Have you ever had those days when you have been feeling fine and then all of sudden you feel crappy? Or, on the opposite, you wake up in a bad mood, but as the day progresses, you are getting better? If you track your mood (or any other psychological state, to that matter) several times a day, there is a way to quantify severity of these “swings”. In this post, I will show how to calculate the “mood swing scores”, and how you can use these scores to learn more about changes in your mood and what causes them.
Statistically speaking, we are looking to measure variability in mood ratings within a single day. There are two simple indicators that you can use for this purpose: variance and range. Let’s look at both, using my October data as an example. I track my mood thrice a day: in the mornings, afternoons and evenings, so my daily records look like this:
The range is the easiest measure that you can calculate on paper or in Excel. It is simply a difference between the highest and the lowest mood ratings on that day. For instance, mood swing score for October 5 is 8.49 – 5.30 = 3.19, and the Excel formula for that cell would be MAX(B6:D6)- MIN(B6:D6). The main advantage of using range is that the scores are easy to understand and interpret. The higher the score, the more severe was change in the mood. The main disadvantage of this method, however, is that it captures differences between the most “extreme” ratings on that day, but completely ignores any other ratings in between. In my example, the swing score took into account difference in the mood in the afternoon and evening, but completely ignored the morning rating.
The variance, on the other hand, takes into account mood ratings across all three day parts, by looking at their “closeness” to the average mood rating on that day. You can compute variance in Excel using VAR formula. For instance, to compute variance in the mood ratings for October 5, I would use VAR(B6:D6). This results in a mood swing score of 2.73. As you can see, unlike the range-based scores, variance-based scores are more difficult to interpret intuitively. Think of variance as an average “distance” of all mood ratings on that day from the average rating. The lower the score, the “closer” are all rating to the daily average, and thus, to each other.
Please note that while both methods capture the magnitude of the change in your mood, neither of them shows the direction (negative or positive), and I will cover that kind of analysis in a different post. But ranges and variance based scoring methods are enough to search for meaningful patterns in your mood changes and identify the factors that affect them. Here are just a couple of examples.
Mood Swings and Day of the Week
Using my October data, I computed daily mood swing scores, and then compared averages across the weekdays:
According to this chart, on Tuesdays and Wednesdays my mood tends to be more stable throughout the day, compared to other days of the week.
Mood Swings and Rain
I also compared average mood swing scores on rainy and non-rainy days. As you can see, my mood is less likely to change drastically on rainy days.
Finally, which scoring method to choose is up to you. As the above examples show, both methods capture the mood swings equally well. If you prefer using just one metric, the choice would probably depend on how often you track mood during the day. If you log your mood only twice a day, then the range or even actual difference (e.g., morning mood minus evening mood) would work. If you record your mood three or more times a day, I personally would recommend using the variance. And of course, you can also use these two methods to track “swings” in other psychological and physiological states, like happiness, stress, blood pressure, etc.