20190528 Notebook

library(tidyverse)

Registered S3 method overwritten by 'dplyr':
  method           from
  print.rowwise_df     
Registered S3 methods overwritten by 'ggplot2':
  method         from 
  [.quosures     rlang
  c.quosures     rlang
  print.quosures rlang
── Attaching packages ────────────────────── tidyverse 1.2.1 ──
✔ ggplot2 3.1.1     ✔ purrr   0.3.2
✔ tibble  2.1.1     ✔ dplyr   0.8.1
✔ tidyr   0.8.3     ✔ stringr 1.4.0
✔ readr   1.3.1     ✔ forcats 0.4.0
── Conflicts ───────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()

library(janitor)

Attaching package: ‘janitor’

The following objects are masked from ‘package:stats’:

    chisq.test, fisher.test

wine_ratings <- readr::read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2019/2019-05-28/winemag-data-130k-v2.csv")

Missing column names filled in: 'X1' [1]Parsed with column specification:
cols(
  X1 = col_double(),
  country = col_character(),
  description = col_character(),
  designation = col_character(),
  points = col_double(),
  price = col_double(),
  province = col_character(),
  region_1 = col_character(),
  region_2 = col_character(),
  taster_name = col_character(),
  taster_twitter_handle = col_character(),
  title = col_character(),
  variety = col_character(),
  winery = col_character()
)

clean_names(wine_ratings)

NA

Look at distribution of scores based on evaluator, remove wine if taster name, price, points NA

known_taster <- wine_ratings %>% filter(!is.na(taster_name))   

known_price_taster <- known_taster %>% filter(!is.na(price))

known_points_price_taster <- known_price_taster %>% filter(!is.na(price))

create rating/price ratio

rp_ratio_df <- mutate(known_points_price_taster, rp_ratio = points/price)

tally(group_by(rp_ratio_df, taster_name))

Group by taster and rp ratio

ggplot(rp_ratio_df) +
  aes(x = taster_name) + 
  aes(y = rp_ratio) +
  geom_jitter(alpha = .5, height = 0, width = .25) +
  aes(col = taster_name) + 
  geom_boxplot(alpha = .25) +
  aes(fill = taster_name) 

ggplot(rp_ratio_df) +
  aes(x = taster_name) + 
  aes(y = points) +
  geom_jitter() +
  aes(col = taster_name) + 
  geom_boxplot() 

ratio by country

ggplot(rp_ratio_df) +
  aes(x = country) + 
  aes(y = rp_ratio) +
  geom_jitter() +
  geom_boxplot() 

(country_count <- tally(group_by(rp_ratio_df, country)))

NA

Merge to have count in the data and rename the column

country_count <- merge(country_count,rp_ratio_df)

colnames(country_count)

 [1] "country"               "n"                    
 [3] "X1"                    "description"          
 [5] "designation"           "points"               
 [7] "price"                 "province"             
 [9] "region_1"              "region_2"             
[11] "taster_name"           "taster_twitter_handle"
[13] "title"                 "variety"              
[15] "winery"                "rp_ratio"             

rename n ufos <- ufos %>% rename(spotter.comments = comments)

country_count <- country_count %>% rename(total_count = n)

add factor level of small <100, mid 100-999, large >1000

country_count <- mutate(country_count, producer_level = ifelse(total_count %in% 0:99, "small",
  ifelse(total_count%in% 100:999, "medium",                                    ifelse(total_count %in% 1000:9999, "large",
                                                                                   ifelse(total_count %in% 10000:100000, "massive", "other"
                                                   )))))

Fix factor country_count %>% as.factor(country_count$producer_level, levels=c(“small”, “medium”, “large”, “massive”))

country_count$producer_level <- factor(country_count$producer_level, levels=c("small", "medium", "large", "massive"), ordered=TRUE)

class(country_count$producer_level)

[1] "ordered" "factor" 

tally(group_by(country_count, producer_level))

tally by production level

by_producer_level <- tally(group_by(country_count, producer_level))

ggplot(country_count) +
  aes(x = producer_level) + 
  aes(y = points) +
  geom_boxplot()+
  labs(title= "Meh: Surprisingly similar scores", subtitle ="Variability of Wine Score by Country Representation", x = "Countries Sorted by Review Representation Category", y = "Points") 

small countries points

small_countries <- filter(country_count, producer_level == "small")

ggplot(small_countries) +
  aes(x=country) +
  aes(y=rp_ratio) +
  geom_boxplot()