First Results of My Mood and Emotions Tracking Experiment
In January, I started experimenting with two-dimensional approach to measuring mood. This approach was mentioned in one of the posts on Quantified Self website, but basically, in addition to using the “valence” scale (I feel bad/good), you rate your mood also on dimension of “arousal” (how “hyped” you feel). This weekend, with over 30 days of data, I had a chance to look at how well does this two-dimensional mood metric reflects my state of psyche, and ended up with an awesome visual map of my emotional states which confirmed that I can potentially drop the individual emotions from my self-tracking log.
To track my mood in January, I set up 10-point valence and arousal scales in my rTracker app and used them to log mood every morning, afternoon and evening. At the same time, I was recording the following basic 24 emotions (using yes/no scale):
Affection, Lust, Longing, Joy, Zest, Contentment, Pride, Hope, Relief, Surprise, Irritability, Frustration, Rage, Disgust, Envy, Suffering, Sadness, Disappointment, Shame, Isolation, Pity, Panic, Anxiety, and Boredom.
First, I looked at how often I experienced each of the emotions, and removed those with incidences less than 10:
I then computed average valence and arousal scores for the remaining emotions, “centered” them around the average monthly scores, and mapped in the valence/arousal space using Excel’s scatter plot feature:
The resulting map of emotions clearly demonstrates that the two-dimensional mood metric captures my emotional states relatively well. All the positive and intense emotions like pride, thrill and joy ended up in the upper right quadrant of the space. The negative and intense emotions like irritation, suffering and anxiety were in the opposite, third quadrant. The less intense emotions were close to the middle of arousal scale, but in the appropriate upper and lower parts of the plane, according to their valence.
This month, I will be using three-dimensional approach (also known as PAD model, or Self-Assessment Manikin) to tracking mood. The third dimension, “dominance”, reflects perceived degree of control over the situation. For instance, fear or sadness are submissive emotional state, whereas irritation or anger are dominant. I hope that all three dimensions (valence, arousal, and dominance) will be enough to track my emotional state accurately, and I can drop the individual emotions from my log.
In the next post, I will look at how two dimensions of mood predict happiness and stress. Stay tuned!