Exploring the Data
Earlier in 2019, A large study was released regarding coffee consumption and its health benefits and risks. Research of 347,077 subjects showed that 5 cups of coffee a day is in fact, the magic number. Drinking 5 cups a day can reduce ones risk of heart failure, stroke, and even death. Drinking 6 cups on the other hand, will increase ones risk of heart disease by 22 percent. With that research in mind, it got me thinking; I am quite the coffee drinker. I enjoy a cup every morning, but after that it’s anyones guess. For this project and for my own wellbeing, I decided to track my coffee consumption for two weeks to try and nail down how much coffee I drink on average, why I drink it, and it’s immediate consequences, if any.
These first two visualizations represent coffee consumption, hours spent asleep, and number of significant obligations on each given day (which I thought might also impact the number of coffees I consume). In the first chart, Coffee Consumption Vs. Hours Slept, I wanted to see if there were trends between lack of sleep, and coffee consumption. This quickly turned into a “What came first, chicken or the egg?” scenario. I expected a lack of sleep to increase my coffee consumption the next day, when in fact, the graph shows that I first drank more coffee, which may have in fact caused a sleep deficit the following night. This was the beginning of a vicious cycle, but how did it start and when would it end? That’s when I decided to explore my number of significant obligations such as work, class, weddings, projects due, on each day I was tracking. The second chart shows an increase in coffee consumption on days that I have more significant obligations. To dive deeper into this idea of coffee consumption and lack of sleep being caused by significant obligations, I created the following two charts.
In this pie chart we can see that Thursdays have the most number of obligations on average, and Sundays have the fewest. Hovering over each section of the chart will show us the exact average number of obligations and day of the week. In the circle graph to the right, size and shade of the circle represent average coffee consumption by day of the week. Thursdays, I tend to consume the most coffee. Hovering over the circle will show us the exact average number of 12 oz. coffees consumed. To compare the data on these charts, I encourage the viewer to click on the day of the week in the legend in the top right corner. This will highlight the day on both charts, showing the correlation between number of obligations, and coffees drank. For example, Sundays have the fewest obligations and coffees consumed on average. This brings us full circle to our earlier question of what exactly ends the vicious cycle of coffee consumption and lack of sleep? The answer seems to be fewer obligations on Sundays.
Justifying the Design
For the first graph, I decided on a line graph because it best shows trends over time. I was looking for the lines to mirror each other, with one line peaking and the other dipping, which is exactly what I got. Closer evaluation revealed something I wasn’t expecting though, coffee causing a sleep deficit. I chose to make the second graph a line/bar graph because I wanted to show coffee consumption over time, while revealing the number of obligations on each day of the study. Similar to the first line graph, I chose to make the lines thicker and darker with number of coffees to emphasize their increase in quantity.
The pie chart was a calculated risk. Still, since I’m working with so few variables and exact percentages aren’t the focus here, it seemed like an appropriate route. I did utilize annotations for this chart to help the viewer decipher the material without having to interact with it. For the circle graph, my goal was to have a legible visualization that would allow the viewer to quickly see average coffee consumption by day of the week. I kept the circles grey in unison with the rest of the graphs showing coffee statistics, but made sure that the X axis matched the legend for the pie chart so that points could be highlighted and singled out.
Next Steps
In the future with more time at my disposal, I would like to have had a larger data set to work with. These two weeks were not as routine, with two four-day weekends in a row. I think more data would give me more accurate averages to work with, and bring new findings to the table.
This entry is licensed under a Creative Commons CC0 1.0 Universal Public Domain Dedication license.
