My Journey through Fuqua

Using Apple Health data to talk about my journey through the initial few months of business school

With ever-increasing data privacy concerns, I realized that my phone must definitely be tracking my movements during these initial few months at Fuqua. And within no time I realized I could access everything from the headphone audio levels to my walking step-length. While working with online datasets might lead to more meaningful insights, working with your personal data has a kiddish excitement of its own. Within no time, I was learning how I could access my Apple Health data so that I could visualize what I have been up to and whether I was walking enough steps to offset all the stress eating (we have exams every 6 weeks).

This data starts recording on my first day of the MQM program at The Fuqua School of Business. Over the course of 85 days that span almost 2 academic terms, I walked more than 345 miles and climbed 1185 flights of stairs, with a cumulative step count of ~778000.

I have used box-plots that use my daily average metrics and compare them to the recommended values based on my physical characteristics such as height (represented with the bold red line). The average strolling pace is 2.45 mph (miles per hour) and the brisk-walking pace is 3mph for a 5 feet 11-inch tall adult. Fair to say, my walking pace is not bad considering that the majority of days the average pace is between 2.68mph and 2.88mph.

On the other hand, the above two box-plots do not paint the best picture of my walking habits. Double-Support time is the percentage of time during a walk that both feet are on the ground. A lower value would indicate that I spend more of my walk with my weight on one foot instead of two, which is a sign of better balance. With a recommended percentage value of 0.29, my data suggests slightly above-average Total Double-Support Time. With a better walking posture, I should be able to improve on my balance.

Everyone hopes to reach the magical 10,000 steps everyday mark. With a median of 9160 steps, I accumulated 31 out of 85 days with more than 10,000 steps.

I plotted my average step count every hour of the day, and a simple polynomial function was able to account for 80 percent of the variation in step count by the hour of the day. This graph was especially interesting as it gives me a birds-eye view of my daily schedule. Since my life currently revolves around this master’s program, the following are some interesting insights.

Evidently, I was attending afternoon classes as my step count is drastically lower between 2 pm-4 pm than that of times surrounding it (when I was probably rushing to (and from) campus). Other such crests and troughs are also visible for other class timings. For example, there is a significantly big rise in step count between 7 am and 8 am as I was gearing up to attend 8 am Data Infrastructure lectures (yes, you heard that right). Also, I wish I had a good explanation for having 300 steps (on average) at 3 in the morning or why my step count for 12 noon is similar to that at 12 midnight.

Sundays are like wildcards (highly variable) — either I am out there walking 15,000 steps exploring Durham and the beautiful Duke campus, or I am in bed all day, with the only steps that I take are from my bed to the kitchen and back. Weekdays are usually standard, with most weekdays having around average step counts. There is, however, an exception to this trend — Wednesdays. Wednesdays are the real weekend for us as part of the MQM cohort here at Fuqua School of Business. We have no classes that day and there is a substantial difference in the average step count on Wednesdays from other days. I like to compare this strategy by Fuqua as the afternoon-nap equivalent for our highly demanding weekly schedule.

Through this exploration of my Apple Health data, I learned that Apple may have all my data, but only I can join the dots to narrate the complete story.

You can view the code of parsing the XML Apple Health file with python on my Github and the visualizations on this Tableau Public post.

Thank You!