Tracking versus Logging
Debbie Reasons
2022-05-10
High Level Summary:
My Bellabeat case study was all about how smart fitness devices are used by customers. One aspect of the project that I wanted to revisit was the visual representation of how distances were recorded. To show that users overwhelmingly prefer to track their distances versus logging their distances, I created a plot using data provided by FitBit users. Counts from the processed data of tracked and logged distance inputs are graphed by date, showing that having the smart device track the distance for the user is the clear preference.
Ask – The Question
In my Bellabeat case study using FitBit data, I realized that I wanted to show visually the method users chose to record the distances they traveled during their fitness activities. I tried to plot the information in different layers, and wasn’t able to use the legend feature because of the way I went about it. I knew how I could improve on what I had done in the case study, and wanted to show that here. The answer was for me to process the data additionly using R prior to plotting the results.
Prepare – Finding and Describing the Data
I will be using the daily activity csv file from FitBit Fitness Tracker Data (CC0: Public Domain, dataset made available through Mobius) which can be found on Kaggle. Several files are provided, but I will be focusing only on the daily activity file for this short project. This file contains the daily distances which are reported in columns for total distance, tracked distance, and logged distance, along with other data that I won’t need for this project.
Process – Cleaning and Manipulating the Data
Installing and Loading Common Packages and Libraries
install.packages("tidyverse")## Installing package into '/cloud/lib/x86_64-pc-linux-gnu-library/4.2'
## (as 'lib' is unspecified)library(tidyverse)## ── Attaching packages
## ───────────────────────────────────────
## tidyverse 1.3.2 ──## ✔ ggplot2 3.3.6 ✔ purrr 0.3.4
## ✔ tibble 3.1.7 ✔ dplyr 1.0.9
## ✔ tidyr 1.2.0 ✔ stringr 1.4.0
## ✔ readr 2.1.2 ✔ forcats 0.5.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()install.packages("dplyr")## Installing package into '/cloud/lib/x86_64-pc-linux-gnu-library/4.2'
## (as 'lib' is unspecified)library(dplyr)Loading the CSV file to create data frames
install.packages("readr")## Installing package into '/cloud/lib/x86_64-pc-linux-gnu-library/4.2'
## (as 'lib' is unspecified)library(readr)
dailyActivity_merged <- read_csv("dailyActivity_merged.csv")## Rows: 940 Columns: 15## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): ActivityDate
## dbl (14): Id, TotalSteps, TotalDistance, TrackerDistance, LoggedActivitiesDi...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.glimpse(dailyActivity_merged)## Rows: 940
## Columns: 15
## $ Id <dbl> 1503960366, 1503960366, 1503960366, 150396036…
## $ ActivityDate <chr> "4/12/2016", "4/13/2016", "4/14/2016", "4/15/…
## $ TotalSteps <dbl> 13162, 10735, 10460, 9762, 12669, 9705, 13019…
## $ TotalDistance <dbl> 8.50, 6.97, 6.74, 6.28, 8.16, 6.48, 8.59, 9.8…
## $ TrackerDistance <dbl> 8.50, 6.97, 6.74, 6.28, 8.16, 6.48, 8.59, 9.8…
## $ LoggedActivitiesDistance <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ VeryActiveDistance <dbl> 1.88, 1.57, 2.44, 2.14, 2.71, 3.19, 3.25, 3.5…
## $ ModeratelyActiveDistance <dbl> 0.55, 0.69, 0.40, 1.26, 0.41, 0.78, 0.64, 1.3…
## $ LightActiveDistance <dbl> 6.06, 4.71, 3.91, 2.83, 5.04, 2.51, 4.71, 5.0…
## $ SedentaryActiveDistance <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ VeryActiveMinutes <dbl> 25, 21, 30, 29, 36, 38, 42, 50, 28, 19, 66, 4…
## $ FairlyActiveMinutes <dbl> 13, 19, 11, 34, 10, 20, 16, 31, 12, 8, 27, 21…
## $ LightlyActiveMinutes <dbl> 328, 217, 181, 209, 221, 164, 233, 264, 205, …
## $ SedentaryMinutes <dbl> 728, 776, 1218, 726, 773, 539, 1149, 775, 818…
## $ Calories <dbl> 1985, 1797, 1776, 1745, 1863, 1728, 1921, 203…Because I want to analyze just the frequency of tracking and logging, I will first create two data frames–one for each choice and including only the distances above zero. Here I will change the column names so that I can merge the two frames back together and have the logged and tracked distances in the same column. The information on the method by which the distance was recorded will still be preserved with each entry by introducing a new column called recording method.
tracking_df <- select(dailyActivity_merged,Id, ActivityDate, TrackerDistance) %>%
rename(Distance = TrackerDistance) %>%
filter(Distance > 0) %>%
mutate(recording_method = "tracked")
glimpse(tracking_df)## Rows: 862
## Columns: 4
## $ Id <dbl> 1503960366, 1503960366, 1503960366, 1503960366, 15039…
## $ ActivityDate <chr> "4/12/2016", "4/13/2016", "4/14/2016", "4/15/2016", "…
## $ Distance <dbl> 8.50, 6.97, 6.74, 6.28, 8.16, 6.48, 8.59, 9.88, 6.68,…
## $ recording_method <chr> "tracked", "tracked", "tracked", "tracked", "tracked"…logging_df <- select(dailyActivity_merged, Id, ActivityDate, LoggedActivitiesDistance) %>%
rename(Distance = LoggedActivitiesDistance) %>%
filter(Distance > 0) %>%
mutate(recording_method = "logged")
glimpse(logging_df)## Rows: 32
## Columns: 4
## $ Id <dbl> 6775888955, 6962181067, 6962181067, 6962181067, 70077…
## $ ActivityDate <chr> "4/26/2016", "4/21/2016", "4/25/2016", "5/9/2016", "4…
## $ Distance <dbl> 1.959596, 4.081692, 2.785175, 3.167822, 4.869783, 4.8…
## $ recording_method <chr> "logged", "logged", "logged", "logged", "logged", "lo…distance_recording_method_df <- full_join(tracking_df, logging_df)## Joining, by = c("Id", "ActivityDate", "Distance", "recording_method")head(distance_recording_method_df)## # A tibble: 6 × 4
## Id ActivityDate Distance recording_method
## <dbl> <chr> <dbl> <chr>
## 1 1503960366 4/12/2016 8.5 tracked
## 2 1503960366 4/13/2016 6.97 tracked
## 3 1503960366 4/14/2016 6.74 tracked
## 4 1503960366 4/15/2016 6.28 tracked
## 5 1503960366 4/16/2016 8.16 tracked
## 6 1503960366 4/17/2016 6.48 trackedAnalyze – Identifying Trends and Relationships
Now that the data has been arranged in more of a long format, I can plot tracked and logged data on the same graph using ggplot.
ggplot() +
geom_point(data = distance_recording_method_df, mapping = aes(x = ActivityDate, y= Distance, color = recording_method))+
labs(title = "Tracked and Logged Distances") +
theme(axis.text.x = element_text(angle = 90))
The scatterplot stresses the distance itself more than the method that was used to record that distance, so I will change to a bar plot that uses the same data but plots the count of observations for logged and tracked distances. This should emphasize the method of recording that was chosen.
recording_preference <- ggplot() +
geom_bar(data = distance_recording_method_df, mapping = aes(x = ActivityDate, fill = recording_method))+
labs(subtitle = "Counts of Tracked and Logged Distances", title = "Users Prefer to Track Distance")+
theme(axis.text.x = element_text(angle = 90))
#ggsave("recording_preference.png", recording_preference)
recording_preference