STAT120 - Statistical Computing in R

Homework 4

Part I

The dataset used for this part is Video_Games_Sales_as_at_22_Dec_2016.csv. You will practicing practice tidyverse functions for data manipulation.

  1. Import the dataset into R and call it games. We have been using the following functions to examine basic information about the dataset. Write the pipe version of the same code (i.e., use the same functions, but use the pipe operator to call each function).
head(games)
str(games)
Show solution
games %>% head()
                      Name Platform Year_of_Release        Genre Publisher
1               Wii Sports      Wii            2006       Sports  Nintendo
2        Super Mario Bros.      NES            1985     Platform  Nintendo
3           Mario Kart Wii      Wii            2008       Racing  Nintendo
4        Wii Sports Resort      Wii            2009       Sports  Nintendo
5 Pokemon Red/Pokemon Blue       GB            1996 Role-Playing  Nintendo
6                   Tetris       GB            1989       Puzzle  Nintendo
  NA_Sales EU_Sales JP_Sales Other_Sales Global_Sales Critic_Score Critic_Count
1    41.36    28.96     3.77        8.45        82.53           76           51
2    29.08     3.58     6.81        0.77        40.24           NA           NA
3    15.68    12.76     3.79        3.29        35.52           82           73
4    15.61    10.93     3.28        2.95        32.77           80           73
5    11.27     8.89    10.22        1.00        31.37           NA           NA
6    23.20     2.26     4.22        0.58        30.26           NA           NA
  User_Score User_Count Developer Rating
1          8        322  Nintendo      E
2                    NA                 
3        8.3        709  Nintendo      E
4          8        192  Nintendo      E
5                    NA                 
6                    NA                 
games %>% str()
'data.frame':   16719 obs. of  16 variables:
 $ Name           : chr  "Wii Sports" "Super Mario Bros." "Mario Kart Wii" "Wii Sports Resort" ...
 $ Platform       : chr  "Wii" "NES" "Wii" "Wii" ...
 $ Year_of_Release: chr  "2006" "1985" "2008" "2009" ...
 $ Genre          : chr  "Sports" "Platform" "Racing" "Sports" ...
 $ Publisher      : chr  "Nintendo" "Nintendo" "Nintendo" "Nintendo" ...
 $ NA_Sales       : num  41.4 29.1 15.7 15.6 11.3 ...
 $ EU_Sales       : num  28.96 3.58 12.76 10.93 8.89 ...
 $ JP_Sales       : num  3.77 6.81 3.79 3.28 10.22 ...
 $ Other_Sales    : num  8.45 0.77 3.29 2.95 1 0.58 2.88 2.84 2.24 0.47 ...
 $ Global_Sales   : num  82.5 40.2 35.5 32.8 31.4 ...
 $ Critic_Score   : int  76 NA 82 80 NA NA 89 58 87 NA ...
 $ Critic_Count   : int  51 NA 73 73 NA NA 65 41 80 NA ...
 $ User_Score     : chr  "8" "" "8.3" "8" ...
 $ User_Count     : int  322 NA 709 192 NA NA 431 129 594 NA ...
 $ Developer      : chr  "Nintendo" "" "Nintendo" "Nintendo" ...
 $ Rating         : chr  "E" "" "E" "E" ...

Note: check the output provided by str(). If User_Score and Year_of_Release are factors, run the following code to convert them to numeric vectors.

games$Year_of_Release <- as.numeric(as.character(games$Year_of_Release))
Warning: NAs introduced by coercion
games$User_Score <- as.numeric(as.character(games$User_Score))
Warning: NAs introduced by coercion
  1. The following code calculates the average game sales (in millions of units) in North America (NA), European Union (EU), and Japan (JP) for the games developed by Nintendo. Write the pipe version of the same code (i.e., use the same functions, but use the pipe operator to call each function).
colMeans(subset(games, Developer == "Nintendo")[, c(6,7,8)])
Show solution
games %>%
  subset(Developer == "Nintendo") %>%
  '['(, c(6, 7, 8)) %>%
  colMeans()
NA_Sales EU_Sales JP_Sales 
3.213056 2.324583 1.241806 
  1. Subset the dataset so that the only games you see are those that belong to the “Shooter” genre. Print the data for the first five shooter games.
Show solution
games %>%
  filter(Genre == "Shooter") %>%
  slice(1:5)
                            Name Platform Year_of_Release   Genre  Publisher
1                      Duck Hunt      NES            1984 Shooter   Nintendo
2 Call of Duty: Modern Warfare 3     X360            2011 Shooter Activision
3      Call of Duty: Black Ops 3      PS4            2015 Shooter Activision
4        Call of Duty: Black Ops     X360            2010 Shooter Activision
5     Call of Duty: Black Ops II      PS3            2012 Shooter Activision
  NA_Sales EU_Sales JP_Sales Other_Sales Global_Sales Critic_Score Critic_Count
1    26.93     0.63     0.28        0.47        28.31           NA           NA
2     9.04     4.24     0.13        1.32        14.73           88           81
3     6.03     5.86     0.36        2.38        14.63           NA           NA
4     9.70     3.68     0.11        1.13        14.61           87           89
5     4.99     5.73     0.65        2.42        13.79           83           21
  User_Score User_Count                         Developer Rating
1         NA         NA                                         
2        3.4       8713 Infinity Ward, Sledgehammer Games      M
3         NA         NA                                         
4        6.3       1454                          Treyarch      M
5        5.3        922                          Treyarch      M
  1. Subset the dataset so that the only variables you see are the name of the game, the platform on which the game runs, the year of release, and the genre. Print the data for the first five games.
Show solution
games %>%
  select(Name, Platform, Year_of_Release, Genre) %>%
  slice(1:5)
                      Name Platform Year_of_Release        Genre
1               Wii Sports      Wii            2006       Sports
2        Super Mario Bros.      NES            1985     Platform
3           Mario Kart Wii      Wii            2008       Racing
4        Wii Sports Resort      Wii            2009       Sports
5 Pokemon Red/Pokemon Blue       GB            1996 Role-Playing
  1. Print the data for the top ten games in terms of the total number of units sold worldwide.
Show solution
games %>%
  arrange(desc(Global_Sales)) %>%
  slice(1:10)
                        Name Platform Year_of_Release        Genre Publisher
1                 Wii Sports      Wii            2006       Sports  Nintendo
2          Super Mario Bros.      NES            1985     Platform  Nintendo
3             Mario Kart Wii      Wii            2008       Racing  Nintendo
4          Wii Sports Resort      Wii            2009       Sports  Nintendo
5   Pokemon Red/Pokemon Blue       GB            1996 Role-Playing  Nintendo
6                     Tetris       GB            1989       Puzzle  Nintendo
7      New Super Mario Bros.       DS            2006     Platform  Nintendo
8                   Wii Play      Wii            2006         Misc  Nintendo
9  New Super Mario Bros. Wii      Wii            2009     Platform  Nintendo
10                 Duck Hunt      NES            1984      Shooter  Nintendo
   NA_Sales EU_Sales JP_Sales Other_Sales Global_Sales Critic_Score
1     41.36    28.96     3.77        8.45        82.53           76
2     29.08     3.58     6.81        0.77        40.24           NA
3     15.68    12.76     3.79        3.29        35.52           82
4     15.61    10.93     3.28        2.95        32.77           80
5     11.27     8.89    10.22        1.00        31.37           NA
6     23.20     2.26     4.22        0.58        30.26           NA
7     11.28     9.14     6.50        2.88        29.80           89
8     13.96     9.18     2.93        2.84        28.92           58
9     14.44     6.94     4.70        2.24        28.32           87
10    26.93     0.63     0.28        0.47        28.31           NA
   Critic_Count User_Score User_Count Developer Rating
1            51        8.0        322  Nintendo      E
2            NA         NA         NA                 
3            73        8.3        709  Nintendo      E
4            73        8.0        192  Nintendo      E
5            NA         NA         NA                 
6            NA         NA         NA                 
7            65        8.5        431  Nintendo      E
8            41        6.6        129  Nintendo      E
9            80        8.4        594  Nintendo      E
10           NA         NA         NA                 
  1. Subset the dataset so that the only games you see are those that are played on N64 (Nintendo 64) and that had more than two million units sold in North America (NA_Sales > 2). Print the Name, Genre, and NA_Sales number for these games.
Show solution
games %>%
  filter(Platform == "N64", NA_Sales > 2) %>%
  select(Name, Genre, NA_Sales)
                                   Name      Genre NA_Sales
1                        Super Mario 64   Platform     6.91
2                         Mario Kart 64     Racing     5.55
3                         GoldenEye 007    Shooter     5.80
4  The Legend of Zelda: Ocarina of Time     Action     4.10
5                     Super Smash Bros.   Fighting     2.95
6                       Pokemon Stadium   Strategy     3.18
7                        Donkey Kong 64   Platform     3.33
8                     Diddy Kong Racing     Racing     2.91
9                           Star Fox 64    Shooter     2.78
10                         Pokemon Snap Simulation     2.23
11            Star Wars Episode I Racer     Racing     2.31
  1. Create a new variable that measures the total number of units sold in North America and the European Union combined. Sort the data in terms of this new variable (from most copies to fewest copies), and then print the data for the top ten most sold games in NA and EU. You can print a few variables of your choice instead of all of the variables.
Show solution
games %>%
  mutate(NA_EU_Sales = NA_Sales + EU_Sales) %>%
  arrange(desc(NA_EU_Sales)) %>%
  slice(1:10) %>%
  select(Name, Platform, Genre, NA_EU_Sales)
                        Name Platform        Genre NA_EU_Sales
1                 Wii Sports      Wii       Sports       70.32
2          Super Mario Bros.      NES     Platform       32.66
3             Mario Kart Wii      Wii       Racing       28.44
4                  Duck Hunt      NES      Shooter       27.56
5          Wii Sports Resort      Wii       Sports       26.54
6                     Tetris       GB       Puzzle       25.46
7                   Wii Play      Wii         Misc       23.14
8  New Super Mario Bros. Wii      Wii     Platform       21.38
9      New Super Mario Bros.       DS     Platform       20.42
10  Pokemon Red/Pokemon Blue       GB Role-Playing       20.16
  1. Find information for the top 10 role-playing games in terms of user score. Note that some games have user score recorded as “tbd”, and you need to disregard these games.
Show solution
games %>%
  filter(Genre == "Role-Playing", !is.na(User_Score)) %>%
  arrange(desc(User_Score)) %>%
  slice(1:10) %>%
  select(Name, Platform, Year_of_Release, Genre, User_Score)
                                  Name Platform Year_of_Release        Genre
1                   Breath of Fire III      PSP            2005 Role-Playing
2      Boktai: The Sun is in Your Hand      GBA            2003 Role-Playing
3             Golden Sun: The Lost Age      GBA            2002 Role-Playing
4  Paper Mario: The Thousand-Year Door       GC            2004 Role-Playing
5                     Skies of Arcadia       DC            2000 Role-Playing
6                           Golden Sun      GBA            2001 Role-Playing
7             The Witcher 3: Wild Hunt       PC            2015 Role-Playing
8                              Deus Ex       PC            2000 Role-Playing
9                    Final Fantasy VII       PS            1997 Role-Playing
10            The Witcher 3: Wild Hunt      PS4            2015 Role-Playing
   User_Score
1         9.7
2         9.6
3         9.5
4         9.4
5         9.4
6         9.3
7         9.3
8         9.3
9         9.2
10        9.2
  1. What is the most sold game globally for each of the game genres? You can print a few variables of your choice instead of all of the variables.
Show solution
games %>%
  group_by(Genre) %>%
  slice_max(order_by = Global_Sales, n = 1) %>%
  ungroup() %>%
  select(Name, Genre, Global_Sales)
# A tibble: 13 × 3
   Name                                 Genre          Global_Sales
   <chr>                                <chr>                 <dbl>
 1 ""                                   ""                     2.39
 2 "Grand Theft Auto V"                 "Action"              21.0 
 3 "Super Mario Land 2: 6 Golden Coins" "Adventure"           11.2 
 4 "Super Smash Bros. Brawl"            "Fighting"            12.8 
 5 "Wii Play"                           "Misc"                28.9 
 6 "Super Mario Bros."                  "Platform"            40.2 
 7 "Tetris"                             "Puzzle"              30.3 
 8 "Mario Kart Wii"                     "Racing"              35.5 
 9 "Pokemon Red/Pokemon Blue"           "Role-Playing"        31.4 
10 "Duck Hunt"                          "Shooter"             28.3 
11 "Nintendogs"                         "Simulation"          24.7 
12 "Wii Sports"                         "Sports"              82.5 
13 "Pokemon Stadium"                    "Strategy"             5.45
  1. Which are the top 3 game genres with the highest global sales number, on average?
Show solution
games %>%
  group_by(Genre) %>%
  summarise(
    global_mean = mean(Global_Sales),
    .groups = "drop"
  ) %>%
  slice_max(order_by = global_mean, n = 3)
# A tibble: 3 × 2
  Genre      global_mean
  <chr>            <dbl>
1 ""               1.21 
2 "Platform"       0.933
3 "Shooter"        0.796
  1. How many games are in each of the genres?
Show solution
games %>% count(Genre)
          Genre    n
1                  2
2        Action 3370
3     Adventure 1303
4      Fighting  849
5          Misc 1750
6      Platform  888
7        Puzzle  580
8        Racing 1249
9  Role-Playing 1500
10      Shooter 1323
11   Simulation  874
12       Sports 2348
13     Strategy  683
  1. What are the top 10 games with the largest difference (in absolute value) between critics score and user score? Note that you need to disregard games with NA values for either of the scores (hint: use filter), and you need to multiply the user score by 10 so that the two scores are compared on the same scale. You can print a few variables of your choice instead of all of the variables.
Show solution
games %>%
  filter(!is.na(Critic_Score), !is.na(User_Score)) %>%
  mutate(User_Score_10 = 10 * User_Score) %>%
  mutate(diff = abs(Critic_Score - User_Score_10)) %>%
  arrange(desc(diff)) %>%
  slice(1:10) %>%
  select(Name, Genre, Platform, Critic_Score, User_Score_10)
                             Name    Genre Platform Critic_Score User_Score_10
1      Thrillville: Off the Rails Strategy      PSP           72             6
2       Shrek Swamp Kart Speedway   Racing      GBA           27            89
3             Company of Heroes 2 Strategy       PC           80            21
4  Call of Duty: Modern Warfare 3  Shooter      PS3           88            32
5  Call of Duty: Modern Warfare 3  Shooter     X360           88            34
6  Call of Duty: Modern Warfare 3  Shooter       PC           78            25
7      Thrillville: Off the Rails Strategy      PS2           70            17
8  Call of Duty: Modern Warfare 3  Shooter      Wii           70            18
9             Antz Extreme Racing   Racing       XB           35            87
10          Football Manager Live   Sports       PC           82            31
  1. Repeat 12, but disregard “obscure” games by limiting to games that have sold more than 3 (million) copies globally.
Show solution
games %>%
  filter(!is.na(Critic_Score), !is.na(User_Score), Global_Sales >= 3) %>%
  mutate(User_Score_10 = 10 * User_Score) %>%
  mutate(diff = abs(Critic_Score - User_Score_10)) %>%
  arrange(desc(diff)) %>%
  slice(1:10) %>%
  select(Name, Genre, Platform, Critic_Score, User_Score_10)
                             Name        Genre Platform Critic_Score
1  Call of Duty: Modern Warfare 3      Shooter      PS3           88
2  Call of Duty: Modern Warfare 3      Shooter     X360           88
3                      Diablo III Role-Playing       PC           88
4            Call of Duty: Ghosts      Shooter     X360           73
5            Call of Duty: Ghosts      Shooter      PS3           71
6                         FIFA 14       Sports      PS3           86
7  Call of Duty: Infinite Warfare      Shooter      PS4           77
8                         FIFA 14       Sports     X360           84
9            Call of Duty: Ghosts      Shooter      PS4           78
10                        FIFA 16       Sports     XOne           84
   User_Score_10
1             32
2             34
3             40
4             26
5             26
6             43
7             34
8             42
9             37
10            44
  1. What are the top 10 platforms in terms of number of games released? How many games were released on each of those 10 platforms? What are the average critic scores of the games on the 10 platforms (ignoring games with missing scores)? What was the median release year for games on each of the 10 platforms (ignoring games with missing release year)?
Show solution
platforms <- games %>%
  group_by(Platform) %>%
  summarize(
    number_of_games = n(),
    average_critic_score = mean(Critic_Score, na.rm = TRUE),
    median_release_year = median(Year_of_Release, na.rm = TRUE),
    .groups = "drop"
  ) %>%
  slice_max(order_by = number_of_games, n = 10)
platforms
# A tibble: 10 × 4
   Platform number_of_games average_critic_score median_release_year
   <chr>              <int>                <dbl>               <dbl>
 1 PS2                 2161                 68.7                2005
 2 DS                  2152                 63.8                2008
 3 PS3                 1331                 70.4                2011
 4 Wii                 1320                 62.8                2009
 5 X360                1262                 68.6                2010
 6 PSP                 1209                 67.4                2009
 7 PS                  1197                 71.5                1998
 8 PC                   974                 75.9                2010
 9 XB                   824                 69.9                2004
10 GBA                  822                 67.4                2003