Data Analysis in R: Uncovering Chipotle Order Trends

One of the fastest ways to make data analysis concrete is to work with data you already have intuitions about. Chipotle order data is perfect for this: you know what a burrito bowl is, you understand why Chicken is ordered more than Sofritas, and you can immediately sanity-check the results against your lived experience.

Here’s a practical walkthrough using a real Chipotle order dataset — focusing on the core R operations every analyst should have fluent.

Setting Up

Start with a clean workspace and load ggplot2 for visualization:

# Clear workspace
rm(list = ls())

# Install ggplot2 if needed
# install.packages("ggplot2")
library(ggplot2)

Loading the Data

The dataset contains individual order line items with columns for order ID, item quantity, item name, choice description, and item price.

# Load from file
orders <- read.csv("chipotle.csv", stringsAsFactors = FALSE)

# First look
head(orders)
str(orders)

Sample output:

  order_id quantity                              item_name choice_description  item_price
1        1        1               Chips and Fresh Tomato Salsa               NULL  $2.39 
2        1        1                                   Izze          [Clementine]  $3.39 
3        1        1                       Nantucket Nectar        [Apple]  $3.39 
4        1        1  Chips and Tomatillo-Green Chili Salsa               NULL  $2.39 
5        2        2                        Chicken Bowl   [Tomatillo-Red...]  $16.98
6        3        1                        Chicken Bowl   [Fresh Tomato...]   $10.98

Aggregating by Item

To find which items are ordered most frequently, sum quantities by item name:

item_totals <- aggregate(quantity ~ item_name, data = orders, FUN = sum)

# Sort descending
item_totals <- item_totals[order(-item_totals$quantity), ]

head(item_totals, 10)

                      item_name quantity
Chicken Bowl               761
Chicken Burrito            591
Chips and Guacamole        506
Steak Burrito              386
Canned Soft Drink          351
Chips                      230
Steak Bowl                 221
Bottled Water              162
Chips and Fresh Tomato...  130
Chicken Salad Bowl         123

The Chicken Bowl’s dominance isn’t surprising — it’s Chipotle’s most customizable and calorie-flexible option. What’s interesting is that Chips and Guacamole ranks third by volume, ahead of Steak Burrito. That’s a meaningful data point for menu pricing strategy.

Visualization with ggplot2

A horizontal bar chart makes quantity comparisons easy to read:

# Top 15 items
top15 <- head(item_totals, 15)

ggplot(top15, aes(x = reorder(item_name, quantity), y = quantity)) +
  geom_bar(stat = "identity", fill = "#A8BF87") +
  coord_flip() +
  labs(
    title = "Top 15 Chipotle Items by Quantity Ordered",
    x = NULL,
    y = "Total Units Ordered"
  ) +
  theme_minimal() +
  theme(
    axis.text.y = element_text(size = 9),
    plot.title = element_text(face = "bold")
  )

Key choices in this code:

• reorder() sorts the bars by quantity rather than alphabetically
• coord_flip() turns vertical bars horizontal — better for long item names
• theme_minimal() removes unnecessary chart chrome

Examining Raw Distributions

Sometimes you want to look at the raw quantity values rather than aggregated totals:

# Distribution of quantities per line item
quantities <- sort(orders$quantity)
summary(quantities)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
   1.00    1.00    1.00    1.07    1.00   15.00 

Most orders are single items, but the max of 15 suggests bulk orders — likely catering or group orders. You’d want to handle these differently in an analysis focused on individual customer behavior.

Cleaning Item Prices for Analysis

The item_price column is stored as a string with a $ prefix. To use it numerically:

orders$price_clean <- as.numeric(gsub("\\$", "", orders$item_price))

# Average price by item (top 10)
avg_price <- aggregate(price_clean ~ item_name, data = orders, FUN = mean)
avg_price <- avg_price[order(-avg_price$price_clean), ]
head(avg_price, 10)

This kind of cleaning step — stripping formatting characters to get to usable numeric data — comes up constantly in real-world datasets.

What to Explore Next

This dataset has more depth than a single walkthrough covers:

• Co-occurrence analysis: Which items are most commonly ordered together in the same order?
• Price sensitivity: Do higher-priced items show lower order frequency, and by how much?
• Customization patterns: The choice_description column contains rich free-text data about modifications — parsing it reveals preference patterns
• Order size distribution: Are there identifiable customer segments by order value?

Each of these leads somewhere different. That’s what makes exploratory data analysis useful — the early summary statistics point you toward the questions worth pursuing, rather than forcing you to decide upfront which questions matter.