Full Chapter PDQ Statistics PDQ Series 3Rd Edition Norman Geoffrey R Streiner David L PDF
Full Chapter PDQ Statistics PDQ Series 3Rd Edition Norman Geoffrey R Streiner David L PDF
Full Chapter PDQ Statistics PDQ Series 3Rd Edition Norman Geoffrey R Streiner David L PDF
https://textbookfull.com/product/statistics-for-business-
economics-david-r-anderson/
https://textbookfull.com/product/statistics-for-business-
economics-david-r-anderson-2/
https://textbookfull.com/product/statistics-for-management-and-
economics-11th-edition-g-e-r-a-l-d-k-e-l-l-e-r/
https://textbookfull.com/product/modern-mathematical-statistics-
with-applications-3rd-edition-jay-l-devore/
Essentials of Statistics for Business & Economics 9th
Edition David R. Anderson
https://textbookfull.com/product/essentials-of-statistics-for-
business-economics-9th-edition-david-r-anderson/
https://textbookfull.com/product/an-introduction-to-categorical-
data-analysis-3rd-edition-wiley-series-in-probability-and-
statistics-agresti/
https://textbookfull.com/product/statistics-for-business-
economics-with-xlstat-education-edition-printed-access-card-
david-r-anderson/
https://textbookfull.com/product/matrix-differential-calculus-
with-applications-in-statistics-and-econometrics-3rd-edition-jan-
r-magnus/
https://textbookfull.com/product/hazardous-materials-
chemistry-3rd-edition-armando-toby-bevelacqua-laurie-a-norman/
PD
PrettyDarnedQuick
^
Statistics
Third Edition
Geoffrey R . Norman
David L . Streiner
PDQ* SERIES
ACKERMANN
PDQ PHYSIOLOGY
BAKER, MURRAY
PDQ BIOCHEMISTRY
CASSILETH et al.
CORMACK
PDQ HISTOLOGY
DAVIDSON
INGLE
KERN
PDQ HEMATOLOGY
McKIBBON, WILCZYNSKI
NORMAN, STREINER
STREINER, NORMAN
SCHLAGENHAUF-LAWLOR, FUNK-BAUMANN
SCIUBBA
Third Edition
DAVID L. STREINER
Professor of Psychiatry
University of Toronto
Toronto, Ontario
2003
PMPH USA
NEW HAVEN, CONNECTICUT
CARY, NORTH CAROLINA
PMPHUSA
Publishing to Advance
the Practice of Medicine
All rights reserved. Without limiting the rights under copyright reserved above, no part of this
publication may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any
form or by any means (electronic, mechanical, photocopying, recording, or otherwise), without the
prior written permission of the publisher.
20 21 22 23 24/KING/10 9 8
ISBN-13 978-1-60795-286-2
ISBN-10 1-60795-286-6
e-ISBN 978-1-60795-077-6
Canada
Login Canada
300 Saulteaux Cr.
Winnipeg, MB R3J 3T2
Phone: 1.800.665.1148
Fax: 1.800.665.0103
www.lb.ca
Foreign Rights
John Scott & Company
International Publisher’s Agency
P.O. Box 878
Kimberton, PA 19442, USA
Tel: 610-827-1640
Fax: 610-827-1671
[email protected]
United Kingdom, Europe, Middle East, Africa, Singapore, Thailand, Philippines, Indonesia, Vietnam,
Pacific Rim, Korea, Australia, New Zealand, Papua New Guinea, Fiji, Tonga, Solomon Islands, Cook
Islands, Malaysia
Eurospan Limited
Gray's Inn House
127 Clerkenwell Road
London EC1R 5DB, UK
Tel.: +44(0)20 7240 0856
www.eurospanbookstore.com
Notice: The authors and publisher have made every effort to ensure that the patient care
recommended herein, including choice of drugs and drug dosages, is in accord with the accepted
standard and practice at the time of publication. However, since research and regulation constantly
change clinical standards, the reader is urged to check the product information sheet included in the
package of each drug, which includes recommended doses, warnings, and contraindications. This is
particularly important with new or infrequently used drugs. Any treatment regimen, particularly one
involving medication, involves inherent risk that must be weighed on a case-by-case basis against the
benefits anticipated. The reader is cautioned that the purpose of this book is to inform and enlighten;
the information contained herein is not intended as, and should not be employed as, a substitute for
individual diagnosis and treatment.
This book is dedicated to the tens of thousands of
individuals who purchased copies of the first
and second editions, and to the many who wrote to
tell us how much they enjoyed the book. It is these
people who inspired us to write a third edition.
Thank you.
Preface to Third Edition
The second edition of PDQ Statistics came out 11 years after the
first, and now the third edition appears on the scene only 5 years after
the second. Has the pace of change in statistics grown so much that it’s
halved the time necessary for bringing out a new edition? Not really. The
majority of what we’ve added has been around for a decade or three.
What’s changed is us. Now that all of the kids are grown and out of our
homes (at least, we hope so), we have more time for pursuits other than
dealing with minor and not-so-minor adolescent crises; things like carpentry
(GRN), woodworking (DLS), traveling (both of us), and—when there’s any
time left over—learning some new statistical techniques. So, much of the
new stuff (hierarchical and logistic regression, path analysis, and structural
equation modeling) is what we’ve been playing with these last few years.
The other change is that both computer programs and journals have become
more sophisticated, so it’s not unusual to come across these techniques in
clinical articles. Also, editors and grant review panels have (finally) become
aware of the problems of missing data, so that topic has also been added to
this edition.
G.R.N.
D.L.S.
January 2003
Table of Contents
Cover
PDQ* SERIES
Copyright
Preface to Third Edition
Introduction
Part One Variables and Descriptive Statistics
1 Names and Numbers: Types of Variables
2 Describing Data
Part Two Parametric Statistics
3 Statistical Inference
4 Comparison of Means of Two Samples: The t
Test
5 Comparison among Many Means: Analysis of
Variance
6 Relationship between Interval and Ratio
Variables: Linear and Multiple Regression
7 Analysis of Covariance
8 Variations on Linear Regression: Logistic
Regression, General Linear Model, and
Hierarchical Linear Models
9 Time Series Analysis
Part Three Nonparametric Statistics
10 Nonparametric Tests of Significance
11 Nonparametric Measures of Association
12 Advanced Nonparametric Methods
Part Four Multivariate Statistics
13 Introduction to Multivariate Statistics
14 Multivariate Analysis of Variance
15 Discriminant Function Analysis
16 Exploratory Factor Analysis
17 Path Analysis and Structural Equation Modeling
18 Cluster Analysis
19 Canonical Correlation
20 Reprise
21 Research Designs
Index
Unabashed Glossary
Introduction
Warning: This is not an introductory textbook in statistics. Introductory
textbooks imply that you will go on to intermediate textbooks and then to
advanced textbooks. As a result, introductory textbooks usually deal with
only a few of the topics in the discipline. So if you want to apply your
introductory knowledge of statistics to examining journals, most of the topics
used by researchers won’t have been covered.
The intent of this book is to help you through the results section of a research
article where the numbers are actually crunched, and little asterisks or “p <
.05” values appear as if by magic in the margins, to the apparent delight of
the authors. We think that by reading this book, you won’t actually be able to
do any statistics (actually, with computers on every street corner, no one—
doctor, lawyer, beggarman, or statistician—should have to do statistics), but
you will understand what researchers are doing and may even be able to tell
when they’re doing it wrong. There is an old joke about the three little French
boys, ages 4, 5, and 6, who saw a man and a woman naked on a bed in a
basement apartment. They said:
Four-Year-Old-Boy: Look at that man and that woman in there! They are
wrestling!
The 4-year-old boy knew nothing of lovemaking. The 5-year-old boy had
achieved a conceptual understanding, and the 6-year-old boy understood
lovemaking sufficiently well, presumably without actually having done it, to
be a critical observer. The challenge of this book will be to turn you into a 6-
year-old statistician. So, we will not take the “Introductory Textbook”
approach in this book. Instead, we will expose you to nearly every kind of
statistical method you are likely to encounter in your reading. Our aim is to
help you understand what is going on with a particular approach to analysis,
and furthermore, we hope you will understand enough to recognize when the
author is using a method incorrectly. Along the way, you can leave the
calculator on the back shelf, because it won’t be needed.
We hope that as a result of reading this book you will find the results section
of journal articles a little less obscure and intimidating and thereby become a
more critical consumer of the literature. It would not be surprising if you
emerged with a certain degree of skepticism as a result of your reading.
However, it would be unfortunate if you ended up dismissing out of hand any
research advanced statistical methods simply because of your knowledge of
the potential pitfalls. Keep in mind that the objective of all statistical analyses
is to reveal underlying systematic variation in a set of data, either as a result
of some experimental manipulation or from the effect of other measured
variables. The strategy, which forms the basis of all statistical tests, is a
comparison between an observed effect or difference and the anticipated
results of random variation. Like a good stereo receiver, statistical analysis is
designed to pluck a faint signal out of a sea of noise.
Unfortunately, also like a modern stereo receiver, the statistical methods are
contained in black boxes like SPSS and SAS; prominent stickers proclaim
that they should be opened by qualified personnel only so that it is nearly
impossible to understand the detailed workings of a MANOVA, factor
analysis, or logistic regression program. Finally, to complete the analogy,
these boxes of software seem replete with endless switches and dials in the
form of obscure tests or optional approaches, which may be selected to
execute the programs and report the results.
Let us say ahead of time what statistics cannot do to help place this book in
an appropriate context.
The probability or “p” level associated with any test of significance is only a
statement of the likelihood that an observed difference could have arisen by
chance. Of itself, it says nothing about the size or importance of an effect.
Because probability level is so closely related to sample size, small effects in
large studies can achieve impressive levels of significance. Conversely,
studies involving small numbers of subjects may have too little power to
detect even fairly large effects.
No statistical method can effectively deal with the systematic biases that may
result from a poorly designed study. For example, statistical techniques may
be used to adjust for initial differences between two groups, but there is no
way to ensure that this adjustment compensates exactly for the effect of these
differences on the results of the study. Similarly, no statistical analysis can
compensate for low response rates or high dropouts from a study. We can
demonstrate ad nauseam that the subjects who dropped out had the same age,
sex, marital status, education, and income as those who stayed in, but this is
no guarantee that they would have been comparable to the variables that were
measured in the study. The mere fact that they dropped out implies that they
were different on at least one dimension, namely the inclination to remain in
the study. Finally, no measure of association derived from natural variation in
a set of variables, however strong, can establish with certainty that one
variable caused another. To provide you with some tools to sort out the
different experimental designs, we have included a critical review of the
strengths and weaknesses of several designs in Chapter 18.
Finally, even after you are satisfied that a study was conducted with the
appropriate attention to experimental design and statistical analysis and that
the results are important, there remains one further analysis that you, the
reader, can conduct. Whether you are a researcher or clinician, you must
examine whether the results are applicable to the people with whom you deal.
Are the people studied in the research paper sufficiently similar to your
patients that the effects or associations are likely to be similar? For example,
treatments that have significant effects when applied to severely ill patients in
a university teaching hospital may be ineffective when used to treat patients
with the same, albeit milder, form of the disease that is encountered in a
general practice. Similarly, psychological tests developed on university
undergraduates may yield very different results when applied to a middle-
aged general population. The judgment as to the applicability of the research
results to your setting rests primarily on the exercise of common sense and
reasoning, nothing more.
One last word about the intent of this book. Although we would like think
that a good dose of common sense and an understanding of the concepts of
statistics will enable you to examine the literature critically, we’re going to
hedge our bets a bit. Throughout the book we highlight particular areas where
researchers frequently misuse or misinterpret statistical tests. These are
labeled as “C.R.A.P.* Detectors,” with apologies to Ernest Hemingway,† and
are intended to provide particular guides for you to use in reviewing any
study. In applying your new knowledge, don’t be intimidated by all the tables
of numbers and hifalutin talk. Always keep in mind the advice of Winifred
Castle, a British statistician, who wrote that “We researchers use statistics the
way a drunkard uses a lamp post, more for support than illumination.”
Finally, we hope that you enjoy the book!
G.R.N
D.L.S.
May 1986
There are four types of variables. Nominal and ordinal variables consist of
counts in categories and are analyzed using "nonparametric" statistics.
Interval and ratio variables consist of actual quantitative measurements
and are analyzed using "parametric" statistics.
Let’s have a closer look at the data from this experiment. To begin with, there
are at least two variables. A definition of the term variable is a little hard to
come up with, but basically it relates to anything that is measured or
manipulated in a study. The most obvious variable in this experiment is the
measurement of the extent of psoriasis. It is evident that this is something that
can be measured. A less obvious variable is the nature of treatment—drug or
placebo. Although it is less evident how you might convert this to a number,
it is still clearly something that is varied in the course of the experiment.
There are other ways of defining types of variables that turn out to be
essential in determining the ways the numbers will be analyzed. Variables are
frequently classified as nominal, ordinal, interval, or ratio (Figure 1-1). A
nominal variable is simply a named category. Our clam juice versus placebo
is one such variable, as is the sex of the patient, or the diagnosis given to a
group of patients.
Types of variables
1
4T
Qualitative Quantitative
I V I
Independent
T Dependent
Nominal Interval ( reflux ) ( proton pump
(sex, marital ( 36° - 38°C) inhibitor
status ) versus placebo )
Ordinal Ratio
( stage of cancer, ( pulse rate,
pain rating) vital capacity)
C.R.A.P. DETECTORS
Example 1-1
There are two distinct steps in the process of analyzing data. The first step is
to describe the data by using standard methods to determine the average
value, the range of data around the average, and other characteristics. The
objective of descriptive statistics is to communicate the results without
attempting to generalize beyond the sample of individuals to any other group.
This is an important first step in any analysis. For a reader to understand the
basis for the conclusions of any study, an idea of what the data look like is
necessary.
The second step in some, but not all, studies is to infer the likelihood that the
observed results can be generalized to other samples of individuals. If we
want to show that clam juice is an effective treatment for psoriasis, or that
intelligence (IQ) is related to subsequent performance, we are attempting to
make a general statement that goes beyond the particular individuals we have
studied. The rub is that differences between groups can rarely be attributed
simply to the experimental intervention; some people in the clam juice group
may get worse, and some people in the placebo group may get better. The
goal of inferential statistics is to determine the likelihood that these
differences could have occurred by chance as a result of the combined effects
of unforeseen variables not under the direct control of the experimenter. It is
here the statistical heavy artillery is brought to bear. As a result, most damage
to readers of journals is inflicted by inferential statistics. Most of this book is
devoted to the methods of statistical inference. However, a good idea of what
the data look like is a necessary prerequisite for complex statistical analysis,
both for the experimenter and the reader, so let’s start there.
In either case, these numbers are distributed in some manner among the
various categories or throughout the various possible values. If you plotted
the numbers, you would end up with distributions similar to those shown in
Figures 2-1 and 2-2.
Note that there are a couple of things that can be done to make these figures
more understandable. If we divide the numbers in each category by the total
number of people studied, we are then displaying the proportion of the total
sample in each category, as shown on the right side of each graph. Some
manipulations can be performed on these proportions. For example, to find
the probability in Figure 2-2 that one of our folks is 69 or 70 years old, we
must add up the probabilities in these 2-year categories. We can also address
questions such as “What is the probability that a senior citizen is more than
71 years of age?” by adding the categories above age 71 years.
The basic notion is that we can view the original distribution of numbers as
an expression of the probability that any individual chosen at random from
the original sample may fall in a particular category or within a range of
categories. This transformation from an original frequency distribution to a
distribution of probability is a recurrent and fundamental notion in statistics.
Figure 2-2 Age distribution of 10,000 entrants in senior citizen roller derby.
They conveyed him back by slow stages, seeing this and that
continental sight on his homeward-route, but hardly knowing what
he saw. He was in London again for a week or two in June and July
1832, attended medically in a hotel in Jermyn Street. Brought thence
by sea to Edinburgh, he passed a night, a day, and another night, in a
hotel in St. Andrew Square, in a state of utter unconsciousness; and
on the 11th of July they took him to Abbotsford. On their way thither
through the old familiar scenery he began to recognise places and
objects, and to mutter their names,—Gala Water, Buckholm,
Torwoodlee; and, when they approached Abbotsford itself, and he
caught sight of its towers, he sprang up in such a state of excitement
that they could hardly hold him in the carriage. “Ha! Willie Laidlaw!
O man, I have often thought of you,” were his first words, after his
old friend and amanuensis Laidlaw, who was waiting in the porch,
had assisted the rest in carrying him into the house, and seating him
in a chair in the dining-room. The return of consciousness which this
recognition signified became more and more marked, at least at
intervals, in the two months and ten days through which he still
lingered. He talked with those of his family who were about him,
could be shifted from room to room or even wheeled in a Bath chair
through parts of his grounds, and could listen to readings and seem
to take an interest in them. Once he insisted on being placed at his
writing-table, with paper, pens, and ink before him in the
accustomed order, and wanted to be left to himself; but, when the
pen had been put into his hand and his fingers refused to hold it,
tears trickled down his cheeks, and he gave up the attempt. There
were, as often in such cases of brain-paralysis, some days of almost
frantic vehemence, when it was painful to be near him; but these
were succeeded by a feeble quietude and a gradual ebbing-away of
life. On the 21st of September 1832, with the ripple of the Tweed
heard by those who stood round his bed, Sir Walter Scott, then only
in the sixty-second year of his age, breathed his last.
In the Diary itself the narrative of those closing years of Scott’s
life is broken short at the point where they were bringing him back
from Italy as a dying invalid. The last few months are a total blank in
the Diary; where, indeed, the entries for the later years of the
included seven are scantier and more intermittent than those for the
earlier. But it is not solely as an exact autobiographic record of the
incidents of so many memorable years of a memorable life that the
Diary is now of interest. Implicated in that main interest, and
catching the attention of the reader again and again as he advances
through the pages, are certain recurring particular informations as to
Scott’s character and ways which possess an independent interest,
and may be reverted to separately.
Bound up, for example, with the proofs furnished by the Diary of
Scott’s prodigious literary industry, there is plenty of minute
information as to his habits of composition and his rate of
composition. I do not like that word “composition” in any such
application, thinking it a miserable word for the description of the
process by which a great writer marshals the contents of his mind
and commits them to paper; but the word is current, and may serve
for the nonce. Well, Scott’s rate of composition was about the fastest
known in the history of literature. Of all his predecessors in the
literary history of the British Islands, Shakespeare seems to have
been the likest to him in this particular of fluent facility and swiftness
of production. “His mind and hand went together,” is the well-known
report concerning Shakespeare by his literary executors and editors:
“his mind and hand went together, and what he thought he uttered
with that easiness that we have scarce received from him a blot in his
papers.” One has an impression, however, that Shakespeare, with all
his facility when he had the pen in his hand, had it less constantly in
his hand, was less “eident” in the use of it (as our good northern
phrase goes), than Scott,—whether because he had less actual need
to be “eident,” or because verse, which was Shakespeare’s main
element, is intrinsically more difficult, takes more out of a man in a
given time, and so is less favourable to “eidency,” than the prose
element in which, latterly, Scott worked all but exclusively. At all
events, “eidency” and “facility” taken together, the result, in the mere
matter of quantity, was larger from Scott’s industry than from
Shakespeare’s. But it is with the “facility” that we are now concerned,
and with the proofs of this “facility” which are furnished by the
Journal in particular. The mere look of the handwriting is one of
these,—that rapid currente calamo look, without hesitation, and with
hardly an erasure, stoppage to point, or any such thing, and with the
words almost running into each other in their hurry, which is
familiar to all who have seen facsimile reproductions of any portions
of the copy of Scott’s novels, when they were written with his own
hand, and not dictated. That, however, is a characteristic common to
all his writings; and the specific interest of the Diary in this
connection is that it gives us definite information as to the amount of
writing per day which Scott usually got through in his currente
calamo style. In entry after entry there is note of the number of
pages he had prescribed to himself as a sufficient day’s “task” or
“darg,” with growls when for any reason he had fallen short of it, and
smiles of satisfaction when he had exceeded it; and from one entry
we ascertain that his maximum per day when he was in good vein
was eight pages of his own close manuscript, making forty pages of
the usual type in which his copy was set up by the printers. One can
compute the difference between that rate and any other rate of which
one may happen to have knowledge or experience; but there is no
need to conclude that Scott’s rate is to be passionately desired or
universally aimed at, or that, because it suited Scott, it would suit
others. On the contrary, one sees some disadvantages, even in Scott’s
own case, counterbalancing the advantages of such extreme rapidity.
He was aware of the fact himself; and he once quotes, with some
approbation, an admirable maxim of Chaucer on the subject:—
“There n’ is no werkman, whatsoever he be,
That may both werken well and hastily.”
PART I.—1809–1818
Early in November 1809 two boys walked together from
Ecclefechan in Dumfriesshire to Edinburgh, to attend the classes in
the University there. The distance, as the crow flies, is about sixty
miles; and the boys took three days to it. The elder, who had been at
College in the previous session, and therefore acted as the guide,
generally stalked on a few paces ahead, whistling an Irish tune to
himself. The younger, who was not quite fourteen years of age, and
had never been out of Dumfriesshire before, followed rather wearily,
irritated by the eternal Irish tune in front of him, but mainly given up
to his own “bits of reflections in the silence of the moors and hills.”
The elder of the two boys was a Thomas Smail, afterwards of some
note as a Burgher minister in Galloway; the younger was Thomas
Carlyle.
Of the arrival of the two boys in Edinburgh on the 9th of
November 1809, after their third day’s walk of twenty miles, and of
Carlyle’s first stroll, that afternoon, under Smail’s convoy, through
some of the main streets, to see the sights, one may read in his own
Reminiscences. What he remembered best of that first stroll was the
look of the Old High Street, with St. Giles’s Kirk on one side and the
old Luckenbooths running up the middle in its broadest part, but
chiefly the amazing spectacle to which he was introduced when Smail
pushed open a door behind St. Giles’s Kirk, and he found himself in
the outer house of the Court of Session, amid the buzz of the lawyers
and others walking up and down, with the red-robed judges hearing
cases in their little throned enclosures.
Content with the description of that first stroll, he leaves us to
imagine how, in the first days and weeks of his residence in the city,
he gradually extended his acquaintance with it by further rambles,
and by inspection of this and that interesting to a young stranger.
The task is not difficult. The lodging which Smail and he had taken
between them was, he says, “a clean-looking, most cheap lodging,” in
the “poor locality” called Simon Square. The locality still survives
under that name, though hardly as a square any longer, but only a
poor street, at the back of Nicolson Street, on the left hand as one
goes southwards from the University, and accessible most directly by
an arched passage called Gibb’s Entry. From that obscure centre, by
walks from it in the mornings, and returns to it during the day and in
the evenings, we can see the little Dumfriesshire fellow gradually
conquering for himself some notion of the whole of that Edinburgh
into which he had come. It was the old Edinburgh, of less than
100,000 inhabitants, which we think of so fondly now as the
Edinburgh of Scott before his novels had been heard of and when his
fame depended chiefly on his poems, of Jeffrey in the early heyday of
his lawyership and editorship of the Edinburgh Review, and of the
other local celebrities, Whig and Tory, immortalised in tradition and
in Cockburn’s Memorials.
It was chiefly of the externals of the city that the boy was making
his notes; for the living celebrities, as he tells us, were hardly even
names to him then. Scott and Jeffrey, he says, may have been in the
peripatetic crowd of wigged and gowned lawyers he had seen in the
hall of the Parliament House on the day of his arrival; but the only
physiognomy he had marked there so as to know it again was that of
John Clerk of Eldin. A reminiscence which I have heard from his own
lips enables me, however, to connect his first days in Edinburgh with
the memory of at least one Edinburgh worthy of a still elder
generation. It was on the 18th of December 1809, or just six weeks
after Carlyle’s arrival in Edinburgh, that the well-known Dr. Adam,
Rector of the High School, died; and I have heard Carlyle tell how the
event impressed him, and how he went to see the funeral procession
of the old scholar start from the High School yard at the foot of
Infirmary Street. With a number of other boys, he said, he hung on
by the railings outside, looking in upon the gathered assemblage of
mourners. He seemed to remember the scene with peculiar
vividness; for, after picturing himself as a boy hanging on by the
High School railings, and watching the incidents within, he added,
“Ay me! that moment then, and this now, and nothing but the
rushing of Time’s wings between!”[10] He had a liking to the last for
old Dr. Adam. I have heard him say that any Scotsman who was at a
loss on the subject of shall and will would find the whole doctrine in
a nutshell in two or three lucid sentences of Dr. Adam’s Latin
Grammar; and I had an idea at the time that he had used this brief
precept of Dr. Adam’s little book in his own early practice of English.
At the date of Dr. Adam’s death Carlyle had been for six weeks a
student in the University, with pupils of Dr. Adam among his fellow-
students on the same benches. One can see his matriculation
signature, “Thomas Carlyle,” in his own hand,—a clear and good
boyish hand, differing considerably from that which he afterwards
wrote,—in the alphabetically arranged matriculation list of the Arts
Students of the session 1809–10. It is the sixth signature under the
letter C, the immediately preceding signature being that of a
Dumfries youth named “Irvine Carlyle” (spelt so, and not “Irving
Carlyle,”) of whom there is mention in the Reminiscences. It is clear
that the two Carlyles were drawn to each other by community of
name and county, if not by kin, and had gone up for matriculation
together.
The College of those days was not the present complete
quadrangle, but a chaotic jumble of inconvenient old class-rooms,
with only parts of the present building risen among them, and
finished and occupied. The classes which Carlyle attended in his first
session were the 1st Humanity Class, under Professor Alexander
Christison, and the 1st Greek Class, under Professor George Dunbar.
From an examination of the records I find that among his class-
fellows in both classes were the aforesaid Irving Carlyle, and Lord
Inverurie, afterwards seventh Earl of Kintore, and that among his
class-fellows in the 1st Greek Class was the late venerable Earl of
Wemyss, then Lord Elcho. Neither from the records nor from the
Reminiscences can anything be gathered of the history of the two
classes through the session, or of the place taken in each by the
young Dumfriesshire boy among the medley of his fellow-students,
from 150 to 200 in number. The Latin class-room, we do learn from
the Reminiscences, was a very dark room, so that Professor
Christison, having two students of the name of Carlyle, never
succeeded in distinguishing the one from the other; which was all the
harder, Carlyle thought, because the other Carlyle, Mr. Irving
Carlyle, was not only different physically, being “an older,
considerably bigger boy, with red hair, wild buck teeth, and scorched
complexion,” but was also the worst Latinist in the whole class.
Carlyle himself had been so well grounded in Latin at Annan School
that probably he could have held his own in the class even against
Dr. Adam’s pupils from the Edinburgh High School. To the end of his
life, at all events, he was a fair Latinist. To Greek he never in later life
made any pretence; and whatever Greek he did learn from Dunbar,—
which can have been but small in quantity,—must have faded
through disuse. He retained, however, a high admiration for the
Elementa Linguæ Græcæ of Dr. James Moor of Glasgow,—which
was, I suppose, the Greek grammar then used in Dunbar’s class,—
thinking it the very best grammar of any language for teaching
purposes he had ever seen.
While we know so little of Carlyle’s Greek and Latin studies in
his first University session, it is something to know that he was a
pretty diligent reader of books that session from the College Library.
Having examined a dusty old folio of the library receipts and
outgoings, which chances to have been preserved, I am able to report
that Carlyle had duly paid, before December 1809, his deposit or
security of one guinea, entitling him to take books out, and that, in
that month and the succeeding month of January 1810, he had out
the following books, in parcels or in succession, in the following
order:—Robertson’s History of Scotland, vol. ii.; Cook’s Voyages;
Byron’s Narrative, i.e. “the Hon. John Byron’s Narrative of the Great
Distresses suffered by Himself and his Companions on the Coast of
Patagonia, 1740–6”; the first volume of Gibbon; two volumes of
Shakespeare; a volume of the Arabian Nights; Congreve’s Works;
another volume of the Arabian Nights; two volumes of Hume’s
England; Gil Blas; a third volume of Shakespeare; and a volume of
the Spectator. This is a sufficiently remarkable series of volumes for
a boy of fourteen to have had out from the College library; and other
books from other libraries may have been lying at the same time on
the table in the small room in Simon Square which he shared with
Tom Smail. What is most remarkable is the run upon books of
voyages and travels, and on classic books of English literature, or
books of mere literary amusement, rather than on academic books.
Clearly there had been a great deal of previous and very
miscellaneous reading at Ecclefechan and Annan, with the already
formed result of a passion for reading, and very decided notions and
tastes as to the kinds of books that might be worth looking after. But
how, whether at Ecclefechan or in Annan, had the sedate boy been
attracted to Congreve?
At the close of Carlyle’s first college session in April 1810 he
returned to Ecclefechan. He was met on the road near the village, as
he tells us so touchingly in his Reminiscences, by his father, who had
walked out, “with a red plaid about him,” on the chance of seeing
Tom coming; and the whole of the vacation was spent by him at
home in his father’s house. It is not, therefore, till the beginning of
the session of 1810–11 that we again hear of him in Edinburgh. He
then duly matriculated for his second session, his signature again
standing in the alphabetical Arts matriculation-list immediately after
that of his namesake “Irving Carlyle” (now spelt so). His classes for
this session were the 1st Mathematical Class, under Professor John
Leslie, and the Logic Class, under Professor David Ritchie; and I
have found no note of his having gone back that year, or any other,
for a second course of Latin from Professor Christison. In the 1st
Mathematical Class, consisting of seventy students, he had again
Irving Carlyle on the benches with him; in the Logic Class, consisting
of 194 students, the same Irving Carlyle was one of his fellow-
students, and the late Earl of Wemyss was another. What he made of
the Logic Class we have not the least intimation; and it is only by
inference that we know that he must have distinguished himself in
the Mathematical Class and given evidences there of his unusual
mathematical ability. As before, however, he found variation, or
diversion, from his work for the classes by diligent reading in his
lodgings. Between Saturday the 1st December 1810 and Saturday 9th
March 1811, I find, he took from the University library the following
books in the following order:—Voyages and Travels, the 15th volume
of some collection under that name; a volume of Fielding’s works; a
volume of Smollett; Reid’s Inquiry into the Human Mind; a book
called Scotland Described; two more volumes of Fielding’s works;
Locke’s Essay in folio; another volume of Fielding; a volume of
Anacharsis, i.e., of an English Translation of the Abbé Barthélémy’s
Travels of Anacharsis the Younger in Greece during the middle of the
Fourth Century before the Christian Era; and a volume of some
translation of Don Quixote. His choice of books, it will be seen, is still
very independent. Reid’s Inquiry and Locke’s Essay connect
themselves with the work in the Logic Class; but the other volumes
were evidently for mere amusement. Whether it was still in the
lodging in Simon Square, and with Smail for his chum, that these
books were read, is uncertain. His comradeship with Smail
continued, indeed, he tells us, over two sessions; but the lodging may
have been changed. It was still, doubtless, somewhere near the
University.
For the session of 1811–12 the Matriculation Book is not
alphabetically in Faculties, but general or mixed for the three
Faculties of Arts, Law, and Medicine. There were 1475 students for
those three Faculties conjointly; and “Thomas Carlyle, Ecclefechan,”
appears among them, his matriculation number being 966. That
session, his third at the University, he attended the 2d Greek Class,
under Dunbar, the 2d Mathematical Class, under Leslie, and the
Moral Philosophy Class, under Dr. Thomas Brown. In the Greek
Class, which consisted of 189 students, he had among his class-
fellows the late venerable Sir Robert Christison, Sir Robert’s twin-
brother, Alexander Christison, the late Earl of Wemyss again, his
brother, the Honourable Walter Charteris, a Thomas Murray from
Kirkcudbrightshire, afterwards a well-known citizen of Edinburgh,
the inextinguishable Irving Carlyle, and an Andrew Combe, whom I
identify with the subsequently well-known Dr. Andrew Combe, the
brother of George Combe the phrenologist. In the Mathematical
Class, which numbered forty-six, there were several Dumfriesshire
students besides himself; and it was in this 2d Mathematical Class, if
the tradition is correct, that Carlyle took the first prize,—another
Dumfriesshire youth, who lived in the same lodging with him, taking
the second. I have turned with most interest, in this session, to the
“List of Students attending Dr. Thomas Brown’s Class,” preserved in
the peculiarly neat, small handwriting of Dr. Brown himself. It was
the second session of Brown’s full tenure of the Professorship of
Moral Philosophy in succession to Dugald Stewart, and the fame of
his lectures was at its highest. The class consisted of 151 students;
and among them, besides Carlyle and his inseparable Irving Carlyle,
and a Robert Mitchell and a Paulus Aemilius Irving, both from
Dumfriesshire, there were Duncan McNeill, afterwards Lord
Colonsay, his brother, John McNeill, Sir Andrew Agnew, David
Welsh, afterwards Dr. David Welsh and Professor of Church History,
and a James Bisset from Aberdeenshire, whom I identify with the
late Rev. Dr. Bisset of Bourtie. Some of these were outsiders, already
in the Divinity or Law Classes, who had returned to the Moral
Philosophy Class for the benefit of Dr. Brown’s brilliant lectures,—
notably young David Welsh, who had already attended the class for
two sessions, but was full of enthusiasm for Brown, whose
biographer and editor he was to be in time. Carlyle, I am sorry to say,
was not one of the admirers of the brilliant Brown. Over and over
again I have heard him speak of Brown, and always with mimicry
and contempt, as “a finical man they called Brown, or sometimes
Missy Brown, that used to spout poetry.” This can hardly have been
out of disregard for metaphysics as such, for he had much respect for
Dugald Stewart, the then retired professor. The dislike seems to have
been partly personal, partly to the new kind of highly ingenious
metaphysics which Brown was trying to substitute for the older and
more orthodox Scottish Philosophy of Reid and Stewart. At all
events, it is worthy of note that those brilliant lectures of Thomas
Brown, which James Mill and John Stuart Mill admired so much in
their published form, regarding them as an introduction to much
that is best in modern British Philosophy, had no effect, in their
actual delivery, on the hard-headed young Carlyle, but fell upon him
as mere dazzle and moonshine.
As Carlyle tells us incidentally that he was in Edinburgh in the
summer of 1812, it is to be supposed that he spent less of that
vacation than usual in his Dumfriesshire home. I find also that he
matriculated rather late in our books for the session of 1812–13, his
name not appearing in the first or main matriculation list, but only in
a supplementary list, and then as “Thomas Carlyle, Hoddam,
Dumfriesshire.” His father had by that time given up his trade of
mason, and had left Ecclefechan to try a small farm in the
neighbourhood. The number of students matriculated that year in
the three faculties of Arts, Law, and Medicine, was 1503; and
Carlyle’s matriculation number was 1403. The classes in which he
was enrolled for that session, his fourth and last in Arts, were Leslie’s
2d Mathematical Class (attended a second time, we may suppose, for
such higher instruction as might be fit for very advanced students),
and the Natural Philosophy Class, under Professor John Playfair. In
this last session, accordingly, as a student only of Mathematics and
Physics, with no distraction towards either Classics or Mental
Philosophy, Carlyle may be said to have been in his element. He
worked very hard in both classes, and distinguished himself in both.
My own impression, from talks with him on the subject, is that he
was, by acknowledgment of professors and fellow-students, easily
supreme in both. Leslie’s second class that year numbered but forty-
one students, and it was natural that his most distinguished student
in two previous sessions should now be familiar with him and receive
his especial notice. Certain it is that of all the Professors of
Edinburgh University in Carlyle’s time Leslie was the only one of
whom he spoke always with something of real gratitude and
affection. The affection was mixed, indeed, with a kind of laughing
remembrance of Leslie’s odd, corpulent figure, and odd rough ways;
and he would describe with particular gusto the occasional effects of
Leslie’s persistent habit of using hair-dyes, as when a streak of pink
or green would be observable amid the dark-brown or black on those
less accessible parts of his head where the chemicals had been too
liberally or too rashly applied. But he had a real esteem for Leslie’s
great abilities, and remembered him as a man to whose
mathematical instructions, and to whose private kindness, he owed
much.——A greater Hero with him in Pure Mathematics than even
Leslie, I may mention parenthetically, was the now totally-forgotten
John West, who had been assistant-teacher of Mathematics in the
University of St. Andrews for some time from about 1780 onwards,
and of whom Leslie, Ivory, and all the other ablest mathematicians
sent forth from that University, had been pupils. Of this man, whom
he knew of only by tradition, but whom he regarded as, after Robert
Simson of Glasgow, the most original geometrical genius there had
been in Scotland, I have heard him talk I know not how often. He
would sketch West’s life, from the time of his hard and little-
appreciated labours at St. Andrews to his death in the West Indies,
whither he had emigrated in despair for some chaplaincy or the like;
he would avow his belief that Leslie had derived some of his best
ideas from that poor man; and he expressed pleasure at finding I
knew something of West independently, and had a copy of West’s
rare Elements of Mathematics, published in 1784. That book,
obsolete now, was, I have no doubt, a manual with Carlyle while he
was studying Mathematics in Edinburgh University, as I chance to
know it had been with Dr. Chalmers at St. Andrews in his earlier
mathematical days.——Of Leslie’s colleague, the celebrated Playfair,
formerly in the Mathematical Chair, but since 1805 in that of Natural
Philosophy, Carlyle had a less affectionate recollection personally
than of Leslie. Sharing, I believe, the common opinion of Playfair’s
great merits, and minutely acquainted with the facts of his life, as
indeed he was with the biographies of all persons of any mark with
whom he had come into contact, he rather resented a piece of
injustice which he thought Playfair had done to himself. There were
131 students in the Natural Philosophy Class in 1812–13; and Carlyle,
as he assured me, was single in that whole number for having
performed and given in every one of all the prescribed exercises,
mathematical or other. Another Dumfriesshire student, who came
next to him, had failed in one, and that the most difficult. Naturally,
at the end of the session, he expected that his certificate would
correspond to his distinction in the class; and it was of some
consequence to him that it should. But, when he called at Playfair’s
house for the certificate, and it was delivered to him by a man-
servant, he was a good deal disappointed. The usual form of the
wording for a good student was to the effect that the Professor
certified that so-and-so had attended the class in such and such a
session and had “made good proficiency in his studies.” In Carlyle’s
case there was a certain deviation from this form, but only to the
effect that he had attended the class and that the Professor “had
reason to know that he had made good proficiency in his studies.” I
can remember Carlyle’s laugh as he told me of this delicate
distinction; and I have always treasured the anecdote as a lesson for
professors. They ought to be very careful not only in noting talent on
the benches before them, but also in signifying what they have
noted, if only because, as in Playfair’s case, they may be sometimes
entertaining an angel unawares, and some angels have severe
memories.
We have thus brought Carlyle to the summer of 1813, when he
had completed his Arts course in the University of Edinburgh, and
was in the eighteenth year of his age. Though qualified, according to
the present standard, for the degree of M.A., he did not take it; but in
that he was not in the least singular. In those days hardly any
Edinburgh student ever thought of taking a degree in Arts; as far as
Edinburgh University was concerned, the M.A. degree had fallen into
almost complete disuse; and not till within very recent memory has it