Bellabeat CaseStudy using R
Google Data Analytics Capstone Project
Sami Parkkali
14/01/2023
Introduction
Bellabeat is a high-tech company founded in 2013 that manufactures health-focused smart products for women. The collection of data on activity, sleep, stress, and reproductive health has allowed Bellabeat to empower women with knowledge about their health and daily habits. The company has 5 products: bellabeat app, leaf, time, spring and bellabeat membership. Our team has been asked to analyse smart device data to gain insight into how consumers are using their gadgets. The insights of this analysis will help to guide Bellabeat’s marketing strategy.
Business Task
Analise a given data set and present some recommendations to improve Bellabeat’s marketing strategy.
Key Stakeholders
- Urška Sršen – Co-founder and Chief Creative Officer
- Sando Mur – Co-founder and key member of the Bellabeat executive team
Data
- The data used for this analysis is the Fitbit Fitness Tracker Data available on Kaggle.
- It contains personal tracker data of 30 users who have consented to its submission.
- A survey via Amazon Mechanical Turk was conducted between 03.12.2016 and 05.12.2016.
This data set has 2 main limitations: only 30 users and their data was collected from March to May of 2016 which is a short time frame.
Coding
Loading Packages
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ ggplot2 3.4.0 ✔ purrr 1.0.0
## ✔ tibble 3.1.8 ✔ dplyr 1.0.10
## ✔ tidyr 1.2.1 ✔ stringr 1.5.0
## ✔ readr 2.1.3 ✔ forcats 0.5.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()library(lubridate)## Loading required package: timechange
##
## Attaching package: 'lubridate'
##
## The following objects are masked from 'package:base':
##
## date, intersect, setdiff, unionlibrary(dplyr)
library(ggplot2)
library(tidyr)Importing Datasets
I have already set my working directory.
dailyActivity_merged <- read_csv("Fitbit_Fitness_Tracker/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.Activity <- dailyActivity_merged
HeartRate <- read_csv("Fitbit_Fitness_Tracker/heartrate_seconds_merged.csv")## Rows: 2483658 Columns: 3
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): Time
## dbl (2): Id, Value
##
## 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.Weight <- read_csv("Fitbit_Fitness_Tracker/weightLogInfo_merged.csv")## Rows: 67 Columns: 8
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): Date
## dbl (6): Id, WeightKg, WeightPounds, Fat, BMI, LogId
## lgl (1): IsManualReport
##
## 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.Sleep <- read.csv("Fitbit_Fitness_Tracker/sleepDay_merged.csv")
Calories <- read.csv("Fitbit_Fitness_Tracker/dailyCalories_merged.csv")Formatting Time
class(Sleep$SleepDay)## [1] "character"Sleep$SleepDay=as.POSIXct(Sleep$SleepDay, format="%m/%d/%Y %I:%M:%S %p", tz=Sys.timezone())
Sleep$date <- format(Sleep$SleepDay, format = "%m/%d/%y")
Activity$ActivityDate=as.POSIXct(Activity$ActivityDate, format="%m/%d/%Y", tz=Sys.timezone())
Activity$date <- format(Activity$ActivityDate, format = "%m/%d/%y")
HeartRate$Time=as.POSIXct(HeartRate$Time, format="%m/%d/%Y", tz=Sys.timezone())
HeartRate$date <- format(HeartRate$Time, format = "%m/%d/%y")Exploring and summarising data
n_distinct(Activity$Id)## [1] 33n_distinct(Calories$Id)## [1] 33n_distinct(Sleep$Id)## [1] 24n_distinct(HeartRate$Id)## [1] 14n_distinct(Weight$Id)## [1] 8There is information about 33 users from calories and activity, 24 for sleep and 14 for heart rate. The data frame for weight has only 8 users which is not enough to make conclusions regarding weight and fat percentage.
Let’s see what activity data frame tells us:
Activity %>%
select(TotalSteps,
VeryActiveMinutes,
LightlyActiveMinutes,
SedentaryMinutes) %>%
summary()## TotalSteps VeryActiveMinutes LightlyActiveMinutes SedentaryMinutes
## Min. : 0 Min. : 0.00 Min. : 0.0 Min. : 0.0
## 1st Qu.: 3790 1st Qu.: 0.00 1st Qu.:127.0 1st Qu.: 729.8
## Median : 7406 Median : 4.00 Median :199.0 Median :1057.5
## Mean : 7638 Mean : 21.16 Mean :192.8 Mean : 991.2
## 3rd Qu.:10727 3rd Qu.: 32.00 3rd Qu.:264.0 3rd Qu.:1229.5
## Max. :36019 Max. :210.00 Max. :518.0 Max. :1440.0If we consider these users as more active people than average, 7406 steps as the median value is not satisfactory and could be improved. Despite moderate and intense exercise having more health benefits than walking 10,000-12,000 steps a day, achieving both is something to strive for. The mean sedentary time is 991 minutes (16hours) and the maximum sedentary time is 1440 minutes (24h). This needs to be addressed.
HeartRate %>%
select(Value) %>%
summary()## Value
## Min. : 36.00
## 1st Qu.: 63.00
## Median : 73.00
## Mean : 77.33
## 3rd Qu.: 88.00
## Max. :203.00Average values and there is no information regarding heart rate while exercising or sleeping.
Sleep %>%
select(.) %>%
filter(!complete.cases(.))## data frame with 0 columns and 0 rowsSleep %>%
select(TotalMinutesAsleep) %>%
summarise(mean(TotalMinutesAsleep))## mean(TotalMinutesAsleep)
## 1 419.4673Sleep %>%
select(TotalMinutesAsleep) %>%
summarise(median(TotalMinutesAsleep))## median(TotalMinutesAsleep)
## 1 433The mean sleeping time is just under 7 hours and the median sleeping time is 7h13mins.
Merging Data
merged_data <- merge(Sleep, Activity, by=c('Id', 'date'))
head(merged_data)## Id date SleepDay TotalSleepRecords TotalMinutesAsleep
## 1 1503960366 04/12/16 2016-04-12 1 327
## 2 1503960366 04/13/16 2016-04-13 2 384
## 3 1503960366 04/15/16 2016-04-15 1 412
## 4 1503960366 04/16/16 2016-04-16 2 340
## 5 1503960366 04/17/16 2016-04-17 1 700
## 6 1503960366 04/19/16 2016-04-19 1 304
## TotalTimeInBed ActivityDate TotalSteps TotalDistance TrackerDistance
## 1 346 2016-04-12 13162 8.50 8.50
## 2 407 2016-04-13 10735 6.97 6.97
## 3 442 2016-04-15 9762 6.28 6.28
## 4 367 2016-04-16 12669 8.16 8.16
## 5 712 2016-04-17 9705 6.48 6.48
## 6 320 2016-04-19 15506 9.88 9.88
## LoggedActivitiesDistance VeryActiveDistance ModeratelyActiveDistance
## 1 0 1.88 0.55
## 2 0 1.57 0.69
## 3 0 2.14 1.26
## 4 0 2.71 0.41
## 5 0 3.19 0.78
## 6 0 3.53 1.32
## LightActiveDistance SedentaryActiveDistance VeryActiveMinutes
## 1 6.06 0 25
## 2 4.71 0 21
## 3 2.83 0 29
## 4 5.04 0 36
## 5 2.51 0 38
## 6 5.03 0 50
## FairlyActiveMinutes LightlyActiveMinutes SedentaryMinutes Calories
## 1 13 328 728 1985
## 2 19 217 776 1797
## 3 34 209 726 1745
## 4 10 221 773 1863
## 5 20 164 539 1728
## 6 31 264 775 2035Visualization
ggplot(data=Activity, aes(x = TotalDistance, y = VeryActiveDistance,)) +
geom_point (color = 'darkgreen') + labs(title = "Very Active Distance in relation to Total Distance")
cor(Activity$TotalDistance, Activity$VeryActiveDistance)## [1] 0.7945816There is a moderate correlation between distance in total and very active distance. People may tend to exercise more vigorously if they improve their walking/moving time.
plot( TotalMinutesAsleep ~ TotalTimeInBed,
data = Sleep,
main = "Do people sleep if they go to bed?",
col.main = 'navyblue',
fg = 'blue',
col = 'darkorange',
col.axis = 'navyblue',
col.lab = 'darkorange',
xlab = "TotalTimeInBed",
ylab = "TotalMinutesAsleep"
) 
cor(Sleep$TotalTimeInBed, Sleep$TotalMinutesAsleep)## [1] 0.9304575There is a strong correlation between time spent in bed and sleeping time. Bellabeat may recommend going to bed half an hour earlier to improve sleeping time. Customers could receive points for good quality sleep and get discounts on other Bellabeat’s products.
ggplot(data=merged_data, aes(x = TotalMinutesAsleep, y = LightlyActiveMinutes)) +
geom_point( color = 'darkgreen') + geom_smooth( color = 'red') +
labs(title = 'Sleep and Activity')## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
ggplot(data=merged_data, aes(x = TotalMinutesAsleep, y = FairlyActiveMinutes)) +
geom_point( color = 'darkgreen') + geom_smooth( color = 'red') +
labs(title = 'Sleep and Activity')## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
ggplot(data=merged_data, aes(x = TotalMinutesAsleep, y = VeryActiveMinutes)) +
geom_point( color = 'darkgreen') + geom_smooth( color = 'red') +
labs(title = 'Sleep and Activity')## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
ggplot(data=merged_data, aes(x = TotalMinutesAsleep, y = SedentaryMinutes)) +
geom_point( color = 'darkgreen') + geom_smooth( color = 'red') +
labs(title = 'Sleep and Activity')## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
cor(merged_data$TotalMinutesAsleep, merged_data$SedentaryMinutes)## [1] -0.599394There seems to be no correlation between sleep time and active spent time. Sleep has a negative low correlation with sedentary minutes. However, sleep time should not be the focus to improve an active lifestyle according to this data.
Recommendations for Bellabeat
There are no gender information about the participants which may bias these recommendations as Bellabeat’s target audience is women. Regarding the limited data set analysed, I can suggest the following:
- Increasing average total steps may improve wellness overall. If there is a weekly goal on the app that changes according to previous records, women will walk more.
- Sedentary time is the worst indicator analised which has to be addressed. Notifications to stand up, stretch or go for a small walk in the office may reduce sedentarism.
- Mean sleeping time is barely 7h and there is a strong correlation between time asleep and time spent in bed. As the waking time is often non-negotiable, some kind of app reward to go to bed earlier may increase sleep time and probably its quality.
- More data has to be collected from users to get more direct insights; data collected for women using bellabeat app and its other products would suit the business better.
Session Info
sessionInfo()## R version 4.2.2 (2022-10-31 ucrt)
## Platform: x86_64-w64-mingw32/x64 (64-bit)
## Running under: Windows 10 x64 (build 22621)
##
## Matrix products: default
##
## locale:
## [1] LC_COLLATE=English_United Kingdom.utf8
## [2] LC_CTYPE=English_United Kingdom.utf8
## [3] LC_MONETARY=English_United Kingdom.utf8
## [4] LC_NUMERIC=C
## [5] LC_TIME=English_United Kingdom.utf8
##
## attached base packages:
## [1] stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] lubridate_1.9.0 timechange_0.1.1 forcats_0.5.2 stringr_1.5.0
## [5] dplyr_1.0.10 purrr_1.0.0 readr_2.1.3 tidyr_1.2.1
## [9] tibble_3.1.8 ggplot2_3.4.0 tidyverse_1.3.2
##
## loaded via a namespace (and not attached):
## [1] lattice_0.20-45 assertthat_0.2.1 digest_0.6.31
## [4] utf8_1.2.2 R6_2.5.1 cellranger_1.1.0
## [7] backports_1.4.1 reprex_2.0.2 evaluate_0.19
## [10] highr_0.10 httr_1.4.4 pillar_1.8.1
## [13] rlang_1.0.6 googlesheets4_1.0.1 readxl_1.4.1
## [16] rstudioapi_0.14 jquerylib_0.1.4 Matrix_1.5-1
## [19] rmarkdown_2.19 splines_4.2.2 labeling_0.4.2
## [22] googledrive_2.0.0 bit_4.0.5 munsell_0.5.0
## [25] broom_1.0.2 compiler_4.2.2 modelr_0.1.10
## [28] xfun_0.36 pkgconfig_2.0.3 mgcv_1.8-41
## [31] htmltools_0.5.4 tidyselect_1.2.0 fansi_1.0.3
## [34] crayon_1.5.2 tzdb_0.3.0 dbplyr_2.2.1
## [37] withr_2.5.0 grid_4.2.2 nlme_3.1-160
## [40] jsonlite_1.8.4 gtable_0.3.1 lifecycle_1.0.3
## [43] DBI_1.1.3 magrittr_2.0.3 scales_1.2.1
## [46] cli_3.5.0 stringi_1.7.8 vroom_1.6.0
## [49] cachem_1.0.6 farver_2.1.1 fs_1.5.2
## [52] xml2_1.3.3 bslib_0.4.2 ellipsis_0.3.2
## [55] generics_0.1.3 vctrs_0.5.1 tools_4.2.2
## [58] bit64_4.0.5 glue_1.6.2 hms_1.1.2
## [61] parallel_4.2.2 fastmap_1.1.0 yaml_2.3.6
## [64] colorspace_2.0-3 gargle_1.2.1 rvest_1.0.3
## [67] knitr_1.41 haven_2.5.1 sass_0.4.4