Exploratory Data Analysis in R
Learn to explore a dataset in R with bar charts, histograms, box plots and summary statistics, then apply them to an email spam case study.
A table or database says little about a dataset beyond its size and the kinds of variables it holds. Exploratory data analysis in R looks further with plots and summary statistics, showing which variables seem to move together, which observations look unusual and how values are distributed. The work uses dplyr to prepare data and ggplot2 to draw charts, turning a raw table into findings you can question and explain.
What You Will Achieve
- Read side-by-side bar charts and conditional proportions to judge whether two categorical variables are related
- Select the plot that best fits a numerical variable and compare its distribution across groups
- Describe a distribution’s center, spread and shape with statistics suited to the data
- Spot unusual observations and question a dataset’s integrity before drawing conclusions
Program Overview
Exploring Categorical and Numerical Data
Graphical summaries for one and two variables, with bar charts for categories and histograms, box plots and density plots for numbers.
- Contingency tables and dropping unused levels
- Counts versus proportions in bar charts
- Faceted histograms and binwidth choice
- Plots that show three variables at once
Numerical Summaries of Distributions
Statistics that describe where a distribution sits, how widely it varies and what shape it takes.
- Measures of center
- Measures of variability
- Shape and transformations
- Identifying outliers
Email Spam Case Study
A real email dataset explored with the earlier techniques to see which features relate to spam.
- Character counts and exclamation marks against spam
- Collapsing levels
- Data integrity checks
- Answering questions with chained commands
Certificate Awarded (Paid)
Job Outlook
These exploration skills apply in analysis, reporting and research work that uses R to make sense of tabular data, as the first look at a dataset before deeper statistical work.
Good for: Analysts working in R who want a repeatable way to size up a new dataset with charts and summary statistics.
Keep in mind: Many exercises are interpretation and choice questions about plots and measures, so expect to read output as often as you write R code.
Bottom line: A focused set of charts and summary statistics for a first look at data in R, tied together by an email spam case study.
Frequently Asked Questions
Which R packages are used in the exercises?
The exercises rely on dplyr for manipulating data and on ggplot2 for building charts, two packages that both belong to the tidyverse collection.
What dataset does the case study use?
The closing chapter works with a real dataset of emails, studied to see how features such as the number of characters, exclamation marks and images relate to whether a message is spam.
How does the course handle outliers?
Outliers come up twice. Box plots are used to spot them graphically, and the summary statistics chapter has its own lessons on identifying them, alongside choosing measures for center and spread.
Can I plot more than two variables at once?
Yes. The chapter on numerical data ends with lessons on visualization in higher dimensions, where you build a plot that shows three variables together and then interpret what it reveals.


There are no reviews yet.