An Information-Theoretic Analysis of Adherence to Physical Exercise Routines

<p>One of the most common recommendations in healthcare is to simply form healthy habits, but little research has been done to understand the formation and continuation of a healthy habit that isn’t heavily influenced by an individual’s interpretation. Arizona State University’s WalkIT study aimed to analyze how goal setting and financial reinforcement can influence moderate-to-vigorous physical activity (MVPA) in adults, while using data from accelerometers to alleviate individual bias. In this trial, 512 insufficiently active adults were recruited to wear an accelerometer for 1 year and were then randomly assigned to one of the four study groups. Each group had either a static or adaptive MVPA goal and received financial rewards either based on a predetermined steadily increasing schedule or earned points towards their financial rewards by achieving their set MVPA goals. Data was collected from each of the 512 participant’s accelerometer for every minute of every day for the full duration of the 1-year study indicating their physical exertion levels. In this thesis, that data is analyzed to investigate how habitual participants are with their physical exercise routines. We hypothesize that the more consistent participants are with the time of day they exercise, the more they will exercise. However, with the use of information entropy, a linear regression model, and the R² statistic we disprove this hypothesis. Instead, we find that the more a participant exercises, the less habitual they are with the time of day they exercise. Given how busy our lives tend to be, this aligns with the idea that we are more likely to exercise if we fit it in whenever we have the time to, rather than attempting to stick to exercising solely at a specific time each day that will inevitably become busy when something unexpected pops up. With the natural chaos of the society we live in today, it’s easier to meet goals that have built-in flexibility rather than trying to control the uncontrollable.</p>

Name thesis:degree_name

Master of Science (MS)

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computational and Data Sciences

Year

2021

Author dc:creator

• Foster, Lily

Contributors dc:contributor

• Erik Linstead
• Vincent Berardi
• Rene German

• Habits
• Linear Regression
• Information Entropy
• Physical Exercise Routines
• Machine Learning
• Motivation
• Data Science

Repository record dc:identifier

https://digitalcommons.chapman.edu/cads_theses/12

OAI identifier oai:identifier

oai:digitalcommons.chapman.edu:cads_theses-1011