DATA 607 Fall 2025 – Meetup 2: Visualizing Data

Week Summary

1. Read Chapters 1-4 of R4DS (Work out the problems as you go!)

2. Supplementary Readings- Data Viz 1-5 (Wilke), Calling BS (West and Bergstrom) Proportional Ink and Misleading Axes

3. Lab 1: Airbnb in NYC- submit qmd file and pdf on Brightspace by Sunday Midnight

4. Data Science in Context- sign up if you haven’t yet

5. If you don’t have a working software setup speak up in Slack!!!

Why Make Graphs?

Graphs are the main way we communicate data
Alternative would be summary statistics, but visualizations are much richer.
Big debate within statistics community because summary stats perceived as more “rigorous”
Graphs won the debate- they are the main and most important thing people will remember about your data analyses

Anscombe’s Quartet

Set 1:

x	y
10	8.04
8	6.95
13	7.58
9	8.81
11	8.33
14	9.96
6	7.24
4	4.26
12	10.84
7	4.82
5	5.68

Set 2:

x	y
10	9.14
8	8.14
13	8.74
9	8.77
11	9.26
14	8.10
6	6.13
4	3.10
12	9.13
7	7.26
5	4.74

Set 3:

x	y
10	7.46
8	6.77
13	12.74
9	7.11
11	7.81
14	8.84
6	6.08
4	5.39
12	8.15
7	6.42
5	5.73

Set 4:

x	y
8	6.58
8	5.76
8	7.71
8	8.84
8	8.47
8	7.04
8	5.25
19	12.50
8	5.56
8	7.91
8	6.89

Anscombe Summary Stats:

Set	mean(x)	mean(y)	sd(x)	sd(y)	cor(x,y)
1	9	7.500909	3.316625	2.031568	0.8164205
2	9	7.500909	3.316625	2.031657	0.8162365
3	9	7.500000	3.316625	2.030424	0.8162867
4	9	7.500909	3.316625	2.030578	0.8165214

Are the datasets the basically the same?

Anscombe Visualized:

1D Example

Datasaurus Dozen

Non-Contrived Example

Table shows the results of a linear regression between income inequality and voter turnout, claiming to show a statistically significant negative correlation (Jackman 1980)

Outlier caused relationship

Graphs as Comparisons

Two purposes for graphs:

Understand and communicate the size and direction of comparisons that were already of interest
Discover new patterns

A graph is often all someone will remember about your data analysis, can make the difference between convincing important decision makers or not.

Improve this Graph

Reformulating the Data

Two Possible Stories
- Children had higher accuracy than adults
- Multiple objects is harder than an individual object
Caption suggests Adult vs. Child accuracy

Solution:

Instead of plotting responses, plot accuracy
Plot Accuracy of each group on the same axis

Case Study Solution: Line Plot

ggplot2 Basics

ggplot2 is tidyverse plotting package
Grammar of Graphics
Key figure components
- data maps to aesthetics
- interpreted using geoms
- others: theme, facets, coordinates, stats

Source: DS-Box

Vignette: Transit Costs

Source tidy tuesday Download Vignette Download Data

Categories of Graphs

Avoiding Wrong

Wrong figures are missing information needed to decipher the meaning or contain mathematical errors

Label all axes and provide units if ambiguous
All axes need scales
Aestethic elements accurately represent the data
Legends and labels present when appropriate to identify meaning of other visual elements

Ethics: Principal of Proportional Ink

“The representation of numbers, as physically measured on the surface of the graphic itself, should be directly proportional to the numerical quantities represented.” (1983, p.56) Edward Tufte, The Visual Display of Quantitative Information