Week2 Data Dive - Summaries
Woods
2026-02-02
Step 1: Load in data
library(readr)
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ forcats 1.0.0 ✔ stringr 1.5.2
## ✔ ggplot2 4.0.0 ✔ tibble 3.3.0
## ✔ lubridate 1.9.4 ✔ tidyr 1.3.1
## ✔ purrr 1.1.0## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ 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 errorslibrary(htmltools)
packageVersion("htmltools")## [1] '0.5.8.1'library(readxl)
library(tibble)
library(ggplot2)
# Load in data
coffee <- read_csv("~/Downloads/coffee_analysis.csv")## Rows: 2095 Columns: 12
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (10): name, roaster, roast, loc_country, origin_1, origin_2, review_date...
## dbl (2): 100g_USD, rating
##
## 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.# Show column types
spec(coffee) # double = numeric## cols(
## name = col_character(),
## roaster = col_character(),
## roast = col_character(),
## loc_country = col_character(),
## origin_1 = col_character(),
## origin_2 = col_character(),
## `100g_USD` = col_double(),
## rating = col_double(),
## review_date = col_character(),
## desc_1 = col_character(),
## desc_2 = col_character(),
## desc_3 = col_character()
## )Step 2: Numeric data summaries
summary(coffee[, sapply(coffee, is.numeric)]) ## 100g_USD rating
## Min. : 0.120 Min. :84.00
## 1st Qu.: 4.930 1st Qu.:92.00
## Median : 5.860 Median :93.00
## Mean : 9.323 Mean :93.11
## 3rd Qu.: 8.785 3rd Qu.:94.00
## Max. :132.280 Max. :98.00# Output: Min/Max, Mean & Median (quartiles), 1st & 3rd quartilesnumeric_summary <- sapply(
coffee[, sapply(coffee, is.numeric)],
function(x) c(
min = min(x, na.rm = TRUE),
max = max(x, na.rm = TRUE),
mean = mean(x, na.rm = TRUE),
median = median(x, na.rm = TRUE),
sd = sd(x, na.rm = TRUE),
quantiles = quantile(x, probs = c(0.25, 0.5, 0.75), na.rm = TRUE)
)
)
numeric_summary## 100g_USD rating
## min 0.120000 84.000000
## max 132.280000 98.000000
## mean 9.323313 93.114081
## median 5.860000 93.000000
## sd 11.430659 1.563024
## quantiles.25% 4.930000 92.000000
## quantiles.50% 5.860000 93.000000
## quantiles.75% 8.785000 94.000000# Output: Mean, SD, quantilesStep 3: Categorical column summaries
categorical_summary <- lapply(
coffee[, sapply(coffee, is.factor) | sapply(coffee, is.character)],
function(x) table(x)
)
# Created DataFrame 'categorical_summary'
# Output: Unique values and countsStep 4: Complete Numeric and Categorical Summaries
# Numeric
coffee %>%
summarise(across(
where(is.numeric),
list(
min = ~min(.x, na.rm = TRUE),
max = ~max(.x, na.rm = TRUE),
mean = ~mean(.x, na.rm = TRUE),
median = ~median(.x, na.rm = TRUE),
q25 = ~quantile(.x, 0.25, na.rm = TRUE),
q75 = ~quantile(.x, 0.75, na.rm = TRUE)
)
))## # A tibble: 1 × 12
## `100g_USD_min` `100g_USD_max` `100g_USD_mean` `100g_USD_median` `100g_USD_q25`
## <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 0.12 132. 9.32 5.86 4.93
## # ℹ 7 more variables: `100g_USD_q75` <dbl>, rating_min <dbl>, rating_max <dbl>,
## # rating_mean <dbl>, rating_median <dbl>, rating_q25 <dbl>, rating_q75 <dbl># Categorical
coffee %>%
summarise(across(
where(~is.factor(.x) | is.character(.x)),
~list(table(.x))
))## # A tibble: 1 × 10
## name roaster roast loc_country origin_1 origin_2 review_date
## <list> <list> <list> <list> <list> <list> <list>
## 1 <table[…]> <table[…]> <table[…]> <table[…]> <table[…]> <table[…]> <table[…]>
## # ℹ 3 more variables: desc_1 <list>, desc_2 <list>, desc_3 <list>Questions & Analysis
- Are coffees that tend to receive consistently higher ratings generally more expensive?
‘cor_price_rating’ is correlation between coffee price and rating. 0 = No relationship, 0.25 = Weak, 0.75 = Strong. The correlation between price and rating in this coffee reviews data set is a weak correlation, as the cor_price_rating = 0.259.
coffee %>%
summarise(
avg_price = mean(`100g_USD`, na.rm = TRUE),
avg_rating = mean(rating, na.rm = TRUE),
cor_price_rating = cor(`100g_USD`, rating, use = "complete.obs")
)## # A tibble: 1 × 3
## avg_price avg_rating cor_price_rating
## <dbl> <dbl> <dbl>
## 1 9.32 93.1 0.260- Which type of roast on average receives the highest ratings world wide?
The highest rated coffe roast world wide gathered from this data set is a light roast, earning an average rating of 93.5 out of 100. This tibble does show a strong correlation between rating and price when arranging by type of roast, The two highest rated roasts were the most expensive coffees while dark roasts received the lowest overall rating, making it the cheapest to purchase.
coffee %>%
group_by(roast) %>%
summarise(
avg_price = mean(`100g_USD`, na.rm = TRUE),
avg_rating = mean(rating, na.rm = TRUE),
n = n()
) %>%
arrange(desc(avg_rating))## # A tibble: 6 × 4
## roast avg_price avg_rating n
## <chr> <dbl> <dbl> <int>
## 1 <NA> 42.2 93.5 15
## 2 Light 11.4 93.5 287
## 3 Medium-Light 8.99 93.2 1490
## 4 Medium 7.06 92.3 259
## 5 Medium-Dark 9.51 91.9 39
## 6 Dark 4.46 88.2 5- What places of origin produce the highest rated coffees on average?
This question aims to provide insights into where the highest rated coffee beans are grown. The broader goal is to discover the top regions around the world where coffee is grown, based on aggregated ratings from this data set.
coffee %>%
group_by(origin_1) %>%
summarise(
avg_rating = mean(rating, na.rm = TRUE),
count = n()
) %>%
filter(count >= 5) %>% # avoids tiny samples
arrange(desc(avg_rating))## # A tibble: 88 × 3
## origin_1 avg_rating count
## <chr> <dbl> <int>
## 1 Chiriqui Province 94.9 7
## 2 Gedeo Zone 94.6 5
## 3 Kirinyaga 94.6 5
## 4 Boquete Growing Region 94.6 39
## 5 Piedra Candela 94.6 9
## 6 Kiambu Growing Region 94.5 12
## 7 Sidama Growing Region 94.4 14
## 8 Alto Quiel 94.3 6
## 9 Embu County 94.3 6
## 10 Holualoa 94.3 41
## # ℹ 78 more rows- What kind of trends does this dataset reveal based on how the cost of coffee has fluctuated over time?
The price of coffee appears to be steadily rising. This data set contains coffee reviews from 2017-2022. In 2017, the average cost of coffee was $7.87 and in 2022, the average cost increased to $9.27, after eclipsing an average of $10 throughout 2021. Aggregating the price of coffee by month gives more insight into increased prices over time, although there is higher fluctuation in monthly prices, resulting in the yearly aggegrating providing a more steady analysis.
# Converting the review date to date to give R a full date to process in order to successfully aggregate the average cost of coffee over time
coffee <- coffee %>%
mutate(review_date = as.Date(paste0("01 ", review_date), format = "%d %B %Y"),
year = year(review_date))# Aggregating average coffee price by year
price_by_year <- coffee %>%
group_by(year) %>%
summarise(
avg_price = mean(`100g_USD`, na.rm = TRUE),
coffees = n()
) %>%
arrange(year)
price_by_year## # A tibble: 6 × 3
## year avg_price coffees
## <dbl> <dbl> <int>
## 1 2017 7.87 68
## 2 2018 8.22 297
## 3 2019 9.29 291
## 4 2020 9.43 426
## 5 2021 10.1 543
## 6 2022 9.27 470# Aggregating average coffee price by month and year
price_by_month <- coffee %>%
mutate(year_month = floor_date(review_date, "month")) %>%
group_by(year_month) %>%
summarise(
avg_price = mean(`100g_USD`, na.rm = TRUE),
coffees = n()
) %>%
arrange(year_month)
price_by_month## # A tibble: 61 × 3
## year_month avg_price coffees
## <date> <dbl> <int>
## 1 2017-11-01 6.97 40
## 2 2017-12-01 9.15 28
## 3 2018-01-01 19.4 17
## 4 2018-02-01 5.94 22
## 5 2018-03-01 5.87 22
## 6 2018-04-01 12.2 27
## 7 2018-05-01 6.15 22
## 8 2018-06-01 5.16 20
## 9 2018-07-01 4.82 12
## 10 2018-08-01 9.37 32
## # ℹ 51 more rowsVisual Summary
Q1 Visualization Analysis
# Exploratory Distribution Visuals
ggplot(coffee, aes(x = `100g_USD`)) +
geom_histogram(bins = 30, fill = "steelblue", alpha = 0.7) +
labs(
title = "Distribution of Coffee Prices (per 100g)",
x = "Price (USD per 100g)",
y = "Count"
)
ggplot(coffee, aes(x = rating)) +
geom_histogram(bins = 30, fill = "darkgreen", alpha = 0.7) +
labs(
title = "Distribution of Coffee Ratings",
x = "Rating",
y = "Count"
)
The visualizations above are exploratory bar charts that preview the
‘Coffee Price vs Rating’ scatter plot below. These distributions provide
a brief overview of the variance in price and ratings in the data set as
a whole. The charts show that there is a clutter around less to average
priced coffee, while there is much less expensive coffee consumed
worldwide, where as the coffee ratings offer a bit more balanced
distribution ranging from roughly 88.0 to 98.0 ratings out of 100.
ggplot(coffee, aes(x = `100g_USD`, y = rating)) +
geom_point(alpha = 0.6) +
geom_smooth(method = "lm", se = FALSE) +
labs(
title = "Coffee Price vs Rating",
x = "Price (USD per 100g)",
y = "Rating"
)## `geom_smooth()` using formula = 'y ~ x'
‘Coffee Price vs Rating’ plot above directly answers Q1 by visualizing
the correlation and trend between the variables, showing how ratings
subtly rise in relation to higher priced coffee. This insight leads us
to believe that people are willing to pay for more expensive coffee if
the quality is excellent. This also opens up the ability to explore a
correlation between price and rarity, as rarer coffees create a high
demand, leading to increased price, in addition to the great
quality.
Q2 Visualization Analysis
# Visualizing rating & price trends by type of coffee roast
ggplot(coffee, aes(x = `100g_USD`, y = rating, color = roast)) +
geom_point(alpha = 0.7) +
geom_smooth(method = "lm", se = FALSE) +
labs(
title = "Price vs Rating by Roast Level",
x = "Price (USD per 100g)",
y = "Rating",
color = "Roast"
)## `geom_smooth()` using formula = 'y ~ x'
This scatter plot above highlights trends and potential correlation
between the type of coffee roast and the average rating and price for
each. The trend line for dark roast coffee is an exception as there are
few observations in the data set. With that said, every single trend
line is on a positive slope, trending slightly upward, proving a
correlation does exist between price and rating. In general, as coffee
price increases, so does the rating. The dots represent each individual
observation, and there is an overwhelming number of pink and gold dots,
representing light and medium-light roasts, which tells us that these
two roasts are likely the most popular roasts that are grown, roasted,
and consumed, in turn, receiving many reviews collected by this
dataset.