Embedded Files
What spawned the idea?
What spawned the idea?
Early in 2021, I realized that despite continuing to monitor and track how happy I was, I was not producing or collecting data as to why these were happy days.
Early in 2021, I realized that despite continuing to monitor and track how happy I was, I was not producing or collecting data as to why these were happy days.
Additionally, I began to move away from the goal of achieving greater happiness to instead achieving greater overall well-being. In February 2021, for example, the days in which I was home eating cereal and watching Grey's Anatomy was on average much happier than the day I went hiking, got coffee, and spent time around friends I have not seen for awhile. Long term, I wish to have more days closer to the latter.
Additionally, I began to move away from the goal of achieving greater happiness to instead achieving greater overall well-being. In February 2021, for example, the days in which I was home eating cereal and watching Grey's Anatomy was on average much happier than the day I went hiking, got coffee, and spent time around friends I have not seen for awhile. Long term, I wish to have more days closer to the latter.
Through my happiness rates tracking, I came to understand that even upon averaging moment to moment happiness, it was not directly link to what I was seeking to understand.
Through my happiness rates tracking, I came to understand that even upon averaging moment to moment happiness, it was not directly link to what I was seeking to understand.
I was not in a search to understand what leads to happiness, I was in a search to understand what leads to contentment.
I was not in a search to understand what leads to happiness, I was in a search to understand what leads to contentment.
How did I collect it?
How did I collect it?
Every day, I began to ask myself, "long-term, is today an example of a day you would seek" and I would provide myself a ranking from zero to ten. Zero as in that day is far from what I seek long-term, and ten being close to what I seek long-term.
Every day, I began to ask myself, "long-term, is today an example of a day you would seek" and I would provide myself a ranking from zero to ten. Zero as in that day is far from what I seek long-term, and ten being close to what I seek long-term.
Additionally, I was eager to evaluate the factors that can predict my overall well-being on a given day. Using Martin Seligman's PERMA theory, research gathered in Shawn Achor's book "The Happiness Advantage" and my three years of experience being attentive to how I feel for each fifteen minute interval of my life, I created a series of questions I would answer. The questions are included below, titled, "Well-Being Potential Predictor Questions".
Additionally, I was eager to evaluate the factors that can predict my overall well-being on a given day. Using Martin Seligman's PERMA theory, research gathered in Shawn Achor's book "The Happiness Advantage" and my three years of experience being attentive to how I feel for each fifteen minute interval of my life, I created a series of questions I would answer. The questions are included below, titled, "Well-Being Potential Predictor Questions".
Well-Being Potential Predictor Questions
Well-Being Potential Predictor Questions
How busy were you today overall:
If I were to pick a random time in the day, what's like likelihood you'd remember what you did?
Of the things you did today, what percent of it was spent doing things you love doing?
In general, are you content with the work you are doing everyday? How much does it feel like your "calling"?
Are you proud of what you accomplished today?
Excitement: Upcoming Week
Excitement: Upcoming Months
Excitement: Upcoming Year
Anxiety: Upcoming Week
Anxiety: Upcoming Months
Anxiety: Upcoming Year
Contact with someone virtually
Contact with someone virtually who you somewhat enjoy or more
Contact with someone virtually who you really enjoy or more:
Contact with someone virtually who you absolutely love or more:
Contact with someone virtually who you don't enjoy or less
Surrounded by one person F2F:
Surrounded by one person F2F who you somewhat enjoy or more:
Surrounded by one person F2F who you really enjoy or more:
Surrounded by one person F2F who you absolutely love or more:
Surrounded by one person F2F who you don't enjoy or less:
Surrounded by people (2) F2F:
Surrounded by people (2) F2F who you somewhat enjoy or more:
Surrounded by people (2) F2F who you really enjoy or more:
Surrounded by people (2) F2F who you absolutely love or more:
Surrounded by people (2) F2F who you don't enjoy or less:
Surrounded by people (3 or more) F2F:
Surrounded by people (3 or more) F2F who you somewhat enjoy or more:
Surrounded by people (3 or more) F2F who you really enjoy or more:
Surrounded by people (3 or more) F2F who you absolutely love or more:
Surrounded by people (3 or more) F2F who you don't enjoy or less:
Your location today, have you lived there at least 6 weeks?
Your location today, have you been there physically 3 of the past 6 weeks?
Your location today, have you been there in the past week?
How sick do you feel?
How sleepy do you feel?
How exhausted do you feel?
How hungry do you feel?
How cold do you feel?
Coffee Time - 1
Coffee Time - 2
Coffee Time - 3
Exercise Start Time - 1
Exercise Finish Time - 1
Exercise Start Time - 2
Exercise Finish Time - 2
Tired Start Time - Sleepy
Tired End Time - Sleepy
Tired Start Time - Exhausted
Tired End Time - Exhausted
What is Linear Regression?
What is Linear Regression?
Let’s say you love to run, and want to be able to predict how fast you will run on a given day.
To the right is the data you have collected. For example, on the day you ran a mile time in 8 min, the temperature was 50º, you slept 4 hours, and ate 20 grams of sugar total.
Multiple Linear Regression allows understanding how these different factors interrelate, and whether or not they can predict mile time. The strength of the prediction is denoted by the r-squared value, where a value closer to 1 indicates the factors are strong predictors, and a value closer to 0 indicates weak association between the variables. Additionally, it would produce coefficients, which are used to create an equation such as the one below.
Mile time = .02*(Temp) + .30*(Sleep) + .06*(Sugar)
Linear Regression Models
Linear Regression Models
Factors that Predict Well-Being
Factors that Predict Well-Being
In the model above, the r-squared value is 0.87
In the model above, the r-squared value is 0.6243
Components of Physical Well-Being
Components of Physical Well-Being
To the right are the factors evaluated in terms of sleep and how it affects exhaustion and sleepiness measured throughout the day.
Multiple linear regression found no significant impact on any of these factors as it relates to overall exhaustion or midday exhaustion
Analysis of Results
Analysis of Results
The purpose of this is to evaluate and analyze the factors that contribute to well-being in myself to aid in the design and direction for future work, as well as future iterations of tracking. All comments regarding the results are preliminary and may not hold true with more data, time, or testing.
For days where positive emotion is relatively equal, but well-being rating differs, “Well-Being Linear Regression Model (simplified)” emphasize a simplified version of the specific factors that made it differ. Daily positive emotion values below the 25th or above the 75th percentile of all the positive emotion values were removed at the start of this analysis. This has four predictor variables, with well-being rating as the response variable.
A rating of ten for the predictor, “074-How busy were you today overall?” is “no free time at all”, and a ten rating for “077-In general, are you content with the work you are doing every day? How much does it feel like your "calling"?” is “incredibly content”.
Like the model represented in “Well-being Linear Regression Model with Emotional & Physical Measures as Predictors”, anticipation for the upcoming week and the influence of being surrounded by individuals whom one “really” enjoys is significant in predicting overall well-being. However, when excluding differences in positive emotion from affecting well-being, the “meaning” facet of PERMA theory (positive emotion, engagement, relationships, meaning, accomplishment) has strong significant correlations. Additionally, amount of time one spends dedicated to something—how busy was defined—is a factor that is significant in predicting overall well-being provided equalized positive emotion.
In terms of the impact of sleep, exercise, or diet, on well-being, there are no major association between these variables independently as indicated by the Adjusted R-squared values in the models. For example, when only looking at only nutritional intake as it relates to well-being, there are no significant correlations.
Dietary factors included in this model include water intake (measured in fluid ounces) and caloric intake (measured in kilocalories), and both are on the day (T) in which well-being is being assessed. Dairy intake is a subjective measure on a ten-point scale where a rating of ten is “no dairy consumed”, and a rating of five is described as a “decent amount”. Additionally, this measure is averaged with the values on the previous three days (Y3) prior to day the well-being is being evaluated. For instance, the average of the dairy intake of Day 1, Day 2, and Day 3 would be used as a predictor variable for the response of the well-being rating on Day 4. “Salt” is an objective measure of the average sodium intake in milligrams for two days (Y2) prior to day the well-being is being evaluated.
Physical factors included in his model include a rating of how “exhausted” the individual felt during the day. One important note is the distinction between “exhausted” and “sleepy” where “sleepy” is more of an intense overwhelming “wave” of tired, as “exhausted” is more like how one feels when their body feels as if it needs sleep. Exhaustion ratings are measured on a ten-point scale, where a ten indicates that at no point in the day did the individual feel “exhausted”. The predictor “cold” is a measure of if I felt cold in the environment, not a measure of the actual temperature of the environment, where a rating of ten indicates that at no point in the day did the individual feel “cold”. Finally, the time in which the first coffee beverage of the day was tracked and recorded, where “7:15 AM” would be recorded as 7.25 and “3:30 PM” would be recorded as 15.5.
The combination of these factors continues to highlight the aspects discussed earlier as to the important components of well-being, but also now consider physical and dietary components. As for physical factors, the level of exhaustion experienced within a day is associated with a strong negative correlation on overall well-being for that given day. Additionally, warmth is an important factor, where days in which I felt cold had a negative correlation on that day’s overall well-being. Dietary decisions have a role in the impact on the individual as well, as increased water intake and caloric intake on a given day are directly associated with well-being. Furthermore, average daily intake and sodium intake in the prior days affect the individual as well, where higher sodium and lower dairy intake led to greater well-being.
Distribution of Happiness Rates on Well-Being
Distribution of Happiness Rates on Well-Being
To the left is an example of how distribution of happiness rates was evaluated. For example, on April 1st, 33% of that day was spent at a mood of 5.5.
Analysis has indicated that moments rated at 5.9 and above are strong predictors for overall well-being in a given day, but other components of the distribution have no significant impact. This emphasizes the importance of having very happy moments in the course of a day in order to achieve a higher state of well-being on a given day. Additionally, this highlights how very unhappy moments, or moments opposite of positive emotion are not as substantial.
Page updated
Google Sites
Report abuse