condwanaland / phonewar

Results from an iPhone vs Android Preferences Survey

phonewar

phonewar provides a dataset containing peoples iPhone vs Android preferences based on a survey done by tech magazine ‘The Hustle’ (https://thehustle.co/home/).

Installation

You can install the phonewar from Github with:

remotes::install_github("condwanaland/phonedata")

Example

To use phonewar first load it

library(phonewar)
## basic example code

phonedata contains a single dataframe you can load. It contains information about the respondents phone preferences, their reasons for this preference, the city they come from, and metadata about this city (coordinates, population).

phones <- phonedata
head(phones)
#>   key phone_type
#> 1   1    Android
#> 2   2     iPhone
#> 3   3     iPhone
#> 4   4    Android
#> 5   5     iPhone
#> 6   6     iPhone
#>                                                                                                                                                                                                                                                 reason
#> 1                                                                                                                                                                                                                                                 Cost
#> 2                                                                                                                                                                            Ecosystem work base and connectivity between devices around work networks
#> 3                                                                                                               Ease of use, speed, app coverage, traditionally used iphone, tried to switch but it was harder to use & slow so I moved back to iphone
#> 4 I started with an iPhone. The battery lasted about 2 hours and Apple didn't seem to care. Also, it stopped working in cold weather. That's a problem in Canada. All in all, I find Apple products to be overpriced, confusing and they break easily.
#> 5                                                                                                                                                                                             Too many reasons to enumerate, but basically safety & UX
#> 6                                                                                                                                                                      Ease of use, connected ecosystem across various devices, privacy considerations
#>   considered_switching less_respect_for_other  original_city       city     lat
#> 1                   No                     No        spokane    spokane 47.6671
#> 2                  Yes                     No marbella spain   marbella 36.5167
#> 3                  Yes                     No           cork       cork 51.9000
#> 4                   No                     No         ottawa     ottawa 45.4247
#> 5                   No                     No manchester, uk manchester 53.4794
#> 6                   No                     No        bangkok    bangkok 13.7500
#>         lng        country population
#> 1 -117.4330  united states     412275
#> 2   -4.8833          spain     147633
#> 3   -8.4731        ireland     208669
#> 4  -75.6950         canada     989567
#> 5   -2.2453 united kingdom    2705000
#> 6  100.5167       thailand   17066000

You can check out some graphs of this data

library(dplyr)
library(ggplot2)
library(viridis)
library(tidyr)
phone_sum <- phonedata %>% 
  group_by(phone_type) %>% 
  summarise(n = n()) %>% 
  mutate(freq = n/sum(n))
ggplot(phone_sum, aes(y = phone_type, x = freq, fill = phone_type)) +
  geom_col() +
  geom_text(aes(label = paste0(round(freq, 3), "%"), hjust = 1.2), color = "grey") +
  scale_fill_viridis(discrete = TRUE, option = "turbo")

countries <- phonedata %>% 
  group_by(country, phone_type) %>% 
  summarise(n = n()) %>% 
  mutate(freq = n/sum(n)) %>% 
  filter(n > 1) %>% 
  filter(!is.na(country)) %>% 
  select(-n) %>% 
  pivot_wider(names_from = phone_type, values_from = freq, values_fill = 0) %>% 
  mutate(diff = Android - iPhone)

ggplot(countries, aes(y = country, x = diff)) +
  geom_col() +
  labs(caption = "Negative values indicate preference for iPhone, positive for Android",
       x = "Phone preference percentage") +
  theme_bw()

License

The original collectors of this data was ‘The Hustle’ (https://thehustle.co/home/). The original version of the data can be found here (https://docs.google.com/spreadsheets/d/1UC2I7vO-bS4Qcv0neNRDqJU2qt1Q6ednz8232cmb0_g/edit#gid=430244235).