(eBook PDF) Principles of Biostatistics 2nd

Edition instant download

https://ebooksecure.com/product/ebook-pdf-principles-of-
biostatistics-2nd-edition/

Download more ebook from https://ebooksecure.com

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

(eBook PDF) Fundamentals of Biostatistics 8th Edition

http://ebooksecure.com/product/ebook-pdf-fundamentals-of-
biostatistics-8th-edition/

Fundamentals of Biostatistics 8th Edition (eBook PDF)

http://ebooksecure.com/product/fundamentals-of-biostatistics-8th-
edition-ebook-pdf/

Basic Biostatistics: Statistics for Public Health

Practice 2nd Edition (eBook PDF)

http://ebooksecure.com/product/basic-biostatistics-statistics-
for-public-health-practice-2nd-edition-ebook-pdf/

(Original PDF) Biostatistics for the Biological and

Health Sciences 2nd Edition

http://ebooksecure.com/product/original-pdf-biostatistics-for-
the-biological-and-health-sciences-2nd-edition/
Principles of Biology 2nd Edition (eBook PDF)

http://ebooksecure.com/product/principles-of-biology-2nd-edition-
ebook-pdf/

(eBook PDF) Principles of Pathophysiology, 2nd Edition

http://ebooksecure.com/product/ebook-pdf-principles-of-
pathophysiology-2nd-edition/

(eBook PDF) Principles of Pathophysiology 2nd Edition

http://ebooksecure.com/product/ebook-pdf-principles-of-
pathophysiology-2nd-edition-2/

Basic Biostatistics For Medical And Biomedical

Practitioners 2nd Edition Julien I.E. Hoffman - eBook
PDF

https://ebooksecure.com/download/basic-biostatistics-for-medical-
and-biomedical-practitioners-ebook-pdf/

(eBook PDF) Principles of Cognitive Neuroscience 2nd

Edition

http://ebooksecure.com/product/ebook-pdf-principles-of-cognitive-
neuroscience-2nd-edition/
 Preface vii

Over and above their precision, there is something more to numbers-maybe a

little magic-that makes them fun to study. The fun is in the conceptualization more
than the calculations, and we are fortunate that we have the computer to do the drudge
work. This allows students to concentrate on the ideas. In other words, the computer al-
lows the instructor to teach the poetry of statistics and not the plumbing.

Computing
To take advantage of the computer, one needs a good statistical package. We use Stata,
which is available from the Stata Corporation in College Station, Texas. We find this
statistical package to be one of the best on the market today; it is user-friendly, accu-
rate, powerful, reasonably priced, and works on a number of different platforms, in-
cluding Windows, Unix, and Macintosh. Furthermore, the output from this package is
acceptable to the Federal Drug Administration in New Drug Approval submissions.
Other packages are available, and this book can be supplemented by any one of them.
In this second edition, we also present output from SAS and Minitab in the Further Ap-
plications section of each chapter. We strongly recommend that some statistical pack-
age be used.
Some of the review exercises in the text require the use of the computer. To help
the reader, we have included the data sets used in these exercises both in Appendix B
and on a CD at the back of the book. The CD contains each data set in two different for-
mats: an ASCII file (the "raw" suffix) and a Stata file (the "dta" suffix). There are also
many exercises that do not require the computer. As always, active learning yields bet-
ter results than passive observation. To this end, we cannot stress enough the importance
of the review exercises, and urge the reader to attempt as many as time permits.

New to the Second Edition

This second edition includes revised and expanded discussions on many topics through-
out the book, and additional figures to help clarify concepts. Previously used data sets,
especially official statistics reported by government agencies, have been updated when-
ever possible. Many new data sets and examples have been included; data sets described
in the text are now contained on the CD enclosed with the book. Tables containing exact
probabilities for the binomial and Poisson distributions (generated by Stata) have been
added to Appendix A. As previously mentioned, we now incorporate computer output
from SAS and Minitab as well as Stata in the Further Applications sections. We have
also added numerous new exercises, including questions reviewing the basic concepts
covered in each chapter.

Acknowledgements
A debt of gratitude is owed a number of people: Harvard University President Derek
Bok for providing the support which got this book off the ground, Dr. Michael K. Martin
for calculating Tables A.3 through A.8 in Appendix A, and John-Paul Pagano for
viii Preface

assisting in the editing of the first edition. We thank the individuals who reviewed the
manuscript: Rick Chappell, University of Wisconsin; Dr. Todd G. Nick, University of
Mississippi Medical Center; Al Bartolucci, University of Alabama at Birmingham;
Bruce E. Trumbo, California State University, Hayward; James Godbold, The Mount
Sinai School of Medicine of New York University; and Maureen Lahiff, University of
California, Berkeley. Our thanks to the teaching assistants who have helped us teach the
course and who have made many valuable suggestions. Probably the most deserving of
thanks are the students who have taken the course over the years and who have toler-
ated us as we learned how to teach it. We are still learning.

Marcello Pagano
Kimberlee Gauvreau
Boston, Massachusetts
Contents

l Introduction
1. 1 Overview of the Text 2
1.2 Review Exercises 5
Bibliography 6

2 Data Presentation 7
2.1 Types of Numerical Data 7
2.1.1 Nominal Data 7
2.1.2 Ordinal Data 9
2.1 .3 Ranked Data 10
2.1.4 Discrete Data 10
2.1.5 Continuous Data 11
2.2 Tables 11
2.2.1 Frequency Distributions 12
2.2.2 Relative Frequency 13
2.3 Graphs 15
2.3.1 Bar Charts 15
2.3.2 Histograms 16
2.3.3 Frequency Polygons 18
2.3.4 One-Way Scatter Plots 20
2.3.5 Box Plots 21
2.3 .6 Two-Way Scatter Plots 22
2.3.7 Line Graphs 22
2.4 Further Applications 24
2.5 Review Exercises 30
Bibliography 36

ix
x Contents

3 Numerical Summary Measures 38

3. 7 Measures of Central Tendency 38
3.1.1 Mean 38
3.1.2 Median 41
3.1.3 Mode 42
3.2 Measures of Dispersion 44
3.2.1 Range 44
3.2.2 Interquartile Range 44
3.2.3 Variance and Standard Deviation 46
3.2.4 Coefficient of Variation 48
3.3 Grouped Data 48
3.3.1 Grouped Mean 49
3.3.2 Grouped Variance 51
3.4 Chebychev's Inequality 52
3.5 Further Applications 54
3.6 Review Exercises 59
Bibliography 64

4 Rates and Standardization 66

4. 7 Rates 66
4.2 Standardization of Rates 70
4.2.1 Direct Method of Standardization 72
4.2.2 Indirect Method of Standardization 74
4.2.3 Use of Standardized Rates 75
4.3 Further Applications 84
4.3.1 Direct Method of Standardization 86
4.3.2 Indirect Method of Standardization 86
4.4 Review Exercises 89
Bibliography 95

5 Life Tables 97
5. 7 Computation of the Life Table 97
5.1.1 Column 1 97
5.1.2 Column 2 99
 Contents xi

5.1.3 Columns 3 and 4 101

5.1.4 Column 5 102
5.1.5 Column 6 103
5.1.6 Column 7 103
5.2 Applications of the Life Table 04
5.3 Years of Potential Life Lost 707
5.4 Further Applications 111
5.5 Review Exercises 116
Bibliography 124

6 Probability 125
6. 7 Operations on Events and Probability 125
6.2 Conditional Probability 129
6.3 Bayes' Theorem 1 131
6.4 Diagnostic Tests 135
6.4.1 Sensitivity and Specificity 136
6.4.2 Applications of Bayes' Theorem 136
6.4.3 ROC Curves 140
6.4.4 Calculation of Prevalence 141
6.5 The Relative Risk and the Odds Ratio 144
6.6 Further Applications 149
6.7 Review Exercises 155
Bibliography 160

7 Theoretical Probability Distributions 162

7. 7 Probability Distributions 162
7.2 The Binomial Distribution 164
7.3 The Poisson Distribution 172
7.4 The Normal Distribution 176
7.5 Further Applications 185
7.6 Review Exercises 191
Bibliography 194
xii Contents

B Sampling Distribution of the Mean 196

B. 1 Sampling Distributions 196
8.2 The Central Limit Theorem 197
8.3 Applications of the Central Limit Theorem 198
8.4 Further Applications 204
8.5 Review Exercises 210
Bibliography 213

9 Confidence Intervals 214

9. 1 Two-Sided Confidence Intervals 214
9.2 One-Sided Confidence Intervals 219
9.3 Student's t Distribution 220
9.4 Further Applications 225
9.5 Review Exercises 227
Bibliography 230

J0 Hypothesis Testing 232

10. 1 General Concepts 232
10.2 Two-Sided Tests of Hypotheses 235
10.3 One-Sided Tests of Hypotheses 238
10.4 Types of Error 239
10.5 Power 243
10.6 Sample Size Estimation 246
10.7 Further Applications 249
10.8 Review Exercises 254
Bibliography 257

JJ Comparison of Two Means 259

11. 1 Paired Samples 260
11 .2 Independent Samples 265
11.2.1 Equal Variances 266
11.2.2 Unequal Variances 270
 Contents xiii

77.3 Further Applications 272

77.4 Review Exercises 278
Bibliography 282

J2 Analysis of Variance 285

72. 7 One-Way Analysis of Variance 285
12.1.1 The Problem 285
12.1.2 Sources ofVariation 288
72.2 Multiple Comparisons Procedures 292
72.3 Further Applications 294
72.4 Review Exercises 298
Bibliography 301

13 Nonparametric Methods 302

73. 7 The Sign Test 302
73.2 The Wilcoxon Signed-Rank Test 305
73. 3 The Wilcoxon Rank Sum Test 308
73.4 Advantages and Disadvantages of
Non parametric Methods 3 122
73.5 Further Applications 3 12
73.6 Review Exercises 3 17
Bibliography 321

J4 Inference on Proportions 323

74. 7 Normal Approximation to the Binomial Distribution 324
74.2 Sampling Distribution of a Proportion 325
74. 3 Confidence Intervals 327
74.4 Hypothesis Testing 329
74.5 Sample Size Estimation 330
74.6 Comparison of Two Proportions 332
74.7 Further Applications 335
74.8 Review Exercises 338
Bibliography 341
xiv Contents

l5 Contingency Tables 342

15. 1 The Chi-Square Test 342
15.1.1 2 X 2 Tables 342
15.1.2 r X c Tables 347
15.2 McNemar's Test 349
15.3 The Odds Ratio 352
15.4 Berkson's Fallacy 357
15.5 Further Applications 360
15.6 Review Exercises 366
Bibliography 372

J6 Multiple 2 x 2 Tables 374

16. 1 Simpson's Paradox 374
16.2 The Mantei-Haenszel Method 376
16.2.1 Test of Homogeneity 377
16.2.2 Summary Odds Ratio 381
16.2.3 Test of Association 384
16.3 Further Applications 387
16.4 Review Exercises 393
Bibliography 396

J7 Correlation 398
17. 1 The Two-Way Scatter Plot 398
17.2 Pearson's Correlation Coefficient 400
17.3 Spearman's Rank Correlation Coefficient 404
17.4 Further Applications 407
17.5 Review Exercises 412
Bibliography 414

JB Simple Linear Regression 4 15

18. 1 Regression Concepts 415
 Contents xv

78.2 The Model 420

18.2.1 The Population Regression Line 420
18.2.2 The Method of Least Squares 422
18.2.3 Inference for Regression Coefficients 425
18.2.4 Inference for Predicted Values 428
78.3 Evaluation of the Model 432
18.3.1 The Coefficient of Determination 432
18.3.2 Residual Plots 433
18.3.3 Transformations 435
78.4 Further Applications 438
78.5 Review Exercises 443
Bibliography 447

J9 Multiple Regression 449

79. 7 The Model 449
19.1.1 The Least-Squares Regression Equation 450
19 .1.2 Inference for Regression Coefficients 452
19 .1.3 Evaluation of the Model 453
19.1.4 Indicator Variables 455
19 .1.5 Interaction Terms 457
79.2 Model Selection 458
79.3 Further Applications 460
79.4 Review Exercises 465
Bibliography 469

20 Logistic Regression 470

20.7 The Model 471
20.1.1 The Logistic Function 472
20.1.2 The Fitted Equation 473
20.2 Multiple Logistic Regression 476
20.3 Indicator Variables 478
20.4 Further Applications 481
20.5 Review Exercises 484
Bibliography 487
xvi Contents

21 Survival Analysis 488

2 7. 7 The Life Table Method 489
27.2 The Product-Limit Method 495
21.3 The Log-Rank Test 499
27.4 Further Applications 503
21.5 Review Exercises 511
Bibliography 51 2

22 Sampling Theory 514

22.7 Sampling Schemes 514
22.1.1 Simple Random Sampling 515
22.1.2 Systematic Sampling 515
22.1.3 Stratified Sampling 516
22.1.4 Cluster Sampling 517
22.1.5 Nonprobability Sampling 517
22.2 Sources of Bias 51 7
22.3 Further Applications 520
22.4 Review Exercises 524
Bibliography 525

Appendix A
Tables A-1

Appendix B
Data Sets B-1

Index 1-1
 Introduction

In 1903, H. G. Wells hypothesized that statistical thinking would one day be as neces-
sary for good citizenship as the ability to read and write. Statistics do play an important
role in many decision-making processes. Before a new drug can be marketed, for in-
stance, the United States Food and Drug Administration requires that it be subjected to
a clinical trial, an experimental study involving human subjects. The data from this
study must be compiled and analyzed to determine whether the drug is not only effec-
tive, but safe. In addition, the U.S. government's decisions regarding Social Security
and public health programs rely in part on predictions about the longevity of the na-
tion's population; consequently, it must be able to predict the number of years that each
individual will live. Many other issues need to be addressed as well. Where should a
government invest its resources if it wishes to reduce infant mortality? Does the use of
a seat belt or an air bag decrease the chance of death in a motor vehicle accident? Should
a mastectomy always be recommended to a patient with breast cancer? What factors in-
crease the risk that an individual will develop coronary heart disease? To answer these
questions and others, we rely on the methods of biostatistics.
The study of statistics explores the collection, organization, analysis, and inter-
pretation of numerical data. The concepts of statistics may be applied to a number of
fields that include business, psychology, and agriculture. When the focus is on the bio-
logical and health sciences, we use the term biostatistics.
Historically, statistics have been used to tell a story with numbers. Numbers often
communicate ideas more succinctly than do words. The message carried by the follow-
ing data is quite clear, for instance. In 1979, 48 persons in Japan, 34 in Switzerland, 52
in Canada, 58 in Israel, 21 in Sweden, 42 in Germany, 8 in England, and 10,728 in the
United States were killed by handguns [1]. The power of these numbers is obvious; the
point would be made even if we were to correct for differences in population size.
As a second example, consider the following quotation, taken from an editorial in
The Boston Globe [2]:

Lack of contraception is linked to an exceptionally high abortion rate in the Soviet

Union-120 abortions for every 100 births, compared with 20 per 100 births in
2 Chapter 1 Introduction

Great Britain, where access to contraception is guaranteed. Inadequate support for

family planning in the United States has resulted in 40 abortions for every 100
births-a lower rate than the Soviet Union, but twice as high as most industrialized
nations .

In this case, a great deal of information is contained in only three numbers: 120, 20, and
40. The statistics provide some insight into the consequences of differing attitudes to-
ward family planning.
In both these examples, the numbers provide a concise summary of certain as-
pects of the situation being studied. Surely the numerical explanation of the handgun
data is more illuminating than if we had been told that some people got killed in Japan,
fewer in Switzerland, more in Canada, still more in Israel, but far fewer in Sweden, and
so forth. Both examples deal with very complex situations, yet the numbers convey the
essential information. Of course, no matter how powerful, no statistic will convince
everyone that a given conclusion is true. The handgun data are often brushed away with
the aphorism "Guns don't kill people, people do." This should not be surprising; after
all, there are still members in the Flat Earth Society. The aim of a biostatistical study is
to provide the numbers that contain information about a certain situation and to present
them in such a way that valid interpretations are possible.

l. l Overview of the Text

If we wish to study the effects of a new diet, we might begin by measuring the changes
in body mass over time for all individuals who have been placed on the diet. Similarly,
if we wanted to investigate the success of a certain therapy for treating prostate can-
cer, we would record the lengths of time that men treated with this therapy survive be-
yond diagnosis with the disease. These collections of numbers, however, can display a
great deal of variability and are generally not very informative until we start combin-
ing them in some way. Descriptive statistics are methods for organizing and summa-
rizing a set of data that help us to describe the attributes of a group or population. In
Chapter 2, we examine tabular and graphical descriptive techniques. The graphical ca-
pabilities of computers have made this type of summarization more feasible than in the
past, and a whole new mode of presentation is available for even the most modest
analyses.
Chapter 3 goes beyond the graphical techniques presented in Chapter 2 and in-
troduces numerical summary measures. By definition, a summary captures only a par-
ticular aspect of the data being studied; consequently, it is important to have an idea of
how well the summary represents the set of measurements as a whole. For example, we
might wish to know how long AIDS patients survive after diagnosis with one of the op-
portunistic infections that characterize the disease. If we calculate an average survival
time, is this average then representative of all patients? Furthermore, how useful would
the measure be for planning future health service needs? Chapter 3 investigates de-
scriptive techniques that help us to answer questions such as these.
 I.I Overview of the Text 3

Data that take on only two distinct values require special attention. In the health
sciences, one of the most common examples of this type of data is the categorization of
being either alive or dead. If we denote the former state by 0 and the latter by 1, we are
able to classify a group of individuals using these numbers and then to average the re-
sults. In this way, we can summarize the mortality associated with the group. Chapter
4 deals exclusively with measurements that assume only two values. The notion of di-
viding a group into smaller subgroups or classes based on a characteristic such as age
or gender is introduced as well. We might wish to study the mortality of females sepa-
rately from that of males, for example. Finally, this chapter investigates techniques that
allow us to make valid comparisons among groups that may differ substantially in com-
position.
Chapter 5 introduces the life table, one of the most important techniques available
for study in the health sciences. Life tables are used by public health professionals to
characterize the well-being of a population, and by insurance companies to predict how
long a particular individual will live. In this chapter, the study of mortality begun in
Chapter 4 is extended to incorporate the actual time to death for each individual; this
results in a more refined analysis. Knowing these times to death also provides a basis
for calculating the survival curve for a population. This measure of longevity is used
frequently in clinical trials designed to study the effects of various drugs and surgical
treatments on survival time.
In summary, the first five chapters of the text demonstrate that the extraction of
important information from a collection of numbers is not precluded by the variability
among them. Despite this variability, the data often exhibit a certain regularity as well.
For example, if we look at the annual mortality rates of teenagers in the United States
for each of the last ten years, we do not see much variation in the numbers. Is this just
a coincidence, or is it indicative of a natural underlying stability in the mortality rate?
To answer questions such as this, we need to study the principles of probability.
Probability theory resides within what is known as an axiomatic system: we start
with some basic truths and then build up a logical system around them. In its purest
form, the system has no practical value. Its practicality comes from knowing how to use
the theory to yield useful approximations. An analogy can be drawn with geometry, a
subject that most students are exposed to relatively early in their schooling. Although it
is impossible for an ideal straight line to exist other than in our imaginations, that has
not stopped us from constructing some wonderful buildings based on geometric calcu-
lations. The same is true of probability theory: although it is not practical in its pure
form, its basic principles-which we investigate in Chapter 6---can be applied to pro-
vide a means of quantifying uncertainty.
One important application of probability theory arises in diagnostic testing. Un-
certainty is present because, despite their manufacturers' claims, no available tests are
perfect. Consequently, there are a number of important questions that must be answered.
For instance, can we conclude that every blood sample that tests positive for HIV actu-
ally harbors the virus? Furthermore, all the units in the Red Cross blood supply have
tested negative for HIV; does this mean that there are no contaminated samples? If there
are contaminated samples, how many might there be? To address questions such as
these, we must rely on the average or long-term behavior of the diagnostic tests; prob-
ability theory allows us to quantify this behavior.
4 Chapter I Introduction

Chapter 7 extends the notion of probability and introduces some common proba-
bility distributions. These mathematical models are useful as a basis for the methods
studied in the remainder of the text.
The early chapters of this book focus on the variability that exists in a collection
of numbers. Subsequent chapters move on to another form of variability-the variabil-
ity that arises when we draw a sample of observations from a much larger population.
Suppose that we would like to know whether a new drug is effective in treating high
blood pressure. Since the population of all people in the world who have high blood
pressure is very large, it is extremely implausible that we would have either the time or
the resources necessary to examine every person. In other situations, the population may
include future patients; we might want to know how individuals who will ultimately de-
velop a certain disease as well as those who currently have it will react to a new treat-
ment. To answer these types of questions, it is common to select a sample from the
population of interest and, on the basis of this sample, infer what would happen to the
group as a whole.
If we choose two different samples, it is unlikely that we will end up with pre-
cisely the same sets of numbers. Similarly, if we study a group of children with con-
genital heart disease in Boston, we will get different results than if we study a group of
children in Rome. Despite this difference, we would like to be able to use one or both
of the samples to draw some conclusion about the entire population of children with
congenital heart disease. The remainder of the text is concerned with the topic of sta-
tistical inference.
Chapter 8 investigates the properties of the sample mean or average when re-
peated samples are drawn from a population, thus introducing an important concept
known as the central limit theorem. This theorem provides a foundation for quantifying
the uncertainty associated with the inferences being made.
For a study to be of any practical value, we must be able to extrapolate its find-
ings to a larger group or population. To this end, confidence intervals and hypothesis
testing are introduced in Chapters 9 and 10. These techniques are essentially methods
for drawing a conclusion about the population we have sampled, while at the same time
having some knowledge of the likelihood that the conclusion is incorrect. These ideas
are first applied to the mean of a single population. For instance, we might wish to es-
timate the mean concentration of a certain pollutant in a reservoir supplying water to
the surrounding area, and then determine whether the true mean level is higher than the
maximum concentration allowed by the Environmental Protection Agency. In Chapter
11, the theory is extended to the comparison of two population means; it is further gen-
eralized to the comparison of three or more means in Chapter 12. Chapter 13 continues
the development of hypothesis testing concepts, but introduces techniques that allow the
relaxation of some of the assumptions necessary to carry out the tests. Chapters 14, 15,
and 16 develop inferential methods that can be applied to enumerated data or counts-
such as the numbers of cases of sudden infant death syndrome among children put to
sleep in various positions-rather than continuous measurements.
Inference can also be used to explore the relationships among a number of different
attributes. If a full-term baby whose gestational age is 39 weeks is born weighing 4 kilo-
grams, or 8.8 pounds, no one will be surprised. If the baby's gestational age is only 22
 1.2 Review Exercises 5

weeks, however, then his or her weight will be cause for alarm. Why? We know that birth
weight tends to increase with gestational age, and, although it is extremely rare to find a
baby weighing 4 kilograms at 22 weeks, it is not uncommon at 39 weeks. The study of
the extent to which two factors are related is known as correlation analysis; this is the
topic of Chapter 17. If we wish to predict the outcome of one factor based on the value of
another, regression is the appropriate technique. Simple linear regression is investigated
in Chapter 18, and is extended to the multiple regression setting-where two or more
factors are used to predict a single outcome-in Chapter 19. If the outcome of interest
can take on only two possible values, such as alive or dead, an alternative technique must
be applied; logistic regression is explored in Chapter 20.
In Chapter 21 , the inferential methods appropriate for life tables are introduced.
These techniques enable us to draw conclusions about the mortality of a population
based on a sample of individuals drawn from the group.
Finally, Chapter 22 examines an issue that is fundamental in inference-the con-
cept of the representativeness of a sample. In any study, we need to be confident that
the sample we choose provides an accurate picture of the population from which it is
drawn . Several different methods for selecting representative samples are described.
The notion of bias and various problems that can arise when choosing a sample are dis-
cussed as well. Common sense plays an important role in sampling, as it does through-
out the entire book.

1.2 Review Exercises

1. Design a study aimed at investigating an issue that you believe might influence the
health of the world. Briefly describe the data that you will require, how you will
obtain them, how you intend to analyze the data, and the method you will use to
present your results. Keep this study design and reread it after you have completed
the text.

2. Consider the following quotation regarding rapid population growth [3]:

512 million people were malnourished in 1986-1987, up from 460 mil-

lion in 1979-1981.

(a) Suppose that you agree with the point being made. Justify the use of these
numbers.
(b) Are you sure that the numbers are correct? Do you think it is possible that 513
million people were malnourished in 1986-1987, rather than 512 million?

3. In addition to stating that "the Chinese have eaten pasta since 1100 B.c.," the label
on a box of pasta shells claims that "Americans eat 11 pounds of pasta per year,"
whereas "Italians eat 60 pounds per year." Do you believe that these statistics are
accurate? Would you use these numbers as the basis for a nutritional study?
6 Chapter I Introduction

Bibliography
[1] McGervey, J.D., Probabilities in Everyday Life, Chicago: Nelson-Hall, 1986.
[2] "The Pill's Eastern Europe Debut," The Boston Globe, January 19, 1990, 10.
[3] United Nations Population Fund, "Family Planning: Saving Children, Improving Lives,"
New York: Jones & Janello.
 Data
Presentation

Every study or experiment yields a set of data. Its size can range from a few measure-
ments to many thousands of observations. A complete set of data, however, will not nec-
essarily provide an investigator with information that can easily be interpreted. For
example, Table 2.1 lists by row the first 2560 cases of acquired immunodeficiency syn-
drome (AIDS) reported to the Centers for Disease Control and Prevention [1] . Each in-
dividual was classified as either suffering from Kaposi's sarcoma, designated by a 1, or
not suffering from the disease, represented by a 0. (Kaposi's sarcoma is a tumor that af-
fects the skin, mucous membranes, and lymph nodes.) Although Table 2.1 displays the
entire set of outcomes, it is extremely difficult to characterize the data. We cannot even
identify the relative proportions of Os and 1s. Between the raw data and the reported re-
sults of the study lies some intelligent and imaginative manipulation of the numbers,
carried out using the methods of descriptive statistics.
Descriptive statistics are a means of organizing and summarizing observations.
They provide us with an overview of the general features of a set of data. Descriptive
statistics can assume a number of different forms ; among these are tables, graphs, and
numerical summary measures. In this chapter, we discuss the various methods of dis-
playing a set of data. Before we decide which technique is the most appropriate in a
given situation, however, we must first determine what kind of data we have.

2. J Types of Numerical Data

2. l. l Nominal Data
In the study of biostatistics, we encounter many different types of numerical data. The dif-
ferent types have varying degrees of structure in the relationships among possible values.
One of the simplest types of data is nominal data, in which the values fall into unordered
categories or classes. As in Table 2.1, numbers are often used to represent the categories. In
a certain study, for instance, males might be assigned the value 1 and females the value 0.
7
8 Chapter 2 Data Presentation

TABLE 2.J
Outcomes indicating whether an individual had Kaposi's sarcoma for the first 2560 AIDS
patients reported to the Centers for Disease Control and Prevention in Atlanta, Georgia
00000000 00010100 00000010 00001000 00000001 00000000 10000000 00000000
00101000 00000000 00000000 00011000 00100001 01001100 00000000 00000010
00000001 00000000 00000010 01100000 00000000 00000100 00000000 00000000
00100010 00100000 00000101 00000000 00000000 00000001 00001001 00000000
00000000 00010000 00010000 00010000 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00001000 00000000 00010000 10000000 00000000
00100000 00000000 00001000 00000010 00000000 00000100 00000000 00010000
00000000 00000000 00000100 00001000 00001000 00000101 00000000 01000000
00010000 00000000 00010000 01000000 00000000 00000000 00000101 00100000
00000000 00000000 00000100 00000000 01000100 00000000 00000001 10100000
00000100 00000000 00010000 00000000 00001000 00000000 00000010 00100000
00000000 00000000 00000000 10001000 00001000 00000000 01000000 00000000
00000000 00001100 00000000 00000000 10000011 00000001 11000000 00001000
00000000 00000000 00000000 00000000 01000000 00000001 00010001 00000000
10000000 00000000 01000000 00000000 00000000 01010100 00000000 00010100
00000000 00000000 00000000 00001010 00000101 00000000 00000000 00010000
00000000 00000000 00000000 00000001 00000100 00000000 00000000 00001000
11000000 00000100 00000000 00000000 00000000 00000000 00000000 00001000
11000000 00010010 00000000 00001000 00000000 00111000 00000001 01001100
00000000 01100000 00100010 10000000 00000000 00000010 00000001 00000000
01000010 01000100 00000000 00010000 00000000 01000000 00000001 00000000
01000000 00000001 00000000 10000000 01000000 00000000 00000000 00000100
00000000 00000000 01000010 00000000 00000000 00000000 00000000 00000000
00000000 00000010 00001010 00001001 10000000 00000000 00000010 00000000
00000000 01000000 00000000 00001000 00000000 01000000 00010000 00000000
00001000 01000010 01001111 00100000 00000000 00100000 00000000 10000001
00000001 00000000 01000000 00000000 00000000 00000000 00000000 01000000
00000000 00000000 00100000 01000000 00100000 00000000 00000011 00000000
01000000 00000100 10000001 00000001 00001000 00000100 00001000 00001000
00100000 00000000 00000000 00000000 00000010 01000001 00010011 00000000
00000000 10000000 10000000 00000000 00000000 00001000 01000000 00000000
00001000 00000000 01000010 00011000 00000001 00001001 00000000 00000001
01000010 01001000 01000000 00000010 00000000 10000000 00000100 00000000
00000010 00000000 00000000 00000010 00000000 00100100 00000000 10110100
00001100 00000100 00001010 00000000 00000000 00000000 00000000 00000000
00000010 00000000 00000000 00000000 00100000 10100000 00001000 00000000
01000000 00000000 00000000 00100000 00000000 01000001 00010010 00010001
00000000 00100000 00110000 00000000 00010000 00000000 00000100 00000000
00010100 00000000 00001001 00000001 00000000 00000000 00000000 00000000
00000010 00000100 01010100 10000001 00001000 00000000 00010010 00010000

Although the attributes are labeled with numbers rather than words, both the order and the
magnitudes of the numbers are unimportant. We could just as easily let 1 represent females
and 0 designate males. Numbers are used mainly for the sake of convenience; numerical
values allow us to use computers to perform complex analyses of the data.
Discovering Diverse Content Through
Random Scribd Documents
were from 16 to 24 years of age. Of the Negro women 16
years of age and over engaged in domestic and personal
service, 35.1 per cent, or more than one-third, were
between the ages of 16 and 24. The percentage in the other
age groups of the total number of women 16 years of age
and over engaged in domestic and personal service
decreased by classes. That of Negro women 16 years of age
and over engaged in domestic and personal service
decreased by classes until those 55 years of age and over
constituted only 9.6 per cent of the total number of Negro
women so employed. The modal age of Negro male
domestic workers like that of white male domestic workers
was from 25 to 44 years. The age distribution of domestic
and personal service workers for 1920 is about the same as
that for 1900. Because of the incompleteness of the age
data obtained from the general schedule sent to
employment agencies, they were not used for this study.
The average ages of the 9,976 male and female Negro
domestic and personal service workers of Washington, D.
C., were: 30.5 years for the males and 28.1 years for the
females.
In 1900, among Negro women the percentage of
breadwinners did not show such a marked decline after
marriage as among white women. Of the Negro female
breadwinners 32.5 per cent were married, while only 9.0
per cent of the female breadwinners of all the races were
married. The percentage of married Negro male domestic
and personal service workers is higher than that of married
female workers, while the number of widowed and divorced
is three and one-half times as great among female as
among male domestic and personal service workers. In
1920, 29.4 per cent of all the female domestic and personal
service workers 15 years of age and over were married,
while 70.6 per cent were classed as single, widowed,
divorced, and unknown.
The significance of age grouping and marital condition of
Negro domestic workers in their relation to employers is
borne out by the testimony of experienced employment
agents in New York City, Philadelphia, Baltimore,
Washington, D. C., Chicago, and Detroit. Women domestic
workers between the ages of 20 and 25 are the most
sought after by employers. Those between 25 and 35 years
of age are next in favor. All of the agents testified to the
unpopularity of the young girl domestic worker. She is
employed principally because of the tight domestic labor
market. Employers apparently feel that a majority of the
women beyond the ages of 45 and 50 have become too set
in their ways, somewhat cranky, and largely unable to do
general housework. The most frequent objections of
employers to young girl domestic workers are: They are
untrained and inexperienced; they are unwilling to sleep in;
they are saucy; and their interest in men company causes
them to neglect their work.
The older Negro women in domestic service, realizing that
with their advancing years their possibilities for employment
become less, often hesitate and even fail to give their
correct ages when applying at employment agencies for
positions. For example, a New York City agency registered a
woman who gave her age as 34, but whose written
references, yellowed with age, showed that she had worked
for different members in one family for fifty years.
Frequently an older woman registrant when asked her age
hesitates and ends by saying "just say 'settled woman.'"
In addition to the age situation of Negroes engaged in
domestic service, the marital condition of female domestic
workers furnishes a perplexing problem for both their
employers and themselves. The testimony of employment
agents relative to employers' most commonly registered
objections to hiring married women for domestic service is:
Married women take away food for the support of their
families; married women have so many responsibilities and
problems in their own homes they oftener than not go out
to work with a weary body and a disturbed mind; married
women find it difficult to live and sleep on employers'
premises.
Besides these problems there is apparently a still more
perplexing one for the Negro domestic workers with children
of their own or other dependents, namely, how to provide
proper care and protection for their dependents while they
are away from home at work, especially if the hours are
long. Day nurseries are often mentioned as a possible
solution for this particular problem, but they exist for
Negroes in very few cities of the South. Even in the District
of Columbia with a population of servants and waiters—
servants largely Negroes—totaling 21,444, there is not one
day nursery for Negro children. The other alternative is to
get some elderly woman to take care of a child. The usual
charge made by such a woman for a limited number of
hours during the day is from $5 to $6 a week, the mother
furnishing food for the child. With these two items and
carfare deducted from a mother's weekly wage of $9 there
is little left for other necessities.
The problem of dependents manifests itself also among
widowed and divorced Negro women engaged in domestic
service. The U. S. Employment Office, Washington, D. C.,
registered 9,774 Negro women 15 years of age and over for
domestic service from January, 1920, to May, 1922. Of this
number 5,124 were single, 2,579 were married, 2,071 were
widowed or divorced. Of the widowed or divorced 2,056 had
from 1 to 5 dependents; 79 had from 6 to 10 dependents.
Although no record was made of the number of
breadwinners in each of these families, many of these
widows expressed their weight of responsibilities by
referring to the high cost of living when their children had
no one to look to for support but themselves.
Divorced domestic workers and also unmarried mothers
constitute marital groups that are not all together
negligible. Three of the divorced women sent from the
Washington office had the added problem of finding their
husbands at their respective places of employment after
absences of 5, 2, and 2 years respectively. Among the
5,124 single women registered at the Washington office
there were reported 9 unmarried mothers.
In the District of Columbia there is a Training School for
delinquent Negro girls, a large number of whom go into
domestic service when they are paroled. They are better
trained than the average domestic employee, but since the
Training School requires them to keep their young babies
with them, it is difficult to place them in homes. If they take
a room and attempt to do day work they have the difficult
problem of getting someone to take care of their children.
The marital condition of 471 new applicants for domestic
positions in Indianapolis, Indiana, for 1922, is given in the
following table:
Showing marital condition of 471 women seeking work as
domestic servants— Indianapolis, Ind., 1922
Table II
Widows 63
Separated from husbands 50
Married and living with husbands 238
Divorced 34
Single 85
Unmarried mothers 1

The large proportion of married persons in the table may be

accounted for, in part, by the fact that 51 per cent of the
total number had recently come into Indianapolis from the
adjoining States of Kentucky and Tennessee.
 III. Turnover, Training, and Efficiency of
Negro Domestic and Personal Service
Workers
The increase of 13,738,354, or 14.9 per cent, in the total
population of the United States during the last decade, and
a decrease of 367,667 in the domestic and personal service
occupations population increases the possibilities of
turnover. In 1890, the average tenure of service of a
domestic worker in the United States was less than one and
one-half years.[5] Ten years later the average length of
service of a Negro domestic worker in the seventh ward of
Philadelphia was five years less than one month.[6] Many of
these workers perhaps had been for a long time in the older
families of Philadelphia. Figures for a three-year period,
from 1906 to 1908, show that the modal period of service
of the New York Negro domestic worker was at that time
from six to eleven months.[7] In 1914, among 104 unskilled
Negro workers of St. Louis—cooks, laundresses, porters,
chambermaids, waiters, scrubwomen, manual laborers, and
the like—the greatest frequency for length of service among
the men was from one to three months, and among the
women from three to six months.[8] Six years later the
largest proportion of Negro domestic workers of Gainesville,
Georgia, showed a disposition to remain in one position less
than three months, while the next largest proportion
remained in one position from three to six months.[9]
Some concrete illustrations of the frequency of turnover
may be referred to as further evidence. Nearly two hundred
different women were sent out from the Springfield,
Massachusetts, office for day work in 1915. Two years later
over 500 different workers were sent out from that office,
about 200 of whom were Negro women. Of these 167 white
women and 124 Negro women were placed with employers
less than ten times in 1917; 2 white and 4 Negro women
were sent out from 41 to 50 times; 3 white and 1 Negro
woman were placed fifty times during 1917. In 1918, the
Springfield office reported as having filled 4,000 places with
1,000 women.[10]
In 1920 the United States Employment Office, Washington,
D. C., placed 1,488 different women for day work, all of
whom were Negroes except one. Of these, 458, after being
given permanent positions for every day in the week, were
referred again not over twelve times; 23 of them were sent
out over fifty times, and 5 of them over one hundred times
during the year. General housework was so unpopular
during that year that few would take it. Although the
turnover in day work was greater than that in any other
specific employment handled by the domestic service
section of that office, the 164 cooks remained in one
position on an average of about three months.
There was, however, in the District of Columbia during the
fall and winter of 1920 a decrease in the rate and volume of
turnover for Negro day workers and hotel workers
consequent upon the minimum wage law which became
effective for hotels and restaurants in the spring of 1920.
Many Negro women displaced in the hotels turned to day
work. For this reason added to the normal increase during
the first half of 1921, the number of day workers increased
to 3,115. White workers did not have to apply for day work
because they could secure positions in hotels and
restaurants.
During an unemployment period which extended over the
latter half of 1921 and the first half of 1922, when so many
men were thrown out of work, the day workers increased to
4,615. There were more day workers than there were
positions for them. Consequently the turnover in day work
decidedly decreased, but it increased in general housework.
Many who could not get a sufficient number of days' work to
make ends meet were forced to turn to general housework.
The difficulty of keeping an accurate record of the turnover
in general housework makes the value of figures on
turnover in general housework seem very questionable.
However, the length of service of 1,000 general
houseworkers sent out from the Washington office for the
latter half of 1921 and the first few months of 1922 gives a
fairly accurate picture of the situation at that time.
Length of Service of 1,000 General Houseworkers,
Washington, D. C., 1921-22
Table III
317 remained in one position 1 week or less.
582 remained in one position from one to three months.
101 remained in one position 4 months and over.

Some of the conditions of the turnover at that time are

illustrated by typical cases. One employer advertised for a
general houseworker without laundry, with the privilege of
going at night, and with hours off. Eleven domestic workers
answered the advertisement in person, and no two met at
the house at the same time. The employer engaged every
one of them, to return to work the next morning. Every one
of them gave assurance of being there, but not one of them
came. Another employer with only two in family, very
desirous of securing a general houseworker, called up six
different employment agencies, each of which sent her a
worker. Among the number there was one man. The
employer put every one of them to work, each in a different
room, with the idea of choosing the two best ones at the
end of the day and engaging them for permanent work—
thus assuring herself of securing one worker. She managed
her plan, as she thought, very successfully, but the next
morning she did not have a single worker.
Employers' statements on reference blanks, and reports
from employment agencies indicate that the reasons
generally given by domestic workers for leaving positions
are that they wish a change, or the hours are too long, or
the work is too hard, or the employer is too particular, or
they have no time off. Time off for domestic employees is
no doubt greatly limited. Negro domestic workers, however,
proverbially take Christmas Day and the Fourth of July off,
giving such various excuses for their absence as death in
the family, automobile accidents, and the like. Just after
these two holidays, large employment agencies handling
Negro help are for the most part swamped with applicants.
To some extent turnover in domestic service is linked up
with lack of training and efficiency of domestic workers.
Because of their great need of domestic help, employers
frequently engage persons who are so utterly untrained that
they cannot be retained. There is a tendency on the part of
employers to propose to a domestic worker that each take
the other on a week's trial. Domestic workers are inclined to
refuse such offers on the ground that they are looking for
permanent employment. This suggestion of trying out
domestic workers leads logically to the question of training
and efficiency in domestic service.

Training of White Domestic and Personal Service Workers

Some facts relative to the special opportunities for the

training of white household workers in England and in the
United States may throw some light on the problem of
efficiency. In England, following the World War and under
the ministry of reconstruction, there was created a women's
advisory committee to study the domestic service problem.
Each of the four sub-committees appointed made a report.
Among the advisory committee's final recommendations for
getting the work of the nation's homes done satisfactorily
and reducing waste, were technical training for domestic
help and fixed standards of qualifications for them. This
committee reported that in 1914 there were only ten
domestic service schools in England and Wales, and four of
these were in the London area. During the year of 1922
courses of three months' duration were given at some
technical institutions in England in all branches of household
work and management. This training enabled women to
take the better posts in daily or residential work. Training in
cooking and catering could be had at any technical college
for three or more months as required.
To help meet the serious unemployment situation the
Central Committee on Women's Training and Employment in
cooperation with the Ministry of Labor set up homecraft
training centers in districts where unemployment was most
noticeable. At these centers a course of training was given
for about three months, such as would enable women to
take posts in domestic service at the end of the course.
These classes were most successful. By August, 1922,
about 10,000 women had received the training and the
courses were still continued. These courses were given to
women and girls between the ages of 16 and 35 upon their
signing an agreement to be punctual in attendance, to do
their best in making the classes successful, and stating their
willingness to enter domestic service after receiving their
training.
In the United States, in 1900, there were more illiterate
persons in domestic and personal service than in any other
field of employment except agriculture. The number of
agricultural colleges in the different States for the purpose
of developing improved farming and farmers has increased
since that time, and the Federal Government farm
demonstration agents are actually teaching the citizens on
the plantations where they live and work. Facilities for the
training of domestic help, however, have received little
attention from State or Federal Government, and private
enterprise in this field has been very limited.
Just twenty years ago the Home Economics Committee of
the Association of Collegiate Alumnae of the United States,
through the inspiration of Mrs. Alice Freeman Palmer and
Mrs. Ellen H. Richards, undertook an experiment for
studying at first hand the problems of household labor.
Among the disabilities in domestic service regarded as
fundamental causes of the disfavor in which it was held,
was the low grade of intelligence and skill available among
domestic service workers. This household aid company
committee opened a training center and applied educational
tests to its candidates and undertook to give a course of six
weeks' training to each aid before she was sent out to work.
The number of aids taking the training was small. Mainly
because of a lack of funds, however, the experiment was
given up after a trial of two years. Prior to that time the
civic club of Philadelphia attempted to standardize the work
and wages of domestic workers in that city.
Nearly twenty years later, a committee on household
assistants was organized in New York City by the United
States Employment Service. The committee succeeded in
planning a household occupations course to be given at the
Washington Irving High School, and made efforts to
advertise the course; but since no one registered for it the
committee concluded that the matter would have to be
taken up and pushed by employers before it could succeed.
[11]

Some of these efforts, however, have met with a measure

of success. The Bureau of Household Occupations of the
Housewives' League of Providence, R. I., organized in
November, 1918, has conducted very successful training
classes for domestic workers. No meals or lodging are to be
furnished the household attendants of the Bureau. The
Bureau of Occupations under the auspices of the
Housewives' League of Hartford, Connecticut, has given its
training courses through the generosity of some of its
members. One member taught cooking, another taught
waiting table, another laundry work. Classes were taught in
the homes of some of the members with much success.

Training of Negro Domestic Workers

Available data shows that opportunities for the special

training of Negro domestic workers have been even less
than those for white domestic workers. During the latter
quarter of the 19th century Mrs. L. J. Coppin, of
Philadelphia, maintained a small home for the training of
the Negro domestic workers of Philadelphia. In the
comparatively few social settlements for Negroes there is
meagre opportunity for training in domestic service. The
Domestic Efficiency Association of Baltimore, Maryland, an
organization of employers, has announced its plans for
opening a training school for white and Negro domestic
workers. This Association maintained in 1921 and 1922 a
training school for Negro domestic help, in which special
lessons could be given or general training for one month or
more. A rate of $5 a week for board, lodging, and training
was charged. If an applicant had no money the Domestic
Efficiency Association advanced it on her signing an
agreement to secure her position through the Association
when ready for it, and to repay the debt out of her wages at
the rate of at least $2.50 a week.
The domestic science training given in the public schools
may be a small factor in the efficiency of Negro domestic
workers, but most of the permanent domestic workers do
not go beyond the fifth grade in school and thus do not go
far enough to get an appreciable amount of domestic
science training. Negro workers who go through the high or
normal schools do not enter permanently into domestic
service. This statement is based on the data indicated by
the permanent occupations of 606 Negro graduates of the
Sumner High School, St. Louis, Missouri, of 305 graduates
of Miner Normal School, Washington, in the District of
Columbia, of 15 graduates of the Gainesville, Georgia,
public schools 1917-1919;[12] and on data for students
applying at the Washington, in the District of Columbia, and
the Indianapolis Employment Agencies. Tables IV and V
below set forth these facts.
Occupations of 606 Negro High School Graduates, Sumner
High School, St. Louis, Mo., 1895-1911[13]
Table IV
Occupation Number
Those engaged in, or prepared for, teaching 288
Entered college 49
Clerical work 43
Postoffice clerks 30
Entered business 4
Mechanics 17
Women at home or married 120
Miscellaneous 32
Unknown 23

Although Tables IV and V direct one's attention to the

limited fields of employment for Negro high school
graduates, especially so since clerical and mechanical work,
business and professional service, must be engaged in
almost wholly among Negroes, yet few if any of the 911
graduates have entered domestic service. The young
women graduates of the Gainesville, Georgia, schools 1917-
19, with the exception of three, entered higher institutions
of learning.
In Washington, in the District of Columbia, during the
academic year 1920-22 there were among the 9,976
applicants for domestic work, 17 male and 159 female
students who had attended or were attending high school;
75 female normal school students; 13 male and 126 female
college students. Also in Indianapolis, in 1922, 73 female
high school students and 12 female college students applied
for domestic service. These large numbers of high school,
normal school, and college students seek domestic service
mainly for after-school hours, Saturdays, Sundays, summer
months, and temporarily for earning money to continue
their education, or until they can find other employment.
Occupations of 305 Negro Graduates of Miner Normal
School, Washington, D. C., 1913-1922
Table V
Occupation Number
Teaching in Washington, D. C.:
Elementary 207
Kindergarten 50
Domestic Science 4
Domestic Art 3
Manual Arts 1
Drawing 1
Music 1
Ungraded 1
Teaching in Maryland 8
Teaching in Virginia 2
Teaching in North Carolina 1
Teaching in South Carolina 1
Teaching in New York 1
Substitute teachers in Washington, D. C. 2
Students 5
Government Service 7
Housekeepers 5
Printers 1
Private Music Teachers 1
Physicians 1
Insurance 1
 Y. W. C. A. 1

Table VI shows the grades on leaving school of 8,147 Negro

domestic workers—men and women—of the Washington, D.
C., office; and Table VII shows grades on leaving school of
471 Negro domestic workers, not separated by sex, of an
Indianapolis Employment Office conducted by Flanner
House in that city. Each of these workers was personally
interviewed by the agent at each respective office. The
reported grade of each on leaving school was placed on an
application card which was filed for reference. The
application cards were filled out solely on the testimony of
the applicants. The agent in the Washington office handling
the women did not ordinarily register men except as man
and wife applied at the same time, or a woman sent her
husband to the agent, or a special employer asked the
agent to select male help, or teachers in the Negro schools
sent boys and men who were in search of work. Therefore,
the number of men from the Washington office for whom
grades are given is comparatively small.
In examining Tables VI and VII below one must take into
consideration several factors. In the first place, 81.2 per
cent of the Washington applicants and 73.9 per cent of the
Indianapolis applicants were born in the South where the
standard is not so high as in the North; and many of these
applicants attended school in the rural districts of the South
where the schools were not standardized, and only a few
schools had any domestic science instruction. Then, too, a
large proportion of them left school some years ago when
all of the grades or groups of a school were taught by one
teacher in one room.
Those persons who could not read or write seemed to feel
their illiteracy very keenly. Many of them offered excuses by
saying that the "white folks raised" them; or their parents
died and they had to help the other children; or they were
"sickly," and the like. Those who had never been to school
but could read and write a little were listed as being in the
first grade. One applicant said that she had never been
through any grade but she could read and write and go
anywhere in the city she wished to go. Another one, an
elderly woman, expressed her regrets because she never
had a chance to go to school, but she had learned to read
and write so that she could sign her name instead of simply
"touching the pen" when she was transacting her business.
Grades on Leaving School of 7,975 Female and 172 Male
Negro Domestic Workers from the U. S. Employment
Service, Washington, D. C., 1920-1922
Table VI
Male Female
Illiterate 8 418
1st Grade 5 244
2d Grade 7 436
3d Grade 9 842
4th Grade 17 1,073
5th Grade 31 1,417
6th Grade 28 1,237
7th Grade 25 998
8th Grade 42 1,310

Grades on Leaving School of 387 Negro Domestic Workers,

Irrespective of Sex, Indianapolis, Ind., 1922
Table
VII
1st 2d 3d 4th 5th 6th 7th 8th
Illiterate
Gr. Gr. Gr. Gr. Gr. Gr. Gr. Gr.
21 7 11 22 44 63 51 47 120

The figures show that of a total of 7,975 female applicants

for domestic work in Washington, D. C., 4,430, or 55.5 per
cent, had received school training in the sixth grade or
below; leaving only 29.9 per cent who had seventh or
eighth grade training. Of the 387 applicants for domestic
service in Indianapolis, 168, or 43.3 per cent, had received
school training up to the fifth grade or below; and 219, or
56.7 per cent, had been to the sixth grade or below, leaving
43.3 per cent who had been in the seventh or eighth grade.
The larger proportions of those from higher grades in
Indianapolis may be accounted for by the lesser opportunity
in other occupations as compared with Washington, and by
the smaller number of applicants involved. In short,
domestic service as a regular occupation does not attract
and hold Negro workers of the higher grades of educational
training and intelligence.
In order to understand exactly what is meant by saying that
consideration of certain factors must be taken into account
in any attempt to formulate some idea of the educational
status of the rank and file domestic worker reckoned by his
grade when he left school, some letters, typical of the
educational equipment among the 9,774 domestic workers
(applicants), should be read. These letters were written to
the agent in the Washington, D. C., office by 5th grade
domestic workers.[14]
Many of these domestic workers also showed their lack of
training by their inability to figure out their weekly wages at
the rate of $40, 45, or $50 per month. Such inability often
caused them to feel and say that their employers were
"cheaters." To a considerable number of them, $40 a month
meant $10 a week, and vice versa; $45 a month meant
$11.25 a week, and $50 a month meant $12.50 a week.
They generally secured their pay twice a month—the first
and the fifteenth. However, such an arrangement did not
seem to clarify matters, since they thought of four weeks as
making a month.
Then comes the question of the efficiency of Negro domestic
workers. In Philadelphia, Baltimore, Detroit, Indianapolis,
and Washington, D. C., agents find that employers of
domestic labor, like other employers, do not like to write
down their grievances, but many of them do make
complaints to the agents over the telephone about the
inefficiency of domestic help. Agents in Detroit and
Indianapolis state that Negro domestic workers from the
South—many of them from the farms and untrained,
unaccustomed to Northern methods of domestic work—find
it difficult to give satisfaction. The consensus of opinion of
eleven white and Negro agents in New York City was that
with respect to efficiency there are three distinct types of
domestic workers in New York City. In the first place, comes
the West Indian, who is unaccustomed to domestic work,
and therefore unable to convince himself that he is on that
plane. He makes a more or less inefficient domestic worker.
Then there is the New York Negro who has difficulty in
adjusting himself to domestic duties. The southern Negro,
however, a decidedly different sort of laborer, makes a more
efficient domestic worker than either of the other two types.
Opinions elsewhere also vary. There was a migration of
Negro women domestic workers from Georgia to
Springfield, Massachusetts, in 1916-1917. Many of these
women were very satisfactory employees and compared
favorably with northern born Negro women domestic
workers of that locality, according to the 11th Annual Report
of the Massachusetts Bureau of Statistics. In the United
States Employment Office, Washington, D. C., where all
sorts and conditions of domestic workers were handled,
reports from employers on the efficiency of the new workers
from the South indicated that they were unaccustomed to
modern methods of housework and were less efficient than
northern born workers.
In any attempt to rate the efficiency of Negro domestic
workers by verbal testimonials and written references from
their employers or by wages received or length of service
period of the workers, due consideration must be given to
factors beyond the workers' control. Some of these factors
are differences in the standards of efficiency in the many
homes and the temperament of employers together with
the attitude of some employers toward Negroes generally.
For example, occasionally, a former employer, in sympathy
with the struggles of Negroes and not wishing to hinder an
unsatisfactory worker from securing another position, writes
for her a letter of recommendation. Sometimes another
employer, because of misunderstanding of some sort
between her and the worker, refuses to give any reference
whatever.
In 1890, 57 per cent of 1,005 housekeepers representing
the whole United States found more or less difficulty in
securing efficient help. This probably was an underestimate
of the true condition.[15] In 1901, out of 1,106 domestic
workers from all sections of the United States, 34 per cent
were rated excellent; 37.4 per cent good; 24.8 per cent
fair; 3.8 per cent poor. Although these figures indicate that
96.2 per cent of the total were between excellent and fair,
the Commission's report in summing up the matter states
that according to the testimony of employers of domestic
labor and of employment agents, the character of the
service rendered by domestic laborers is in a large
proportion of cases unsatisfactory. It further states that the
quality of men's work is about the same as that of women's
work.[16]
In New York City, employment agencies send reference
blanks to former employers of domestic workers to be filled
out and returned.[17] These references are kept on file as a
record of the domestic worker's capability, sobriety and
honesty. From 1906 to 1909 efficiency ratings taken from
such blanks for 902 Negro domestic and personal service
workers were as follows: 25.6 per cent very capable; 10.2
per cent fairly capable; 2.2 per cent inefficient, and 2.0 per
cent not stated.[18] One employment agency in this city
made 304 placements of Negro women domestic workers
during January, 1923. According to those workers'
references from their former employers 93.3 per cent were
capable or fairly capable and honest. This high degree of
efficiency among domestic workers from this one office is
due probably to the fact that this office with its limited staff
of secretaries makes no attempt to handle the evidently
inexperienced workers. The other employment agencies in
New York and Brooklyn visited in 1923 spoke favorably of
the quality of service rendered by domestic workers in these
cities, according to their reports from employers.
Opinions of employers are not conclusive evidence of the
efficiency or inefficiency of workers, but they throw
considerable light upon the question. Written references are
more or less held in disfavor by the Washington, D. C.,
employers of domestic labor because they feel that
domestic workers sometimes write their own references.
This is true to a limited extent. Many of the workers come
from small towns and rural sections where the employers of
domestic labor do not use elegant stationery, the best
English, and the most correct spelling in writing references
for domestic workers who leave for the cities. Such
references do domestic workers coming to Washington, D.
C., more harm than good.
However, domestic workers are more and more seeking
written references on leaving their places of employment
because they are beginning to realize that such are
generally required by employers. Often a former employer
has moved away from the city, is in Europe, or has died,
when the domestic worker needs most to refer to her. A
prospective employer usually doubts that such an excuse, if
given, is true. Of course, some workers do try to take
advantage in this way, but most of them are not so unwise.
Types of written and oral testimonials of employers of
domestic labor in Washington, D. C., are also informing.[19]
In cases where three or more employers testified to the
efficiency or inefficiency of a worker, the word "efficient,"
"inefficient," or "poor" was written across the bottom of his
application card. The following table in some measure
represents in detail the character of service reported to the
United States Employment Service, Domestic Section.
Summary of Testimonials of Former Employers of 9,976
Wage Earners Engaged in Domestic Personal Service,
Washington, D. C., January 1920-May 1922

Efficient Fairly Efficient Inefficient

Table VIII
No. Per ct. No. Per ct. No. Per ct.
Male 90 44.6 94 46.5 11 19.4
Female 3,008 30.8 4,543 46.5 1,892 .05
Total 3,098 37.7 4,637 46.5 1,903 9.7

No Report
No. Per Cent.
Male 7 .03
Female 331 .03
Total 338 .03

In this table 44.6 per cent of the males as over against 30.8
per cent of the females are reported as being efficient, while
19.4 per cent of the females and only 0.05 per cent of the
males are listed as inefficient. This should not lead to the
conclusion that the male Negro domestic workers of
Washington, D. C., were more efficient than the female
Negro domestic workers of that city, since the 202 male
domestic workers do not represent the rank and file. They
represent men of family responsibilities, and students
working their way through high school and college. Both of
these groups had a more or less definite responsibility and
aim in doing domestic work and therefore were more
willing, at least for a time, to accommodate themselves to
conditions obtaining in it. The office received no report
concerning .03 per cent of the workers. Occasionally both
employer and employee were so well pleased with each
other that neither was heard from unless the office in its
follow-up work discovered the happy situation.
The opinion of employers that 19.4 per cent of 9,773 Negro
female domestic workers of Washington, D. C., were
reported inefficient does not, without other data, justify this
as a scientific conclusion. Some typical examples of their
inefficiency are interesting.[20] The inefficiency is due in
large measure to pure ignorance which for the most part is
the sequel to lack of opportunity and training. For example,
the older type cook, who cannot read and write, finds it
difficult, if not impossible, to carry all the different modern
salad and dessert combinations in her memory and cannot
supplement her instructions by the use of literature on
domestic science.
Employment agencies in Chicago in 1923, moreover, have
hardly told the whole truth in giving the following figures on
the efficiency of 200 female domestic workers and 200 male
domestic workers: Women, satisfactory 175, or 87.5 per
cent; unsatisfactory 10, or 5.0 per cent; neither satisfactory
nor wholly unsatisfactory 15, or 7.5 per cent. Men,
satisfactory 125, or 62.5 per cent; unsatisfactory 45, or
22.5 per cent; neither 30, or 15 per cent.
Efficient domestic workers apparently regret that they are in
an occupational group representing such a high degree of
ignorance and inefficiency. They sometimes take pride in
saying that they have never worked for poor people. Such a
class of workers is represented by a Washington, D. C.,
domestic worker who gave as her former employers Mrs.
John Hays Hammond, Mrs. Arthur Glasgow, Senator
Beveridge, Senator Guggenheim, and President Wilson. She
took pride in the fact that she could even show anyone a
piece of the president's wedding cake.
Honesty in domestic service is so closely associated with
efficiency that practically no reference for a domestic
worker is complete without some statement about this
qualification. In 1890 Miss Salmon raised a serious question
with regard to the honesty of Negro domestic workers in the
South. Her question was based on answers received from
schedules sent to employers of that section.[21] In 1901,
92.6 per cent of 583 domestic labor employers representing
the whole United States testified that their employees were
honest and responsible. Most employment bureaus were
also agreed upon the general honesty of domestic workers.
[22] In 1899 the Philadelphia Negro domestic worker of the
Seventh Ward was described as purloining food left from the
table but as having the balance in his favor in regard to
honesty.[23] In 1906 opinions of former employers of 902
Negro wage-earners in domestic and personal service in
New York City were that 91.3 per cent were honest; 7.1 per
cent were either honest or fairly so; 0.6 per cent were
dishonest, and no statement was given for 1.0 per cent.[24]
Out of 9,638 Negro domestic workers reported upon for
Washington, D. C, between the years of 1920-1922, only .2
per cent were rated by their former employers with
assurance as being dishonest; 90.4 per cent were listed as
being honest. There were various answers for the 9.4 per
cent. Some did not remain long enough to have judgment
passed upon them. Others were in a doubtful class but with
no proof against them, and the like. This low percentage of
dishonesty eliminates the tradition of taking food except in
seven cases. The seven cases of food taking are included
because they were directly reported and regarded by the
employers as dishonest. Some employers, according to their
own statement of the case, do not regard taking food left
from the table as stealing, although such is against the will
of the employer. According to the southern tradition of a low
wage and taking food to piece it out, domestic workers are
still virtually expected to follow this custom.
200 women and 200 men domestic workers of Chicago have
the following record for honesty: Women, honest, 199, or
99.5 per cent; dishonest, 1, or 0.5 per cent; men, honest,
197, or 98.5 per cent; dishonest, 3, or 1.5 per cent.
Employment agents in other leading cities already
mentioned have very little complaint against the honesty of
Negro domestic workers except in the matter of taking food.
Their explanation of the psychology of such dishonesty is as
given above.

IV. Wages, Hours, and Specific Occupations

While wages and efficiency are in some degree related in
domestic service, there is the custom of paying the "going
wage" for specific occupations, irrespective of efficiency.
Wages vary, of course, in different sections of the country
and in different localities. Occasionally attempts are made
to grade such laborers. One employment bureau, in
Indianapolis, for example, divides its day workers into
grades A and B with respective wages of 30 cents and 25
cents an hour for each grade.
Two other questions current in the problem of wages in
domestic service, both of which seem to be slowly lending
themselves to adjustment, are the payment of weekly
wages instead of bi-weekly or monthly wages, and equal
pay for equal work irrespective of whether a man or a
woman, a Negro or a white employee, does the work. Bi-
monthly payment in domestic service has come to be the
custom due largely to the convenience of the employer, and
to the possibility of weekly wages increasing the turnover. A
domestic worker often leaves unceremoniously as soon as
he gets his first pay. However, workers claim that the
custom of bi-weekly or monthly pay inconveniences them
since they cannot arrange to pay their rent, or purchase
clothing and other necessities on that basis.
The question of equal pay for Negro domestic workers does
not enter the domestic service wage problem of the South
because Negroes pre-empt this field in that section.
Although the scarcity of domestic labor seems to be settling
this matter in other sections of the country, it still persists
in some measure. Twenty-five years ago Miss Eaton
discovered that Negro butlers on Rittenhouse Square,
Philadelphia, received on an average $36.90 a month, while
white butlers were getting from $40 to $45 a month.[25]
In the fall of 1921, during the period of labor depression,
eleven of the Washington, D. C., clubs and expensive
boarding houses attempted to make a change from Negro
to white chambermaid-waitresses at an increase of $10 a
month for each worker. The four clubs that succeeded in
making the change discharged their white chambermaid-
waitresses after one week each and re-employed Negroes
at the old wage of $35 a month. One of the successful
employers felt that, inasmuch as the white servants were
no more satisfactory than the Negro workers, she had just
as well keep the Negroes and pay them less.
When the minimum wage law for women and minors of
Washington, D. C., recently declared unconstitutional by the
United States Supreme Court, went into effect, practically
all of the hotels and restaurants in that city immediately
discharged Negro workers and took on white ones. Some of
the managers told the agent at the United States
Employment Bureau that they were making the change
because white servants were more efficient than Negro
workers. Other managers, some of whom had used Negro
labor for more than fifteen years, simply said that $16.50 a
week was too much to pay Negroes, and, therefore, wished
white workers instead. The few hotels and restaurants that
retained Negroes as a rule put them on a much shorter
working week than 48 hours, thus reducing their pay.
Boarding houses and institutions such as private schools,
sanatoria, and the like, that offer excuses and fail to pay
workers should be mentioned in this connection. The
manager of one such boarding house in Washington, D. C.,
was sued by a worker who won her case because other
unpaid laborers testified against the manager. The
superintendent of a small private school in that city—also
among such paymasters—had repeatedly been reported to
the Minimum Wage Board which forced her to pay the Negro
women day workers. After a few months of such experience
she changed her help and began to employ men, over
whom the Minimum Wage Board had no jurisdiction.
The wages of Negro domestic workers today are
considerably higher than they were in past decades, as is
shown by a comparison of figures in past periods for the
Continental United States and for selected cities with figures
in 1920-1923. For the twenty-five years prior to the World
War there had been only a slight wage increase in domestic
and personal service. During the World War there was a
considerable increase in wages for both male and female
domestic workers, the increase for the latter being larger
than that for the former. Since the World War wages for
such workers have fallen to some extent but not anywhere
near the pre-war level.
The following tables, with one exception, show the wage
changes at different ten-year periods over a range of 30
years. In Table IX the figures from the Boston Employment
Bureau illustrate the fact that the average weekly wages for
female domestic workers of Boston were decidedly higher
than elsewhere in the country. This table also makes clear
the fact that wages for men were considerably higher than
those for women.
Average Daily and Weekly Wages in Selected Domestic
Service Occupations, 1889-1890[26]
 Table IX
Weekly Wages for Weekly Wages for
Occupation
the United States Boston, Mass.
Women
Cooks $3.72 $4.45
Cooks and laundresses 3.39
Chambermaids 3.39 3.86
Waitresses 3.19 3.7
Second girls 3.16 3.7
Chambermaids and waitresses 3.10
Parlor maids 3
General servants 2.91 3
Men
Coachmen $7.84
Coachmen and gardeners 6.54
Butlers 6.11
Cooks 6.08
Women Daily Wages
Laundresses .82
Seamstresses 1.01
Men
Gardeners 1.33
Chore-men .87

Table X below gives average wages for selected domestic service occupations in the
United States for a decade later than the figures of Table IX. The slight variation in
the figures of Table X from those of Table IX may be due to probable error incident
to the collection of the data or to some other factor. The indications of these two
tables, however, with ten years intervening between the compilation of the data, are
that wages probably had changed very little, if any.
Average Weekly Wages for Selected Domestic Service Occupations in the United
States, 1900[27]
Table X
Occupation Average Weekly Wage
Women
General houseworkers $3.28
Cooks 3.95
Waitresses 3.43
Other specialists 3.54
Men
For all domestic service occupations 6.03
Women
For all domestic service occupations 3.51

In comparison with the two preceding tables, Table XI below gives wages for
domestic service in Philadelphia for about the same period. The weekly wages range
higher than for the country as a whole. The lower wages in the southern border and
middle sections of the United States have reduced the average for the country below
that for this eastern city in which also special conditions may have operated to bring
such wages above the general level.
Average Weekly Wages of Negro Domestic Workers of Philadelphia, 1896-1897[27]
 Table XI
Occupation Average Weekly Wage
Women
General worker $3.24
Janitress 4.06
Chambermaid-laundress 3.58
Cook-laundress 4.00
Laundress 4.04
Lady's maid 3.63
Chambermaid and waitress 3.17
Waitress 3.31
Women
Chambermaid 3.17
Child's nurse 3.35
Errand girl 2.00
Cook 4.02
Men
General worker 5.38
Valet 8.00
Cook 6.17
Waiter 6.14
Coachman 8.58
Butler 8.24
Bellboy 2.61

Table XII which follows is drawn from The Negro at Work in New York City, and
shows the modal wage groups for specific occupations in domestic and personal
service, New York City, 1906-1909. Although data for New York City are not typical
of the entire country, these are the only available figures for this period, and they
may indicate the trend of wages in domestic personal service in that section. In
comparison with the preceding Table of Wages in Philadelphia, the increase in wages
in New York City may be due to differences of conditions in the two cities rather than
to any general increase or decrease in wages.
Modal Wage Groups for Selected Occupations, 1906-1909[28]
Table XII
Occupation Range of Modal Wage
Female
Switchboard operator $4.00-4.99
Chambermaid 4.00-4.99
Chambermaid-cook 5.00-5.99
Chambermaid-laundress 5.00-5.99
Chambermaid-waitress 4.00-4.99
Kitchenmaid 4.00-4.99
Cook 5.00-5.99
Cook and general worker 5.00-5.99
Cook-waitress 4.00-4.99
Cook-laundress 5.00-5.99
Errand girl Less than 4.00
General houseworker 4.00-4.99
Laundress 4.00-4.99
Lady's maid 4.00-4.99
Parlor maid 4.00-4.99
Nurse Less than 3.00
 Pantry girl 4.00-4.99
Waitress 4.00-4.99
Dishwasher 4.00-4.99
Male
Bellman Less than 4.00
Butler-cook 5.00-5.99
Waiter 5.00-5.99
Butler 5.00-5.99
Coachman 5.00-5.99
Cook 5.00-5.99
Elevator operator 5.00-5.99
Furnaceman 5.00-5.99
Gardener 4.00-4.99
Hallman and doorman 4.00-4.99
Houseman 5.00-5.99
Janitor 5.00-5.99

The last decade embraces the World War when wages in domestic and personal
service were at their maximum. The following tables for selected cities present
graphically the increase in wages for male and female domestic workers and the
slight increase in wages of females over that of males. These tables also show how
wages vary in different sections of the country. Although these figures are for 1920,
and the first quarter of 1921, the decline in wages generally did not begin until the
fourth quarter of 1920, and it was not so pronounced in domestic and personal
service as in many other occupational groups, and was scarcely appreciable in
domestic service until the middle of 1921.
Tables XIII, XIV, XV, XVI, and XVII indicate that although wages in domestic and
personal service among Negroes have fallen somewhat, they are still far above
those of pre-war times. They also show that since the War there has been
considerable decline in rates paid men for day work in New York City and
Washington, D. C., but very little decrease in the rates for women day workers in
either of the two cities.
Any analysis of these tables must take into consideration that female day workers in
the cities included in the tables receive their carfare and at least one meal; cooks,
general houseworkers, waiters and waitresses, housemen, mothers' helpers, some
kitchen help, part-time workers and nurses receive their meals and, in many
instances, their quarters.
In this table wages for clerical workers, factory workers, laborers, truckers,
butchers, etc., are given in comparison with the wages of domestic and personal
service workers. For example: a stenographer receives $18 a week, while a cook
receives from $18 to $25 a week and board; a factory girl receives from 25 cents to
30 cents an hour, while a day worker in domestic service receives $22 a week, and a
cook receives $25 a week and board.
Weekly Wages of 118 Negro Men in Domestic Service by Specified Occupations, New
York City, 1920-1921[29]
Table XIII
Occupations Number Employed Weekly Wages
Cleaners 3 $ .50 per hour
5 3.00 per day
Cooks 2 15.00-17.99
 3 18.00-19.99
3 25.00 or more
Dishwashers 2 10.00-12.99
4 13.00-14.99 and meals
1 15.00-17.99
11 18.00-21.99
1 26.00
Doormen 1 38.50 and meals
3 (monthly) 40.00-79.00
Elevator operators (apt. house) 1 under 10.00
1 10.00-12.99
11 15.00-17.99
1 18.00-21.99
Elevator and switchboard operators 6 14.00
6 17.00
1 18.00
Firemen (apt. house) 1 3.00 per day
1 20.00-24.99
1 20.00 and board
1 30.00
Janitors (apt. house) 1 (monthly) 20.00 and apartment
1 (monthly) 30.00 and keep
1 (monthly) 40.00 and keep
1 (monthly) 60.00 and keep
Assistant janitors (apt. house) 1 10.00-12.99
1 15.00 and room
Porters-apartment houses 1 16.00
6 18.00-20.99
Waiters 3 under 10.00
18 (exclusive
of tips) 15.00-17.99
6 18.00-20.99
7 10.00-11.99

Range of Weekly Wages of 754 Negro Women in Domestic and Personal Service,
Specified by Occupations, New York City, 1920-1921[30]
Table XIV
Occupations Number Employed Wages
General houseworkers 5 under $ 9.00
706 10.00-18.00
Chambermaids 1 under 9.00
Chambermaids-waitresses 7 12.00-18.00
Cooks 6 15.00-21.00
Kitchen helpers 8 12.00-17.00
2 under 9.00
Mothers' helpers and Nurses 9 10.00-15.00
Nurses (practical) 3 15.00-21.00
Waitresses 5 12.00-14.00

Range of Daily and Weekly Wages of 1,565 Male and Female Negro Domestic and
Personal Service Workers, Washington, D. C., 1920
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about testbank and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebooksecure.com