The Art of Data Science Roger D.

Peng pdf
download

[Link]
peng/

Download more ebook from [Link]

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

R Programming for Data Science Roger D. Peng

[Link]
roger-d-peng/

Communicating with Data: The Art of Writing for Data

Science Nolan

[Link]
writing-for-data-science-nolan/

Statistics: The Art and Science of Learning from Data

[RENTAL EDITION] Coll

[Link]
learning-from-data-rental-edition-coll/

Educational Assessment in a Time of Reform Standards

and Standard Setting for Excellence in Education 1st
Edition Coert Loock

[Link]
of-reform-standards-and-standard-setting-for-excellence-in-
education-1st-edition-coert-loock/
gRPC Microservices in Go (MEAP V08) Hüseyin Babal

[Link]
meap-v08-huseyin-babal/

From Asia-Pacific to Indo-Pacific 1st Edition Robert G.

Patman

[Link]
pacific-1st-edition-robert-g-patman/

Embedded Software System Testing: Automatic Testing

Solution Based on Formal Method 1st Edition Yongfeng
Yin

[Link]
automatic-testing-solution-based-on-formal-method-1st-edition-
yongfeng-yin/

Collectors, Commissioners, Curators: Studies in

Medieval Art for Stephen N. Fliegel 1st Edition Elina
Gertsman

[Link]
studies-in-medieval-art-for-stephen-n-fliegel-1st-edition-elina-
gertsman/

Public Apology Between Ritual and Regret Symbolic

Excuses on False Pretenses or True Reconciliation Out
of Sincere Regret 1st Edition Daniël Cuypers

[Link]
regret-symbolic-excuses-on-false-pretenses-or-true-
reconciliation-out-of-sincere-regret-1st-edition-daniel-cuypers/
EcoGothic Gardens in the Long Nineteenth Century
Phantoms Fantasy and Uncanny Flowers 1st Edition Sue
Edney

[Link]
nineteenth-century-phantoms-fantasy-and-uncanny-flowers-1st-
edition-sue-edney/
The Art of Data Science
A Guide for Anyone Who Works with
Data

Roger D. Peng and Elizabeth Matsui

This book is for sale at
[Link]

This version was published on 2018-06-22

This is a Leanpub book. Leanpub empowers authors and

publishers with the Lean Publishing process. Lean
Publishing is the act of publishing an in-progress ebook
using lightweight tools and many iterations to get reader
feedback, pivot until you have the right book and build
traction once you do.

© 2015 - 2018 Roger D. Peng and Elizabeth Matsui

Also By These Authors
Books by Roger D. Peng
R Programming for Data Science
Exploratory Data Analysis with R
Executive Data Science
Report Writing for Data Science in R
The Data Science Salon
Conversations On Data Science
Mastering Software Development in R

Books by Elizabeth Matsui

The Data Science Salon
Special thanks to Maggie Matsui, who created all of the artwork
for this book.
Contents

1. Data Analysis as Art . . . . . . . . . . . . . . . . . . 1

2. Epicycles of Analysis . . . . . . . . . . . . . . . . . . 4
2.1 Setting the Scene . . . . . . . . . . . . . . . . . 5
2.2 Epicycle of Analysis . . . . . . . . . . . . . . . 6
2.3 Setting Expectations . . . . . . . . . . . . . . . 8
2.4 Collecting Information . . . . . . . . . . . . . 9
2.5 Comparing Expectations to Data . . . . . . . 10
2.6 Applying the Epicycle of Analysis Process . 11

3. Stating and Refining the Question . . . . . . . . . 16

3.1 Types of Questions . . . . . . . . . . . . . . . . 16
3.2 Applying the Epicycle to Stating and Refin-
ing Your Question . . . . . . . . . . . . . . . . 20
3.3 Characteristics of a Good Question . . . . . 20
3.4 Translating a Question into a Data Problem 23
3.5 Case Study . . . . . . . . . . . . . . . . . . . . . 26
3.6 Concluding Thoughts . . . . . . . . . . . . . . 30

4. Exploratory Data Analysis . . . . . . . . . . . . . . 31

4.1 Exploratory Data Analysis Checklist: A Case
Study . . . . . . . . . . . . . . . . . . . . . . . . 33
4.2 Formulate your question . . . . . . . . . . . . 34
4.3 Read in your data . . . . . . . . . . . . . . . . . 35
4.4 Check the Packaging . . . . . . . . . . . . . . 36
4.5 Look at the Top and the Bottom of your Data 39
CONTENTS

4.6 ABC: Always be Checking Your “n”s . . . . . 40

4.7 Validate With at Least One External Data
Source . . . . . . . . . . . . . . . . . . . . . . . . 45
4.8 Make a Plot . . . . . . . . . . . . . . . . . . . . 46
4.9 Try the Easy Solution First . . . . . . . . . . . 49
4.10 Follow-up Questions . . . . . . . . . . . . . . 53

5. Using Models to Explore Your Data . . . . . . . . 55

5.1 Models as Expectations . . . . . . . . . . . . . 57
5.2 Comparing Model Expectations to Reality . 60
5.3 Reacting to Data: Refining Our Expectations 64
5.4 Examining Linear Relationships . . . . . . . 67
5.5 When Do We Stop? . . . . . . . . . . . . . . . 73
5.6 Summary . . . . . . . . . . . . . . . . . . . . . . 77

6. Inference: A Primer . . . . . . . . . . . . . . . . . . 78
6.1 Identify the population . . . . . . . . . . . . . 78
6.2 Describe the sampling process . . . . . . . . 79
6.3 Describe a model for the population . . . . . 79
6.4 A Quick Example . . . . . . . . . . . . . . . . . 81
6.5 Factors Affecting the Quality of Inference . 84
6.6 Example: Apple Music Usage . . . . . . . . . 86
6.7 Populations Come in Many Forms . . . . . . 89

7. Formal Modeling . . . . . . . . . . . . . . . . . . . . 92
7.1 What Are the Goals of Formal Modeling? . 92
7.2 General Framework . . . . . . . . . . . . . . . 93
7.3 Associational Analyses . . . . . . . . . . . . . 95
7.4 Prediction Analyses . . . . . . . . . . . . . . . 104
7.5 Summary . . . . . . . . . . . . . . . . . . . . . . 111

8. Inference vs. Prediction: Implications for Mod-

eling Strategy . . . . . . . . . . . . . . . . . . . . . . . 112
8.1 Air Pollution and Mortality in New York City113
8.2 Inferring an Association . . . . . . . . . . . . 115
8.3 Predicting the Outcome . . . . . . . . . . . . 121
CONTENTS

8.4 Summary . . . . . . . . . . . . . . . . . . . . . . 123

9. Interpreting Your Results . . . . . . . . . . . . . . . 124

9.1 Principles of Interpretation . . . . . . . . . . 124
9.2 Case Study: Non-diet Soda Consumption
and Body Mass Index . . . . . . . . . . . . . . 126

10. Communication . . . . . . . . . . . . . . . . . . . . . 145

10.1 Routine communication . . . . . . . . . . . . 146
10.2 The Audience . . . . . . . . . . . . . . . . . . . 147
10.3 Content . . . . . . . . . . . . . . . . . . . . . . . 149
10.4 Style . . . . . . . . . . . . . . . . . . . . . . . . . 152
10.5 Attitude . . . . . . . . . . . . . . . . . . . . . . . 153

11. Concluding Thoughts . . . . . . . . . . . . . . . . . 154

About the Authors . . . . . . . . . . . . . . . . . . . . . . 156

1. Data Analysis as Art
Data analysis is hard, and part of the problem is that few
people can explain how to do it. It’s not that there aren’t
any people doing data analysis on a regular basis. It’s that
the people who are really good at it have yet to enlighten us
about the thought process that goes on in their heads.
Imagine you were to ask a songwriter how she writes her
songs. There are many tools upon which she can draw. We
have a general understanding of how a good song should
be structured: how long it should be, how many verses,
maybe there’s a verse followed by a chorus, etc. In other
words, there’s an abstract framework for songs in general.
Similarly, we have music theory that tells us that certain
combinations of notes and chords work well together and
other combinations don’t sound good. As good as these
tools might be, ultimately, knowledge of song structure
and music theory alone doesn’t make for a good song.
Something else is needed.
In Donald Knuth’s legendary 1974 essay Computer Program-
ming as an Art1 , Knuth talks about the difference between
art and science. In that essay, he was trying to get across
the idea that although computer programming involved
complex machines and very technical knowledge, the act of
writing a computer program had an artistic component. In
this essay, he says that

Science is knowledge which we understand so

well that we can teach it to a computer.
1 [Link]
Data Analysis as Art 2

Everything else is art.

At some point, the songwriter must inject a creative spark
into the process to bring all the songwriting tools together
to make something that people want to listen to. This is a
key part of the art of songwriting. That creative spark is
difficult to describe, much less write down, but it’s clearly
essential to writing good songs. If it weren’t, then we’d have
computer programs regularly writing hit songs. For better
or for worse, that hasn’t happened yet.
Much like songwriting (and computer programming, for
that matter), it’s important to realize that data analysis is an
art. It is not something yet that we can teach to a computer.
Data analysts have many tools at their disposal, from linear
regression to classification trees and even deep learning,
and these tools have all been carefully taught to computers.
But ultimately, a data analyst must find a way to assemble
all of the tools and apply them to data to answer a relevant
question—a question of interest to people.
Unfortunately, the process of data analysis is not one that
we have been able to write down effectively. It’s true that
there are many statistics textbooks out there, many lining
our own shelves. But in our opinion, none of these really
addresses the core problems involved in conducting real-
world data analyses. In 1991, Daryl Pregibon, a promi-
nent statistician previously of AT&T Research and now of
Google, said in reference to the process of data analysis2
that “statisticians have a process that they espouse but do
not fully understand”.
Describing data analysis presents a difficult conundrum.
On the one hand, developing a useful framework involves
characterizing the elements of a data analysis using abstract
2 [Link]
proceedings-of-a-forum
Data Analysis as Art 3

language in order to find the commonalities across differ-

ent kinds of analyses. Sometimes, this language is the lan-
guage of mathematics. On the other hand, it is often the very
details of an analysis that makes each one so difficult and
yet interesting. How can one effectively generalize across
many different data analyses, each of which has important
unique aspects?
What we have set out to do in this book is to write down
the process of data analysis. What we describe is not a
specific “formula” for data analysis—something like “ap-
ply this method and then run that test”— but rather is a
general process that can be applied in a variety of situ-
ations. Through our extensive experience both managing
data analysts and conducting our own data analyses, we
have carefully observed what produces coherent results and
what fails to produce useful insights into data. Our goal is to
write down what we have learned in the hopes that others
may find it useful.
2. Epicycles of Analysis
To the uninitiated, a data analysis may appear to follow a
linear, one-step-after-the-other process which at the end,
arrives at a nicely packaged and coherent result. In reality,
data analysis is a highly iterative and non-linear process,
better reflected by a series of epicycles (see Figure), in which
information is learned at each step, which then informs
whether (and how) to refine, and redo, the step that was
just performed, or whether (and how) to proceed to the next
step.
An epicycle is a small circle whose center moves around
the circumference of a larger circle. In data analysis, the
iterative process that is applied to all steps of the data
analysis can be conceived of as an epicycle that is repeated
for each step along the circumference of the entire data
analysis process. Some data analyses appear to be fixed and
linear, such as algorithms embedded into various software
platforms, including apps. However, these algorithms are
final data analysis products that have emerged from the
very non-linear work of developing and refining a data
analysis so that it can be “algorithmized.”
Epicycles of Analysis 5

Epicycles of Analysis

2.1 Setting the Scene

Before diving into the “epicycle of analysis,” it’s helpful

to pause and consider what we mean by a “data analysis.”
Epicycles of Analysis 6

Although many of the concepts we will discuss in this

book are applicable to conducting a study, the framework
and concepts in this, and subsequent, chapters are tailored
specifically to conducting a data analysis. While a study
includes developing and executing a plan for collecting
data, a data analysis presumes the data have already been
collected. More specifically, a study includes the devel-
opment of a hypothesis or question, the designing of the
data collection process (or study protocol), the collection
of the data, and the analysis and interpretation of the data.
Because a data analysis presumes that the data have already
been collected, it includes development and refinement of a
question and the process of analyzing and interpreting the
data. It is important to note that although a data analysis is
often performed without conducting a study, it may also be
performed as a component of a study.

2.2 Epicycle of Analysis

There are 5 core activities of data analysis:

1. Stating and refining the question

2. Exploring the data
3. Building formal statistical models
4. Interpreting the results
5. Communicating the results

These 5 activities can occur at different time scales: for

example, you might go through all 5 in the course of a day,
but also deal with each, for a large project, over the course of
many months. Before discussing these core activities, which
will occur in later chapters, it will be important to first
Epicycles of Analysis 7

understand the overall framework used to approach each

of these activities.
Although there are many different types of activities that
you might engage in while doing data analysis, every aspect
of the entire process can be approached through an iterative
process that we call the “epicycle of data analysis”. More
specifically, for each of the five core activities, it is critical
that you engage in the following steps:

1. Setting Expectations,
2. Collecting information (data), comparing the data to
your expectations, and if the expectations don’t match,
3. Revising your expectations or fixing the data so your
data and your expectations match.

Iterating through this 3-step process is what we call the

“epicycle of data analysis.” As you go through every stage
of an analysis, you will need to go through the epicycle to
continuously refine your question, your exploratory data
analysis, your formal models, your interpretation, and your
communication.
The repeated cycling through each of these five core activi-
ties that is done to complete a data analysis forms the larger
circle of data analysis (See Figure). In this chapter we go into
detail about what this 3-step epicyclic process is and give
examples of how you can apply it to your data analysis.
Epicycles of Analysis 8

Epicycles of Analysis

2.3 Setting Expectations

Developing expectations is the process of deliberately think-

ing about what you expect before you do anything, such
as inspect your data, perform a procedure, or enter a com-
mand. For experienced data analysts, in some circumstances,
developing expectations may be an automatic, almost sub-
conscious process, but it’s an important activity to cultivate
and be deliberate about.
For example, you may be going out to dinner with friends
at a cash-only establishment and need to stop by the ATM
to withdraw money before meeting up. To make a decision
about the amount of money you’re going to withdraw, you
have to have developed some expectation of the cost of
dinner. This may be an automatic expectation because you
dine at this establishment regularly so you know what the
Another Random Scribd Document
with Unrelated Content
 Bill Ryan.

"Blacked his shoes," muttered Bill, turning away and suppressing

a grin. "Oh, Lord! and a clean shave, too."
By this time the teller and the cashier had stopped hand-
shaking, and the latter was pushing John in through the office door
toward us.
"The rest of the fellows want to shake hands with you, John,"
said he, slapping the disconcerted father on the back.
This was John's ordeal. Two emotions were visibly struggling in
him. He knew perfectly, poor fellow, his incompetence, and the
consequent slight estimate in which we (being only plodding
accountants, with no very exalted criterion to judge men by) held
him. Nevertheless, this morning it must have been plain to him a
new factor had entered into his position among us—one, moreover,
which, quite irrespective of his ability or inefficiency as a commercial
automaton, entitled him to a positive measure of respect from us.
Diffident, however, and totally lacking in self-confidence as he was,
how was he to break through the old barriers of contempt and
derision which we held out against him, and demand of us, and
enforce from us the payment of this new obligation?
It was a task, truly, which seemed to require more courage and
power over others than the little man possessed, and very much
depended on an initial success. One could see that he felt this
himself; for as he walked toward us (his knees perceptibly shaking,
in spite of the unusual length of his strides) he shifted his eyes from
side to side; and when they did rest on one of us for a moment,
there was in their weak, watery blue an appeal rather than a
command.
Ted was the first to meet him. He gripped his hand hard and
cordially, and looked straight into his face.
"Mighty glad to hear it, John," he said.
John flushed, and his eyes brightened and he held on fast.
"Thanks, Ted, thanks!" he stammered, much moved.
But, as their hands parted, Ted smiled. It was not meant
unkindly; but Ted, who was a cocky, self-assured chap, and
something of a sport, too, never seemed able to look seriously at the
affair for more than a second at a time.
John's courage, which had begun to rise, left him instantly; and
he quite lost his self-control. He was white as he took the limp hand
Bill stretched out to him.
"Congratulate you, John," the latter said, frigidly, and the
"Thanks, Bill, thanks," of the reply was all of a tremble.
 Suddenly, however, a new feeling seemed to come over John,
and this was indignation—indignation at himself, and anger at the
man before him. He reddened, and stood erect again, and dropped
Bill's hand; and, without a word, turned to me.
I don't recollect what I said to John. Perhaps I said nothing. I
remember only I was thinking, "You'd have lost that bet, Bill, if you'd
taken it."
 II

Shortly after, Al Williams, who was John's next in rank, came in;
but I did not notice his greeting as I was busy over by the window
filing checks. Then, at nine, we opened up, and the regular routine
of work began.
Nine-thirty was my time for starting off with the morning's
collections, drafts on tradespeople, post-office orders, protested
checks and the like. I was very anxious that the president should
arrive before I left, for I was particularly curious to see how John
would take his congratulations, and in what spirit they would be
offered.
John had entered the bank as clerk when the president was
teller—almost twenty years ago—and had worked under him ever
since. Both men at first sight impressed one as of a type very
common in this bustling country of ours. Small, nervous men, with
light, drooping mustaches, and excitable ways, they both were. To
each of them the touch of silver, or the smell of dirty bills, or the
holding of a pen between the fingers was but the signal for a certain
set of reactions on the accuracy of which his claim to usefulness in
this world depended. Mere machines one might call them both, but
there was a vast difference between them nevertheless. For, while
John's nervousness was the nervousness of dissipated force, the
president's was that of concentrated alertness and precision and
celerity. John was a very poor machine, indeed, and as like as not to
go wrong and become tangled up in his own mechanism. Habinger,
on the other hand, was a very perfect one, and it was a saying in
the bank that he could foot a column with every wink of his eye. His
every pen-stroke, too, was an ultimatum, and stood on the books as
it was first written, without blot or erasure.
 So John (who had no other standards to measure men by but
those of the ledger and the time-lock) had made an idol of the
president. In his worship he was not only sincere and fervent, but
entirely without jealousy; for whatever egotism he might have had
to start with must long ago have been knocked out of him by the
successions of selfish and ambitious clerks he had seen pass beyond
and above him; and, as Bill had so cruelly hinted, by his ten years of
unfruitful married life.
It was, then, a real pleasure for John, on days when business
was rushing, to have Habinger unceremoniously shove him aside at
the counter, and in fifteen minutes dispose of a long row of
customers whom the hapless teller had suffered to gather there.
At these times John would stand behind the president, and look
over his shoulder with wonder like a little child's on his face; and
when the work was finished, his "Thank you, sir; thank you!" was
uttered in a tone of glad gratitude quite unalloyed, even by the
consciousness of Bill's sneering whispers at his back, or by the sly
smile of the next depositor, as he handed over his bills and checks.
So, as I said, I wished greatly to be in the bank when the
president came; and with this purpose I lingered a moment over my
time at the check-file, pretending to be very much occupied.
John's eye this morning, however, was as sharp as Habinger's;
and, as the pointer of the clock above his head marked five minutes
past the half hour, he called out brusquely,
"Hi, Jimmy! Time you were gone; and a heavy clearing this
morning, too, so you want to be back early."
His manner was authoritative, and I rose hastily, and reluctantly
commenced sorting out my collections and memoranda. But just
then Habinger came in, and with a quick brush of my arm, I swept
my papers on the floor directly behind John.
 I don't know whether John fathomed my design or not; but he
was down by me on the floor in an instant; and before I had
touched one of them, the papers were gathered up and stuffed into
my pocket-book.
"Now, off with you!" he cried, and gave me a shove, and then
turning, met the president's outstretched hand.

Ted comfortably settled himself.—Page 447.

 "Mr. Makeator," the latter began, and this was all I heard, for I
was heartily ashamed of my impertinence.
However, those two words and the glance I could not help
throwing back were enough. John's face was flushed again, but this
time with joy and pride; for never before had the president thus
publicly called him by his last name. Indeed, of all the shrewd things
Habinger ever said, I believe this was the shrewdest, and I would
have given Bill ten to one on that bet had I thought he would take it.
I made a mess of my work that morning I know—was fined two
dollars at the clearing for a wrong subtraction; forgot to call for a
couple of drafts I had left at Shan's—the liquor dealer—the day
before; mislaid a registered letter; and entered Boston remittances
in the New York book. My thoughts while out of the bank were on
John, and while in the bank my eyes were on no one else.
Indeed, there was a fascination in watching the little teller work.
He never made more mistakes, perhaps, in his life; but he detected
every one of them instantly. He had squeezed his sponges dry in two
hours, and, not thinking to have me moisten them again, simply wet
his fingers in his mouth, and thumbed his bills and scraped his silver
all unconscious of any inconvenience.
He was perpetually on the go, dabbling in everyone else's work,
but never losing his head. He ordered us around as if he were
president and directors all in one. Once, I recollect, when a ten-
dollar roll of quarters fell and split on the floor, he told Bill
peremptorily to pick them up, without so much as a "please," or
turning around to see if he were obeyed—which he was, and
promptly, too.
As for Bill, at first he simply sat dumfounded on his stool, and
watched John open-mouthed. But John found him out in a jiffy,
tossed him a handful of pass-books, which Bill took without a remark
and proceeded to balance forthwith.
 John's conversation over the counter was of a line with his
actions.

John would stand behind the president.—Page 450.

"Mornin', Mr. Bemis, mornin'! Hot day? Yes, I should say so.
Good deposit this morning. Business picking up? Yes?—Eh?—Ah.—
Yes!—yes, thanks, sir! yes, doin' splendid sir, splendid!—Coughlin to
pitch this afternoon?—yes, going to call her Margaret, sir—my wife's
name—Ah, this check here, sir? Call & Co. $123.75. Well—Eh—Hi,
Jim!" (this in a whisper to me, and handing me the doubtful check
under the counter). "Telephone down to the 'Third' and see if that's
good—yes, ten pounds seven ounces, sir. Let's see, $443 in bills I
make it only."
Then, as I came back from the telephone, "All right, Mr. Bemis.
Wanted to make sure, you know; $123.75? Born at half-past two
exactly. Good-morning——
 "But, Mr. Bemis! Oh, Mr. Bemis! Wait a moment, please. Forgot
to indorse this, I guess? Yes? All right. Feeling fine myself, sir—first
rate. Yes. Good-morning."
 III

Mr. Young insisted on John's leaving early for lunch, and staying
as long as he pleased.
"Well, George, I will," said the latter, "because my work's all
done up ahead of time—B & A bills counted" (and he pointed to a
heap of vile-smelling green-backs, neatly sorted into little packages,
and lying at one side) "and pay-rolls all made up. And I'll stay, too, if
they want me."
However, he was back sharp at half-past one; and came in with
two packages of dry goods under his arm. These he hurriedly
secreted in a cupboard under the counter, though from the solicitude
with which he handled them, I judged that he would almost have
preferred to lock them up in the vault. Bill (and the occurrence did
not seem strange then) made no comments of any sort during this
proceeding.
Then he went out back, put on his linen jacket, and in a moment
replaced Al at the counter, and the hustling commenced again.
From now on, until a quarter to four, John did all the talking. The
rest of us were too much occupied in obeying his orders. I never
knew him so voluble. He must first tell us about the state of affairs
at home. Everything was doing finely; baby lusty and thriving, wife
in good spirits, and "almost strong enough to get up," nurse scarcely
needed, doctor still less. Then a list of his congratulations, and an
account of Mrs. Makeator's visitors during the morning; and finally,
as the choicest bit of news, and typical of the generally satisfactory
condition of things, his wife's declaration that he must not bother
about her and the baby, but go up to the park with the rest of us
and see the ball-game.
 All this was gone over a dozen times to us; and once, at least, to
every customer whom he knew. While telling it, too, he thumbed his
bills, checked off deposit tickets, received telephone messages from
me, and directed the answering of them; bossed Bill, Ted, and Al
about, as he had never done before, and never once asked the
cashier's or president's advice on any topic—a circumstance entirely
new in our experience of him.
At a quarter to four we were ready to strike a balance. Al, with
the result of his half of the figuring (with which John's counter-book
should agree), stood peering over the little man's shoulder. Bill, by
force of habit mainly (for he looked forlorn enough), was behind
John on the other side. Ted and I pressed up close, too; and Mr.
Young sat at his table quietly, watching the group of us.
At these times John was generally very nervous; and frequently
the mere consciousness of having all of us at his back flustered him
so that he could not make his last deduction correctly. But his hour
of triumph was now at hand, and he knew it and rose manfully to
the occasion. He worked imperturbably and without the slightest
trace of annoyance; nor was there the least hesitancy in the rapid
tappings of his pen; and he made his footings with a decision which
showed how thoroughly confident he was of the correctness of his
calculations.
When he was done he said, "All right, Al. How is it?" and Al read
off his balance.
John jotted it down in pencil beside his own, and subtracted.
"847.43," he said.
"Over?" asked the cashier from his table.
"No; short, George," and without waiting to prove his own work,
John jumped over to Bill's clearing-books, and began footing them.
 Bill and Ted and Al (Bill in front) pounced on John's book; but
they had barely time to put pencil to it before John cried out:
"Seven less on the credit!" and Bill had the pleasure of
correcting his own mistake on John's book.

He knew perfectly, poor fellow, his incompetence.—Page 449.

"Footing?" he queried, moodily.

"Yes," said John, without looking up, as he ran his pencil down
another row of figures.
 "Seven cents on the 'first' footing," he called again, almost
immediately. "More on the credit this time, Bill. Makes it just 850.50,
doesn't it?"
Bill, however, did not answer, but edged out between Ted and Al,
and went to work on his own pass-books, errors in which did not
appear "in cash."
Then John called to me to check the listing while he read off the
clearings. Here, again, we found two mistakes, an inversion and
another, which reduced the discrepancy to about fifty dollars.
This time John did not announce the mistake (for, with his
growing assurance, all desire for public vindication and acquittal had
left him), but went and "fixed" it himself.
Ted and Al had, long since, given up the search at the counter;
and the latter, who was entering into the fun of the moment, cried
laughingly, after going over my draft-registers,
"All right, Jim!" and then to Mr. Young, "Bill didn't take your bet,
did he, George?"
"No," laughed the cashier in his turn, and added, "Better help us
on 'cash,' Bill. Your books will wait."
Bill, crestfallen, marched over to his clearing-books, and gazed
sheepishly at the corrections on them in John's handwriting.
John then discovered an error in Al's "Redemption" letter, which
Al good-humoredly acknowledged; and, shortly after, one "on" Mr.
Young himself.
This left us only a few cents out, so the cashier cried, "All right,
boys. Let her go. You'll see seven innings of the game if you hurry."
 John jotted it down.—Page 453.

"Why, aren't you going, too, George?" exclaimed John with

evident disappointment. "I wanted to treat you this afternoon," and
he pulled out of his pocket one of the new bank-notes which had
come in just a few days ago.
 It was too hot for talk.—Page 455.

"Sorry, John, but I've got some back work I must make up. And
I want you to stay, too, Jim, and slice the rest of these green-backs.
Habinger was late in signing them, you know."
This was, in some measure, a fresh disappointment to John (as
it was a very great one to me), for I could see he wanted to take the
whole of us. Al and Ted had already accepted.
Finally he went up to Bill and asked timidly and expectantly:
"You're going, too, aren't you Bill?"
But Bill refused the offer snarlingly, and mumbling something in
a priggish tone about playing golf, left the bank without another
word.
Five minutes later the others had gone, too. As they went out,
John cried:
"I'll stop in on the way back to get my wheel. You'll be here,
George?"
"Yes. See you later, John. Good luck to you."
 IV

The afternoon was broiling. The sun came in, scarcely checked
by the yellow shades; fell on and soaked into the smooth, varnished
surfaces of the desks and tables, and turned the iron of the big vault
into a sort of storage battery of heat. Even the electric fan in the
president's office, which we had placed on top of the telephone
closet (as near us as its length of wire would allow), gave but little
relief.
Both of us were working in our shirt-sleeves, but the sweat
stood on our brows, and my fingers were so sticky I could scarcely
handle my bills. It was too hot for conversation even; so the only
sounds in the room were the snipping of my shears, the crisp
fluttering of the fresh, new bills as they fell one by one on the table;
and the snapping of rubber bands as the cashier went over bundle
after bundle of the bank paper, on the security of which all our
positions depended.
As I said, it was too hot for talk; and besides, I had plenty to
engross me in my own thoughts—which were about John, of course.
I began by thinking how profitable it would be to the bank if
John might only have a baby every day; and then, as this was out of
the question, fell to calculating how long this one that had just
arrived would continue to work the same beneficial influence on her
father's actions.
Presently, however, my ideas became more serious; and at last
so serious that they brought about a reaction in the shape of a
suspicion that perhaps I had been making too much out of the
incident, after all. So I determined to get Mr. Young's opinion on the
subject, if I could; and was just framing my first interrogatory, when
the telephone rang.
 "Claflin National?" said the voice.
"Yes."
"The cashier in? Young, isn't it?"
"Yes. Yes, he's in."
"See him a moment?"
"Yes."
I was certain that the voice did not belong to any of the bank
employees in town; and yet it was familiar.
"Someone to see you, sir," I said, trying all the while to place the
voice; and then, the resemblance suddenly dawning on me,
"Is Spencer John's family physician?"
"Yes. Why?" and the cashier started.
"I think it's he at the 'phone."
The cashier was in the telephone closet almost five minutes.
When he came out he was white, and it was plain he had been
undergoing very strong emotions, though the worst of them was
evidently passed.
He began hurriedly gathering his notes together.
"Put up your work, Jim," he said. "We must lock up as soon as
possible. John's baby's dead."
The news hardly took me by surprise. I foresaw from the first
that it was something pretty bad. So I simply commenced doing as I
was told.
"He wants me to tell him," began the cashier after a moment.
"The doctor?"
 "Yes, and," looking at the clock, "he'll be back any minute now,
and, perhaps, Jim——"
"I'd best be going?"
"Yes. I'll fix up, and—My God, it's sad!—and be down early to-
morrow, Jim."
"John won't be here, I suppose."
"I hope not; but there's no telling. At any rate he won't—hustle
—to-morrow as—he did to-day. I was thinking of that."
So, as I left the bank, I found that the question I was going to
put the cashier as the telephone rang, had been answered, after all.
A Stork's Nest, Dordrecht, Holland—12-inch Lens.
Stork's Nest—Telephoto Lens.
 TELEPHOTOGRAPHY

By Dwight L. Elmendorf
Illustrated by the Author's Photographs

J UST when the telescope was invented is not known, but it is

certain that Galileo was the first to direct his toward the heavens
early in the seventeenth century. His instrument consisted of a long
tube with a convex lens at one end and a concave ocular at the
other. A modified form of this instrument still obtains in the ordinary
opera and field glasses, which are binocular Galilean telescopes; and
a single barrel of a field-glass is practically the telephoto lens of to-
day.
Whenever anything is so far away that we cannot see it
distinctly, we make use of a field-glass or telescope, which produces
a magnified image of the object so that we are able to perceive what
the unaided eye could not. In a similar manner the telephoto
attachment enlarges the image formed by the ordinary lens in the
camera. To produce on a photographic plate an image that fairly
resembles what our eyes see, requires a lens of much longer focus
than is generally used, and a camera that would permit the use of
such a lens would be unwieldy and too cumbersome for a peripatetic
photographer, and simply impossible for a mountain-climber. The
telephoto lens overcomes this difficulty by producing the effect of a
lens of long focus in a very compact camera.
 It would be interesting to know who first applied this form of
lens to a camera for the purpose of photographing distant objects.
In 1890, while experimenting with the lenses from an old field-glass,
I discovered that a dim yet distinct image of St. Patrick's Cathedral
spires was formed in my camera, although the Cathedral was
eighteen blocks away. After making several exposures with this
combination of lenses I became convinced that with lenses of the
best possible optical construction wonderful results might be
attained. Having previously purchased a telescope with a three-and-
a-half inch lens of sixty inches focus (with the idea of attaching it to
a long box-camera as a photographic lens for the purpose of making
photographs of distant terrestrial objects, as astronomers
photograph heavenly bodies). I found that the field-glass
combination of lenses yielded an image nearly as large as that
produced by the telescope lens, and that too with a camera only one
third the length of the other.
 Milan Cathedral from Opposite Corner of Piazza.

Becoming deeply interested in this line of investigation I called

upon a celebrated lens maker in London and learned that he had
manufactured what he called a "Compound Telephoto Lens"
consisting of a portrait lens with a small negative or concave lens
adjusted at a suitable distance back of it. This instrument was too
large and cumbersome for my small camera, and shortly afterward a
negative lens, with a rack and pinion mounting, was manufactured
of such a size that it could be attached to any fine rectilinear lens of
suitable focus, although in some cases special corrections are
necessary.
This is called the "Telephoto Attachment," and was employed in
making the telephoto illustrations here shown. The tube is 3¼
inches long and 1½ inch in diameter. When this lens is attached to
the ordinary lens the time of exposure is necessarily increased,
because only a few of the rays of light which diverge from the
positive or ordinary lens pass through the negative lens to the plate.
This is a serious drawback, for it not only debars one from using it
upon moving subjects, but also increases the liability of the image to
be blurred by vibrations of the camera. In order to obtain the best
results the camera must be very rigid. Most of the cameras and
tripods of to-day are too light and unstable for telephotography.
The method of using the telephoto attachment is very simple,
but requires very great care, particularly in the matter of focussing.
Suppose that an exposure has been made in the ordinary way upon
a certain object; the lens is then removed from the camera front and
screwed into the tube of the telephoto attachment, forming a small
telescope; the whole combination is then put back on the camera as
if it were the ordinary lens. Upon the ground glass or focussing
screen will be seen an enlarged image which may be made sharp or
distinct by adjusting the focus by means of the rack and pinion
movement on the telephoto tube, just as a field glass is adjusted to
suit the eyes of the observer. If greater amplification be desired it is
obtained by moving the front of the camera, holding the lenses
farther from the ground-glass and then readjusting the focus as
before. It will be seen from this that the attachment forms a lens of
variable focus, changeable at the pleasure of the operator within the
limits of the camera.
Roof and Dome, Milan Cathedral—Telephoto Lens, from Same Corner of
Piazza.

Some of the attachments on the market require a camera with a

very long bellows, because the difference between the foci of the
negative and positive lenses is not great enough to give ample
power unless the combination is several feet from the plate. With my
own attachment, eight inches from the plate the image is equal to
that formed by an ordinary lens of twenty-four inches focus; while at
twenty-four inches from the plate it is equivalent to that of a lens of
sixty-four inches focus.
The camera used in making the accompanying illustrations takes
a plate measuring four by five inches, and the bed allows an
extension of twenty-four; and when closed for transportation the box
measures seven by seven by six-and-a-half inches.
 Of all my experiences in photography none were so
unsatisfactory as my attempts on mountain scenery with an ordinary
lens. This was especially true of the photographs of the Alps made
while tramping through that heavenly tramping ground, Switzerland.
The small camera made the mountains look like little humps of rocks
and snow, and all the views made from a great elevation seemed to
be like photographs of the waves of the ocean, smoothed out flat.
These results caused me to experiment in the direction of telescopic
work with the camera.
It is often the case that grand mountains appear at their best
only from some point so distant that the ordinary lens can produce
little or nothing of the desired effect.
One of the most charming views in Switzerland is the evening
view of the Jungfrau as seen from the Höheweg or promenade at
Interlaken, about sixteen miles from the mountain. With her robes of
dazzling white she rises majestically above the Lauterbrunnen Thal
to a height of nearly fourteen thousand feet. Upon several former
occasions I had endeavored to photograph this queen of the
Bernese Oberland, but did not succeed until I used the telephoto
attachment. The two illustrations of this view [pp. 462-63] were
made from the same standpoint on the Hoheweg, one with the
ordinary lens, the other with the telephoto attachment added to the
lens, no change being made in the camera at all. It is a pleasure to
note the wonderful detail in the telephotograph, and not only that,
the mountain seems to rise, giving the impression of abruptness
which rarely if ever is obtained with an ordinary lens. I suppose
something of this result might have been obtained with the ordinary
lens had I been up in a balloon at an elevation of about four
thousand feet and about three miles from the Jungfrau. The pictures
of this mountain taken from the Wengern Alp do not give this
beautiful effect.
 Façade of the Cathedral of Florence, from Sidewalk.

This is especially the case with Popocatepetl in Mexico, a

beautiful volcanic cone rising gradually above the plateau about ten
thousand feet, its snow-capped summit being over seventeen
thousand feet above sea-level. Having tried in vain from several
places near by, I finally succeeded in obtaining a fair view of it from
the roof of the Hotel Jardin in the city of Puebla, about thirty miles
or more from the peak [pp. 466-67]. Desiring to take the only train
for Oaxaca, leaving Puebla at 5.30 in the morning, I was compelled
to photograph the mountain rather early, and the atmosphere was
not at that time in the best condition, so that the reader would have
needed a field-glass to see the mountain clearly. To obtain good
results with the telephoto attachment a clear atmosphere is a sine
qua non.
Not only does this apply to mountain subjects but to many
others alike. What remarkable pictures of the naval battle of
Santiago, the chase of the Cristobal Colon, or the gallant rescue of
the despairing Spaniards from their burning ships, might have been
obtained from the battle-ship New York, with a lens of this
description, even at long range! I believe it will be of inestimable
value for the purpose of securing views of the batteries and
fortifications of an enemy's harbor, which might be done at a safe
distance from their guns.
Central Rose Window, Cathedral of Florence, from Sidewalk—Telephoto
Lens.

While this attachment is of great value in photographing things

miles away, it is even more useful in obtaining photographs of choice
bits of landscape which are on the opposite side of a river or lake,
and are just beyond the working capacity of an ordinary lens. Odd
things are always turning up at unexpected moments, and are
frequently just out of reach.
Mosaic over Central Door, Cathedral of Florence, from side of Baptistery
—Telephoto Lens.

A particular instance of this kind is illustrated in the two views

made of a stork's nest [p. 457] which I happened to see while
sauntering along one of the picturesque old canals near Dordrecht in
Holland. Of course the nest was on the wrong side of the canal, and
a nearer approach was impossible without a ducking; so one view
was made with a twelve-inch lens and then the telephoto was used,
although not much was expected, for there was a stiff Holland
breeze blowing, which is not conducive to perfect results, and
moreover the storks seemed inclined to greater activity than well-
behaved birds of this species generally exhibit. The exposure was
almost instantaneous and the result a surprise to the operator.
The Jungfrau from the Höheweg, Interlaken, Switzerland (sixteen miles
distant).

Another example of the curious uses to which this lens may be

put is seen in the illustration of the beautiful memorial column at
West Point [p. 468]. The general view was taken at a distance of
about three hundred feet in the ordinary way and then a telephoto
was made of the bronze figure of Victory which surmounts the
column.
Inaccessible parts of fine architecture offer an endless series of
subjects for telephoto work, where remarkable results may be
obtained. The cathedral at Milan, since the removal of the buildings
which formerly obstructed the view, now appears to great advantage
when viewed from the opposite side of the piazza. The two views of
this beautiful structure [pp. 458-59] were made from a second story
window on the opposite side of the piazza. I chose this point of view
because of the enormous dimensions of the building. I first used the
ordinary lens, obtaining the general view, and then
telephotographed various portions of it.
The cathedral at Florence is so shut in by adjacent buildings that
it must receive other treatment. The vast amount of work upon the
façade is lost to the casual observer because of the propinquity of
the baptistery, which completely destroys the effect of this wonderful
mosaic. Standing on the sidewalk, as far from the façade as the
other buildings would permit, I made the general view of it with a
lens of four inches focus, then retreating still farther, till a corner of
the baptistery began to interfere, I used the telephoto attachment
on the central rose-window, the camera being about a hundred
yards from it [pp. 460-61]. Then taking a position beside the
baptistery I telephotographed the mosaic over the central door. It
will be noticed that in the telephotographs there is less distortion
than in the ordinary view, for although the rose-window is over a
hundred feet above the pavement it was photographed from such a
distance that only a slight inclination of the camera was necessary,
and the picture appears as if taken from an elevation, whereas it
was actually made from the sidewalk. The delicate carving and
mosaic work about the central door are distinctly brought out, and it
is one of the best examples of telephoto work the attachment has
made.
 The Jungfrau from the same Standpoint (sixteen miles distant),
Telephoto Lens.

At Venice one turns instinctively toward the grand Piazza, the

Mecca of many a traveller as well as of the Venetians themselves. St.
Mark's Cathedral offers many studies for the camera, and for many
years the glass mosaics upon the upper part of the front of the
building were a perplexing problem to me, for the balcony was too
near and the pavement below was too far away for successful work
with the ordinary lens; and if taken from a near position below they
were so distorted as to be useless. The problem was not solved till
the advent of the telephoto attachment, which procured the studies
with ease. After making a picture of the whole front of the cathedral
from the centre of the piazza in the ordinary way, the camera was
moved a little to the right, so that one of the large flag-poles would
not interfere, and the upper left-hand mosaic, representing the
"Descent from the Cross," was telephotographed [pp. 466-67]. The
result is about the same as that which might have been obtained in
the ordinary way from the top of a scaffold fifty feet high and about
forty feet from the mosaic.

A Cocoanut-tree, St. Kitts, British West Indies.

 Cocoanuts on the Tree, St. Kitts, British West Indies—Telephoto Lens.

As all the illustrations mentioned were made with the idea of

reproducing them as lantern slides, which are only about three
inches square, they do not indicate the full power of the attachment
in a single case. Therefore I placed my camera near a window in one
of my rooms and photographed the row of dwellings across the back
yards [p. 465]. The actual distance from the camera to the first
dwelling is one hundred and thirty feet; to the chimney, one hundred
and fifty-four feet. Then, after putting on the telephoto attachment
and extending the front of the camera as far as the bellows would
permit, I telephotographed one of the chimneys on the third house,
the only change in the camera being a slight inclination so that the
chimney would be in the centre of the plate. This picture, when
compared with that taken with the ordinary lens, shows an
enlargement of nearly sixteen diameters, which is considerably more
than the capacity or power of a very large field-glass.