A Cohort of Self-Experimenters: Lessons Learned from N=1 Personal Informatics Experiments

Personal informatics, the trend in collecting and analyzing data about one’s own self, has been heavily influenced by the involvement of technology in our daily lives. In this paper we aim to understand what happens when novice self-experimenters are provided with a structured lesson in experimental design. We conducted a month-long study on self-experimentation, where twenty students in a seminar performed a self-experiment of their choice. The students were prepared with background readings and lessons on statistical analysis and experimental design. Their experiments were designed in a structured manner: a specified number of variables to track and a set duration for the study. We also analyze videos of self-experimenters from the Quantified Self community, and compare them to the methods and outcomes used by the students. We find pitfalls that both students and self-motivated Quantified-Selfers experience, such as a too short duration of the study and insufficient planning of the set up of the experiment. Based on these findings, we propose an iterative self-experiment design method that addresses those pitfalls. We also discuss broader implications for future self-experimenters and designers of tools for self-experimentation.