Google Play Store mobile app data
Maria Seo
2023-09-09
setwd("/Users/mariaseo/Desktop/MSQM/R/mydata")
library(readr)
apps <- read_csv("/Users/mariaseo/Desktop/MSQM/R/mydata/apps.csv")## New names:
## Rows: 9659 Columns: 14
## ── Column specification
## ──────────────────────────────────────────────────────── Delimiter: "," chr
## (10): App, Category, Installs, Type, Price, Content Rating, Genres, Last... dbl
## (4): ...1, Rating, Reviews, Size
## 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.
## • `` -> `...1`#create subset with app categories of interest
apps_subset <- subset(apps, Category == "BUSINESS" |
Category == "FINANCE" |
Category == "DATING" |
Category == "EDUCATION" |
Category == "EVENTS" |
Category == "SOCIAL" |
Category == "FOOD_AND_DRINK" |
Category == "GAME")Summary statistics
#summary
summary(apps_subset)## ...1 App Category Rating
## Min. : 187 Length:2429 Length:2429 Min. :1.000
## 1st Qu.: 1667 Class :character Class :character 1st Qu.:4.000
## Median : 5640 Mode :character Mode :character Median :4.300
## Mean : 5180 Mean :4.199
## 3rd Qu.: 8129 3rd Qu.:4.500
## Max. :10835 Max. :5.000
## NA's :358
## Reviews Size Installs Type
## Min. : 0 Min. : 0.00 Length:2429 Length:2429
## 1st Qu.: 37 1st Qu.: 6.70 Class :character Class :character
## Median : 2555 Median : 19.00 Mode :character Mode :character
## Mean : 368980 Mean : 26.65
## 3rd Qu.: 55256 3rd Qu.: 39.00
## Max. :78158306 Max. :100.00
## NA's :299
## Price Content Rating Genres Last Updated
## Length:2429 Length:2429 Length:2429 Length:2429
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
##
## Current Ver Android Ver
## Length:2429 Length:2429
## Class :character Class :character
## Mode :character Mode :character
##
##
##
## Which app type received the highest reviews?
#consolidate categories
apps_subset$Category[apps_subset$Category == "BUSINESS"] <- "Business & Finance"
apps_subset$Category[apps_subset$Category == "FINANCE"] <- "Business & Finance"
apps_subset$Category[apps_subset$Category == "DATING"] <- "Dating"
apps_subset$Category[apps_subset$Category == "EDUCATION"] <- "Education"
apps_subset$Category[apps_subset$Category == "EVENTS"] <- "Events & Social"
apps_subset$Category[apps_subset$Category == "SOCIAL"] <- "Events & Social"
apps_subset$Category[apps_subset$Category == "FOOD_AND_DRINK"] <- "Food"
apps_subset$Category[apps_subset$Category == "GAME"] <- "Game"
#generate bar chart
library(ggplot2)
app_rating <- ggplot(apps_subset, aes(x=Category, y=Rating)) +
geom_bar(fun="mean", stat="summary", fill="#d6d6d6") +
theme(panel.background = element_blank(), axis.line = element_line(colour = "black"))
app_rating## Warning: Removed 358 rows containing non-finite values (`stat_summary()`).
Although there isn’t a significant difference, education apps received higher reviews than others. Games and event & social apps also received high ratings. On the other hand, dating apps had the lowest ratings, although the difference wasn’t substantial.
Which app type received the most reviews?
# Load the ggplot2 and scales packages
library(ggplot2)
library(scales)##
## Attaching package: 'scales'## The following object is masked from 'package:readr':
##
## col_factorlibrary(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.2 ✔ stringr 1.5.0
## ✔ forcats 1.0.0 ✔ tibble 3.2.1
## ✔ lubridate 1.9.2 ✔ tidyr 1.3.0
## ✔ purrr 1.0.2## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ scales::col_factor() masks readr::col_factor()
## ✖ purrr::discard() masks scales::discard()
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorsapps_subset <- apps_subset %>%
mutate(Reviews = as.numeric(Reviews))
##apps_subset <- apps_subset %>%
## mutate(Reviews = Reviews / 1000)
# Set the plot size options
options(repr.plot.width =9, repr.plot.height =9)
# Create the ggplot graph
app_reviews <- ggplot(apps_subset, aes(x=Category, y=Reviews)) +
geom_bar(fun="mean", stat="summary", fill="#d6d6d6") +
theme(panel.background = element_blank(), axis.line = element_line(colour = "black")) +
labs(
title = "Average Number of Reviews by Category",
x = "App Category",
y = "Average Reviews"
) +
scale_y_continuous(labels = label_number_si(accuracy = 1))## Warning: `label_number_si()` was deprecated in scales 1.2.0.
## ℹ Please use the `scale_cut` argument of `label_number()` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.# Display the graph
app_reviews
Event & social apps received the highest number of reviews, followed by game apps, suggesting there are the most popular. Dating apps received the lowest number of reviews.
What type of apps do people download the most?
# Step 1: Remove the "+" sign
apps_subset$Installs <- gsub("\\+", "", apps_subset$Installs)
# Step 2: Remove Commas
apps_subset$Installs <- gsub(",", "", apps_subset$Installs)
# Changing to numeric
apps_subset <- apps_subset %>%
mutate(Installs = as.numeric(Installs))
#generate bar chart
app_download <- ggplot(apps_subset, aes(x = reorder(Category, -Installs, FUN=mean), y = Installs)) +
geom_bar(fun="mean", stat="summary", fill="#d6d6d6") +
theme(panel.background = element_blank(), axis.line = element_line(colour = "black")) +
labs(
title = "Average Number of download by Category",
x = "App Category",
y = "Average download"
) +
scale_y_continuous(labels = label_number_si(accuracy = 1))
app_download
Event & social apps and game apps were downloaded the most, further suggesting that they are the most popular. Dating apps are the least downloaded.
Findings
Many people downloaded event & social and game apps, giving them high ratings and writing the most reviews. It’s interesting that business & finance apps had a relatively high number of downloads but received fewer reviews compared to other apps.