MSc work. I’m really proud of her work on that project, because it was a very difficult study to run, but she was nevertheless able to complete the study and analyze the data with an interesting set of results.

The context for Lisa’s work is that we are increasingly able to track and monitor person data in a wide variety of contexts. Our watches are able to track the steps we take, while other tools help us to monitor and track exercise and fiteness goals, while still others can track our spending habits and so forth. With all this data though, the problem is that most people do not know how to analyze the data in order to generate generate meaningful, actionable insights. That is, we have all this data, but we don’t know what to do with it.

Lisa’s project focused on trying to jump start this reflection process, and it does so by leveraging other people. The basic idea for the project comes from the observation that, “It is often easier to see flaws in someone else’s essay than in one’s own essays.” Similarly, our hope was that it was easy to find new ideas for exploration while looking at someone else’s data than when looking at one’s own data.

Lisa’s study was set up in such a way that people would go through several weeks of personal data collection (on whatever data they were interested in collecting) in cohorts. Within this cohort, on a weekly basis, they would be asked to review someone else’s data (within the cohort), and return that review to the owner. Our thinking was that this review activity would have two different effects: (1) first, one would get new ideas about one’s own data (from someone else) – a new perspective, perhaps; (2) second, based on review of someone else’s data, one would be able to see one’s own data from a new, fresh perspective. (This is the basic idea behind “peer evaluation” in teaching/learning circles.)

In the end, we definitely saw both of these effects, where people:

  • Gained new insight due to the work of others. For instance, there were sometimes suggestions for how to engage in the practice of data collection (e.g. how to count things, or how to think about things). Other times, there were suggestions about new actions one could take on the data (e.g. if one had plateaued on the way toward fitness goals).
  • Found new ways of thinking about their own data. This was sometimes the direct result of seeing and reviewing someone else’s data. For instance, some participants developed new ways of collecting, categorizing or thinking about their data after having seen someone else’s data collection process.

And while this process seems promising, the major challenge with itt is that process-wise, it is very difficult to execute. Lisa had a hard time recruiting participants, and then ensuring that they stayed on task for the duration of the study. Nevertheless, the approach shows promise that is worth investigating further.

You can read more about the study in our GROUP 2016 paper.