Graphical Data Analysis with R Antony Unwin

install download

https://textbookfull.com/product/graphical-data-analysis-with-r-
antony-unwin/

Download more ebook instantly today - get yours now at textbookfull.com

We believe these products will be a great fit for you. Click
the link to download now, or visit textbookfull.com
to discover even more!

Data Analysis with R 2nd Edition Tony Fischetti

https://textbookfull.com/product/data-analysis-with-r-2nd-
edition-tony-fischetti/

Clinical trial data analysis with R and SAS Second

Edition Chen

https://textbookfull.com/product/clinical-trial-data-analysis-
with-r-and-sas-second-edition-chen/

Analysis of Correlated Data with SAS and R Mohamed M.

Shoukri

https://textbookfull.com/product/analysis-of-correlated-data-
with-sas-and-r-mohamed-m-shoukri/

Uncertainty analysis of experimental data with R 1st

Edition Benjamin David Shaw

https://textbookfull.com/product/uncertainty-analysis-of-
experimental-data-with-r-1st-edition-benjamin-david-shaw/
Data analysis using hierarchical generalized linear
models with R 1st Edition Youngjo Lee

https://textbookfull.com/product/data-analysis-using-
hierarchical-generalized-linear-models-with-r-1st-edition-
youngjo-lee/

Statistical Analysis of Financial Data: With Examples

In R 1st Edition James Gentle

https://textbookfull.com/product/statistical-analysis-of-
financial-data-with-examples-in-r-1st-edition-james-gentle/

R in Action Data Analysis and Graphics with R Bonus ch

23 ONLY 2nd Edition Robert Kabacoff

https://textbookfull.com/product/r-in-action-data-analysis-and-
graphics-with-r-bonus-ch-23-only-2nd-edition-robert-kabacoff/

Introduction to Data Science Data Analysis and

Prediction Algorithms with R 1st Edition By Rafael A.
Irizarry

https://textbookfull.com/product/introduction-to-data-science-
data-analysis-and-prediction-algorithms-with-r-1st-edition-by-
rafael-a-irizarry/

Textual Data Science with R Chapman Hall CRC Computer

Science Data Analysis 1st Edition Mónica Bécue-Bertaut

https://textbookfull.com/product/textual-data-science-with-r-
chapman-hall-crc-computer-science-data-analysis-1st-edition-
monica-becue-bertaut/
 Statistics
The R Series

Graphical Data Analysis with R

Graphical Data Analysis with R shows you what information you
can gain from graphical displays. The book focuses on why you draw
graphics to display data and which graphics to draw (and uses R to
do so). All the datasets are available in R or one of its packages and
the R code is available at rosuda.org/GDA.
Graphical Data
Graphical data analysis is useful for data cleaning, exploring data
structure, detecting outliers and unusual groups, identifying trends
and clusters, spotting local patterns, evaluating modelling output,
Analysis with R
and presenting results. This book guides you in choosing graphics
and understanding what information you can glean from them. It
can be used as a primary text in a graphical data analysis course
or as a supplement in a statistics course. Colour graphics are used
throughout.
Features
• Concentrates on why graphics are drawn and what they reveal
• Emphasises the value of drawing a variety of different graphics
• Uses real datasets in R to show how graphical data analysis
works in practice
• Supplies R code for all the graphics on the author’s website
• Includes a set of exercises in each chapter to facilitate hands-on
learning
Antony Unwin is a professor of computer-oriented statistics and data
analysis at the University of Augsburg. He is a fellow of the American
Statistical Society. His research focuses on data visualisation,
especially in interactive graphics.
Unwin

Antony Unwin

K25332

w w w. c rc p r e s s . c o m

K25332_cover.indd 1 3/4/15 3:38 PM

 Graphical Data
Analysis with R

K25332_FM.indd 1 3/4/15 4:04 AM

 Chapman & Hall/CRC
The R Series

Series Editors
John M. Chambers Torsten Hothorn
Department of Statistics Division of Biostatistics
Stanford University University of Zurich
Stanford, California, USA Switzerland

Duncan Temple Lang Hadley Wickham

Department of Statistics RStudio
University of California, Davis Boston, Massachusetts, USA
Davis, California, USA

Aims and Scope

This book series reflects the recent rapid growth in the development and application
of R, the programming language and software environment for statistical computing
and graphics. R is now widely used in academic research, education, and industry.
It is constantly growing, with new versions of the core software released regularly
and more than 6,000 packages available. It is difficult for the documentation to
keep pace with the expansion of the software, and this vital book series provides a
forum for the publication of books covering many aspects of the development and
application of R.
The scope of the series is wide, covering three main threads:
• Applications of R to specific disciplines such as biology, epidemiology,
genetics, engineering, finance, and the social sciences.
• Using R for the study of topics of statistical methodology, such as linear and
mixed modeling, time series, Bayesian methods, and missing data.
• The development of R, including programming, building packages, and
graphics.
The books will appeal to programmers and developers of R software, as well as
applied statisticians and data analysts in many fields. The books will feature
detailed worked examples and R code fully integrated into the text, ensuring their
usefulness to researchers, practitioners and students.

K25332_FM.indd 2 3/4/15 4:04 AM

 Published Titles

Stated Preference Methods Using R, Hideo Aizaki, Tomoaki Nakatani,

and Kazuo Sato
Using R for Numerical Analysis in Science and Engineering, Victor A. Bloomfield
Event History Analysis with R, Göran Broström
Computational Actuarial Science with R, Arthur Charpentier
Statistical Computing in C++ and R, Randall L. Eubank and Ana Kupresanin
Reproducible Research with R and RStudio, Christopher Gandrud
Introduction to Scientific Programming and Simulation Using R, Second Edition,
Owen Jones, Robert Maillardet, and Andrew Robinson
Nonparametric Statistical Methods Using R, John Kloke and Joseph McKean
Displaying Time Series, Spatial, and Space-Time Data with R,
Oscar Perpiñán Lamigueiro
Programming Graphical User Interfaces with R, Michael F. Lawrence
and John Verzani
Analyzing Sensory Data with R, Sébastien Lê and Theirry Worch
Parallel Computing for Data Science: With Examples in R, C++ and CUDA,
Norman Matloff
Analyzing Baseball Data with R, Max Marchi and Jim Albert
Growth Curve Analysis and Visualization Using R, Daniel Mirman
R Graphics, Second Edition, Paul Murrell
Data Science in R: A Case Studies Approach to Computational Reasoning and
Problem Solving, Deborah Nolan and Duncan Temple Lang
Multiple Factor Analysis by Example Using R, Jérôme Pagès
Customer and Business Analytics: Applied Data Mining for Business Decision
Making Using R, Daniel S. Putler and Robert E. Krider
Implementing Reproducible Research, Victoria Stodden, Friedrich Leisch,
and Roger D. Peng
Graphical Data Analysis with R, Antony Unwin
Using R for Introductory Statistics, Second Edition, John Verzani

Advanced R, Hadley Wickham

Dynamic Documents with R and knitr, Yihui Xie

K25332_FM.indd 3 3/4/15 4:04 AM

K25332_FM.indd 4 3/4/15 4:04 AM
 Graphical Data
Analysis with R

Antony Unwin
University of Augsburg
Germany

K25332_FM.indd 5 3/4/15 4:04 AM

CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2015 by Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Version Date: 20150224

International Standard Book Number-13: 978-1-4987-1524-9 (eBook - PDF)

This book contains information obtained from authentic and highly regarded sources. Reasonable
efforts have been made to publish reliable data and information, but the author and publisher cannot
assume responsibility for the validity of all materials or the consequences of their use. The authors and
publishers have attempted to trace the copyright holders of all material reproduced in this publication
and apologize to copyright holders if permission to publish in this form has not been obtained. If any
copyright material has not been acknowledged please write and let us know so we may rectify in any
future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying, microfilming, and recording, or in any information stor-
age or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copy-
right.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222
Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that pro-
vides licenses and registration for a variety of users. For organizations that have been granted a photo-
copy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
Contents

Preface xi

1 Setting the Scene 1

1.1 Graphics in action . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 What is Graphical Data Analysis (GDA)? . . . . . . . . . . . . . . 5
1.4 Using this book, the R code in it, and the book’s webpage . . . . . 14

2 Brief Review of the Literature and Background Materials 19

2.1 Literature review . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

2.2 Interactive graphics . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3 Other graphics software . . . . . . . . . . . . . . . . . . . . . . . 21
2.4 Websites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.5 Datasets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.6 Statistical texts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3 Examining Continuous Variables 27

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.2 What features might continuous variables have? . . . . . . . . . . . 29
3.3 Looking for features . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.4 Comparing distributions by subgroups . . . . . . . . . . . . . . . . 44
3.5 What plots are there for individual continuous variables? . . . . . . 46
3.6 Plot options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.7 Modelling and testing for continuous variables . . . . . . . . . . . 48

4 Displaying Categorical Data 53

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.2 What features might categorical variables have? . . . . . . . . . . . 56
4.3 Nominal data—no fixed category order . . . . . . . . . . . . . . . 57
4.4 Ordinal data—fixed category order . . . . . . . . . . . . . . . . . 62
4.5 Discrete data—counts and integers . . . . . . . . . . . . . . . . . 66
4.6 Formats, factors, estimates, and barcharts . . . . . . . . . . . . . . 70
4.7 Modelling and testing for categorical variables . . . . . . . . . . . 71

vii
viii

5 Looking for Structure: Dependency Relationships and Associations 75

5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2 What features might be visible in scatterplots? . . . . . . . . . . . 77
5.3 Looking at pairs of continuous variables . . . . . . . . . . . . . . . 78
5.4 Adding models: lines and smooths . . . . . . . . . . . . . . . . . . 83
5.5 Comparing groups within scatterplots . . . . . . . . . . . . . . . . 86
5.6 Scatterplot matrices for looking at many pairs of variables . . . . . 88
5.7 Scatterplot options . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.8 Modelling and testing for relationships between variables . . . . . 94

6 Investigating Multivariate Continuous Data 99

6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.2 What is a parallel coordinate plot (pcp)? . . . . . . . . . . . . . . . 100
6.3 Features you can see with parallel coordinate plots . . . . . . . . . 102
6.4 Interpreting clustering results . . . . . . . . . . . . . . . . . . . . 106
6.5 Parallel coordinate plots and time series . . . . . . . . . . . . . . . 108
6.6 Parallel coordinate plots for indices . . . . . . . . . . . . . . . . . 112
6.7 Options for parallel coordinate plots . . . . . . . . . . . . . . . . . 115
6.8 Modelling and testing for multivariate continuous data . . . . . . . 127
6.9 Parallel coordinate plots and comparing model results . . . . . . . 127

7 Studying Multivariate Categorical Data 131

7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

7.2 Data on the sinking of the Titanic . . . . . . . . . . . . . . . . . . 132
7.3 What is a mosaicplot? . . . . . . . . . . . . . . . . . . . . . . . . 133
7.4 Different mosaicplots for different questions of interest . . . . . . . 140
7.5 Which mosaicplot is the right one? . . . . . . . . . . . . . . . . . 148
7.6 Additional options . . . . . . . . . . . . . . . . . . . . . . . . . . 149
7.7 Modelling and testing for multivariate categorical data . . . . . . . 151

8 Getting an Overview 155

8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

8.2 Many individual displays . . . . . . . . . . . . . . . . . . . . . . . 159
8.3 Multivariate overviews . . . . . . . . . . . . . . . . . . . . . . . . 162
8.4 Multivariate overviews for categorical variables . . . . . . . . . . . 168
8.5 Graphics by group . . . . . . . . . . . . . . . . . . . . . . . . . . 170
8.6 Modelling and testing for overviews . . . . . . . . . . . . . . . . . 174
 ix

9 Graphics and Data Quality: How Good Are the Data? 177

9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

9.2 Missing values . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
9.3 Outliers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
9.4 Modelling and testing for data quality . . . . . . . . . . . . . . . . 193

10 Comparisons, Comparisons, Comparisons 197

10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

10.2 Making comparisons . . . . . . . . . . . . . . . . . . . . . . . . . 199
10.3 Making visual comparisons . . . . . . . . . . . . . . . . . . . . . 202
10.4 Comparing group effects graphically . . . . . . . . . . . . . . . . 208
10.5 Comparing rates visually . . . . . . . . . . . . . . . . . . . . . . . 212
10.6 Graphics for comparing many subsets . . . . . . . . . . . . . . . . 214
10.7 Graphics principles for comparisons . . . . . . . . . . . . . . . . . 216
10.8 Modelling and testing for comparisons . . . . . . . . . . . . . . . 218

11 Graphics for Time Series 223

11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223

11.2 Graphics for a single time series . . . . . . . . . . . . . . . . . . . 224
11.3 Multiple series . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
11.4 Special features of time series . . . . . . . . . . . . . . . . . . . . 233
11.5 Alternative graphics for time series . . . . . . . . . . . . . . . . . 237
11.6 R classes and packages for time series . . . . . . . . . . . . . . . . 238
11.7 Modelling and testing time series . . . . . . . . . . . . . . . . . . 238

12 Ensemble Graphics and Case Studies 243

12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

12.2 What is an ensemble of graphics? . . . . . . . . . . . . . . . . . . 246
12.3 Combining different views—a case study example . . . . . . . . . 248
12.4 Case studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252

13 Some Notes on Graphics with R 257

13.1 Graphics systems in R . . . . . . . . . . . . . . . . . . . . . . . . 257

13.2 Loading datasets and packages for graphical analysis . . . . . . . . 258
13.3 Graphics conventions in statistics . . . . . . . . . . . . . . . . . . 258
13.4 What is a graphic anyway? . . . . . . . . . . . . . . . . . . . . . . 259
13.5 Options for all graphics . . . . . . . . . . . . . . . . . . . . . . . 261
13.6 Some R graphics advice and coding tips . . . . . . . . . . . . . . . 264
13.7 Other graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
13.8 Large datasets . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
13.9 Perfecting graphics . . . . . . . . . . . . . . . . . . . . . . . . . . 273
x

14 Summary 275

14.1 Data analysis and graphics . . . . . . . . . . . . . . . . . . . . . . 275

14.2 Key features of GDA . . . . . . . . . . . . . . . . . . . . . . . . . 276
14.3 Strengths and weaknesses of GDA . . . . . . . . . . . . . . . . . . 276
14.4 Recommendations for GDA . . . . . . . . . . . . . . . . . . . . . 277

References 279
General index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Datasets index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Preface

Graphical Data Analysis is useful for data cleaning, exploring data structure, de-
tecting outliers and unusual groups, identifying trends and clusters, spotting local
patterns, evaluating modelling output, and presenting results. It is essential for ex-
ploratory data analysis and data mining. There are several fine books on graphics us-
ing R, such as “ggplot2” [Wickham, 2009], “Lattice” [Sarkar, 2008], and “R Graph-
ics” [Murrell, 2011]). These books concentrate on how you draw graphics in R. This
book concentrates on why you draw graphics and which graphics to draw (and uses
R to do so).
The target readership includes anyone carrying out data analyses who wants to
understand their data using graphics. The book can be used as the primary textbook
for a course in Graphical Data Analysis or as an accompanying text for a statistics
course. Prerequisites for the book are an interest in data analysis and some basic
knowledge of R.
The main aim of the book is to show, using real datasets, what information graph-
ical displays can reveal in data. Seeing graphics in action is the best way to learn
Graphical Data Analysis. Gaining experience in interpreting graphics and drawing
your own data displays is the most effective way forward.
The graphics shown in the book are a starting point. Sometimes more graph-
ics could have been drawn, and alternative graphics could always have been drawn.
Readers may have their own ideas of how best to present certain features of the
datasets. Although each graphic reveals information contained in its dataset, it is
likely that in every case there is more to be discovered. It is certainly one of the aims
of each analysis to find out as much as possible about the data. The graphics are not
drawn for their own sake, they are drawn to reveal and convey information.
A central idea underlying this book is that many graphics should be drawn. The
aim should not have to be to draw a single graphic that summarises everything that
can be said about the data. That is too difficult, if not impossible. The aim is to find
a number of graphics, maybe even a large number of them, where each contributes
something to the overall picture. Just as many photographs of the same object taken
from different angles in different lights make it easier for us to grasp a whole object,
datasets should be visualised in many different ways.
The emphasis is on exploring datasets first and on presenting results second.
Graphical Data Analysis is about using graphics to find results. One way to think
about this is to imagine you are looking at a new package in R and it uses a dataset
you are not familiar with for the examples in the help. What does the dataset look
like? How would you go about finding out what features it has, and how that might
affect the use of the methods in the package? What information can you find graphi-

xi
xii Graphical Data Analysis with R

cally in the data that a modelling approach should also find? What graphical displays
are there that help you understand the results of other people’s models, such as the
examples given on the help page? This presupposes an active interest on the part
of the reader. Roland Barthes, the French structuralist, referred to readerly texts and
writerly texts. In a writerly text the reader takes an active role in the construction of
meaning. I hope the readers of this book will take an active role in thinking about
what graphics show, what information can be gleaned from them, and why they were
chosen.
As every dataset used is available in R or one of its packages, information about
them can usually be found on the relevant help page, including which variables of
what types are involved and how big the dataset is. Ideally there should be a descrip-
tion of why and how it was collected, with references to original sources. Context is
important for interpreting results and you have to know your dataset and its prove-
nance. A well-developed sense of curiosity is very helpful in data analysis.
Graphical Data Analysis is an attractive way of working with data. It encourages
you to look at many different aspects and to investigate in many different directions.
You can be surprised by what you uncover and even by which graphic turns out to be
most effective in revealing information. Your results are easy to show to others and
are easy to discuss with others.
For any result found graphically, we should try to check what statistical sup-
port there is for it, just as we use graphics to review the results of our statistical
modelling. Graphical Data Analysis and more traditional statistical approaches com-
plement each other very well and we should take advantage of this.

Acknowledgements
No book on R should omit thanking Robert Gentleman, Ross Ihaka, and all the
many R contributors. They have made analysis of data much easier for the rest of
us. Thanks also to Hadley Wickham for all his R packages (sometimes referred to as
the Hadleyverse), especially for ggplot2, and to Yihui Xie for knitr, a major help in
keeping this book in order. Particular thanks are due to Bill Venables for his words
of wisdom and for R advice and code. If any of the book’s code looks elegant, then
it must be Bill’s, and if it looks clumsy, it is certainly mine.
Dennis Freuer, Urs Freund, Katrin Grimm, Harold Henderson, Ross Ihaka, Kary
Myers, Alexander Pilhöfer, Maryann Pirie, Friedrich Pukelsheim, Christina Sanchez,
Günther Sawitzki, Rolf Turner, Chris Wild, and Aisen Yang read one or more chap-
ters and made many helpful suggestions for improvement, some of which I have been
able to adopt. John Kimmel was an encouraging and efficient publisher, who organ-
ised several constructively critical reviewers, including Di Cook, Michael Friendly,
and Ramnath Vaidyanathan. I would also like to thank the Statistics Department at
the University of Auckland for a stimulating and sociable environment in which to
work on this book during my sabbatical.
Preface xiii

Finally I would like to thank my family for never asking me when the book would
be finished and for many other kindnesses.

Augsburg, December 2014 Antony Unwin

1
Setting the Scene

‘What is the use of a book’, thought Alice, without pictures or conversations?’

Lewis Carroll

1.1 Graphics in action

Female

0
Male

0
160 180 200 220
Speed (km/hr)

FIGURE 1.1: Histograms of speeds reached at the 2011 World Speed Skiing Champi-
onships. Source: www.fis-ski.com. There were more male competitors than females,
yet the fastest group of females were almost as fast as the fastest group of males. The
female competitors were all either fast or (relatively) slow—or were they?

1
2 Graphical Data Analysis with R

The code for producing Figure 1.1 is:

library(ggplot2); library(ggthemes)
data(SpeedSki, package = "GDAdata")
ggplot(SpeedSki, aes(x=Speed, fill=Sex)) + xlim(160, 220) +
geom_histogram(binwidth=2.5) + xlab("Speed (km/hr)") +
facet_wrap(~Sex, ncol=1) + ylab("") +
theme(legend.position="none")

The 2011 World Speed Skiing Championships were held at Verbier in Switzer-
land. Figure 1.1 shows histograms of the speeds reached by the 12 female and 79
male competitors. As well as emphasising that there were many more competitors in
the men’s competition than in the women’s, the plots show that the fastest person was
a man and that a woman was slowest. What is surprising (and more interesting) is
that the fastest women were almost as fast as the fastest men and that there were two
distinct groups of women, the fast ones and the slow ones. There also appear to be
two groups of men, although the gap between them is not so large. All of this infor-
mation is easy to see in the plots and would not be readily apparent from statistical
summaries of the data.
A little more investigation reveals the reason for the groupings: There are actu-
ally three different events, Speed One, Speed Downhill, and Speed Downhill Junior.
Figure 1.2 shows the histograms of speed by event and gender. We can see that Speed
One is the fastest event (competitors have special equipment), that no women took
part in the Downhill, and that there was little variation in speed amongst the Juniors.
The reason for the two female groups is now clear: They took part in two different
events. The distribution of the men’s speeds is affected by the inclusion of speeds for
the Downhill event and by the greater numbers of men who competed. It is interesting
that there is little variation in speed amongst the 7 women who competed in the Speed
One event, compared to that of the 39 men who took part. The women were faster
than most of the men.
The code for the plots takes a little getting used to. On the one hand the informa-
tion would still have been visible with less coding, although perhaps not so clearly.
Setting sensible scale limits, specifying meaningful binwidths, and aligning graphics
whose distributions you want to compare one above the other with the same size and
scales all help.
On the other hand, the plots might have benefitted from more coding to make
them look better: adding a title, choosing different colours, or specifying different
tick marks and labelling. That is more a matter of taste. This book is about data anal-
ysis, primarily exploratory analysis, rather than presentation, so the amount of coding
is reduced. Sometimes defaults are removed (like the legends) to reduce unnecessary
clutter.
The Speed Skiing example illustrates a number of issues that will recur through-
out the book. Graphics are effective ways of summarising and conveying informa-
tion. You need to think carefully about how to interpret a graphic. Context is im-
portant and you often have to gather additional background information. Drawing
several graphics is a lot better than just drawing one.
Setting the Scene 3

ggplot(SpeedSki, aes(Speed, fill=Sex)) +

geom_histogram(binwidth=2.5) + xlab("Speed (km/hr)") +
ylab("") + facet_grid(Sex~Event) +
theme(legend.position="none")

Speed Downhill Speed Downhill Junior Speed One

8

Female
4

0
8

Male
4

0
160 180 200 160 180 200 160 180 200
Speed (km/hr)

FIGURE 1.2: Histograms of speeds in the 2011 World Speed Skiing Championships
by event and gender. There was no Speed Downhill for women. The few women
taking part in the fastest event, Speed One, did very well, beating most of the men.

1.2 Introduction
There is no complex theory about graphics. In fact there is not much theory at all, and
so the topic is not covered in depth in books or lectures. Once the various graphics
forms have been described, the textbooks can pass on to supposedly more difficult
topics such as proving the central limit theorem or the asymptotic normality of max-
imum likelihood estimates.
The evidence of how graphics are used in practice suggests that they need more
attention than a cursory introduction backed up by a few examples. If we do not
have a theory which can be passed on to others about how to design and interpret
informative graphics, then we need to help them develop the necessary skills using a
range of instructive examples. It is surprising (and sometimes shocking) how casually
graphics may be employed, more as decoration than as information, more for reasons
of routine than for reasons of communication.
It is worthwhile, as always, to check what the justly famous John Tukey has to
say. In his paper [Tukey, 1993] he summarised what he described as the true purpose
of graphic display in four statements:

1. Graphics are for the qualitative/descriptive—conceivably the semiquantitative—

never for the carefully quantitative (tables do that better).
4 Graphical Data Analysis with R

2. Graphics are for comparison—comparison of one kind or another—not for ac-

cess to individual amounts.
3. Graphics are for impact—-interocular impact if possible, swinging-finger impact
if that is the best one can do, or impact for the unexpected as a minimum—-but
almost never for something that has to be worked at hard to be perceived.
4. Finally, graphics should report the results of careful data analysis—-rather than
be an attempt to replace it. (Exploration—to guide data analysis—-can make
essential interim use of graphics, but unless we are describing the exploration
process rather than its results, the final graphic should build on the data analysis
rather than the reverse.)
These are all excellent points, although the last one implies he is emphasising graph-
ics displays for presentation. It is notable that Tukey writes of ‘the final graphic’ as
if there might only be one. In practice he commonly used more than one.
Data analysis is a complex topic. Conclusions are seldom clear-cut, and there are
often several alternative competing explanations. Graphics are part of the data analy-
sis process; both the choice of graphics displays and how they are viewed can have an
important influence on any conclusions drawn. The subject matter of the study will
also affect how any graphics are interpreted. A positively correlated collection of
points in a scatterplot may be taken as evidence of a useful association (e.g., income
as a function of years of experience) or as evidence of insufficient agreement (e.g.,
where two methods of measuring the same quantity are compared). In the first case
an outlier may be unusual, but nevertheless perfectly plausible, while in the second
it may be a clear indication of a faulty measurement.
To appreciate how much can be revealed by even the simplest graphics, it is useful
to think of all the different forms a graphic might take and what they might tell you
about the data. Consider a barchart showing the frequencies of the three categories
of a variable:

• The bars could all be the same height (as you might expect in a scientific study
with three groups).

• The bars might have slightly different heights (possibly suggesting some missing
values in a scientific study).
• One of the bars might be very small, suggesting that that category is either rare
(a particular illness perhaps) or not particularly relevant (support for a minor
political party).
• The bars might not follow an anticipated pattern (sales in different regions or the
numbers of people with various qualifications applying for a job).
• ...

There is literally no limit to the number of possibilities once you take into account
the different settings the data may have come from. This means that you need to gain
Setting the Scene 5

experience in looking at graphics to learn to appreciate what they can and cannot
show.
As with all statistical investigations it is not only necessary to identify potential
conclusions, there has to be enough evidence to support the conclusions. Tradition-
ally this has meant carrying out statistical tests. Unfortunately there are distinct limits
to testing. A lot of insights cannot easily be directly tested (Does that outlying clus-
ter of points really form a distinctive group? Is that distribution bimodal?) and even
those that can be require restrictive assumptions for the tests to be valid. Additionally
there is the issue of multiple testing. None of this should inhibit us from testing when
we can, and occasionally a visually tentative result can be shown to have such a con-
vincingly small p-value that no amount of concerns about assumptions can cast much
doubt on the result. The interplay of graphics with testing and modelling is effective
because the two approaches complement each other so well. The only downside is
that while it is usually feasible to find a graphic which tells you something about the
results of a test, it is not always possible to find a test which can help you assess a
feature you have discovered in a graphic.

1.3 What is Graphical Data Analysis (GDA)?

It is simplest to see what GDA can do by looking at a few examples. These exam-
ples are all just initial looks at the datasets to give the flavour of GDA, not complete
analyses. In each case we will draw graphical displays of a dataset to reveal some
of the information in the data. Graphics are good for showing structure and for com-
municating results. They are generally easier to interpret than tables (which are good
for providing exact values) or statistical reports (which are good for giving estimates
and formal comparisons) and convey more qualitative information.
For each example more than one graphic has been drawn. In general it is always
better to draw many graphics, offering many different views, to ensure you get as
much information out of a dataset as you can. This is part of the open-ended nature
of GDA; it is a process, in which you pursue multiple ideas in parallel, just as any
investigative process should be. As a result, a graphical data analysis of these exam-
ples in practice would involve drawing a lot more graphics, checking to see if there
are other features of interest, comparing different versions of various graphics to see
which ones work best, and finally settling on a group of graphics to summarise the
analysis.
Drawing graphical displays of data is not about selecting the one best graphic,
it is about selecting the best set of graphics. In the past drawing graphics required
extensive effort and producing one graphic by hand took a lot of time, so it made
sense to draw only a few. Nowadays we can draw graphics very quickly, many of
the basic options are sensibly chosen by software defaults, and the overall quality
is extremely high. It is still a good idea to be selective about which graphics you
keep and especially about which graphics you show to others, but the main aim is
6 Graphical Data Analysis with R

to uncover information and there is every reason to draw more graphics rather than
fewer when doing GDA.
With presentation graphics you prepare one graphic for many potential viewers.
You need experience in deciding which graphic to present and expertise in how to
draw it well. With GDA you prepare many graphics for one viewer, yourself, and
your aim is to uncover the information hidden in the data. You need expertise in
choosing a set of informative graphics and experience in interpreting graphics.

The Iris dataset

The dataset contains information on three species of iris. There are 50 plants for
each species and measurements in cms for four attributes of each of the 150 plants.
Originally the dataset was used by Fisher to illustrate linear discriminant analysis
[Fisher, 1936]. (It has been used for all manner of other analyses since.) Figure 1.3
shows a default histogram of petal length, one of the four attributes. There appear to
be two clearly different groups in the dataset.

ggplot(iris, aes(Petal.Length)) + geom_histogram()

20
count

10

0
2 4 6
Petal.Length

FIGURE 1.3: A histogram of petal lengths from Fisher’s iris dataset. The data divide
into two distinct groups.

We can look at a plot of the two petal attributes together, petal length and petal
width. Figure 1.4 shows that there is a very strong relationship between these two
attributes, providing further convincing evidence of at least two distinct groups of
flowers. The colouring by species shows that the lower group are all setosa, that the
upper group is made up of both versicolor and virginica flowers, and that these two
groups are moderately well separated by their petal measurements.
The iris dataset is so well known that many readers will be familiar with this in-
formation. Imagine, however, that you wanted to present this information to someone
who did not already know it. Are there better ways than simple graphics?
Setting the Scene 7

library(ggthemes)
ggplot(iris, aes(Petal.Length, Petal.Width, color=Species)) +
geom_point() + theme(legend.position="bottom") +
scale_colour_colorblind()

2.5 ! !!

! !

!!!! ! ! ! !

! ! !

!!!! ! !

2.0 !!!! ! !

!! ! !

!! ! !! ! ! !

! !

! ! ! !

1.5
Petal.Width

! !!! !!!

! ! !!! !

! !!!!!!!

!! ! ! !

! !!

1.0 ! ! ! !!

0.5 !

! !!! !

!!! !

! !!!!!! !

! !!

0.0
2 4 6
Petal.Length

Species ! setosa ! versicolor ! virginica

FIGURE 1.4: A scatterplot of petal lengths and petal widths from Fisher’s iris dataset
with the flowers coloured by species. The two variables are highly correlated and
separate setosa clearly from the other two species. The colours used do not reflect
the real colours of the species, which are all fairly similar.
8 Graphical Data Analysis with R

Student Admissions at UC Berkeley dataset

The data concern applications to graduate school at Berkeley for the six largest de-
partments in 1973 classified by admission and sex. One of the reasons to study the
data was to see whether there was any gender bias in the admission of students.
Figure 1.5 shows separate barcharts for the three variables (department, gender, and
whether admitted or not).

library(gridExtra)
ucba <- as.data.frame(UCBAdmissions)
a <- ggplot(ucba, aes(Dept)) + geom_bar(aes(weight=Freq))
b <- ggplot(ucba, aes(Gender)) + geom_bar(aes(weight=Freq))
c <- ggplot(ucba, aes(Admit)) + geom_bar(aes(weight=Freq))
grid.arrange(a, b, c, nrow=1, widths=c(7,3,3))

750
2000 2000
count

count

count
500

1000 1000
250

0 0 0
A B C D E F Male Female AdmittedRejected
Dept Gender Admit

FIGURE 1.5: Numbers of applicants for Berkeley graduate programmes in 1973 for
the six biggest departments. The departments had different numbers of applicants.
Overall more males applied than females and fewer applicants were admitted than
rejected.

It is immediately apparent that different numbers applied to the departments,

ranging from about 600 to 900, that there were more male applicants than female
ones, and that more applicants were rejected than admitted. None of this is exactly
rocket science and all these comments could equally well have been derived from
summary statistics (with the additional benefit of knowing that the reported range of
departmental applicant numbers was actually from 584 to 933).
Graphics make these points directly and give an overview that is easier to remem-
ber than sets of numbers. Graphics are good for qualitative conclusions and often that
is what is primarily wanted. Of course, precise numbers may be useful as well, and
the two approaches complement one another.
The ordering of the departments is alphabetic, based on the coding used, and
other orderings may be more informative. The widths of the plots were controlled
to avoid wide bars, a common issue with default charts. The vertical scales might
also have been made equal for all three charts (they are close for the last two by
chance) to reflect that they display the same cases. Then the differences between the
departments would have been downplayed.
Setting the Scene 9

The main aim of the study was to examine the acceptance and rejection rates
by gender. For the six departments taken together the acceptance rate for females
was just over 30% and for males just under 45%, suggesting that there may have
been discrimination against females. Results by department are shown in Figure 1.6,
where the widths of the bars are proportional to the numbers in the respective groups..
In four of the six departments females had a higher rate of acceptance. This is an
example of Simpson’s paradox.

library(vcd)
ucb <- data.frame(UCBAdmissions)
ucb <- within(ucb, Accept <-
factor(Admit, levels=c("Rejected", "Admitted")))
doubledecker(xtabs(Freq~ Dept + Gender + Accept, data = ucb),
gp = gpar(fill = c("grey90", "steelblue")))

Accept

Rejected

Admitted
Male Fe Male F Male Female Male Female Ma Female Male Femal Gender
A B C D E F Dept

FIGURE 1.6: Acceptance rates at Berkeley by department and gender. Overall

males were more likely to be accepted, but four of the six departments were more
favourable to females. A doubledecker plot has been used to show the numbers of
each gender applying to the departments. Relatively fewer females applied to depart-
ments A and B, while relatively fewer males applied to departments C and E.
10 Graphical Data Analysis with R

Pima Indians diabetes dataset

This dataset has been used often in the machine learning literature and it can
be found on the web in the excellent UCI library of machine learning datasets
[Bache and Lichman, 2013]. The version from R used here, Pima.tr2 in the MASS
package, is a training dataset of size 300 with zero values recorded as NA’s (a stan-
dard missing value code). There are six continuous variables and default histograms
are shown for all of them in Figure 1.7.
The distributions of three variables (plasma glucose, blood pressure, and body
mass index) look roughly symmetric. The variable skin thickness has at least one
outlier and the diabetes pedigree function distribution is skew, possibly with outliers.
The age histogram shows that most women were young with an age distribution like
half of a classical age pyramid.

data(Pima.tr2, package="MASS")
h1 <- ggplot(Pima.tr2, aes(glu)) + geom_histogram()
h2 <- ggplot(Pima.tr2, aes(bp)) + geom_histogram()
h3 <- ggplot(Pima.tr2, aes(skin)) + geom_histogram()
h4 <- ggplot(Pima.tr2, aes(bmi)) + geom_histogram()
h5 <- ggplot(Pima.tr2, aes(ped)) + geom_histogram()
h6 <- ggplot(Pima.tr2, aes(age)) + geom_histogram()
grid.arrange(h1, h2, h3, h4, h5, h6, nrow=2)

25
20
30 20

15
15
20
count

count

count

10
10
10
5 5

0 0 0
50 100 150 200 50 75 100 0 25 50 75 100
glu bp skin

25 50
60

20 40

15 40 30
count

count

count

10 20
20
5 10

0 0 0
20 30 40 50 0.0 0.5 1.0 1.5 2.0 2.5 20 40 60
bmi ped age

FIGURE 1.7: Histograms of the six continuous variables in Pima.tr2. There are a
few possible outlying values. Two of the variables have skew distributions.
Setting the Scene 11

Rather than drawing a lot of histograms to get an impression of the variables, it

takes up less space to draw boxplots. Since R automatically uses the same scale for
all boxplots in a window, they have to be standardised in some way first and this can
be achieved using the scale function. Figure 1.8 shows the result with the outliers
coloured in red for emphasis.

library(dplyr)
PimaV <- select(Pima.tr2, glu:age)
par(mar=c(3.1, 4.1, 1.1, 2.1))
boxplot(scale(PimaV), pch=16, outcol="red")

!
6

!
4

! !
! ! ! !
!
! ! !
! !
!
!
2
0
!2

glu bp skin bmi ped age

FIGURE 1.8: Scaled boxplots of the six continuous variables in a version of the Pima
Indians dataset in R. A couple of big outliers are picked out, as is a low outlier on bp
(blood pressure). The distributions of the last two variables, ped (diabetes pedigree)
and age, are skewed to the right.

There are several outliers (including a couple of extreme ones) and boxplots are
better for showing that than histograms. The last two variables are clearly not sym-
metric. Two facts should be borne in mind: The boxplots are not to be compared with
one another, drawing them all together in one plot is primarily a time and space sav-
ing exercise. The scaling just transforms each variable to have a mean of zero with a
standard deviation of one and nothing more, so the same points are identified as out-
liers as would be in the equivalent unscaled plots. This display tells us a little about
the shapes of the distributions, but not much, and nothing about the missing values
in the data, a potentially important feature. Of course, the histograms told us nothing
about the missing values either. Plots for missing values are discussed in §9.2.
The two sets of displays, histograms and boxplots, have given us a lot of infor-
mation about the variables in the dataset. A scatterplot matrix, as in Figure 1.9, tells
us even more.
Exploring the Variety of Random
Documents with Different Content
Smith, of Tennessee. lie died in Andrew county, .Missouri, in 1847.
She also died in Missouri. They hail nine children, of whom seven are
still livinir, ,1. A. Stanton being their c)ldest child. Mr. Stanton was
reared in Missouri. His father died when ho was si.Nteeii years old,
and he was then cast iijion his own r;\sourcos. He obtained work on
the Missouri river as a steamboat hand, and in the spring of 1852
started to cross the plains, having a horse and mule and working his
passage. The party with which he traveled were harassed by the
Indians, but reached their destination in safety, arriving in Oregon on
tlui loth of Octolier. He came to Marion county, and for a time
worked tor his uncle, near Silverton. August 29, 1854, he married
Miss Harriet Moser, who was born in Indiaiia, September 13, 183ti,
daughter of Joseph .Moser, of (Torman descent. Somi after their
marriage, Mr. Stanton purchased a farm 1 of 210 acres in Marion
county, on which they resided from 1S()3 to 1877. They then bought
a place on French prairie, where they lived three years, and from
there, after selling out, they came to Clackamas county. Hero they
purchased a farm of Itio acres, and on it they have since resided. Mr.
and Mrs. Stanton have had the following named children: Mary Ellen,
who died when two years old; Isabella, wife of William Murry, resiiles
at Grande ll(jude; ■losephinis wife of 11. S. Thomas, resides on
Butte creek, in Marion county; Jennie, wife fif F. RIackby, die I in her
twenty-third year; William Franklin, who resides in Clackamas
county; Jasper N., who lives near hia father; Henjamin at home;
.Viuanda, wife of William Woodcock, lives lujar her parents; John A.,
Willard, and IJlewford. Mr. Stanton is in politics a Democrat. He is a
reliable, upright citizen, and a good representative of the ()regon
pioneers. g T^Jtft .1,1 ?
 Vi2'i IIISTOHY OF OlWdoN. I'litiM't'd the ministry of tlio
Metlioditit Cliiiruli. lie ci'DsxMJ the jihiins with his fuiiiily i" 1^0-4 witli
one ox teiiin, i'.\|it'iiiMu^i]ii^ llic usual iiiri(ii'iits of tliat iiKidi!
ol'traveh l.amiiiii^ in Orc{fon ill tlu; aiitmnn, in' IdcutiMi liis iloimtioii
claim of ii^idacius lu'ar Starr's I'oint, now Iviiown as Monroe, I'cnton
county, lie tijiMi joiucd tlie Ort'ffoM ("orii'crfueo, anil lor twunty- five
const'iuitive years was enjraj^ed in church work, lit) served one year
us I'l-esidinj^ lOldor of the '.•'ortland District, and in 187.S was
deloifate to tiie General Conference at Cincinnati. Siilfering from a
severe full, folhiweil hy a stroke of ]iaralysis, he was olilioed to retire
in 1880, and now resides near llalsey, Linn county. Isaac W. ijained
his education at Philomath College, Henton county. He then entered
the medical department of Willamette University, ^raduatiiig
tlicrefi'om in 187'.), and commenced jiractice at .Monroe. In 1878 he
removed to IJrowiisville where he has since followed a fjeneral
practice and has Iniilt u|) an extensive patronnije. Jn 1882 he
boufrlit one-half interest in the di'Uff store of Cr. C. lilakely, and
continued until 1884. when Blakely sold his interest to T. Forsyth,
and the firm Starr & Korsyth ojierated the store until 18S7, when the
Doctor sold out. In 1889 he started the City Druo Store in
partnership with (i. (.'. Sranard, and in Feliruary, 18UI), lioiiirht the
Stanard interest and has since continueil alone. lie owns the store
projierty, 22 X 110 feet, ami his residence on Main street. He was
married in Brownsville, in 1882, to Miss Clara ISishop, a native of
Linn eoiinty "nd the daughter oi VV. II. Bishop, a pioneer (d' 1854.
To this union has lieen aiided two chilr, where nuiMV of the company
contracted cholera, which was so jircvalent upon the plains that year.
A brother-in-law of Mr. Wortman and several other members of the
party die
 l/ISTOin dh' allKdoN. i9sa Clmrles V. rt'cuived his oducntion
at Ort'f^on VAty At tliu ai^i) of ui;;liti'iMi liu LMitcred liin fiitlii-r's
tstnjiloy iind Kiili^tMjiieritly liuciime li pMi'tiuM' in tliu liiisiiichH,
coiitiniiiMg iii« hiicIi until 188(1. II(^ tluMi tiisposfil ot' liir- iiiteit't*t,
went to l'liilu(lcl|)liin and entnird tiio .ItdTerson Medical Oolli'^'e,
where he ffnuhiiitcd in Miirch, 1881, with till' ik'j^ree of M. I).
Rituniini; to Oreijoii, he I'oinnieneed |)iiiclic'e at Monroe, lieuton
county, wliere lie reniained tor tlireo years. Then, ill order to furtlier
prepare himself tor his life '.'. nk, he went to Kurope and pii>^ed
two and a half yearit in the University of Vienna and the hospitals of
Iterlin and London, Htudyiiig under the in8triietir)n of private tiitorB
in ordei' to make more rapid advanceinent. lieturning to I'ortland in
the fall of IS'.M), he opened an otlice for jreneral practice and, with
his hroad knouledfje and piactical exjierii'nce, in liotli UKMlicine and
surgery, his reputation as a ekillcii physician was (jiiickly estahlished,
and his success achieved from the very start. The Doctor is a
memher of the V. & A. M., K. of P. and B. l\ O. E. lie is a frequent
lecturer hefore the medical f the following Septemher; tliey had a
safe trip, and oidy ono of the company died on the way. The lather
purchased '■JO acres of land, on which ho liviMJ one wintn , In the
spring he ^dd the tract, and located on another farm of 320 acres of
tiovernnient land; there he huilt a home, and cultivated the lancl for
two years; celling it at the end of that time.; he next purchased (IT'
» acres near Macleay. When lliinm il. Smith started out to meet the
respiin^ibilitii'- of life, he wa> first I'uiployed hy the month, Inil in
Octoher, 1^.")4, lie went to \ reka, Si-kiyoii county, California, and
engaged in placer mining; ho followed this business witii lair success
for three years, his hest day liringiuf* him #100. He next went to
the Idaho mines, where he met with divided fortunes, hut came
away with enough money to buy 200 acres of choic(< land in the
Waldo Hill district; there he huilt a li -e, improved the laiiy emigrated
to Missouri in 1882, where Mr. Scott attained prominence and served
his county ' in the State Senate for six years, and was also j elected
Lieutenant-Oovernor of the State. With j his wife and six children he
started for Califori Ilia in the spring of Is-l.'), the train of ox teams I
nuniberiug about fifteen wagons and fifty people,
 1334 niSTOKY OF OHKnON. Mr. Scott lieinj{ ulfcti'il fiipliiiM
(il' tlie Iraiii. After six iiiuiitlis of triiM'l llioy IiiimIoiI iit Siittiu'V fort,
and passed tiin winter, and in tlio sjiriiif^ of IS4I) sold llieir waj^ons,
and uitii llicir cattle and puci< lirrseu went liy Imlian trail to Oregon,
direet to Vani Hill eoimlv, and passed the winter with Joe Watts, in
tin,' spring of 1847 tliey returned to Lane county, and localtMJ on
ti4()acres in the forks of the Willaniette river, and ent jiroductive
farms of this beautiful valley. He paid for his tirst purchase S17 an
acre, since which ho has, from time to lime, as his means permitted,
added to his original purchase, until he now owns, in one body, 404
acres. His last purchase of 188t acres cost him §39 an acre, and the
whole is now valued at $75 an acre, and is ■ not for sale. Mr. and
Mrs. Scott have three children, all born in Vani Hill county: James
Winticld, William and Sarah lOlizabeth, all of whom are graduates of
the McMinnville ('olieire, and reHect credit on their educators and
the State of their birth. Mr. Scott is Democratic in politics, and takes
a very great interest in all the officers of the county and in the
schools. He holds the office of School Clerk, and, in 1890, received
the nomination for County Commissioner. He is a member of the A.
O. U. W., and takes a prominent interest in the Cxrange. He and his
wife and three of the children are ineitibers of the liaptistOhiiAh, of
which church he has been for the last fifteen years a Deacon, and
most active in its behalf. Heaided in building the church edifice and
also in erecting the college building, both of which are ornaments to
the filace, as well us worthy objects of care, diifnsing, as they
<1<>. knowledge and i.'orality throughout the conitnuiuty. He has
also erected for himself and family a comfortable residence in
McMinnville, on iin
 Htsroiir Oh' (iitKdos. 18'JS •MtilllKMl to !■> Vvllt-li, ill iinty,
OreI. Ill' riiar'lio Ik iilno iiilili! lady. • ■'ii-iii rmnr lii'Of! yoi«rN, wlifch
lid 1 vnry oti■aiii raiMer. ar^ru yiold Ito Day. ii'in a very it^ pi'fsctnt
H' \V( irked •iiiir it and ^' it witli a .riaiii and unt, it has |ir()diurtiv(i
lid t'cir Mh 'di ill! has, luTinittcd, il ho now last piirI acre, and ere,
and is ililren, all Winlield, whom are e, and roe State of and takes ii"s
of the file oUico sived the and takes Idren are ill cliundi I'eacon, in
Iniildi-'tin^ the •nainents of care, i>'orality famil_j' . on iin a
attractive Bite, botli hoiiee and siirroiindin^fH !)ein^f sn^jj^chtive of
cidliire and taste. Thus, honutftly and iierheverinj^ly, he haa carved
out his fortune from the nafiinil advantaM;es of this lieailtifiil and
productive eommonweallh,thaM which there was never a more
j^dorions one in the worhl. In acconiplishini.' this, however, he has
never foi'feited the ^ood-will of his nei(^id)orrt, hut enjoys,
iindispntodly, their lushest respect and eBteeui. fiCOKGP: W.
t^TAVKH.oneof the foremost developers ■'' ;ne ajjrienltiiial husinivss
in the j.'rrrgoii, as manager of the .1. I.asiirer. Mr. and Mrs. Staver
have tour intelligent and promising children: Fraiddin, Ida Thor|),
.lohu Frederick and Klla Hannah. Mr. Staver is intensely public-
spirited, as is witnessed by liis wide and varied coiiiu^ction with
nnnerous social, ecliicational, charitable and religions institutions. He
was one of the most active organizers of the (xartield Fost, (i. \. li.,
in which he takes a prominent part. He is a trustee of both the
Willamette and Portland Fniversities, being elected trustee of the
fiu'iner in 1887, and of the latter in 1891, having been a prime
mover in the hitter's (U'ganizatioii. As a useful member of the First
Methodist Kjtiscopal Church, has served as Trustee and President of
the Board for many years. He was for four years President of the
Portlatid Methodist Hospital, during which time their new hospital
was erected at Sunnyside. For six years he a 'ted as president of the
Pacific Christian Advocate, until the pajier was taken in charge. May,
1892, by trie pajier was taken in ci tile Meihodist Conference.
 laao HlSTOHY OF OHKOON. m I \}m 'M 1 hi ; !;lk: It Ik
i^riililviiif; til litiil u MKiii uIki ciiii anil u ill turn iiHiilcfnnn luoiicy
inakiiij; tn K*f.ssioiiH tooot'ten i liliiiil (Mir siirjit to tin; greatiii'
pD.ssesnioiis liovoii'l, onlv atliiiimlile l)_v l'iilfilliiii( our iliity to iiiankiiiil
ami (iod. fOllN V. STKKFE.Washipl.iiil.kMat North l'(ji'tlaiiil, ouiirt and
operates oiu; of the laruent sliip yai'dn of tlio Xortliwost. He was lioi'ii
in (ii riiiaiiy in 18-t"^, ami spunt his i;arly youth upon a t'ann. He
was ajjpictiticcd to a sliipl)uildcr at Iiiih(!(' to loarii the siiip-
cai'pcnlers' tiade, and after four years of servieo he entered the
Prussian navy-yard at Daiitzic, where he followed his vocation until
November, ISd'l. He then ship|)ed as carijeiiter upon the (iernian
hri^antine Volador tor Alanzaiiithi. but left the vessel at N'alparaiso
in .Inly, 1S65, and re-shipped iijion the Aineriean iiarze Dolphin f'.)r
l^altiinore. lie next sailed iijion the Anieriean sliip t'owpor for Japan,
and, after visiting the important |)orfs, he returned to the United
States with ;iS4C!hiiininen, who were landed at ISan I'Vanciscii in
.lune, 18(i8. Mr. Steffeii then followed his trasels upon the eoant.
The .Vlliaiice. Vrillipan and the tug boat I'ilot are still in active
service. In 1890 the (Jeorgia Oaks was built upon Cceur d'Alene
lake, in Idaho, for passengei' service. Mr. Steffen was married in
I'ortland in 1873 to Miss Laura K. (iallosvay, ami they art; now the
parents of four children: Lorena. V.W-a May, Freiiirick K. and
Margneriti;. Mr. .'^teflen has taken an active interest in the Masonic
fraternity, ill wliicli he lias taken the thirty-second degree. For twenty
years he has been engaged in the art and industry of shiplniilding.
and has carried the business to a high degree of perfection. A dull
season is a stranger to him. and the fact that he is always pressed
with work is the best evidenct of his skill and the iiiteirrity and
honesty of his methods. -^'-« ^^^■'T^ ICIIAEL STEFFEN, one of
Portland's reliabh" merchant tailors, is a native of the city of
Milwaukee, Wisconsin, born on May 18, 184:4. His father, Jo.seph
Steft'en, was born in Prussia, Germany, and married Miss Ann
Steffen. a native of (Termany. having his own name liul no relation.
They came to the United States in 18;?!' and settled tirst in Detroit,
Michigan, then in Cleveland, then in Chicago, Illinois, and finally they
removed to Sheboygan, Wisconsin, where he died. They had eight
children, tive of vvhtjiii are now living. Michael Steffen was raised in
Sheboygan and went to school there, lie learned the trade of tailor in
Milwaukee, and .Vpril 3S, I8rt2, res|ioiided to the call of President
Lincoln for troops, enl'.sting in ('oinpany (-i. iMghteentIi United
States Infantry, and fought three years in the Union army.
Department of the Potomac. He was in the battles of Hoover's (iap,
Ciiickamauga and at the siege of Chattanooga, and was one of the
heroes of Missionary Uidge, where the good army of the Republic
wrote one (rf the most brilliant pages in the history of tlie country
and coveveci themselves with glory. Aftin' the battle they pursued
the enemy to Kinirtrold. Here Mr. Stetfen was taken with ill-])ox and
remained at (Jluvt.tanooga the sn '.mmm^
 HH msm. lll.^TiH;) OF i>l;h'ilil\. 1*57 aii'l Ocean 'i.ltcr. Si-a
.1111(1. All Millllt'il 1k'"^^^('ls ll|>()ll Mil tlu' tllfr in isyo iir d'AltMie
until till- .'Idkc (if till! wiir. Me cam.' out ol llio j^jifiit >trifi' wirlioiit a
^oratcli, ami was iKMiDcably clis{'h;ii'i;t'\: l{t)l)ert.s, the leadinj^
cliitliinj^ lionne of l\>rtlan(l. Since then lie lias cut lor and lirei
connected with the most nroinincnt house- c^f Poi'tlaiid. In August.
l8iU, lu ii|iened lii« present merchant tailor husines!;, where he has
experienced workmen ami a good stock of imported cloths. Mr.
Stert'en has had lonj; experience in the Inisiness and is reliable, as
well a» artistic in his work. lie is liiglily deserving of the nice business
that, he enjoys from the best citizens. He wv.s married in 1880 to
Mary < >"l)ea of Massachusetts, and they have a d(»liglitfiil home.
They have a daughter whom they have named Irene. Mr. StefTen is
rather independent in jiolitics and is a worthy inembei' of the (I. A. li.
—rg^i ■m:^^^^^ UTRTIS C. STROXG, M. I)., came to the
Northwest Territory, formerly known as Oregon, in lS4rit, and now
rejiresents the medical profession us the oldest continuous
jiractitioner of the city of I'ortland. Ili.« father, Judge William Strong.
deceased, for inanvyears filled a prominent place in the pulilic alfairs
of Oregon and Washington Territory lie was descended in the
.seventh geinM-ation from KMer John Strong, of Xortlianipton,
Massachusetts, who arrived in New England in ICJO. Judge Strong
was born at St. Albans, Vermont, July 5, 1817. He was giadnated
from Yale in 1S38, and afterward went to Cleveland. Ohio, where lie
taught school and studied law. After his admission to the bar he
be;;an the Flractice of his ])rofession in this city, where he ived until
.lannary, 1850. He was married at (ioshen, Connecticut, October 15,
1840, to Miss Lucrctia Robinson, a daughter of William Pitt Kobinsoii,
of Xew Haven. Illinis. In 1849 he was appointed As.-oeiato Justice of
the Supreme Court of Oregon Territory, and with his family
embarked, in January. 1850. on board the I'nited State:" ship
Sup|)ly The Supreme Court was held in those days at Oregon City,
then tht capital of the Territory, and upon its removal to Salem Judge
Strong and Chief .lustice Nelson declineil to admit the lei;ality of the
act, and the court conlinned lo sit at Oregon City. Tli(> judges
recei\cMl their share of critieii-m, and e\en abiiNC, but nniinlained
their position with firiiiness and dignity. I and, tluuigli the advoi ite-nf
the lemovid linally triiimplied by congressional action, no luu; today
thinks ot i|uesfii)ning the integrily or abilit\ (if the eonrt I'y the
division of the Terr it "ry in ISoJi .linh'e Strong became a eiti/.iMi of
Washington TerriI tory. In 1858 he was appointed (Jnitecl States
District .lii'ige for Wasliini,'ton Territory, and reimiiried on the bench
until 18(tl. He removed with his faiiiiiy to I'ortlaiid in iK*!:.', ami
there resided unt;', his death, .\pril 1uriMg the last four years of Lis
life he reliii(]iii;-he(l his professio'ial diitie~. From l.Sti2 until the
dissoluf ion o'' the Oivgow •■team Navigation (Company he muh its
legjii art. Iser. Hu was one of tlie prime •overs is the organization of
the Library Assi^Mrttion of this city, and was for teveral years its
vice-president. He was president of the liar .Association during its
brief continuance as an active lioily, a position to which he was
ent:"lei|, both by seniority and rank in his profession He was an
ardent friend of the Indian, one for whom the red muii IwkI great
respect and wh'se eoutis«'ls were alwaj* heeded. For two years
previous to his death In? lived a widower, his cvcellent wife having
been called to the other lite .November '>. iHH-i. They were the
parents of eiglii'' chlld.iu'n. two of whom died in infancy: Dr. Curtis C
Stroiii:, Frederick K. and Thomas N . lawyer-; Williinii il., Ellen and
Caroline, all living in Portland. The Judge's career as a lawyer wa»>
a most successful one, and as an :!dviser his ci>nii-el was highly
esteemed. Curtis ('.Strong was bom in Clevehii.d. OL ". December
~!, Is48, but, having been bnmjjht to the Pacific coast during his
infancy, his »^rlie>'t recollection is of ( atlilamet. wlnre he lived
intrtut fourteen years of age. In 18()2 h* caii.tie to Portland, and
attended the city sel«*»l- luitil 1867, when he entered Marietta
Cwileijt; at Marietta, Ohio. The president d' rhi» iiistitmtion. Prof.
Isaac j\ndrews, was ;i c<>n«t»ction of the family. He remaine! there
until ls('>9, when he went to liutf'alo. .Xew York, ami in 1870 he
began the study of ine«licine under tiie jiri^ceptorship of Dr. Milton
(r. Porter. He then took two courses of lectures in the medicHJ
department of the Fnivers-tj of Ruflalo, and # \^\\^X''i
 ■n Vi-M HlarOHY OF (iumo\. f! lij .'iffcrwanl c'ltorod Belloviic
Hospital .Nfediriil (Uilli'^rc. New York, from whieli \w was j/raclimti'il
ill 187:2. IJctiiniinj^ to I' been horn of the union: Alice, Frederick H.,
Kohert H. and Stuart II. l)i-. Strong Is a lueinlier of the Masonic
order, lieloiigiiig to lodge No. 55, A. F. & A. M. He is one of the
directors of the Oregon I'ioneer Association, and is jiresident ol the
Coliimhia River Centennial Celehration Siiciety. He is the owner of
coiisideralile valiialile I'eal estate in Portland, and, while he invests
hie surplus means in this way, his attention is Dover diverte
 KIMH iiisTonr OF intKnoN rKii ler to S. H I if wai^oiiri trip
wi^^ a ■(f(J in tlif irifr |)iintied ij,' of l«5-i tfr. Veatch creti, north 5
liis anc('.-.ti>rs softlcd in tliat State |ir('\iiiu- in tlie licvdlntiiin. He
niarriiMl ALi.-.s Mai-y iin.-.-. till! dauiflitcr of the l{ev. AloxaiuK-r
li^l^s. Of four cliiliircn horn to tiieni, two survive: Manly Levi, anil I.
L.. the hitter heiiif? the cuhjeet oi' tliis bketeli, who was l)orn in thi'
sanu' State, and who was the ehie.-t ehild. Ili^ motiier died wlion he
was six years old. and his lather married Miss Mariraret Sclireve. of
Illinois. To this union were added nine eliildren, of whom seven are
livinir. Ilis father continued to reside on his larm until the time of his
death, and J[r. Smith was reared anil educateil in that State. He
learned tlie tnnle of earjienter and cahinet-maker. and in Illinois he
euffairi'd in contractino- and huihiincr. Later he was in the furniture
liusiness. until the civil war hurst U[)on the country; when the
demand for men lieeanie great; in 18t)3, to fill up the hrolicn raidis
of the Union army, he enlisted in Com ])any H, ^'iuety-fourtll
V\)lunteer Infantry, ami he served in Mi.-souri, and participated in
the hattle of S|irinnHeld. He was at the takinir of Vickshuro;, was
then sent to N'ew Orleans, and fi'om there to Hrownsville, Texas,
and assisted ill the ca|)ture of Fort Morgan and Mohile. They then
retired to (Jalveston, and were there when news of the surrender of
Lee arrived. Mr. Smith was sick in tlus summer of this year, and the
result of it was that his hearint; was impaired, and he never fully
regained it. When he was niu.ster*^d out he I'c.turned home and
resumed his husiness, and renniii.ed there until 1870. In that year
he went to Forest Grove, and engaged in contracting and huildinju;.
lie huilt the Ladies' Hall of the I'acitie University, a splendid edifice,
and also huilt many of the hest residences in the city and all s
Mjirgaret Matl..rs, of ()liiay is a .-uc cessful teacher of music; I'Ved
lioss is a clerk in his father's store, and Carrie is at her lunne with
her pa.'-euts. .Mrs. Smith, who ha^ he.-n his faithful wife for forty-
thiee years, is still livinir. Mr. Smith is a Kepuhlican in politics, and
has sei'ved on the City Council, and has hecn ! twice elected to the
State Legislature, lie is a charter meiidier of the (i. \. K. Post in
l''ore.st Grove, and is also a worthy memher of tlu' Masonic order. He
is a valued memher of lie Methodist ('hurch of which he has heen a
Trustee, Steward, (Jlass-le.ailer, and Sutniay-school Superintendi'iit.
He was one of the numher who served on the huilding comniitt(M'
when they erected a new church. It cannot hi' said that Mr. Smith
has lived for himself alone, as he has heen all his life dtivoted to the
hest interests of the town in which he has lived for so long. He has
made in.'iiiy friends hy his ujiright, honest, manly way of performing
the husiness iiitrusted to him. fT. SMITH was horn in Monroe county,
Tennessee, in 18;-34-, whei-e his gi'and» father settled prior to the
Uevoltitionary war, and after defending the country of his ado|)tioii
against Great ISritaiii, he settliMl to an agricultural life, and lived to
the n\>K old ago of ninety-four years. The father of our siihiect sold
out his interests in Tennessee in 1852, and with his family emigrated
to (irand Prairie, (ireeno county, Missouri, and there engaged in
farming, hut was more extensively interested ill the stock husiness
until his death. Nine cliildren were horn to this union, P. T. Smith
heiiig the youngest ehild. He was rjared upon the fiirin, and early
inclined toward the stock husiness, in which he was cliiefly engage'!
after reaching Missouri. In )8.");i he was purchasing agent for
Messrs. Whitlock & Steward, and. in 18.")4. with a hand of 1,(KKI
head of cattle, they set out to cross the plains to California, our
suhject heing ont of the drivers, rnd also detailed i.s hunter for the
party, thus v orking his way across. The trip was inaile in four
monthsand twelvedays, arriving at i
 h I I'ilaao IlLSTfUlV OF OUEOON. S.-icruim'tito valley,
where!, iil'tcr six iiioiillirt in l';itti'tiiiii;, tiiestoc'k wiirt Hidi] (ml. Soiiii
after liisiirI'ivnl Mr.Sniitli wont to the mines on nettle Ornfk, Shitstii
county, ami in that hjcality lie followed jihicer inininii for ahout four
years, witlioiit t-triliini; iniv rieh " leiul." lie then went to the I'rasei'
river mines in Uritish C'oliimliia, but lieeause of tlu! larj^e
emir;ration, the rivi^ was ovcr.-toekiMl, prices of livino; were very
hio;li, ami sniiject rclMrned to ()ri!i;on. lie then pnrchii.-iMi l.")t}
acres of the 1). S. Sonthmayd donation claim near the present town
of St. John's Town, and "Misraf^jed in the loirainif business, which
he followed successfully for thirteen years, investiiio his savino-s in
other lands until 'he owned lt;,0()() acres, 'lie also built 400 feet of
dock upon thii river, and I'ngaj^ed in the wood biisliit^ss, supplying
the river steamers with fuel. Aliout 1872 he began tradino- in stock,
aud starteil a dairy of twenty cows, wdiich was gradually increased
to sixty cows, and i.s still contiiHied under the manaj^ement of his
son, Henry W. Mr. Smith is also enijaired in raisiuix horses, and some
very choice stock. He has reduced his lauded interests to 1,000
acres of farm land and rnnge for his horses and cattle. Beside his
dock he owns forty rods of valuable water front helow St. ilohn's. He
laid (df Smith's addition to the town of St. John's; built his resideiute
there in about 187(j, aud there he still resides. Ho was married in
Multnomah county, in isr)2, to Miss Mary J. Wiude, a daughter of
John Wiiide, wlio crossed the plains in lS5;i, and took iii) his donation
claim in Multnonudi county, lo the above union has been added nine
children, seven of whom survive: Henry W., Fraidi. Sarah (i.. Mary A.,
Etlie L., Philip A. and hi/zie 1'. Mr. Smith has taken no lively interest
in jiolitics. e,\cept to help maintain honest elections, and as School
Trnateoto aid in advaticiufr ''ducntional facilities. His life has been
one of activity, push and entorjii'ioe. and liis success is but the just
reward of intelligent, honest labor. IWKN SrMMKKS, Lieutenant-
Colonel of the First Regiment, Oregon National :ing detained in San
Francisco, hec«iise of the ice on the (.'olumbia river,heilid not arrive
in I'ortland until February. Not liking the city, ho retnrr»«Ki to
I'hicago, but again visited San Francisco, and ther(! (Migaged ii\ a
paint and oil manufactory as cK'rk, and remained thei-e until
February,
 MU /iisninr (•!'• (H!h:i,(>\ Vi:n ) their luarHiiil tliiiro iifiicliiro
ol' to ( Miiciiifo, iioi"ili)iilii>ii 57. wiien ho ■thiM- wiirld. ih-eii. ( )wcii
ily w.i^ li/t't. h(iiin;h liiit Ills iiiDthor [i) tlin nt re■i, and after 'ingiigud
at em Toniiesetly in skirtho Oonf'ed0 St. Louis, 18(55 went care of
the 3 northern , 1 S(i5, they in(ii><. mr.="" licli="" he="" i=""
went="" entrained="" in="" during="" the="" u="" serviee="" r=""
siilferstant="" to="" tlie="" etive="" inease="" eontra=""
buildings="" and="" b="" tlui="" ice="" on="" rotnnnxl=""
ancifco.="" mid="" ii="" iinu="" factory="" feliriiary="">, wlien lie
reliinuMi to I'nrthiiid, Oregon.and foniicd a jiarlner!-iiiii witii iiis
iirntlu'iin-liiw, J.C. ( ''ds, under llie tirni name of ( Hds A: Siiniinei'K.
I'iiey rented ii sniall iiilev, 5x:i(l ft Lieutenant, .\ Iter the orgauiziition
of the company known as tlw; First Regiment, ( ). N'. G., tile Veteran
(4iiBrd was enlisted as Company K, of that regiment. In 188.5
Lieutenant Siiniiners was appoiiited lieginiental .Vdjntant of the First
Battalion, O. S. M., ami upon the organization of the First Uegiment
O. N.(t., in .Inly. 18^7, under the new law. Adjutant Summers was
elected Lieutenant-Ojlonel of the First Regiment. ami was
una.iiniously reelected . I iilv 9, 1801. In 188t) (%)lonel Sum'mers
was elected by the Republican party to the 14fh session of the
Oregon iA'gislatiire, and it was through his personal efforts that the
bill for the Oregon National (iiiard became a law. He was also
instrumental in the passage of the badge bill, wiiich makes it a
misdemeanor for an impostor to wear the Imdge or lutton of the (t.
A. R. (_!iiloiiel ."^iiiiimers was a nii'iiil' 'r of llie Ti^er Kngiiic
Cnnipanv. .No. o, ol the old vidiinteer lire : department and I'or three
and one half years I served as treieiit ioils labiir when in'rformed
with jieisislent clelermiiiatioii. iSIU'RV Slim/rZ, -me of the most
relia//KiK ''le and enterprising business men of Dallas. ()reg
 V-l-A'i UlsrOUY OF UliliUON. \'-\ 1 i ('1 liriiit>. .111(1 wim'il
now In- clicap lit >c20() |M!I' Hci't!. Ill his yoiiHi Mr. Sluiltz Ifuriu'il
tlic triirpciitfrK' tnulc, iinil iliiriiig tlii' yi"ii's (.)!' liis rt'si(li'iicc ill Diillii-i
liiif. ii!^ II ciJiitnictoi- mill liuildi'r, lirc'i 11 |iriiiiiiiit'iit tiicliir ill iiilvunciiii;
the i)08t iritt'i'i!.-ts of i\w town. Wt.' 1ms Imilt sevural nice
rcsiilcMci'n for himself, unci at this writiiiff (!S'.l:i)is liiiililiiii^ iiiiotlu'i-
tine one. IIo Iwis iiImi liiiilt mill sold niiiiierons residences. He is ii
stockholder in, mid one of the Uiiilders of the imye woiilen factory
now beiiii; completed ill Dallas, he having superintended il8 erection.
Another child, lOva, was Imrn to Mr. and ^^r8. 8hnltz after their
arrival in Dallas. Their uldet-t child, (iertrlide. is now thi^ wife of
Frank Seahriiii;, and reside^ in Doiii^ias county, < >rei^'Mi,
(icor^iaiia died in her eighteenth year, njlii i.N the wife of R. 11. Me,
('nrf:er, and Kva, of Enjii'iie llavter, tiotli livinji; in Dallas, .lolm ('. is
niarrieii and is also settled in Dallas. Sani'ord is married and resides
in the eastern [.'ii't of ( M'egon. Mr. Shiiltz is a ineinher of the I. O.
O. V. in liotli its hnuiclii's, and in [lolitics lie is a l!epulilicaii. .gaATlLVN
KOONTS SITTON, who came Wl to Oregon as early as l.s4i{, was
horn in Fnltoii, Callaway county. Missouri, Septeiiilier 2, l^-J."). His
father, Fn.nklin Sitton, was a native of Tennessee, of Seottisli
ancestry, who eaine to America early in the history of the colonies. II
is j/raiidfatlier. dell'ery Sitt.m, was horn near Nashvillf, Tennessee,
and during the Ilevolutiouiiry struggle aided the colonies by
forwarding horses to the American forces as iieeiled. AEr. Sit'on's
father married ^liss Reheeca .\iistiii. who was born in N'irginia, and
brought up in ivenlncky, the daughter of .John Austin, and believed
to be of (ierinan descent, or, at leiust, partly so. liy that marriage
there were tive children, only two of whom, however, are now living
Mr. Sitton, whose name introduces this sketch, was the eldtfit sou of
his father's family. He was brought up in his native State until bis
seventeenth vear, when, with two other young men, he started on a
tri[i to Oregon, and while on the journey, he. in some unexplained
manner, attained the .sobri'Tiit, tut! Will yon let a inaii ilieT' He then
sent for a boat and Indians, and sent I'rown and Cox to Vancouver,
to the care of Dr. Uurkley, where he was taken care of for four
weeks, when he had siitiicieiitly recovered to take care of hiimielf.
Cox stayed with him two weeks. The Doctor's charges were o'lv $~0.
which he said the recovered man could jii'.v when he was al)le. Mr.
Sitton says that too much jiraise cannot be given Drs. ^[cLoiighlin
and Rarkley for their kindness to the iininigraiits. The young pioneers
had left their home in Missouri May 2, l84!i, and Mr. Sitton arrived at
(;)regon (!ity October 15 following. IHs comrades had been sent, as
stated, to Vancouver, and there he ..tood, a big, rjigged boy, bare to
his kuees, the legs of his trousers having been worn otf acainst the
low brush as he came over the mountains; his animals were lean
and worn out by the privati(uis of a long journey; and there he stood
leaning against a log, in this great and comparatively uninhabited
country, when old Mr. (Jertimin walked up to him and asked, " Do
you know Doc. Sitton?" Jle replied, "I am Sitton.'" -'Well, then,
C(Mne to my house and stay with me." Mr. Sitton thought it would
not he right to go without, telling hint that he had no money. The
reply to that was; "Never mind; none of us have any money here.
Your comrades have beeti here and have been sent back to
Vancouver, to the doctor, and they told us you were coming, and to
look after you." So they went to the house together; and as young
Sitton sat by the lire, n'ith his legs bare to his kuees, a stranger
asked : '■ Is that all the pants you have?" Sitton replied: ".My other
(dothes are at Vancouver.'' The man replied: " I will give you the
cloth for
 tllsTOIiV OF iiliKiluy. lj:i:i u jiiiir if yiin ciiii t;et tlit'iii iimdc;'"
iiiiil tlio lady ot tliL' house siiid: "I will make tlicm fur }'oii;" aiul \i\
iiiglit tlie next day lie had his new ti'uuser.-: on; and there was no
ehai'ge lor any ol' it! Thin kind-hearted hidy Miiil tohiui: •'My son,
you eim go into the jiantry whenever yuil like, and eat all the l>read
itrid meat you want. I crossed the plains last y(!»r, and I know how
hungry you are." This was in aiidition to his regular meals, and is
told to ^h()W tlu^ kind ot j)eo[)le the early pioneers of Orey-ou
wore. Doc. 8ittun could never forget these kindnesses, and they
lielped to nuike hiiu tlu> great generousliearteil man he is; and hri8s
i'riscilla Rogers, a native of Indiana, auddan;;literof Lewis Rogers, an
hotiored Oregon pioneer of lS4. They had a teakettle without a
cover, a Hudson's I'.ay frying-pan with a long handle; and her
parents gave lici' UHHiey with which to buy a coH'ee pot; and with
this stinted supply of hoiiseholil goods this worthy yiniii^ couple .-
tiu'leil out in life, ha]ipy iti each otlier'.^ love. In the autnnni of iSlK,
lured by the discovery of gold in ('alifornia, Mr. Sitton went to the
niiiics, and remained there from September UTitil i 'hristnias, on the
north fork of the .\mericaii /iver, iKMUg (piite Blie.c(^ssful; he and
two others tcirether took out six ounces id' ifold one day, and in thi^
course of live months Doc had accumulated .. lie has never sought
or held oflice, but has devoted his energies to his tine farm, which
has grown to be very valuable. During his fortynine years' residence
in Oregon, his honest and industrloii- life and noiihi generosity have
won for him hosts of friends, and every one who knows him speaks
well of Doc Sitton, tlie [lionecr of 1843. The children by his first wife
were: Charles E., I)i>rn July 0, 1848, iliod April 111, 1890; Amanda
Klle'n, born January 23, 1800, died in infancy; (Caroline E., born
October 21, ISol, and married L. C. Rogers; Ora Ann, born October
23, 1854, married John McCiillah, and dill April 4,1881; Ella \V., liorn
February Ifl, i.85S, died March 9 followini;; Harry W., born August
11, 1859; Noah H., April 29, 18(53; Fred D„ Fehriiay 2;i. lS(i5;
Eldridge D., September 23, IHCu. 11 :
 f ViHi iiiyruuY OH' (i/ih'uoy. .Iiitic ;J2, IHO'.I. tin- I'liitlil'iil
and lnvili^ iinil iiiihil^fi'Ml iiiotlicr (lictl, ami Mr iSittcm was Ik; rt'l'l of
the wife til' lii> vmitli, .mmI Icl't with ii lar^t,' yoiiMf^ laiiiily to imic
fur. On tlic last (lay of .lamiary, ISTl.lu iiiai'iicii Mih. Maiy 1-aiii.tiiliii.
llic daiij^fhtiT of Micliacl ami Suna Slifllcy, iionoriMl ( )rfj;oii iiioiieerti
of 1848. iMi'h. Siltuii had two I'liildrcii \mm ll.Ll.WI WAU-.VCF T1IA^
FK, ix(iovi-rnor of the Stiite of Orcffon, is a native of New York Stiite,
and was liorn m i.nii!i, l.iviiiirtiton cotinty, of that State, on Jnly 15.
1827. His father (iidcouThayur. was a nativt: of Uliode Ishind. His
[ifiternal ^'ran**=mm'-'^M^4 'IIOMAS THOMPSON, one of Umatilla
county's most successful young farmers, fp.v was born in Dundas
eonnty, Ontario, December ~il, 18()0. He was reared to farm life and
received his educal ion in tlie common .schools of his county. Of the
nine children born to James and Sarah Thompson, our subject is the
eldest. The father died in .878, but his wife still survives and resides
with her son in Oregon. Our subject came to the ITiiited States,
settled ill Fmatilla county, and as he was a poor boy, hired out for
wages, doing whatever came in his way. Finally he secured a
position as ■Hi
 iBiTTiTriiMiiiiiiriniiiiM iiisrour or n/.'/tv/o.v. i-.';i5 forcinaii «if
tlio (^niififriictiuM tniiiis. while tlic Ort'f^oii liivnr iV Niivijratioii
niilniHii \vii> liciiifi Imilt, coiitiiHiinij; witli tlic cciiiiiiiiiiy Im' two yeiu's,
(liiriiiii; wliicli tiiiitt lii^ liml siivcd (mkhi^Ii to purclmt-c ItlO ivcrt'K of
Imid, on wliieh lio iiiiiiiLMiiiitely iriovod. In adiJitioii, li(! iTiited oiieiiall'
K(n;tinii of laiiil, adjoiiiiiif; liis rariii, on whicli 111" I'iiist'd wlioat So .-
iico.cnsfii! was ho at tliis tliat lu' Wiis soon aldu lo add to his
possessions, and now owns 'd'M acres of j^ood t'armintt hind, on
whicii heaveniires aliont 0,000 liubhols of wlieat per year. After u ftnv
years Mr. Thompson einharkedon u new ventnre, that of horse
i)roedin}r and lias become an iinporltM' of (;iv-^>- ^MEDEP] M.
SMITH, the president and manager (jf the Oregon Pottery Company
of Portland, one of the largest manufacturing |)lants of the
Northwost, was born in Geneva, Fayette county, Pennsylvania,
August 8. 1H30. His father. Freeunm Sitiith, was a native of New
Jersey, a inechanic and shipbuilder, who, after settling in Penur-
vlvania, built the lirst sleimlioal whi(di ran niion the .Monongahida
river. His wife was Mai-j^aret llillerbrand, who wa^ a luitive of
Marjlaml. In 1842 tlui large family, coiif-isting of laiher, mother and
iiiiu^ children, of whom .\ni('(li'(^ was the yonngesl, emigrated to
Ceclar Raiiids, Iowa, where .Mr. Smilhengaged in farming and
buililing boats suitable for na\igating on tho Cedar river. The last one
of these, the lllack j Hawk, was used as a transportation boat on the
U])per Mississipjii during thgiment, lu^ was mustere
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com