ISE Data Analytics for Accounting 2nd

Edition Vernon Richardson Professor

Visit to download the full and correct content document:
https://ebookmeta.com/product/ise-data-analytics-for-accounting-2nd-edition-vernon-r
ichardson-professor/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

ISE Data Analytics for Accounting 3rd Edition Vernon

Richardson Professor

https://ebookmeta.com/product/ise-data-analytics-for-
accounting-3rd-edition-vernon-richardson-professor/

Data Analytics for Accounting, 3rd Edition Vernon J.

Richardson

https://ebookmeta.com/product/data-analytics-for-accounting-3rd-
edition-vernon-j-richardson/

Accounting Information Systems, 4e ISE Vernon J.

Richardson

https://ebookmeta.com/product/accounting-information-
systems-4e-ise-vernon-j-richardson/

Introduction to Managerial Accounting (ISE HED IRWIN

ACCOUNTING) 9th Edition Peter C. Brewer Professor

https://ebookmeta.com/product/introduction-to-managerial-
accounting-ise-hed-irwin-accounting-9th-edition-peter-c-brewer-
professor/
Financial Accounting, 11e ISE Robert Libby

https://ebookmeta.com/product/financial-accounting-11e-ise-
robert-libby/

Managerial Accounting, 17e ISE Ray H. Garrison

https://ebookmeta.com/product/managerial-accounting-17e-ise-ray-
h-garrison/

Data Analytics Models and Algorithms for Intelligent

Data Analysis 2nd Edition Thomas A Runkler

https://ebookmeta.com/product/data-analytics-models-and-
algorithms-for-intelligent-data-analysis-2nd-edition-thomas-a-
runkler/

Data-Enabled Analytics: DEA for Big Data Joe Zhu

https://ebookmeta.com/product/data-enabled-analytics-dea-for-big-
data-joe-zhu/

ISE Advanced Financial Accounting 13th Edition Theodore

E. Christensen

https://ebookmeta.com/product/ise-advanced-financial-
accounting-13th-edition-theodore-e-christensen/
 page i

Data Analytics for Accounting

SECOND EDITION

Vernon J. Richardson
University of Arkansas,
Baruch College

Ryan A. Teeter
University of Pittsburgh

Katie L. Terrell
University of Arkansas
 page ii

DATA ANALYTICS FOR ACCOUNTING

Published by McGraw-Hill Education, 2 Penn Plaza, New York, NY

10121. Copyright © 2021 by McGraw-Hill Education. All rights
reserved. Printed in the United States of America. Previous editions
© 2019. No part of this publication may be reproduced or distributed
in any form or by any means, or stored in a database or retrieval
system, without the prior written consent of McGraw-Hill Education,
including, but not limited to, in any network or other electronic
storage or transmission, or broadcast for distance learning.

Some ancillaries, including electronic and print components, may not

be available to customers outside the United States.

This book is printed on acid-free paper.

1 2 3 4 5 6 7 8 9 LWI 24 23 22 21 20

ISBN 978-1-260-57109-7
MHID 1-260-57109-2

Cover Image: Rawpixel.com/Shutterstock

All credits appearing on page or at the end of the book are
considered to be an extension of the copyright page.

The Internet addresses listed in the text were accurate at the time of
publication. The inclusion of a website does not indicate an
endorsement by the authors or McGraw-Hill Education, and McGraw-
Hill Education does not guarantee the accuracy of the information
presented at these sites.

mheducation.com/highered
 page iii

Dedications

My wonderful daughter, Melissa, for your

constant love, encouragement and support.
—Vern Richardson

My wife, Erin, and children, Sylvia and

Theodore.
—Ryan Teeter

To my co-author, friend, and colleague, Vernon

Richardson. Thank you for inviting me to be on
this textbook journey. And thank you for your
guidance and patience—I’m thrilled to be a
part of your team!
—Katie Terrell
 page iv

Preface
Data Analytics is changing the business world—data simply
surrounds us! So much data is available to businesses about each of
us—how we shop, what we read, what we buy, what music we listen
to, where we travel, whom we trust, where we invest our time and
money, etc. Accountants can create value by addressing
fundamental business and accounting questions using data analytics.
All accountants must develop data analytic skills to address the
needs of the profession in the future. Data Analytics for Accounting,
2e recognizes that accountants don’t need to become data scientists
—they may never need to build a data repository or do the real
hard-core Data Analytics or learn how to program a computer to do
machine learning. However, there are seven skills that analytic-
minded accountants must have to be prepared for a data-filled
world, including:
1. An analytics mindset—recognize when and how Data Analytics can
address accounting questions.
2. Data scrubbing and data preparation—comprehend the process
needed to extract (query), clean and prepare the data before
analysis.
3. Data quality—recognize what is meant by data quality, be it
completeness, reliability, or validity.
4. Descriptive data analysis—perform basic analysis to understand the
quality of the underlying data and their ability to address the
business question.
5. Data analysis through data manipulation—demonstrate ability to
sort, rearrange, merge, and reconfigure data in a manner that
allows enhanced analysis.
6. Problem solving through statistical data analysis—identify and
implement an approach that will use statistical data analysis to
draw conclusions and make recommendations on a timely basis.
7. Data visualization and data reporting—report results of analysis in
an accessible way to each varied decision maker and his or her
specific needs.
Consistent with these skills, it’s important to recognize that Data
Analytics is a process. The process begins by identifying business
questions that can be addressed with data, extracting and testing
the data, refining our testing, and finally, communicating those
findings to management. Data Analytics for Accounting, 2e describes
this process by relying on an established data analytics model called
the IMPACT cycle1
1. Identify the question.
2. Master the data.
3. Perform test plan.
4. Address and refine results.
5. Communicate insights.
6. Track outcomes

page v
Adapted from Win with Advanced Business Analytics: Creating
Business Value from Your Data, by Jean Paul Isson and Jesse S.
Harriott.

The IMPACT cycle is described in the first four chapters and then
the process is illustrated in audit, managerial accounting, financial
accounting and tax in Chapters 5-9, adding an all-new tax chapter to
Data Analytics for Accounting, 2e. In response to instructor
feedback, Data Analytics for Accounting, 2e now also includes two
new project chapters, giving students a chance to practice the full
IMPACT model with multiple labs that build on each other.
Data Analytics for Accounting, 2e emphasizes hands-on practice.
Students are provided with hands-on instruction (e.g., click-by-click
instructions, screenshots, etc.) on datasets within the chapter; within
the end-of-chapter materials; and in the labs at the end of each
chapter. Throughout the text, students identify questions, extract
and download data, perform testing, and then communicate the
results of that testing.
The use of real-world data is highlighted by using data from
LendingClub, College Scorecard, Dillard’s, the State of
Oklahoma, as well as other data from our labs. In particular, we
emphasize the rich data from Dillard’s sales transactions that we
use in more than fifteen of the labs throughout the text (including
Chapter 11).
Data Analytics for Accounting, 2e also emphasizes the various
data analysis tools students will use throughout the rest of their
career—Microsoft Excel, Microsoft Access (including SQL), Tableau
(free student license), IDEA (free student license), and Weka (free
student license). Using multiple tools allows students to learn which
tool is best suited for the necessary data analysis, data visualization,
and communication of the insights gained—for example, which tool
is easiest for internal controls testing, which is best for analysis or
querying (using SQL) big datasets, which is best for data
visualizations, and so on.

1Jean Paul Isson and Jesse S. Harriott, Win with Advanced Business Analytics:
Creating Business Value from Your Data (Hoboken, NJ: Wiley, 2013).
 page vi

About the Authors

Vernon J. Richardson

Vernon J. Richardson is a Distinguished Professor of Accounting

and the G. William Glezen Chair in the Sam M. Walton College of
Business at the University of Arkansas and a Visiting Professor at
Baruch College. He received his BS, Master of Accountancy, and MBA
from Brigham Young University and a PhD in accounting from the
University of Illinois at Urbana–Champaign. He has taught students
at the University of Arkansas, Baruch College, University of Illinois,
Brigham Young University, Aarhus University, and University of
Kansas and internationally at the China Europe International
Business School (Shanghai), Xi’an Jiaotong Liverpool University, and
the University of Technology Sydney.
Dr. Richardson is a member of the American Accounting
Association. He has served as president of the American Accounting
Association Information Systems section. He previously served as an
editor of The Accounting Review and is currently an editor at
Accounting Horizons. He has published articles in The Accounting
Review, Journal of Information Systems, Journal of Accounting and
Economics, Contemporary Accounting Research, MIS Quarterly,
International Journal of Accounting Information Systems, Journal of
Management Information Systems, Journal of Operations
Management, and Journal of Marketing. Dr. Richardson is also the
author of McGraw-Hill’s Accounting Information Systems textbook.

Ryan A. Teeter

Ryan A. Teeter is a Clinical Assistant Professor of Accounting in the

Katz Graduate School of Business at the University of Pittsburgh. He
teaches accounting information systems, auditing, and accounting
data analytics. Prior to receiving his PhD in accounting information
systems from Rutgers University, he worked at Google in Mountain
View, California. He has since worked with internal audit
organizations at Siemens, Procter & Gamble, Alcoa/Arconic, and
FedEx, helping to develop robotic process automation programs and
data analytic solutions.
Dr. Teeter is a member of the American Accounting Association
and has published articles in the Journal of Strategic Technologies in
Accounting and Issues in Accounting Education. He has received
grant funding for data analytics research from PwC.
Katie L. Terrell

Katie L. Terrell is an instructor in the Sam M. Walton College of

Business at the University of Arkansas. She received her BA degrees
in English literature and in the Spanish language from the University
of Central Arkansas and her MBA from the University of Arkansas.
She expects a doctoral degree by 2020. She has taught students at
the University of Arkansas; Soochow University (Suzhou, China); the
University College Dublin (Ireland); and Duoc UC, a branch of the
Catholic University of Chile (Vina del Mar, Chile).
She is a member of the American Accounting Association and has
published a Statement on Management Accounting for the Institute
of Management Accountants on managing organizational change in
operational change initiatives. Terrell was named the 2019 Business
Professional of the Year (Education) by the national Beta Alpha Psi
organization. She has recently been recognized for her innovative
teaching by being the recipient of the Mark Chain/FSA Teaching
Award for innovative graduate-level accounting teaching practices in
2016. She has worked with Tyson Foods, where she held various
information system roles, focusing on business analysis, project
management for ERP implementations and upgrades, and
organizational change management.
 page vii

Acknowledgments
Our sincere thanks to all who helped us on this project.
Our biggest thanks to the awesome team at McGraw-Hill
Education, including Steve Schuetz, Tim Vertovec, Allie Kukla, Fran
Simon, Kevin Moran, and Sarah Wood.
Our thanks also to each of the following:
The Walton College Enterprise Team (Paul Cronan, Ron Freeze,
Michael Gibbs, Michael Martz, Tanya Russell) for their work helping
us get access to the Dillard’s data.
Shane Lunceford from LendingClub for helping gain access to
LendingClub data.
Marcia Watson, University of North Carolina–Charlotte; Ryan
Baxter, Boise State University; Antoinette Smith, Florida International
University; and Lorrie Metzger, University at Buffalo for their
accuracy check and review of the manuscript.
In addition, the following reviewers and classroom testers who
provided ideas and insights for this edition. We appreciate their
contributions.

Amelia Annette Baldwin

University of South Alabama
Dereck Barr-Pulliam
University of Wisconsin–Madison
Heather Carrasco
Texas Tech University
Elizabeth Felski
State University of New York at Geneseo
Chris C. Hsu
York College, City University of New York
Venkataraman Iyer
University of North Carolina at Greensboro
Andrea S. Kelton
Middle Tennessee State University
Brandon Lock
Baruch College, CUNY
Sharon M. Lightner
National University
Margarita Maria Lenk
Colorado State University
Partha Mohapatra
California State University, Sacramento
Uday Murthy
University of South Florida
Kalana Malimage
University of Wisconsin–Whitewater
Bonnie Morris
Duquesne University
Kathy Nesper
University at Buffalo
Ali Saeedi
University of Minnesota Crookston
Karen Schuele
John Carroll University
Drew Sellers
Kent State University
Joe Shangguan
Robert Morris University
Vincent J. Shea
St. John’s University
Marcia Watson
University of North Carolina at Charlotte
Liu Yang
Southeast Missouri State University
Zhongxia Ye
University of Texas, San Antonio
Qiongyao (Yao) Zhang
Robert Morris University

Vernon Richardson
Ryan Teeter
Katie Terrell
 page viii

Key Features
Emphasis on Skills: Working through the IMPACT cycle
framework, students will learn problem assessment, data
preparation, data analysis, data visualization, control contesting,
and more.
Emphasis on Hands-On Practice: Students will be provided
hands-on learning (click-by-click instructions with screenshots) on
datasets within each chapter, within the end-of-chapter materials,
and in the labs and comprehensive cases.
Emphasis on Datasets: To illustrate data analysis techniques and
skills, multiple practice datasets (audit, financial, and managerial
data) will be used in every chapter. Students gain real-world
experience working with data from LendingClub, Dillard’s,
College Scorecard, the State of Oklahoma, as well as financial
statement data (via XBRL) from Fortune 100 companies.
Emphasis on Tools: Students will learn how to conduct data
analysis using Excel Access (including SQL), Tableau (free student
license), IDEA (free student license), and Weka (free student
license). Students will compare and contrast the different tools to
determine which are best suited for basic data analysis and data
visualization, which are easiest for internal controls testing, which
are best for SQL queries, and so on.
©Tableau Software, Inc. All rights reserved.
 page ix

Main Text Features

Chapter Maps
These maps provide a guide of what we’re going to cover in the
chapter as well as a guide of what we’ve just learned and what’s
coming next.

Chapter-Opening Vignettes
Because companies are facing the new and exciting opportunities
with their use of Data Analytics to help with accounting and business
decisions, we detail what they’re doing and why in our chapter-
opening vignettes.

Learning Objectives
We feature learning objectives at the beginning of each chapter.
Having these learning objectives provides students with an overview
of the concepts to be taught in the chapter and the labs.

Progress Checks
Periodic progress check questions are posed to the students
throughout each chapter. These checks provoke the student to stop
and consider the concepts presented.
 page x

End-of-Chapter Materials
Answers to Progress Checks
Allow students to evaluate if they are on track with their
understanding of the materials presented in the chapter.

Multiple Choice Questions

Quickly assess student’s knowledge of chapter content.
Discussion Questions
Provide questions for group discussion.

page xi

Problems
Challenge the student’s ability to see relationships in the learning
objectives by employing higher-level thinking and analytical skills.

Labs
Give students hands-on experience working with different types of
data and the tools used to analyze them. Students will conduct data
analysis using Excel, Access (including SQL), Tableau, IDEA, XBRL,
and Weka.

Comprehensive Cases
Use a real-life Big Data set based on Dillard’s actual company data.
This dataset allows students to build their skills and test their
conclusions across concepts covered in each chapter. The
Comprehensive Cases can be followed continuously from the first
chapter or picked up at any later point in the book; enough
information is provided to ensure students can get right to work.
 page xii

Data Analytics for Accounting,

2e Content Updates
General Updates for the 2nd Edition
Added additional End-of-Chapter Multiple Choice Questions and
Problems throughout the text.
Significantly revised many End-of-Chapter Problems for availability
and auto-grading within Connect.
Revised and added many new Discussion Questions in most
chapters.

Chapter by Chapter Updates

Specific chapter changes for Data Analytics for Accounting, 2nd
Edition, are as follows:

Chapter 1
Updated the opening vignette and statistics on Alibaba sales and
use of e-commerce.
Updated the statistics and screenshots for Lending Club Analysis.
Revised Connect questions for problems and labs.

Chapter 2
Improved and clarified the discussion of relational databases,
including updated figures.
Expanded the discussion of different RDBMS (Access, SQLite, and
SQL Server).
Improved discussion of Excel and SQL. The brief introduction to
how to use SQL now has its own place in a dedicated appendix at
the end of the text, and it has been vastly expanded to teach
beginners how to write queries.
Expanded the discussion on data quality.
Added a brief discussion of ETL v. ELT.
Improved labs for clarity and a better learning experience,
particularly Labs 2-1, 2-2, and 2-4.

Chapter 3
Reorganized chapter structure to follow the descriptive, diagnostic,
predictive, and prescriptive approaches to Data Analytics.
New exhibits and examples to illustrate analytics approaches.
Removed previous edition flowchart for model selection.
Additional explanation and examples of each of the methods and
approaches.
Improved labs for clarity.

Chapter 4
Updated the opening vignette.
Improved the discussion on the differences between qualitative and
quantitative data and the discussion of the normal distribution.
Improved and clarified how to select a visualization based on the
four chart types (qualitative vs. quantitative and declarative vs.
exploratory).
page xiii
Updated the discussion on the Gartner Quadrant to take
into account Gartner’s January 2019 analysis of BI tools (focusing
on Excel and Tableau).
Extended the discussion on written and spoken communication.
Added a lab to work with visualizing data and creating dashboards
in Power BI to interactively compare the tool with Tableau.

Chapter 5
Expanded discussion on the modern data environment.
Included additional examples of the Audit Data Standard.
Improved and clarified content to match the focus on descriptive,
diagnostic, predictive, and prescriptive analytics.
New labs (5-1 and 5-2) that have students transform data using a
common data model.
Improved existing labs.

Chapter 6
Clarified chapter content to match the focus on descriptive,
diagnostic, predictive, and prescriptive analytics.
Improved labs.

Chapter 7
Clarified chapter content and provided additional new exhibits and
examples, such as variance analysis.
Improved labs.

Chapter 8
Reorganized chapter content to focus on financial statement
analysis using descriptive, diagnostic, predictive, and prescriptive
approaches.
Added new content on common size and ratio analysis.
Improved discussion of XBRL data.
 Improved XBRL dataset (in Lab 8-4), accessible via Microsoft
Access and included options to do analysis in Excel.

Chapter 9
All-new chapter on tax analytics, including examples of tax data,
tax analysis, tax planning, and tax visualizations.

Chapter 10
All-new basic project chapter that explores the order-to-cash and
procure-to-pay cycles from different user perspectives.

Chapter 11
All-new advanced project chapter, estimating sales returns at
Dillard’s with three question sets highlighting descriptive and
exploratory analysis, hypothesis testing, and predictive analytics.

page xiv

Appendixes
Several all-new appendixes have been added to ease the lab
experience and introduce tools used or mentioned throughout the
text:
Appendix A: Basic Statistics Tutorial.
Appendix B: Accessing the Excel Data Analysis Toolpak.
Appendix C: Excel (Formatting, Sorting, Filtering, and PivotTables).
Appendix D: SQL Part 1. This tutorial introduces the SQL language
for extracting data and explains the following SQL syntax: SELECT,
FROM, INNER JOIN, ON, WHERE, GROUP BY, HAVING, ORDER BY.
Appendix E: SQLite. We have added SQLite files as an option for
each lab that uses Microsoft Access. This lab explains how to
download SQLite and how to use the tool.
Appendix F: Power Query. This appendix contains a short tutorial
on transforming data using Power Query. How to access data files
on the University of Arkansas’ remote desktop is also discussed.
Appendix G: Tableau.
Appendix H: SQL Part 2: On the heels of learning Tableau, students
learn about more complex joins—LEFT and RIGHT.
Appendix I: Power BI.
Appendix J: Dillard’s ER Diagram.
Appendix K: Data Dictionaries.
Another random document with
no related content on Scribd:
is: Why this shift occurs? Surely it is more rational to adopt a theory
which does not require such a supposition? If the sensitive matter
acted upon by the yellow-blue rays be always of the same chemical
composition, the shift cannot occur. It might, perhaps, be allowed
that one shift was practicable, but, unfortunately, the shifts must
become numerous when the cases of partial colour blindness are to
be accounted for, and this would necessitate a constantly varying
chemical composition of this matter, and of that acted upon by the
red-green rays.
Again, in the extinction of the spectrum, the red and the green
sensations in quantities to neutralize one another should be
extinguished nearly together, even allowing for what physiologists tell
us is the case, that the breaking down, or dissimulation, of cell tissue
continues longer than its building up, but we find a large difference
between the two. As already indicated, the luminosity curve of the
feeble spectrum favours the theory of Hering being that here we only
have the white-black sensation, and naturally the persistency curves
must be scored in its favour. But the cases of B. C. and M., it seems
to me, cannot be explained by the theory without any undue straining
or assumptions. If we try and fit the cases of colour blindness due to
tobacco scotoma to the theory, we find that in many cases yellow is
not recognised, though blue is invariably. If the blue be active, the
yellow should also be so.
And here I may remark that it has been assumed that the two
classes of colour blindness are due to different causes. A question to
ask ourselves is whether all colour blindness may not have been
caused originally by disease. In the congenital form, it is true, no
disease of the retina is traceable in the eye, and it is usually
hereditary, but it does not follow that the want of response of the
perceiving apparatus to certain sensations may not have been due
to what, for want of a better expression, I may call an hereditary
partial paralysis of the perceiving apparatus. If this be so, we have a
connecting link between the two classes, and then a perfect theory
should explain both classes on the same grounds. The suspicion
that the monochromatic vision of P. and Q. might possibly be due to
disease before birth, owing to the behaviour of their eyes under
certain conditions, would then be explicable. I have no desire to
press this view, though it seems to me to be one which is not out of
all reason, taking analogies from other defects which are hereditary.
It has been usually accepted that the fields for blue and yellow in
the eye are approximately the same, as are those of the green and
red, and this has been taken as showing the interdependence
between the two pairs according to the Hering theory. It has already
been pointed out that the question of extent of fields requires still
further investigation beyond that which it has received, and
measures made by the method given on page 208 seem to cast a
doubt as to whether this interdependence can be upheld. It will be
noticed that the fields do not extend proportionately on the nasal and
temporal sides (see also Fig. 3). It should also be remarked that the
order of extent of field for the different colours does not follow the
same order as their disappearance. A point that is sometimes raised
in favour of Hering’s theory is the negative image formed after the
eye is fatigued by looking at bright red or bright green. The negative
images (see page 30) are said to be the complementary of these
colours. The Young theory tells us that the red or the green
sensation suffers fatigue by one or other colour, and that when the
eye subsequently rests on a grey surface the other two sensations
are chiefly stimulated and cause the complementary colour. It is said
that it is easier to produce a negative green image than a negative
red image, and the adherents of Hering tell us that this is due to the
fact that destructive action is more readily carried out than
constructive. In the Young theory, it is held that the green sensation
is always mixed with white, whilst the red is fairly pure, and thus, for
equal luminosities, the surplus green sensation is much less
stimulated than the red, which offers a consistent explanation of this
fact. There are several other minor difficulties in the way of accepting
Hering’s theory as it stands from a physical point of view, but we
need not discuss them now.
The final sensation curves for the spectrum colours on the Young
theory are still under consideration, and are not definitely fixed,
though the observations made have been very numerous. Recently
Helmholtz, in the last edition of his “Physiological Optics,” has
calculated, from Kœnig’s observations, that no one of the three
sensations is singly stimulated by any colour, even at the extreme
ends of the spectrum, and he makes the three fundamental
sensations vary considerably from those given in these pages. Every
colour he states is considerably mixed with white light. The
calculations by which he arrived at this conclusion are of a
complicated nature, and I think if he had had besides the colour
equations of Kœnig, the luminosities and the extinction measures
before him, there might have been a modification of his views, for
these last give evidence to the contrary.
There is a possible modification of the Young theory which would
account for a good many of the phenomena that are unaccounted for
by it in its present form, though it may raise new difficulties in the
minds of some. Let us suppose that each of the three sensations
were compounded of fundamental light and of colour in fixed and
definite proportions, and not in the same proportion in each; and
further that the apparatus in the eye which was responsible for each
sensation had two functions, one of which was to respond to the
fundamental light sensation and the other to the colour. One
essential difference between this modification of the Young theory
and that of Hering is that, whilst in the latter the white sensation is a
sensation distinct from the colour sensations, in the former it is a
definite part of them. The fact that the sensation of colour is lost
before the sensation of light is one of the greatest significance, and
any theory to be accepted must offer a reasonable explanation of it.
If the modification suggested be made, it accounts for the existence
of this residuum of light equally as well as Hering’s theory, and
without its drawback. It is not hard to imagine the apparatus which
gives rise to two sensations, on the assumption of different kinds of
atomic motion, induced by the ether motion, or at least three kinds
are possible. When extinction of colour is made, the ether vibrations
would have sufficient energy to induce but one kind of motion; and
when all light was extinguished from the same ray, they would not be
capable of inducing any sensible motion whatever. In the case of
Miss W., who saw all colours as white, it might be that disease had
entirely prevented the first kind of motion in all three sensations, and
that in P. and Q. the red and green sensations were absent or
paralysed in their entirety, whilst the blue sensation was left in full
operation. In B. C. the blue and red sensations would be similarly
absent, leaving the green sensation unchanged. The coincidence of
their persistency and luminosity curves would then indicate that the
proportions of fundamental light and colour remained the same
throughout. Other examples and considerations seem to indicate
that the proportion of colour to fundamental light is greatest in the
red sensation, next in the green, and least in the blue. This would
explain why with increasing intensities blue appears white sooner
than green, and much sooner than red. The proposed modification
would also offer the necessary explanation as to the disappearance
of colour from the field.
Looking at colour vision from what I may call an evolutionary
point of view, the “light-colour” theory commends itself as probable.
There are many reasons for thinking that the visual sensation first
evolved was that of light, subsequently followed by that of colour.
The first evolved colour sensation would appear to have been the
blue, and the last the red. The discussion of this hypothesis would
carry me beyond my limits, and I must leave it thus baldly expressed
for your consideration.
For my own part, whatever theory of colour sensations may
prove to be the right one, I lean strongly to the idea that the cause of
vision will be found in chemical action, induced by the impact of the
different wave-lengths of light falling on sensitive matter. A white
substance may absorb all the wave-lengths found in the spectrum,
and if it have three sets of molecules, one of which has an atom or
atoms vibrating with the same period as the waves of light which
show a maximum for one sensation and another for another, and so
on, the requirements for the colour sensations are met. It may be
that the sensitive part of the retina is like a photographic plate, but
with this essential difference—that the sensitive material is
constantly changing. A photographic plate receives an impression
which is not recognisable by the eye, though it can be shown that a
change in the material does take place during the impact of light, by
electrical and other means. When the eye receives an impression of
light, Dewar has shown that in this case also a current of electricity is
generated. Recent published experiments of my own have
demonstrated that with a low intensity of light, the chemical change
that occurs in a photographic salt is by no means proportionate to
that which takes place with a greater intensity. In the eye, too, there
is a limit of sensibility to very feeble light. Again, the curves of the
stimulation of the colour sensations to the spectrum are closely of
the same form as the curves of sensitiveness of the various sensitive
salts used by photographers. These are analogies and, of course,
must not be pressed too far. There must be such a complexity in the
sensitive material in the eye, both chemical and physiological, that it
may be that the changes induced by light on the sensitive surface of
the retina have to be considered from both aspects. The purely
chemical change is naturally that to which a physicist is most prone
to incline, and his bias must be discounted, as must also that of the
physiologist.
 APPENDIX.
The following is extracted from Maxwell’s paper.
The following table contains the means of four sets of
observations by the same observer (K.):—

Table IV. (K.)

44·3 (20) + 31·0 (44) + 27·7 (68) = W.
16·1 (28) + 25·6 (44) + 30·6 (68) = W.
22·0 (32) + 12·1 (44) + 30·6 (68) = W.
6·4 (24) + 25·2 (36) + 31·3 (68) = W.
15·3 (24) + 26·0 (40) + 30·7 (68) = W.
19·8 (24) + 35·0 (46) + 30·2 (68) = W.
21·2 (24) + 41·4 (48) + 27·0 (68) = W.
22·0 (24) + 62·0 (52) + 13·0 (68) = W.
21·7 (24) + 10·4 (44) + 61·7 (56) = W.
20·5 (24) + 23·7 (44) + 40·5 (60) = W.
19·7 (24) + 30·3 (44) + 33·7 (64) = W.
18·0 (24) + 31·2 (44) + 32·3 (72) = W.
17·5 (24) + 30·7 (44) + 44·0 (76) = W.
18·3 (24) + 33·2 (44) + 63·7 (80) = W.

X.—Reduction of the Observations.

By eliminating W from the equations above by means of the
standard equation, we obtain equations involving each of the
fourteen selected colours of the spectrum, along with the three
standard colours; and by transposing the selected colour to one side
of the equation, we obtain its value in terms of the three standards. If
any of the terms of these equations are negative, the equation has
no physical interpretation as it stands; but by transposing the
negative term to the other side it becomes positive, and then the
equation may be verified.
The following table contains the values of the fourteen selected
tints in terms of the standards. To avoid repetition, the symbols of the
standard colours are placed at the head of each column:—

Table VI.

Observer (K.) (24) (44) (68)

44·3 (20) = 18·6 + 0·4 + 2·8
16·1 (28) = 18·6 + 5·8 - 0·1
22·0 (32) = 18·6 + 19·3 - 0·1
25·2 (36) = 12·2 + 31·4 - 0·8
26·0 (40) = 3·3 + 31·4 - 0·2
35·0 (46) = - 1·2 + 31·4 + 0·3
41·4 (48) = - 2·6 + 31·4 + 3·5
62·0 (52) = - 3·4 + 31·4 + 17·5
61·7 (56) = - 3·1 + 21·0 + 30·5
40·5 (60) = - 1·9 + 7·7 + 30·5
33·7 (64) = - 1·1 + 1·1 + 30·5
32·3 (72) = + 0·6 + 0·2 + 30·5
44·0 (76) = + 1·1 + 0·7 + 30·5
63·7 (80) = + 0·3 - 1·8 + 30·5

Mr. James Simpson, formerly student of Natural Philosophy in

my class, has furnished me with thirty-three observations taken in
good sunlight. Ten of these were between the two standard colours,
and give the following result:—
33·7 (88) + 33·1 (68) = W.
The mean errors of these observations were as follows:—
Error of (88) = 2·5; of (68) = 2·3; of (88) + (68) = 4·8; of (88) - (68) =
1·3.
The fact that the mean error of the sum was so much greater
than the mean error of the difference, indicates that in this case, as
in all others that I have examined, observations of equality of tint can
be depended on much more than observations of equality of
illumination or brightness.
From six observations of my own, made at the same time, I have
deduced the “trichromic” equation—
22·6 (104) + 26 (88) + 37·4 (68) = W (2)
If we suppose that the light which reached the organ of vision
was the same in both cases, we may combine these equations by
subtraction, and so find
22·6 (104) - 7·7 (88) + 4·3 (68) = D (3)
where D is that colour, the absence of the sensation of which
constitutes the defect of the dichromic eye.
The sensation which I have in addition to those of the dichromic
eye is therefore similar to the full red (104), but different from it in
that the red (104) has 7·7 of green (88) in it which must be removed,
and 4·3 of blue (68) substituted. This agrees pretty well with the
A
colour which Mr. Pole describes as neutral to him, though crimson
to others. It must be remembered, however, that different persons of
ordinary vision require different proportions of the standard colours,
probably owing to differences in the absorptive powers of the media
of the eye, and that the above equation (2), if observed by K., would
have been
23 (104) + 32 (88) + 31 (68) = W (4)
and the value of D, as deduced from these observers, would have
been
23 (104) - 1·7 (88) - 1·1 (68) = D (5)
in which the defective sensation is much nearer to the red of the
spectrum. It is probably a colour to which the extreme red of the
spectrum tends, and which differs from the extreme red only in not
containing that small proportion of “yellow” light which renders it
visible to the colour blind.

A
Philosophical Transactions, 1859, Part I., p. 329.

From other observations by Mr. Simpson the following results

have been deduced:—

Table A.

(88) (68) (88) (68)

(99·2 +) = 33·7 1·9 100 (96) = 108 7
31·3 (96) = 33·7 2·1 100 (92) = 120 5
28 (92) = 33·7 1·4 100 (88) = 100 0
33·7 (88) = 33·7 0 100 (84) = 61 11
54·7 (84) = 33·7 6·1 100 (82) = 47 21
71 (82) = 33·7 15·1 100 (80) = 34 33
99 (80) = 33·7 33·1 100 (78) = 22 47
70 (78) = 15·7 33·1 100 (76) = 10 59
56 (76) = 5·7 33·1 100 (72) = 1 92
36 (72) = 0·3 33·1 100 (68) = 0 100
33·1 (68) = 0 33·1 100 (64) = 0 83
40 (64) = 0·2 33·1 100 (60) = 3 60
55·5 (60) = 1·7 33·1
(57 -) = 0·3 33·1

In the table on the left side (99·2 +) means the whole of the
spectrum beyond (99·2) on the scale, and (57 -) means the whole
beyond (57) on the scale. The position of the fixed lines with
reference to the scale was as follows:—
A, 116; a, 112; B, 110; C, 106; D, 98·3; E, 88; F, 79; G, 61; H, 44.
 The values of the standard colours in different parts of the
spectrum are given on the right side of the above table, and are
represented by the curves of Fig. 9, Plate II., where the left-hand
curve represents the intensity of the “yellow” element, and the right-
hand curve that of the “blue” element of colour as it appears to the
colour blind.
The appearance of the spectrum to the colour blind is as
follows:—
From A to E the colour is pure “yellow,” very faint up to D, and
reaching a maximum between D and E. From E to one-third beyond
F towards G the colour is mixed, varying from “yellow” to “blue,” and
becoming neutral or “white” at a point near F. In this part of the
spectrum the total intensity, as given by the dotted line, is decidedly
less than on either side of it, and near the line F, the retina close to
the “yellow spot” is less sensible to light than the parts further from
the axis of the eye. This peculiarity of the light near F is even more
marked in the colour blind than in the ordinary eye. Beyond F the
“blue” element comes to a maximum between F and G, and then
diminishes towards H, the spectrum from this maximum to the end
being pure “blue.”
The results given above were all obtained with the light of white
paper, placed in clear sunshine. I have obtained similar results when
the sun was hidden, by using the light of uniformly illuminated
clouds, but I do not consider these observations sufficiently free from
disturbing circumstances to be employed in calculation. It is easy,
however, by means of such observations, to verify the most
remarkable phenomena of colour blindness, as, for instance, that the
colours from red to green appear to differ only in brightness, and that
the brightness may be made identical by changing the width of the
slit; that the colour near F is a neutral tint, and that the eye in viewing
it sees a dark spot in the direction of the axis of vision; that the
colours beyond are all blue of different intensities, and that any
“blue” may be combined with any “yellow” in such proportions as to
form “white.” These results I have verified by the observations of
another colour-blind gentleman, who did not obtain sunlight for his
observations; and as I have now the means of carrying the requisite
apparatus easily, I hope to meet with other colour-blind observers,
and to obtain their observations under more favourable
circumstances.

Measurements of Colour Fields.

Some experiments in the measurement of the colour fields in the
horizontal direction with the pure spectrum colours will help to show
what importance is to be attached to the luminosity of the colour and
the size of the spot of light with which the observations are made. A
yellow and a blue of the spectrum were taken of such hues that
when mixed they formed a patch of white light similar to the electric
light. Their luminosities were measured, and the yellow found to be
1·6 of the light of an amyl-acetate lamp or 1·28 standard candles;
the blue was 1/24 of this luminosity. The fields for these two colours
were measured by automatically throwing spots of each colour
separately on a white card which moved round a centre over which
the eye was placed. The light was subsequently diminished to ½, ¼,
and ⅛ of the above values, and readings again made. The following
results were obtained with a spot of ·7 inch diameter:—

Light
Yellow. Blue.
Nasal side. Temporal side. Nasal side. Temporal side.
Full 33° 45° 35° 45°
½ 24° 36° 26° 38°
¼ 18° 24° 22° 32°
⅛ 11° 15° 19° 30°

With a spot of ·3 inch diameter the following were obtained:—

Full 24° 32° 21° 27°

½ 17° 28° 16° 22°
¼ 13° 16° 14° 20°
⅛ 8° 10° 13° 16°
 It will be evident how the field contracts as the light is diminished
in brightness, and also that the blue field does not diminish equally
with the yellow field, but is more persistent. Again, it will be noticed
that the luminosity of the blue, for the same extent of field to be
covered, has to be much lower than for the yellow.
The diminished area of the spot of light also diminishes the field,
and the same order of diminution of field is obtained as with the
larger spot.
Another set of experiments, made with the same aperture of slit
passed through the spectrum, and the field taken at different points,
give the following results:—

Spectrum scale.
Nasal side. Temporal side.
(See Fig. 41, Page 210.
58·6 18° 35°
54·6 27° 46°
50·6 33° 47°
46·6 25° 30°
42·6 21° 21°
38·6 17° 17°
34·6 22° 30°
30·6 25° 33°
26·6 33° 40°
22·6 37° 44°
18·6 28° 40°
14·6 22° 34°
8·6 20° 30°

Here we see that although the luminosity of the colour spots

varies at the spectrum luminosity, the fields do not vary
proportionally; when the luminosities of the green, yellow and red are
made equal, the fields become nearly equal on the nasal side. The
field for the blue, however, then becomes vastly larger than that for
the others, showing a peculiarity which is very remarkable.
 Fig. 41.

Spectrum Scale.

Recently published experiments on colour fields have been so

largely based on the exigencies of the Hering theory, that it is
somewhat difficult to decide their significance from any other aspect.

Table I.—Luminosity Curves for the Normal Eye (see Fig. 20).

I. II. III. IV. V.

Scale Wave- Outside Yellow Fovea
number. length. yellow spot. spot. centralis.
64 7217
63 7082 .. 1
62 6957 1 2 2
61 6839 2 4 4
60 6728 3·5 7 8
59 6621 7·5 12·5 15·5
58 6520 12·5 21 24
57 6423 19 33 37·5
56 6330 27·5 50 60
55 6242 35 65 77
54 6152 43 80 90
53 6074 52·5 90 97
52 5996 61·0 96 100
51 5919 71·0 99 100
50 5850 79·0 100 98
49 5783 84 99 95
48 5720 85 97 90
47 5658 83·5 92·5 85
46 5596 81·0 87 79
45 5538 77·0 81 72·5
44 5481 72.5 75 66
43 5427 68·0 69 59
42 5373 62·5 62·5 51
41 5321 57 57 45
40 5270 52 50 40
39 5221 46 42·5 32
38 5172 41·5 36 27·5
37 5128 37·5 29·5 22·0
36 5085 33·5 24 18
35 5043 30·0 18·2 14
34 5002 26·5 14·2 10
33 4963 24 10·5 8·4
32 4924 21 8·5 6·5
31 4885 18·5 7·0 5·5
30 4848 16·5 5·5 4·0
29 4812 14·5 4·7 3·5
28 4776 13·0 4·0 3·0
27 4742 11·5 3·5 2·0
26 4707 10·5 2·8 2·4
25 4675 9·4 2·3 2·1
24 4639 8·2 1·82 1·9
23 4608 7·3 1·6 1·5
22 4578 6·3 1·4
21 4548 5·7 1·2
 20 4517 5·0 1·08 1·0
19 4488 4·5 ·94
18 4459 4·0 ·86
17 4437 3·6 ·78
16 4404 3·1 ·70
15 4377 2·7 ·62 ·62
14 4349 2·3 ·56
13 4323 2·1 ·50
12 4296 1·9 ·45
11 4271 1·65 ·40
10 4245 1·4 ·34
9 4221 1·2 ·30
8 4197 1·0 ·26
7 4174 ·88 ·22
6 4151 ·75 ·18
5 4131 ·63 ·16
4 4106 ·50 ·14

Tables II. and III.—Curves of Luminosity of a Partially Red-

Blind and of a Partially Green-Blind Person (see Fig. 23).

Scale Wave- Luminosity.

number. length. Red-blind. Green-blind.
64 7217 0 1
62 6957 1 2
60 6728 2 7
58 6520 6 21
56 6330 12 50
54 6152 26 80
52 5996 49 96
50 5850 70 98
48 5720 77 93
46 5596 77 83
44 5481 70 70
 42 5373 61 55
40 5270 47 40
38 5172 34 27
36 5085 23 18
34 5002 14 10
32 4924 8·5 5·5
30 4848 5·5 3·0
28 4776 4·0 2·5
26 4707 2·7 2·0
24 4639 1·8 1·8
22 4578 1·35 1·4
20 4517 1·1 1·1

Table IV.—Luminosity of Spectrum Reduced in Intensity, so

that D = 1/132·5 Amyl Lamp 1 Foot Distant (see Fig. 25).

Persistency
Mean reading, P. and Q.’s
Scale Mean curve for the
reduced to 100 readings, 100
number. reading. centre of the
max. max.
eye.
55·6 ·5 ·6 2 2
53·6 5·5 7·0 3·6 3·6
51·6 13 16·7 8 8
49·6 23 29·7 22 22
47·6 40 50·0 44 44
45·6 57 71·2 69 69
43·6 70 87·5 93 93
41·6 79 98·7 100 99·5
39·6 78 97·5 99·5 98·5
37·6 74 92·5 96 93
35·6 66 82·5 89 84
33·6 55 68·7 77·5 71
31·6 44·5 55·2 61 53·5
29·6 35 43·7 45·5 36·5
 27·6 24 30·0 33·5 24
25·6 17 21·7 25 16
23·6 13 16·7 18 10
21·6 10 12·5 13 8
19·6 8 10·0 9·5 6
13·6 3 3·7 4·2 3
9·6 2 2·5 2·5 2

Table V.—Limit of Colour Vision (see Fig. 26.).

Luminosity of the
rays when each
Mean reading of
colour
the colour limit of
Luminosity of the disappears, each
Scale Wave- the spectrum D,
ordinary ray having the
Number. Length. being 1 amyl
spectrum. original
lamp in
luminosity of 1
1/10000ths.
amyl lamp in
1/100000ths.
61 6839 120 4 48·0
60 6728 67 7 46·9
58 6520 26 21 54·6
56 6330 13 50 65·0
54 6152 9·5 80 76·0
52 5996 9·0 96 86·4
50 5850 9·0 100 90·0
48 5720 9·0 97 87·3
44 5481 9·5 75 71·3
40 5270 10·5 50 52·5
36 5085 12·5 24 30·0
32 4924 18 8·5 15·3
28 4776 32 4·0 12·8
24 4639 55 1·8 12·0
20 4517 90 1·08 9·7
16 4404 160 ·70 11·2
 12 4296 250 ·45 11·0
8 4197 400 ·26 10·4
4 4106 700 ·14 9·8

Table VI.—Extinction by Central Portion of Normal Eye (see

Fig. 28).

I. II. III. IV. V. VI.

E.
Reduction
L.
of original Persistency curve
Scale Wave- Luminosity
luminosity in (E × L) / 100 650 / E
number. length. of original
millionths to (Maximum = 100).
beam.
cause
extinction.
64 7217 55,000
63 7082 30,000 1 300·0
62 7957 15,000 2 300·0
61 6839 7500 4 300·0
60 6728 3750 7 262·5
59 6621 1900 12·5 237·5 ·34
58 6520 1050 21 220·5 ·62
57 6423 650 33 214·5 1·0
56 6333 380 50 190·0 1·71
55 6242 272 65 176·8 2·38
54 6152 196 80 156·0 3·32
53 6074 140 90 126·0 4·64
52 5996 97 96 93·12 6·70
51 5919 57 99 56·43 11·40
50 5850 35 100 35·0 18·6
49 5783 24 99 23·76 27·1
48 5720 17 97 16·49 38·2
47 5658 12·6 92·5 11·65 51·6
46 5596 10·2 87 8·87 63·7
45 5538 8·6 81 6·97 75·6
44 5481 7·4 75 5·55 87·8
43 5427 6·7 69 4·62 97·0
42 5373 6·55 62·5 4·09 99·5
41 5321 6·5 57 3·705 100
40 5270 6·55 50 3·27 98·5
39 5221 6·65 42·5 2·83 97·5
38 5172 6·85 36 2·46 95·0
37 5128 7·2 29·5 2·12 90·0
36 5085 7·6 24 1·82 81·3
35 5043 8·15 18·2 1·48 80·0
34 5002 8·8 14·2 1·25 74·0
33 4963 10·2 10·5 1·07 63·0
32 4924 11·6 8·5 ·988 56·0
31 4885 13·6 7·0 ·952 47·7
30 4848 16·3 5·5 ·896 40·0
29 4812 20·5 4·7 ·963 31·7
28 4776 26·0 4·0 1·040 25·0
27 4742 31·0 3·5 1·085 20·9
26 4707 38·5 2·8 1·078 16·9
25 4674 46·0 2·3 1·058 14·1
24 4639 56·0 1·82 1·019 11·6
23 4608 67·0 1·6 1·072 9·7
22 4578 80 1·4 1·120 8·41
21 4548 95 1·2 1·140 7·22
20 4517 107 1·08 1·156 6·1
19 4488 124 ·94 1·165 5·23
18 4459 140 ·86 1·204 4·64
17 4437 160 ·78 1·228 4·1
16 4404 180 ·70 1·260 3·60
15 4377 200 ·62 1·240 3·25
14 4349 220 ·56 1·232 2·95
13 4323 240 ·50 1·200 2·7
12 4296 270 ·45 1·215 2·4
11 4271 300 ·40 1·200 2·18
10 4245 335 ·34 1·139 1·94
9 4221 375 ·30 1·125 1·73
8 4197 430 ·26 1·118 1·51
7 4174 490 ·22 1·078 1·32
6 4151 510 ·18 ·918 1·27
5 4131 640 ·16 1·024 1·01
4 4106 750 ·14 1·050 0·86

Table VII.—Extinction by Whole Eye (see Fig. 28).

I. II. III. IV. V. VI.

E.
Reduction
L.
of original Persistency curve
Scale Wave- Luminosity (E × L)
luminosity in 650 / E
number. length. of original / 160
millionths to (Maximum = 100).
beam.
cause
extinction.
38 5172 6·9 41·5 2·86 94·2
37 5128 7·1 37·5 2·66 91·6
36 5085 7·4 33·5 2·48 87·8
35 5043 7·7 30·0 2·31 84·4
34 5002 8·0 26·5 2·12 81·2
33 4963 8·4 24·0 2·02 77·5
32 4924 8·8 21·0 1·85 73·8
31 4885 9·4 18·5 1·74 69·2
30 4848 10·0 16·5 1·65 65·0
29 4812 10·7 14·5 1·55 60·6
28 4776 11·5 13·0 1·49 56·5
27 4742 13·0 11·5 1·49 50·0
26 4707 14·5 10·5 1·52 44·8
24 4639 18·5 8·2 1·52 34·1
22 4578 23·0 6·3 1·45 28·3
20 4517 30·0 5·0 1·50 21·7
18 4459 39·0 4·0 1·56 16·7