The global smart device market has revolutionized the way people interact with technology and their way of living. Fitbit devices, which are electronic gadgets that one puts on to monitor their exercise, sleep, and heart rate patterns, have improved health outcomes for many women with chronic health conditions. Even more, it has been found that 10,000+ steps can lower risks of developing multiple chronic conditions. No wonder doctors recommend this inspiring technology for the well-being of others. To know more about the wonderful benefits of a Fitbit device, click here.
In this case study, Bellabeat, a high-tech manufacturer of health-focused products for women has asked my team and me to find fitness data with the provided data set in hopes to create opportunities that will benefit the company. The data set acquired for this analysis portrays thirty women and their data while wearing their Fitbit devices. You can view the data set by clicking here. In addition, the team was asked to choose R, which is a programming language that helps data analysts process, interpret, and visualize large amounts of data. During this analysis we will discover how these women were using their devices and how their usage can help the company discover new trends.
Creating visualizations in R requires installing and loading
packages. In this particular case study, I will be loading the
tidyverse package along with ggplot2, which is
a package for creating data or statistical visualizations. As we
discussed earlier, Fitbit devices are used for many
purposes across the globe. The devices have it all to monitor any type
of data regarding health. However, one might doubt this type of
technology and question the abilities of these devices to record large
amounts of data on a single charge. Certainly, it didn’t disappoint as
one of the women in the study reached an outstanding 28.03 mile distance
while wearing her device, pretty impressive! To create this
visualization in R, input the following code into your R Console:
ggplot(data = daily_activity_df) +
geom_point(mapping = aes(x = TrackerDistance, y = TotalSteps, alpha = 1 / 10), color = "hotpink3") +
labs(title = "Built for endurance", subtitle = "Overall distance(miles) covered during trial period")
Fitbit devices are water proof. This hint came from
an observation during my analysis. Running the max()
function helped my team and me discover the amount of calories each
woman produced. In R, the max() function returns the
maximum value within each variable in the data set. Here, running the
code for our variable Calories resulted in a woman burning
4,900 calories in a single day. I must say that Fitbit
devices are great for recording the sweatiest of workouts. click
here to find out more. The result can be derived by running the
following code in RStudio:
max(daily_calories_df$Calories)
## [1] 4900
To figure out the average for this data set, I used the
mean() function. Note that inside the parentheses the data
set and variable must be separated by a $ symbol, like
this:
mean(daily_steps_df$StepTotal)
## [1] 7637.911
To improve Fitbit sales, the team has taken measures
to inform stakeholders at Bellabeat of our findings. In addition, the
team has suggested that customers who purchase Fitbit
devices get an opportunity to claim rewards for each exercise they
perform on the app or each item they buy at one of our locations. They
can then use these rewards to unlock items such as our
Spring water bottle, which uses technology to track daily
water intake, or a $10 to $15 discount for any of our featured items.
This will create an incentive to bring in more women looking to try our
products, which creates additional growth for the company in the long
run.
Bellabeat’s Fitbit devices are here to stay as these
devices have improved the life of millions of women around the world. It
is no secret why Bellabeat is becoming a top player in the smart device
market. With the help of R and RStudio, the team and I were able to
analyze the given data set and generated statistical results. By loading
the tidyverse package along with ggplot2,
transforming the data into a visual was as easy as 1-2-3.