Intermediate Data Visualization with ggplot2
Add statistics, control coordinates and facet plots in R’s ggplot2, then learn which chart types mislead readers and what to draw instead.
Explanatory plots in ggplot2 do more than display raw data. Here you compute statistics as the plot is drawn, including smoothed trends, quantiles and confidence intervals, set coordinates and scales so values read accurately, and split a dataset into panels for comparison. The last part looks at chart types the data viz community commonly discourages, such as dynamite plots, heat maps and pie charts, their pitfalls and the alternatives.
What You Will Achieve
- Overlay trend lines, quantiles and confidence intervals that ggplot2 calculates as it draws
- Keep axes honest when data is lognormal, carries units or repeats in cycles
- Compare subsets of a dataset side by side in labeled, ordered panels
- Judge when a bar chart or heat map misleads and redraw the data in a clearer form
Program Overview
Statistical Layers in ggplot2
Goes past plain plotting to draw statistics inside the plot, with linear models, descriptive and inferential summaries, and custom functions.
- Stats with geoms and smoothing
- Grouping variables and stat_smooth options
- stat_sum and quantile plots
- Stats outside geoms and position objects
Coordinates and Facets
Uses coordinate layers for lognormal, unit-bearing and periodic data, then splits plots into wrapped or gridded panes with custom labels.
- Zooming, aspect ratios, expand and clip
- Log-transformed scales, double and flipped axes
- Polar coordinates, pie charts and wind rose plots
- Facet grids, wrapping, labels, order and margin plots
Data Visualization Best Practices
Reviews plot types the data viz community discourages, the pitfalls they carry and the alternatives worth using.
- Bar plots, dynamite plots and position dodging
- Heat maps and their alternatives
- Suppression of the origin and color blindness
- Typical plotting problems
Certificate Awarded (Paid)
Job Outlook
These skills apply wherever R is used to explore data and present findings, in analysis reports, research write-ups and any chart meant to explain results to other people.
Good for: R analysts who want their ggplot2 charts to explain data, adding statistical summaries, multi-panel views and sounder design choices.
Keep in mind: Everything is done in R code with ggplot2, and the four chapters center on statistics layers, coordinates, facets and the critique of common plot types.
Bottom line: A focused ggplot2 follow-on that sets new plotting layers beside a critical look at which charts mislead readers.
Frequently Asked Questions
Which statistics can ggplot2 draw in this course?
The first chapter applies a range of linear models, descriptive and inferential statistics such as the mean, standard deviation and confidence intervals, plus sums, quantiles and custom functions, all calculated while the plot is built.
What kinds of data do the coordinates lessons handle?
They focus on lognormal datasets, variables with units and periodic data. Along the way you work with zooming, aspect ratios, log-transformed scales, double and flipped axes, polar coordinates, pie charts and wind rose plots.
What does the best practices chapter cover?
It reviews three plot types the data viz community commonly discourages: heat maps, pie charts and dynamite plots. It shows their pitfalls and how to avoid them, and also looks at bar plots built on aggregated data, suppression of the origin and color blindness.
Does the course build up to a custom plotting function?
By the end you have the tools to write a custom plotting function that explores a large dataset with statistics and clear visuals. Custom functions are taught among the statistical methods of the first chapter.


There are no reviews yet.