Google Data Analytics Case Study -Bellabeat
Google Data Analytics Case Study -Bellabeat
Embedded Files
In this case study - presented by Coursera's Google Data Analytics certification program - I used outside smart device data from Fitbit users to evaluate business practices, determine trends and theorize solutions to maximize usage of the Bellabeat app and its most prominent "Leaf" device.
In this case study - presented by Coursera's Google Data Analytics certification program - I used outside smart device data from Fitbit users to evaluate business practices, determine trends and theorize solutions to maximize usage of the Bellabeat app and its most prominent "Leaf" device.
I then determined steps moving forward based on these findings!
I then determined steps moving forward based on these findings!
Within this study, I found that users of the Fitbit device...
Took an average of 6,638 daily steps
Traveled an average of 5.49 miles
Spent an average of 991 minutes in a sedentary state
Spent an average of 21 minutes being very active, and burned an average of 2,304 calories.
Using the Loess method, I found a low correlation and negative relationship between steps taken and sedentary minutes.
Those that are not as active & take less steps on a regular basis burn less calories!
FINDING #1: MOVEMENT IS THE KEY! By encouraging Bellabeat users to simply get out and maintain a fitness/walking schedule, they are making their lives healthier.
Throughout my analysis, there were additional correlations between caloric deficit, lightly active AND fairly active methods of movement.
Throughout my analysis, there were additional correlations between caloric deficit, lightly active AND fairly active methods of movement.
FINDING #2: Therefore, any movement is good movement! Though encouraged, users of the Leaf device do not have to stick to a strenuous workout plan; in place of them, opt for much more moderate, low-intensity habits (e.g. - walking the dog, taking the stairs over the elevator, taking a stroll to your nearest supermarket) aid in calorie burning.
The Loess Method does as said - shows trends in data plots that contain less than 1000 points. I found that there are trends that move in an upward direction; calories are burned when there is any consistent movement.
All three of these scatterplots using geom() functions show several correlation between sleep and activity level but relay one single truth...
FINDING #3: That those who logged their sleep data and have high & moderate levels of activity often get the preferred hours of nightly sleep (~7-8 hours). Encouraging Bellabeat users to track their sleep data on a regular basis and inspire them to move (while also notifying them of when they are oversleeping/not sleeping enough) aids in the maximization of healthiness.
The final visualization we have shows the number of times users logged their data into their Fitbit devices, as the chart suggests.
The middle of the week showed the most data tracked, while Friday to Monday showed the least.
Users opted to track their fitness mid-weekly than consistently throughout the week.
FINDINGS #4 & 5: Additionally, I found that - through running several functions to determine the actual amount of participants of this survey (33), only 24 of them logged their sleeping data as well. Users must be encouraged to track their data on a nightly basis!
Page updated
Google Sites
Report abuse