Data Science: Data Visualization Boot CampRelationshipScatter Plot

Chuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhDChuck Cartledge, PhD

24 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 202024 January 2020

1/24

2/24

Type Sample data Hands on Q & A Conclusion References Files

Table of contents (1 of 1)

1 TypeUsesGeneral considerations

2 Sample data

3 Hands on

4 Q & A

5 Conclusion6 References7 Files

3/24

Type Sample data Hands on Q & A Conclusion References Files

A definition

“Sometimes referred toas dot or symbol graph.Point graphs are a family ofgraphs that display quantita-tive information by means ofpoints represented by sym-bols such as dots, circles,or squares (called plot orplotting symbols). Whenthe points are connected bylines, the graph is generallyreferred to as a line graph.”

R. L. Harris [2]

4/24

Type Sample data Hands on Q & A Conclusion References Files

R supplied data set (1 of 2)

Included in the R package ggplot2.

“This dataset contains a subset of the fuel econ-omy data that the EPA makes available on . It contains only models which hada new release every year between 1999 and 2008 - this wasused as a proxy for the popularity of the car.”

H. Wickham [3]

library(ggplot2)

?mpg

head(mpg)

Resulting in:

5/24

Type Sample data Hands on Q & A Conclusion References Files

R supplied data set (2 of 2)

# A tibble: 6 x 11

manufacturer model displ year cyl trans drv cty hwy fl class

1 audi a4 1.8 1999 4 auto(l5) f 18 29 p comp

2 audi a4 1.8 1999 4 manual(m5) f 21 29 p comp

3 audi a4 2 2008 4 manual(m6) f 20 31 p comp

4 audi a4 2 2008 4 auto(av) f 21 30 p comp

5 audi a4 2.8 1999 6 auto(l5) f 16 26 p comp

6 audi a4 2.8 1999 6 manual(m5) f 18 26 p comp

6/24

Type Sample data Hands on Q & A Conclusion References Files

More recent mileage data

Downloaded from:https://www.fueleconomy.gov/feg/download.shtml

Described at: https://www.fueleconomy.gov/feg/ws/index.shtml#vehicle

We will:

1 Extract csv data from a zip file (39,865 rows)

2 Select certain makes (attempt to replicate the sample data)

3 Display different data for selected makes/models

https://www.fueleconomy.gov/feg/download.shtmlhttps://www.fueleconomy.gov/feg/ws/index.shtml#vehiclehttps://www.fueleconomy.gov/feg/ws/index.shtml#vehicle

7/24

Type Sample data Hands on Q & A Conclusion References Files

The first codes (1 of 3)

8/24

Type Sample data Hands on Q & A Conclusion References Files

The first codes (2 of 3)

rm(list=ls())

library(ggplot2)

data(mpg, package="ggplot2")

mpg_select

9/24

Type Sample data Hands on Q & A Conclusion References Files

The first codes (3 of 3)

g + geom_point(aes(col=manufacturer))

g + geom_jitter(aes(col=manufacturer))

g + geom_jitter(aes(col=manufacturer)) +

geom_smooth(aes(col=manufacturer), method="lm", se=FALSE)

10/24

Type Sample data Hands on Q & A Conclusion References Files

The second codes (1 of 4)

11/24

Type Sample data Hands on Q & A Conclusion References Files

The second codes (2 of 4)

rm(list=ls())

library(ggplot2)

saveFileName

12/24

Type Sample data Hands on Q & A Conclusion References Files

The second codes (3 of 4)

labs(subtitle="mpg: Displacement vs City Mileage",

title="Scatter chart",

x="Engine displacement",

y="City miles per gallon",

color="Manufacturer")

g + geom_point()

g + geom_point(aes(col=make))

g + geom_jitter(aes(col=make))

g + geom_jitter(aes(col=make)) +

geom_smooth(aes(col=make), method="lm", se=FALSE)

g + geom_jitter(aes(col=make)) +

geom_smooth(aes(col=make), method="lm", se=TRUE)

13/24

Type Sample data Hands on Q & A Conclusion References Files

The second codes (4 of 4)

14/24

Type Sample data Hands on Q & A Conclusion References Files

The third codes (1 of 3)

Idea and data from[1].

15/24

Type Sample data Hands on Q & A Conclusion References Files

The third codes (2 of 3)

rm(list=ls())

library(ggplot2)

data

16/24

Type Sample data Hands on Q & A Conclusion References Files

The third codes (3 of 3)

labs(title="Assessed value vs. square footage",

subtitle="Selected zip codes in the Seattle, Washington area",

x="Square footage",

y="Assessed value"

)

g + geom_point()

g + geom_jitter()

g + stat_binhex(colour="white") +

scale_fill_gradient(low="white", high="black")

g + geom_point(alpha = 0.1) + geom_density2d(colour="white")

17/24

Type Sample data Hands on Q & A Conclusion References Files

The fourth codes (1 of 3)

Idea and data from[1].

18/24

Type Sample data Hands on Q & A Conclusion References Files

The fourth codes (2 of 3)

rm(list=ls())

library(ggplot2)

data

19/24

Type Sample data Hands on Q & A Conclusion References Files

The fourth codes (3 of 3)

subtitle="Selected zip codes in the Seattle, Washington area",

x="Square footage",

y="Assessed value"

)

g + stat_binhex(colour="white") +

scale_fill_gradient(low="white", high="blue") +

facet_wrap("ZipCode")

20/24

Type Sample data Hands on Q & A Conclusion References Files

Hands-on exercises

1 The legend is off to the right in the margin. Move the legendinto the upper right area of the plot.

2 Move the title to the center of the plot.

21/24

Type Sample data Hands on Q & A Conclusion References Files

Q & A time.

Q: Why was Stonehengeabandoned?A: It wasn’t IBM compatible.

22/24

Type Sample data Hands on Q & A Conclusion References Files

What have we covered?

Scatter plots are:

Very quick and easy to puttogetherUsed to show 2, possibly relateddata setsEasily overwhelmed with toomany datapointsRelies on human ability to detectpatterns, or relationships in theplotted data

Usually the first plot forexploratory data analysis.

Next: Bubble charts (getting 3D out of 2D)

23/24

Type Sample data Hands on Q & A Conclusion References Files

References (1 of 1)

[1] Peter Bruce and Andrew Bruce,Practical Statistics for Data Scientists, O’Reilly Media, Inc.,2017.

[2] Robert L. Harris,Information Graphics: A Comprehensive Illustrated Reference,Oxford University Press, 2000.

[3] Hadley Wickham, ggplot2: Elegant Graphics for Data Analysis,Springer-Verlag New York, 2009.

24/24

Type Sample data Hands on Q & A Conclusion References Files

Files of interest

1 Code snippet to createimages in this presentation

2 Extract Federal fuel data

## First codesrm(list=ls())

library(ggplot2)data(mpg, package="ggplot2")

mpg_select