Full download test bank at ebook textbookfull.

com

Statistics For The Behavioural

Sciences An Introduction To
Frequentist And Bayesian Approaches

CLICK LINK TO DOWLOAD

https://textbookfull.com/product/statistics-
for-the-behavioural-sciences-an-introduction-
to-frequentist-and-bayesian-approaches-2nd-
edition-riccardo-russo/

textbookfull
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Reasoning with Data An Introduction to Traditional and

Bayesian Statistics Using R 1st Edition Jeffrey M.
Stanton

https://textbookfull.com/product/reasoning-with-data-an-
introduction-to-traditional-and-bayesian-statistics-using-r-1st-
edition-jeffrey-m-stanton/

Fundamental Approaches to Software Engineering

Alessandra Russo

https://textbookfull.com/product/fundamental-approaches-to-
software-engineering-alessandra-russo/

Introduction to Biological Physics for the Health and

Life Sciences 2nd Edition Kirsten Franklin

https://textbookfull.com/product/introduction-to-biological-
physics-for-the-health-and-life-sciences-2nd-edition-kirsten-
franklin/

Modern statistics for the social and behavioral

sciences : a practical introduction Second Edition.
Edition Wilcox

https://textbookfull.com/product/modern-statistics-for-the-
social-and-behavioral-sciences-a-practical-introduction-second-
edition-edition-wilcox/
The Bayesian way: introductory statistics for
economists and engineers First Edition Nyberg

https://textbookfull.com/product/the-bayesian-way-introductory-
statistics-for-economists-and-engineers-first-edition-nyberg/

An introduction to nonparametric statistics First

Edition Kolassa

https://textbookfull.com/product/an-introduction-to-
nonparametric-statistics-first-edition-kolassa/

Statistics for the behavioral sciences Gravetter

https://textbookfull.com/product/statistics-for-the-behavioral-
sciences-gravetter/

Statistics for the Life Sciences 5th Edition Samuels

https://textbookfull.com/product/statistics-for-the-life-
sciences-5th-edition-samuels/

Foundations and Applications of Statistics An

Introduction Using R 2nd Edition Randall Pruim

https://textbookfull.com/product/foundations-and-applications-of-
statistics-an-introduction-using-r-2nd-edition-randall-pruim/
Statistics for the Behavioural Sciences

This accessible textbook is for those without a mathematical background (just some
notions of basic algebra are sufficient) and provides a comprehensive introduction to all
topics covered in introductory behavioural sciences statistics courses. It includes plenty
of real examples to demonstrate approaches in depth based on real psychology experi-
ments utilising the statistical techniques described.
New content in this thoroughly updated second edition includes an introduction to
Bayesian statistics which complements the coverage of Classical/Frequentist statistics
present in the first edition. It also offers practical details on how to perform analyses
using JASP – a globally employed, freely downloadable statistical package. The updated
eResources also feature a range of new material including additional exercises so read-
ers can test themselves on what they have learned in the book.
This timely and highly readable text will be invaluable to undergraduate students of
psychology and research methods courses in related disciplines, as well as anyone with
an interest in understanding and applying the basic concepts and inferential techniques
associated with statistics in the behavioural sciences.

Riccardo Russo is a Professor of Psychology at the University of Essex, UK, and the
University of Pavia, Italy. His research interests vary in applied and theoretical areas of
cognitive psychology and cognitive neuroscience.
Statistics for the
Behavioural Sciences
An Introduction to Frequentist
and Bayesian Approaches
Second Edition

Riccardo Russo
Second edition published 2021
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
and by Routledge
52 Vanderbilt Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2021 Riccardo Russo
The right of Riccardo Russo to be identified as author of this work has been asserted
by him in accordance with sections 77 and 78 of the Copyright, Designs and Patents
Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or utilised
in any form or by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying and recording, or in any information
storage or retrieval system, without permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or registered
trademarks, and are used only for identification and explanation without intent to
infringe.
First edition published by Routledge 2003
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Names: Russo, Riccardo, author.
Title: Statistics for the behavioural sciences : an introduction to
Frequentist and Bayesian approaches / Riccardo Russo.
Description: Second edition. | Abingdon, Oxon ; New York, NY : Routledge,
2021. | Includes bibliographical references and index. |
Identifiers: LCCN 2020024195 (print) | LCCN 2020024196 (ebook) |
ISBN 9781138711488 (hardback) | ISBN 9781138711501 (paperback) |
ISBN 9781315200415 (ebook)
Subjects: LCSH: Psychometrics.
Classification: LCC BF39 .R82 2021 (print) | LCC BF39 (ebook) | DDC
300.1/5195–dc23
LC record available at https://lccn.loc.gov/2020024195
LC ebook record available at https://lccn.loc.gov/2020024196

ISBN: 978-1-138-71148-8 (hbk)

ISBN: 978-1-138-71150-1 (pbk)
ISBN: 978-1-315-20041-5 (ebk)

Typeset in Bembo
by KnowledgeWorks Global Ltd.
Visit the eResources: https://www.routledge.com/9781138711501
To my wonderful wife and son, Sandra and
Alexander, my family.
Contents

Preface edition I xiv

Acknowledgements edition I xvii
Preface edition II xviii
Acknowledgements edition II  xix

0 Mathematics and algebra: A rapid-mini review 1

0.1 Operators and symbols 1
0.2 Orders of operations 2
0.3 Dealing with fractions 2
0.4 Variables, constants and equations 3
0.5 Graphs and equations 4
0.6 How to solve an equation with one unknown 4

1 Introduction and basic concepts  5

1.1 Why is statistics useful in the behavioural sciences? 5
1.2 Simple example of statistical testing 5
1.3 Measurement scales 10
1.4 Descriptive and inferential statistics 13
1.5 What is an experiment? 14
1.6 Correlational studies 15
1.7 Irrelevant variables 16

2 Descriptive statistics 18
2.1 Organising raw data 18
2.2 Frequency distributions and histograms 18
2.3 Grouped data 21
2.4 Stem-and-leaf diagrams 23
2.5 Summarising data 24
2.6 Measures of central tendency: Mode, median, and mean 25
2.7 Advantages and disadvantages of mode, median, and mean 27
2.8 A useful digression on the ∑ notation 30
viii Contents
2.9 Measures of dispersion (or variability) 31
2.10 Further on the mean, variance, and standard deviation
of frequency distributions 38
2.11 How to calculate the combined mean and the combined variance
of several samples (Web only content) 39
2.12 Properties of estimators 39
2.13 Mean and variance of linearly transformed data 42
2.14 Using JASP for data analysis: Descriptive statistics 45

3 Introduction to probability 52
3.1 Why are some notions of probability useful? 52
3.2 Some preliminary definitions and the concept of probability 52
3.3 Venn diagrams and probability 55
3.4 The addition rule and the multiplication rule of probability 57
3.5 Probability trees 59
3.6 Conditional probability 60
3.7 Independence and conditional probability 63
3.8 Bayes’ theorem 64

4 Introduction to inferential statistics 68

4.1 Inferential statistics and probability 68
4.2 The Classical/Frequentist approach to inferential statistics 69
4.3 How the inferential statistic process operates in Frequentist
statistics 70
4.4 Reducing the risk of false positives 72
4.5 The risk of making false negative errors 73
4.6 Estimating the magnitude of the size of the parameter associated
to the theory 75
4.7 Confidence intervals and inferential statistics 75
4.8 The Bayesian approach to inferential statistics 76
4.9 Odds, probabilities and how to update probabilities 78
4.10 Chickenpox or Smallpox? This is the dilemma. Bayesian inference
in practice 80
4.11 The Bayes Factor:The Bayesian equivalent of significance
testing 82
4.12 The Bayes Factor in practice 83
4.13 Computing the Bayes Factor and interpreting its function in statistical
inference 88
4.14 Estimating the magnitude of the size of the parameter associated
to the theory: Credible intervals 90
4.15 Frequentist and Bayesian approaches to statistical inference:
A rough comparison 91
 Contents ix
5 Probability distributions and the binomial distribution 93
5.1 Introduction 93
5.2 Probability distributions 94
5.3 Calculating the mean (μ) of a probability distribution 96
5.4 Calculating the variance (σ2) and the standard deviation (σ)
of a probability distribution 98
5.5 Orderings (or permutations) 99
5.6 Combinations 102
5.7 The binomial distribution 102
5.8 Mean and variance of the binomial distribution 105
5.9 How to use the binomial distribution in testing hypotheses:
The Frequentist approach 107
5.10 The sign test 111
5.11 Further on the binomial distribution and its use in hypothesis
testing (Web only content) 112
5.12 Using JASP to conduct the binomial test (Frequentist approach) 112
5.13 The Bayesian binomial test 117
5.14 Using JASP to conduct the binomial test (Bayesian approach) 119
5.15 The selection of the prior 122

6 Continuous random variables and the normal distribution  124

6.1 Introduction 124
6.2 Continuous random variables and their distribution 124
6.3 The normal distribution 127
6.4 The standard normal distribution 130
6.5 Hypothesis testing and the normal distribution:The Frequentist
approach 133
6.6 Type I and Type II errors 135
6.7 One-tailed and two-tailed statistical tests 139
6.8 Hypothesis testing and the normal distribution:
The Bayesian approach 141
6.9 Using the normal distribution as an approximation of the binomial
distribution (Web only content) 142

7 Sampling distribution of the mean, its use in hypothesis testing

and the one-sample t-test (Frequentist approach) 143
7.1 Introduction 143
7.2 The sampling distribution of the mean and the Central Limit
Theorem 143
7.3 Testing hypotheses about means when σ is known 145
x Contents
7.4 Testing hypotheses about means when σ is unknown:The Student’s
t-distribution and the one-sample t-test 146
7.5 Two-sided confidence intervals for a population mean: Estimating
the size of the population mean 150
7.6 A fundamental conceptual equation in Frequentist data analysis: Magnitude
of a significance test = Size of the effect × Size of the study 156
7.7 Statistical power analysis: A brief introduction and its application
to the one-sample t-test 158
7.8 Power calculations for the one-sample t-test 163
7.9 Using JASP to conduct the one sample t-test (Frequentist
approach) 166

8 Comparing a pair of means: The matched- and the

independent-samples t-test (Frequentist approach) 168
8.1 Introduction 168
8.2 The matched-samples t-test 168
8.3 Confidence intervals for a population mean 171
8.4 Counterbalancing 172
8.5 The sampling distribution of the difference between pairs of means
and the independent-samples t-test 173
8.6 The independent-samples t-test 175
8.7 An application of the independent-samples t-test 177
8.8 Confidence intervals for the difference between two population
means 178
8.9 The robustness of the independent-samples t-test 179
8.10 An example of the violation of the assumption of homogeneity
of variances (Web only content) 181
8.11 Ceiling and floor effects 181
8.12 Matched-samples or independent-samples t-test:Which of these
two tests should be used? 183
8.13 A fundamental conceptual equation in data analysis: Magnitude of
a significance test = Size of the effect × Size of the study 184
8.14 Power analysis for the independent-samples and the paired-samples
t-test 186
8.15 Using JASP to conduct the paired and the independent sample
t-test (Frequentist approach) 189

9 The Bayesian approach to the t-test 195

9.1 Introduction 195
9.2 An illustration of how to calculate the Bayes Factor for the one-sample
t-test case 196
9.3 Credible intervals (i.e. the Bayesian version of Frequentist
confidence intervals) 200
 Contents xi
9.4 Using JASP to perform the one-sample t-test and the selection
of the distribution to model your prior 205
9.5 JASP in practice:The Bayesian one-sample t-test 210
9.6 JASP in practice:The Bayesian paired-samples t-test 212
9.7 JASP in practice:The Bayesian independent-samples t-test 216
9.8 Bayesian t-test using Dienes’ calculator 219

10 Correlation 222
10.1 Introduction 222
10.2 Linear relationships between two continuous variables 222
10.3 More on linear relationships between two variables 224
10.4 The covariance between two variables 227
10.5 The Pearson product-moment correlation coefficient r 228
10.6 Hypothesis testing on the Pearson correlation coefficient r 229
10.7 Confidence intervals for the Pearson correlation coefficient 232
10.8 Testing the significance of the difference between two independent
Pearson correlation coefficients r 232
10.9 Testing the significance of the difference between two nonindependent
Pearson correlation coefficients r 233
10.10 Partial correlation 235
10.11 Factors affecting the Pearson correlation coefficient r 237
10.12 The point biserial correlation rpb 240
10.13 The Spearman Rank correlation coefficient 244
10.14 Kendall’s coefficient of concordance W 245
10.15 Power calculation for correlation coefficients 246
10.16 Power calculation for the difference between two independent
Pearson correlation coefficients r 247
10.17 Using JASP to perform correlation analyses (Frequentist
approach) 248
10.18 Using JASP to perform correlation analyses (Bayesian
approach) 250

11 Regression  255
11.1 Introduction 255
11.2 The regression line 255
11.3 Linear regression and correlation 259
11.4 Hypothesis testing on the slope b 260
11.5 Confidence intervals for the population regression slope β 262
11.6 Further on the relationship between linear regression and
Pearson’s r: r2 as a measure of effect size 262
11.7 Further on the error of prediction 263
11.8 Why the term regression? 265
xii Contents
11.9 Using JASP to conduct a linear regression analysis
(Frequentist approach) 265
11.10 Using JASP to conduct a linear regression analysis
(Bayesian approach) 267

12 The chi-square distribution and the analysis

of categorical data 270
12.1 Introduction 270
12.2 The chi-square (χ2) distribution 270
12.3 The Pearson’s chi-square test 272
12.4 The Pearson’s χ2 goodness of fit test 273
12.5 Pearson’s χ2 test used in assessing how well the distribution
of a set of data fits a prescribed distribution (Web only content) 274
12.6 Further on the goodness of fit test (Web only content) 274
12.7 Assumptions underlying the use of Pearson’s χ2 test 275
12.8 Compacting a set of data for the goodness of fit test 275
12.9 Pearson’s χ2 test and the analysis of 2 × 2 contingency tables 276
12.10 Further on the degrees of freedom and the calculation of the
expected frequencies for any contingency table 279
12.11 The analysis of R × C contingency tables 280
12.12 One- and two-tailed tests 281
12.13 How to measure the strength of the association between variables
in a contingency table 282
12.14 A fundamental conceptual equation in data analysis: Magnitude
of a significance test = Size of the effect × Size of the study 285
12.15 The odds ratio and the analysis of 2 × 2 contingency tables 286
12.16 An important note on the inclusion of non-occurrences
in contingency tables 288
12.17 The analysis of contingency tables using JASP (Frequentist
approach) 289
12.18 The analysis of contingency tables using JASP (Bayesian
approach) 291

13 Statistical tests on proportions (Web only content) 295

13.1 Introduction
13.2 Statistical tests on the proportion of successes in a sample
13.3 Confidence intervals for population proportions
13.4 Statistical tests on the difference between the proportions of successes
from two independent samples
13.5 Confidence intervals for the difference between two independent
population proportions
13.6 Power calculation for a single proportion
 Contents xiii
13.7 Power calculation for the difference between two independent
proportions
13.8 Statistical tests on the difference between nonindependent proportions
of successes (McNemar test)

14 Nonparametric statistical tests (Web only content) 295

14.1 Introduction
14.2 The Wilcoxon matched-pairs signed-ranks test
14.3 The Wilcoxon rank-sum test

Appendix 296
References  303
Index  304
Preface edition I

Personal computers that are fully loaded with the latest software packages that allow
us to perform more or less complicated statistical analyses are readily available. Hence,
it has become “easier” for a user to enter a set of data into a computer and obtain the
correct answer. However, this process is not as straightforward as it may appear.
In order to feed the computer with all your data and obtain the right answer, you first
need to know what you want to do with your data and what it is the most appropriate
way to analyse them. If you do not have a clear understanding of the rationale that
underlies each statistical procedure you need to use, then is very likely you will have
a poor understanding of the output obtained from any statistical package (even when
the output may contain the right answer!). Given these premises, the approach taken
in this book has been to introduce and explain statistical concepts and the application
of statistical techniques in a clear and detailed manner. It is fundamental to understand
why and how specific statistical analyses should be performed. A good understanding
of statistics is preferable to learning a set of procedures by rote, or to learning how to
press the right button on the computer. Furthermore, notice that without a relatively
good understanding of the process of statistical hypothesis testing, most of the articles
published in psychology journals will be almost incomprehensible.
In a nutshell, the main aim of this book is to help the reader understand the basic
concepts of statistics. This book provides full and exclusive coverage of the material
usually taught in first year statistics courses which are part of undergraduate degree
schemes in the behavioural sciences. It provides a clear, step-by-step introduction to
basic statistical techniques with plenty of examples, each discussed in depth, based on
psychology studies which utilise the statistical techniques described. The presentation
of these techniques aims to provide a conceptual understanding of the basic concepts
of elementary statistics. The book is written in such a way that fundamental concepts,
e.g., the logic of statistical inference, are often recapitulated and repeated several times
in different contexts, thus providing an adequate basis for an in-depth understanding
and retention of key statistical concepts. Given the introductory emphasis adopted here,
more advanced statistical techniques such as Analysis of Variance, Multiple Regression,
and Multivariate Methods are not covered in this textbook.
It may be surprising that after having mentioned the use of statistical packages in
carrying out statistical analyses no commitment to any particular statistical package has
been made in this book. This unbiased approach has been chosen because it does not
force instructors into using any, maybe unwanted, specific statistical package. Learning
how to use a statistical package is easier than learning statistics. This statement has not
been made to scare the reader, but just to warn that learning statistics is a challenging
 Preface edition I xv
but also a rewarding experience. If you find some concepts difficult to grasp immedi-
ately, do not worry, since this feeling is quite common (I have experienced it on a fairly
regular basis!). Just read the relevant section(s) more than once. On the other hand,
learning to use a statistical package is relatively easy but not necessarily very rewarding:
It is enough to consult its manual.
The textbook is composed of 13 chapters. Chapter 1 is introductory. It provides a
convincing argument, using a simple example, as to why some statistical knowledge
is useful to students in the behavioural sciences. Furthermore, some basic concepts of
statistics and research methodology are presented. Chapter 2 describes the main statis-
tical techniques employed to summarise and describe sets of data. You will notice that
a few mathematical formulae are given. Mathematical formulae cannot be avoided in
statistics, and a minimal knowledge of basic algebra will help to understand them. Some
formulae will appear more complicated than other, but in all cases appropriate steps
have been taken to make these formulae easy to understand. Chapter 3 introduces some
concepts of probability. Since most of the book describes inferential statistics techniques
(alternatively called statistical hypothesis testing techniques), and given that the decision
process performed in statistical inference is based on the probability of obtaining a set
of data assuming that certain conditions are true, Chapter 3 describes some important
concepts of probability which are relevant to clearly understanding the process of sta-
tistical inference.
Chapters 4 and 5 introduce the concept of probability distributions and their use in
the process of statistical hypothesis testing. The appropriate application of this process
to sets of data is paramount to answer questions of the type: “Is a particular type of
therapy effective?” or “Is intentional learning more effective than incidental learning?”.
Chapter 4 mainly concentrates on probability distributions of discrete variables, in par-
ticular the binomial distribution, while Chapter 5 is mainly devoted to probability
distributions of continuous variables, in particular to the normal distribution. Both
chapters give examples of the application of probability distributions in the process of
statistical hypothesis testing. Moreover, Chapter 5 describes the types of errors that can
occur in this process.
Chapter 6 describes the use of the Pearson’s chi-square test in the analysis of categor-
ical data. This chapter also introduces the concept of effect size and the use of indexes
of effect size to express the strength of the relationships between variables. Chapter 7 is
dedicated to the process of statistical inference on proportions. This chapter describes
the necessary tools to answer questions of the type: “If in a random sample of 2000 UK
voters you find that 32% vote Labour, what is the proportion of people in the whole
population of UK voters that vote Labour?”.
Chapters 8 and 9 are devoted to the process of statistical hypothesis testing on the
mean or pairs of means (i.e., z and t tests). The central limit theorem and the concept
of sampling distribution of the mean are also presented. Great emphasis is not only
given to the classic binary decision strategy involved in the process of hypothesis
testing (i.e., can I reject the null hypothesis?), but also to the calculation of confi-
dence intervals. This process uses sampled data to estimate ranges of values that have
a relatively high probability to include the unknown true population mean (you will
learn reading this book that inferential statistics is used to estimate population param-
eters while working on the data collected from relatively small samples). Chapter 10
is dedicated to the nonparametric equivalents of the independent-samples and of the
matched-samples t-tests.
xvi Preface edition I
Chapter 11 describes how to calculate the strength of the linear relationship between
two continuous variables (i.e., the Pearson’s index of correlation r). It also describes
the process to calculate indexes to measure the correlation between variables where at
least one variable is not measured on a continuous scale. In some cases researchers may
not only be interested in assessing the strength of the linear relationship between two
continuous variables, but they may also be interested in predicting the values of one
variable on the basis of their knowledge of a second variable. This can be obtained by
calculating the simple linear regression equation as described in Chapter 12. Finally,
Chapter 13 provides an introduction to power analysis for most of the statistical tests
described in the text. One of the aims of power analysis is to provide researchers with
tools aimed to reduce, when planning a research, the risk of failing to reject a null
hypothesis when this should in fact be rejected.
If one looks at the table of contents it can be observed that the above description
provides only a brief summary of the contents of this book. In fact, each chapter con-
tains more material than that described above. Overall, some of the details and some
of the material presented may be outside the scope of a standard first year statistics
course which is part of undergraduate degree schemes in the behavioural sciences. But
this book could also be useful for “eager” undergraduates who would like to broaden
their knowledge of statistics, or also to postgraduate students looking to “brush up” on
the basics. In saying this, the relative modularity of the textbook is such that it allows
instructors to easily select those sections that are more suitable for the requirements of
the course they teach. For example, if Chapter 7 would not be considered strictly rele-
vant to a specific course, its omission will not affect the understanding of the remaining
part of the book. Similarly, some sections within the various chapters can be omitted
without jeopardising the understanding of later sections.
Finally, since most of the book is dedicated to hypothesis testing using inferential
statistics techniques, the figure below provides a decision tree to select the statistical test
more appropriate for the data to be analysed.
Acknowledgements edition I

I would like to thank the people who have read various versions of this manuscript and
made suggestions and corrections. These include Thom Baguley, David Clark-Carter,
Naz Derakshan, Val Cronin, and Linda Murdoch. I am very grateful for all the help that
I have received from the named sources and others, although, of course, any remaining
errors are the author’s alone.
Preface edition II

This revised edition intends to retain the philosophy that embraced the first edition. It
is fundamental and empowering to understand why and how statistical analyses should
be performed. A good understanding and appreciation of statistics is preferable to trying
to learn a set of procedures by rote, or learning when to press the right button on the
computer. To fulfill the above aim I believe it is useful to complement the previous edi-
tion with some new additions. In particular, this edition will add Bayesian procedures
to the description of Classical/Frequentist statistical inferential procedures (i.e. those
usually presented in introductory statistics textbooks). Bayesian approaches to infer-
ential statistics are becoming more popular in the behavioural sciences and of course
beyond, thus it is useful to introduce these to you, the reader. Unlike classical statistics,
performing Bayesian inferential statistics almost invariably requires the support of com-
puting power. To this aim I will mostly, but not exclusively, refer to the JASP statistical
package, which is freely available and easy to download (https://jasp-stats.org). This
package also comes with a freely downloadable manual as well as other support materi-
als and is, I think, rather user-friendly, enabling statistical analyses to be performed in
both classical and Bayesian formats. When I refer to other apps or packages to support
statistical analysis, this will be indicated in the text. Importantly, these are also free
to use!
In this second edition, rather than having a complete chapter dedicated to Power
analysis, as was the case in the first edition, the material covered in that chapter has been
distributed and merged into the chapters presenting the statistical tests for which power
analysis procedures have been provided. Some sections from the first edition relative
to topics that tend to be infrequently covered in introductory undergraduate modules
have been moved to dedicated web pages rather than be directly available in the printed
version. Signposts in the printed version of the textbook will alert the reader about the
related material solely available on the website. Some problems will be added so that
the reader can self-test, if desired, her/his understanding of the material previously pre-
sented. These will also be provided on dedicated web pages. Spreadsheets of the data
analysed using JASP are also available on the web as well as three spreadsheets used in
the exemplification of the computation of Bayes Factors (Chapter 5, 6 and 9). Finally,
some basic knowledge of mathematics (mostly algebra) is required to better understand
and engage with the content of this book. Therefore, to provide a useful refresh or an
in-house introduction (depending on the readers’ need) of the basic concepts of alge-
bra, a chapter zero has been added to this revised edition so that you have the necessary
information at hand when reading the subsequent chapters.
Acknowledgements edition II

I would like to thank the people who have read various versions of this second edi-
tion and made suggestions and corrections. These include Kimberly Dudley-McEvoy,
Denise Wallace and Kathryn Buchanan. I am very grateful for all the help that I have
received from the named sources and others, although, of course, any remaining errors
are the author’s alone.
0 Mathematics and algebra
A rapid-mini review

In statistics the manipulation of numbers is rather common to say the least, thus it is
worth refreshing some handy basic mathematics that may have been forgotten. If you
think the content of this chapter is familiar to you, then simply skip it and move directly
to Chapter 1.

0.1 Operators and symbols

Let see how some basic symbols and operators work:

Operator Meaning Example

= Equal to 9=9
≠ Not equal to 8≠9
> Greater than 9>4
≥ Greater than or equal to Age ≥ 18 years
< Smaller than 23 < 77
≤ Smaller than or equal to Length ≤ 100 cm
+ Add 9 + 9 = 18
− Subtract 9−9=0
± Add or Subtract 7 ± 1 corresponds to
7 + 1 = 8 and 7 − 1 = 6
× or () () or . Multiply (9)(6) = 54
10
÷ or or / Divide = 2.5
4
( )2 Square ( 3)2 = 32 = 3 × 3 = 9
( x )n A number to the power of ( 3)4 = 34 = 3 × 3 × 3 × 3 = 81
exp(x) Exponential e x , e.g., e 2 = 2.718 2 = 7.389056
where e is the base of the natural logarithm, i.e., 2.718 (to the third decimal point)
ln(x) Natural logarithm ln(7.389056) = 2

The natural logarithm of a number x is the exponent to which e, i.e., 2.718 limited to
the third decimal point, must be raised, to produce that number x

Square root 9 = ±3
|⋯| Absolute value |−7| = 7    |7| = 7

(i.e., remove the sign if negative)

2 Mathematics and algebra: A rapid-mini review
0.2 Orders of operations
When a sequence of operations is performed some operators have order of precedence
over others. First consider what is within brackets; then the square and the square root
operators (or higher orders like “elevate a number to the power of x”, or “take the cubic
root of x”); then multiplication and division, and finally at the bottom of the prece-
dence list there are additions and subtractions. Two acronyms used to remember this
order of precedence are:

BODMAS
Brackets first followed by Order then by Division and Multiplication (going left to
right) then Addition and Subtraction (going left to right) (notice that Order means “to
the power of ”);
or

PEDMAS
Parentheses first then Exponents (e.g., Powers, Square, and Cubic roots) then
Multiplication and Division (left to right), and Addition and Subtraction (left to right).
Thus, for example:

(6 − 3 ) × 7 − 2 = (6 − 9) × 7 − 2 = −3 × 7 − 2 = −21 − 2 = −23
2

Notice also the results of the following operations with respect to the obtained signs:

− × + = − ( and + ×− = − ) , e.g., −7 × 4 = −28

− / + = − ( and + /− = − ) , e.g., −8 / 4 = −2

− × − = +, e.g., −7 × ( −4 ) = 28

− / − = +, e.g., −8 / −4 = 2

The last two examples capture, numerically, the fact that, when speaking or writing, a
double negation leads to an affirmation!
From the above formulae it follows that if you want to change the sign of a number
(e.g., −15 into 15) you just multiply this number by −1. In this case we would obtain,
−1( −15) = 15.

0.3 Dealing with fractions

Adding and multiplying fractions can be done either by calculating each ratio and then
4 3
adding these up (e.g., + = 0.444 + 0.75 = 1.194 to the third decimal place). However,
9 4
a more elegant approach that does not require to round any value in the addenda (i.e.,
the elements of the sum) is to find first a common multiple denominator of the fractions
and then apply the suitable multiplications to the numerators to be then added up. The
 Mathematics and algebra: A rapid-mini review 3
final fraction is obtained by adding the numerators and divide this sum by the common
denominator. Applying this procedure to the above example will clarify this process:
4 3 4 × 4 3 × 9 16 27 43
+ = + = + = .
9 4 9 × 4 9 × 4 36 36 36
To find the product of fractions you need, first, to multiply the numerators, then the
denominators, and finally take the ratio of the obtained numerator over the obtained
4 3 4 × 3 12 1
denominator. For example, × = = = .
9 4 9 × 4 36 3
The above procedures apply also when you have more than two fractions. For
instance,
4 3 4 4 × 3 × 4 48 4
× × = = = .
9 4 5 9 × 4 × 5 180 15

0.4 Variables, constants and equations

The notation standardly used in formulae allows their application to several circum-
stances. Hence the use of letters as signposts, within formulae, to indicate that a num-
ber, that needs to be inserted in a particular position within a formula is either fixed or
variable. In general, letters like x, y, and z, are used to indicates variable values, while
letters like a, b, and c are used to indicate that a numerical value is constant within a for-
mula. Often, formulae used in statistics take the form of an equation. An equation indi-
cates that two things are equal. So, for example, we could have the following equation:
y = a + bx
This means that the value that y can take is equal to a constant number a plus another
constant number b multiplied by a variable value x. Using mathematical terminology,
we say that, in the above equation, y is a function of x. So, for instance, if we may decide
that a = 2 and that b = 1.2, our equation becomes
y = 2 + 1.2x
If we then take, just to try, some possible values that x may take (e.g., −1, 0, and 2), we
will see the corresponding values of y.

y = 2 + 1.2 × ( −1) = 0.8, so if x = −1 then y = 0.8

y = 2 + 1.2 × ( 0 ) = 2, so if x = 0 then y = 2

y = 2 + 1.2 × ( 2 ) = 4.4, so if x = 2 then y = 4.4

Consider now the following equation:

y= ∑x
Here we have an operator Σ (sigma), which indicates you need to add all the values that
x can take. So, y is equal to the sum of all the possible values of x. Assuming that x can
take the values of 5, 10, and 22, then
y = 5 + 10 + 22 = 37.
4 Mathematics and algebra: A rapid-mini review

Figure 0.1 Graph of the equation y = 2 + 1.2x.

0.5 Graphs and equations

Using a graph, it is possible to display the y values of an equation as a function of the
values taken by x (thus is said that y is a function of x). To this aim, it is possible to
use the, so called, Cartesian, orthogonal axes, i.e., that join at a 90° angle, to identify
the position of a point in a two-dimensional space. In a nutshell, it looks like giving the
coordinates of a point while plain battle ship. So for an equivalence (or better function)
like y = 2 + 1.2x, we obtain the following graph (here we display a limited set of values
for both x (horizontal axis) and y (vertical axis) variables). So, for example, when x = 0
then y = 2 + 1.2(0) = 2 (see Figure 0.1).
This procedure, if applied to all possible values of x (i.e., in principle from minus
infinite to plus infinite) will determine a set of values of y that if interpolated generate
a straight line. The above equation, in its more general form y = a + bx, is the equation
representing a straight line in a two dimensional space.

0.6 How to solve an equation with one unknown

Given the following equation: 47x − 22 = 18, what is the value of x? In order to find
its value you need to perform some algebraic manipulations. Firstly, you take “−22” to
the right of the equal symbols (thus it becomes “22” – remember that when you move
a number to the other side of an equation its sign changes): 47x = 18 + 22.
1 1 1
Then you multiply both sides of the equation by so we obtain × 47x = (18 + 22 )
40 47 47 47
thus x = . The above types of algebraic manipulation will lead you to the solution of
47
any equation where there is a single unknown variable.
After this brief brush up of some useful mathematics, let us now move to the main
topic of this book: Statistics!
1 Introduction and basic concepts

1.1 Why is statistics useful in the behavioural sciences?

To say the least, it is quite common that a large number of students starting an under-
graduate degree in psychology are not enthusiastic about having to study statistics and
research methodology. It is highly possible that this lack of enthusiasm originates, at
least in part, from the unappreciation of the relevance that statistics plays in the scien-
tific understanding of human behaviour. The aim of this introduction is to provide a
convincing argument, using a simple example, as to why some statistical knowledge is
useful to students in the behavioural sciences.

1.2 Simple example of statistical testing

While revising for the end of year exams a university student notices that the material
is better remembered when some classical music is played in the background, while less
information is retained when no background music is played. This observation is used
by this student as a basis to support their theory that listening to classical music while
studying facilitates the memorisation of the exam material. As a consequence, he/she
suggests the classmates to adopt the same strategy when revising. If you were one of the
classmates would you be so impressed by your friend’s theory and follow the sugges-
tion? Probably not, but why not?
One objection that could be raised is that classical music may work for your friend,
but its effect may not generalise to other people. After all, not everybody likes classical
music; so there is the possibility that some students may find that this music interferes
with their studying. Therefore, in order to see if the effect of listening to classical music
on memory for exam material is not specific to your friend, a relatively large number of
students should be tested. A second objection refers to the way in which the assessment
of the two methods of study was made. More precisely what kind of evidence was used
to claim that the music method led to better retention than the no music method? It is
likely that your classmate had a feeling that they remembered more information after
listening to classical music, but no formal assessment was made. However, it would
be appropriate to have a more accurate way of measuring the amount of information
retained under the two study conditions. Basically, it is important to obtain some sort
of measurement of the phenomenon under investigation.
Overall it appears the theory suggesting that listening to classical music while stud-
ying helps memorisation could potentially be correct, but the evidence used to support
this theory is not compelling. Hence, more people should be tested, and a clear way
6 Introduction and basic concepts
to measure the amount of information retained is fundamental to assess if the effect of
listening to classical music while studying leads to an increment in the amount of infor-
mation memorised compared to a condition where no music is played.
Thus, let us imagine implementing a study to provide a better evaluation of what we
could name the “classical music theory”. As remarked above, to have a clearer assess-
ment of the effect of classical music on memory for exam material, we will need to test
a sizeable number of people (how large this number should be is discussed in later chap-
ters when statistical power will be introduced). At this point a further decision needs to
be made. Should the same group of people be tested in both study conditions (i.e., with
and without exposure to classical music)? Or should two different groups of people be
used so that, within each group, people are tested in only one of the two conditions?
These two approaches have both advantages and disadvantages that will be discussed
later in the book. For the moment, and without questioning why, we use the second
approach. Therefore, one group of students will be given some material to study for a
given amount of time during which classical music will be played in the background. A
second group will receive the same material to be studied in the same amount of time,
but no music will be played in the background. If classical music is beneficial we should
find that the group exposed to classical music should retain more information than the
group not exposed to music while studying.
Before discussing how we can measure memory for the material studied, some
issues about the selection of the people participating in the study need to be addressed.
Participants should be allocated to each of the two groups in a way that no bias is intro-
duced in the evaluation of the two study techniques. Assuming, for simplicity, that
30 university students participate in the study, it would be inappropriate, for example,
for the best students to be allocated to the music condition while the poorest students
to be allocated to the no music condition. Under these circumstances it is likely that the
classical music group would remember more information than the no classical music
group simply because the students in the music group are more capable, and not because
of the music factor. A study like this would lack internal validity, i.e., its results do not
represent what we think they should represent. A way to circumvent the above problem
would be to randomly allocate the participating students to the two conditions of the
study, and, possibly, to have the same number of students in each group. In doing this,
roughly the same number of able and poor students should, in principle, be included in
each group so that these individual differences should be comparable between the two
groups, and, therefore, students’ ability should not bias the result of the study. Random
allocation is fundamental because it reduces the risk that there is a confound between
students’ ability and the conditions of the study (e.g., that the better students are selec-
tively allocated to the classical music group).
Before discussing how to measure students’ memory for the exam material, a further
digression on the selection of the participants to the study is in order. We said earlier
that 30 university students should be selected to take part in the study. Thus we study
a sample, and not the entire population of university students, because it would be
unrealistic, very costly, and time consuming to test the entire population. However, we
would like the results obtained with the selected sample to generalise to the entire rele-
vant population. In order to generalise the results of our study to the entire population
of university students, our selected sample should accurately reflect the characteristics of
the entire population. If this occurs, it is then said that the study conducted has external
validity. However, if a sample is biased, then the results obtained are biased too, thus
 Introduction and basic concepts 7
these cannot be generalised to the entire population being studied. For example, if all of
our selected students are very able and all likely to obtain “A” marks in their exams, we
will have a biased sample because only a small percentage of students perform so well.
The optimal way to obtain an unbiased sample is to draw a random sample from the
entire population (using procedures similar to those used in the National Lottery draw).
In this way every member of the population has equal probability of being included in
the sample, thus the sample obtained should accurately reflect the characteristics of the
population of university students. However, random sampling from the entire popula-
tion is not practically possible in most real research. Nevertheless, it is important that
samples are reasonably representative of the populations we want to generalise to (for
more details on sampling procedures see Upton & Cook, 1997). Obviously, the extent
to which the sample obtained is not representative of the entire relevant population
leads to a limitation in the generalisation of the results that can be obtained in a study.
For example, if we study how families with a monthly income up to £1500 per capita
spend their money, which we may have difficulty in generalising the results to the pop-
ulation of families with a monthly income up to £3000 per capita.
Summarising, to test any psychological theory or hypothesis we need to study and
measure some behaviour in a sample of subjects, and in doing this we aim to draw
conclusions on the entire population from which the sample is taken. Therefore, while
studying the behaviour of a relatively small sample we want to be able to generalise
the results obtained to the entire relevant population. As an addendum it should be
said that populations are not necessarily finite. A population could be the collection of
a potentially infinite number of items (e.g., the set of all possible e-mails that people
could potentially write). In this case, it would be impossible to list all the elements in
the population.
Let us now consider how we can measure the retention of the studied material. For
example, each group of 15 students have to study a prose passage for 1 h, the content
of which is comparable to exam material where 60 important pieces of information
are highlighted. Remember that we sampled 30 students, of which 15 to be exposed
to classical music while studying and 15 not exposed to any background music. A con-
venient index, but not necessarily the best one to measure the retention of the to-be-
learned information, is the number of key pieces of information correctly recalled 24 h
after the study phase. Hence, the relevant information we will obtain is going to be the
“number of key pieces of information retrieved by each student”, and this measure will
be used to decide if students retain more information when they listen to classical music
while studying compared to a no background music condition.
A reasonable approach to decide on the above matter would be to suggest that classi-
cal music is beneficial if the students in the “classical music” group recalled more units
of information than the students in the “no music” group. Unfortunately, at first, the
obtained data will only be an unorganised set of 15 memory scores for each of the two
groups of students who participated in the study, and no useful information is likely
to be extracted from these numbers without further manipulations. Some order needs
to be imposed on these data to be able to decide which of the two groups of students
recalled more information. For example, as a first step, the performance of the students
in each group could be ranked in order. This would permit the inspection of the mem-
ory scores for each group, ordered from the lowest to the highest. A useful piece of
information is also the average number of items recalled in each group (see Table 1.1a).
A quick glance at the rank ordered data seems to suggest that more items are recalled in
8 Introduction and basic concepts
Table 1.1a F
 ictitious data on the effect of listening to classical music while studying on the
retention of the material to be learned. In Table 1.1a data are ordered within each
studying condition (where I correspond to the rank of the lowest scoring student
and XV to the highest). In Table 1.1b data are ordered irrespective of the studying
condition

Average
Group Rank within each group and score score

I II III IV V VI VII VIII IX X XI XII XIII XIV XV

A (with classical music) 28 30 33 35 36 38 39 40 42 44 45 45 48 49 50 40.1
B (with no music) 25 28 32 34 35 37 38 40 40 42 43 43 44 45 47 38.2

the classical music group (i.e., for almost any rank more items are recalled in the classical
music than in the no music group). Moreover, on average, more items are recalled in
the classical music group. It seems, therefore, we could answer the original question and
conclude that, when classical music is played while studying, more information about
exam material is remembered compared to a condition when no background music is
played.
This conclusion may, however, be premature. In fact, various reasons suggest that
the type of inference used above to decide on the memory effect of classical music is
incorrect. First of all, taking a closer look at the rank ordered data might indicate that
the performances in the two groups are comparable. Apart from a small number of
relatively extreme scores, which appear to be larger in the classical music group, the
majority of scores are comparable in the two groups (see Table 1.1b where all the scores
are ordered from the smallest to the largest).
The second important point to make is that, despite the average memory scores
being numerically different, this difference could have occurred simply by chance and
not because the two groups in the study were treated differently. To illustrate this con-
cept, imagine that 30 more students are randomly sampled from the same population
of university students, and that these students are randomly assigned to two groups.
These are conveniently labelled Group A and Group B. However, unlike the previous
study, both groups are now asked to commit to memory the relevant material while
there is no music in the background for either group. Table 1.2 displays the rank ordered
performances of the two groups. There are some noticeable similarities between the
data displayed in Table 1.1a and Table 1.2. The most striking features are that in both
tables the average performance of Group A is numerically larger than the performance
of Group B, and that the average performance of the A groups is almost identical
in the two tables. Thus the results obtained in the second study are very similar to
those previously found despite no classical music ever being played in the background.
Therefore, these results cast doubts on the validity of the previous conclusion regarding

Table 1.1b

Group Score

A (with classical music) 28 30 33 35 36 38 39 40 42 44 45 45 48 49 50

B (with no music) 25 28 32 35 37 38 40 40 42 43 43 44 45 47
34
Another random document with
no related content on Scribd:
Kamakura, which was to be taken by storm. Meeting no opposition
worthy of the name, they reached the sea-shore at Enoshima, only
three miles from their destination. Here an unexpected difficulty
presented itself. Although it was ebb-tide the sea ran extremely high,
so that the road along the shore was submerged beneath angry
waves, which washed even the steep mountain spurs from whose
base they were generally separated by several hundred feet of hard,
sandy soil. The only way to reach Kamakura seemed to be for them
to retrace their steps by a long détour round the other side of the
mountain. This course would entail loss of time and labor; but the
commander-in-chief was less disturbed on that account than by the
fear that this accident might be misconstrued into an evil omen, and
consequently dampen the enthusiasm of his followers. For at the last
moment the Hōjō had changed their tactics. They all at once showed
unbounded honor and veneration for the Emperor whom they had
placed upon the throne three years before, and took care to have
their new-born loyalty proclaimed far and wide. They knew, as they
said, that the gods from whom the Mikados sprang would never
allow sacrilegious hands to be laid on their descendants, and in the
changes which had been made the Hōjō had merely been
instruments ordained to work out the divine will by placing the
sceptre in the hands of him who could wield it to the best advantage
of the country. Numbers of Hōjō spies had found their way into
Nitta’s camp, where only the evening before some of them had been
discovered and executed; and those undetected would certainly
make the most of this unexpected impediment, causing it to be
regarded as a sign of the displeasure of the gods and an augury of
evil.
 SEIZING A DISGUISED HŌJŌ SPY IN NITTA’S CAMP.

Going down to where the waves washed his feet and frequently
covered him with their spray, Nitta looked out over the sea long and
earnestly. The white-crested billows came and went, but the last-
comer showed no sign of lagging behind those that had preceded it.
Sennoske, seeing the anxiety depicted on his chief’s face, ventured
to step up to him and give his opinion. He had been brought up in a
seaport town, with a harbor which was none of the best, where he
had been out at all times, and had learned to know the tides and the
signs of the weather. Therefore he felt that he could now take it upon
himself to assert that in a few hours, or a day at most, the flood must
subside. Nitta, although he appeared absorbed in thought, and
scarcely to listen to his lieutenant, eagerly drank in every word the
latter said; and, placing confidence in it, performed one of those acts
which, trifling as they may appear, are yet emanations of genius.
Exerting a purely moral force, such actions are often all-powerful in
shaping the destinies of men and nations, by giving decision to the
vacillating, courage to the timid, a feeling of strength which is
equivalent to real force to the weak, and making heroes of those who
possess manly qualities in any degree.
Taking his sword, of matchless workmanship and ornamentation,
glittering in the sunlight which reflected itself in the burnished gold,
Nitta cast it out into the waters as far as a warrior’s arm could throw
it, asking the gods to accept it as a peace-offering, and calling upon
them to give their divine aid to the army which now marched to
uphold the vested and inherited rights of their descendants, the
Emperors of the country. In the name of the Mikado, driven from the
throne into a lonely exile, he prayed them to bid the angry waves
subside and give free passage to those who came to avenge his
despoilers and punish the sacrilege. It was a powerful invocation,
and its effect was thorough and instantaneous, creating general and
unbounded enthusiasm. If any man there had misgivings, it would
have been dangerous for him to express them; and when next
morning the waters had subsided, leaving the road free and open, all
lingering doubts as well as the hopes of the Hōjō sympathizers
vanished as the dew of the night that disappeared in the morning
sun.
As they approached Kamakura, they found a large portion of the
enemy’s army posted before the walls; so fierce, however, was the
onslaught of Nitta’s troops that the ranks of the besieged were at
once broken, and they fled for protection into the city. But the
assailants were close upon them, and followed in such numbers that
the gates could not be shut against them. From ward to ward, from
street to street, from house to house, the Hōjō were driven, in spite
of their persistent and desperate struggles, and before nightfall the
city was taken.
 This battle decided the war and the fate of the Hōjō. The honest
portion of their adherents consisted merely of those who had been
dazzled and won over by the spell of absolute power exercised for a
considerable time; and this spell once broken, they fully recognized
their folly as well as their guilt, and were ready to sue for and accept
pardon on any terms. It was expected that the principal men of the
defeated clan, together with such of their partisans and abetters as
were too deeply implicated to hope for forgiveness, would form into
roving bands and flee to distant inaccessible parts of the empire; but
as they were now without physical or moral support, a general
engagement was no longer to be feared. In view of this it was
decided by Nitta to dismiss at once a part of the army, so that
divisions of the retiring troops might be employed in protecting their
several provinces from the marauders who should flee thither, and in
hunting them down and bringing them to punishment.
NITTA PRAYING TO THE GODS BEFORE THROWING HIS SWORD INTO THE
SEA.

One of the provinces to which it was believed that the defeated

would be certain to resort in large numbers was Idzu, lying to the
west of the Tokaido, beyond the Hakone Mountain-range, on the
road to Kuwana. Here the great Yoritomo had in his youth found an
asylum from his Taira foes. It required coolness and courage, a
steady eye and a firm foot to track foes into these Idzu fastnesses,
over these rugged, pathless mountains, where Nature alone offered
almost insurmountable obstacles. Nitta gave a proof to the country of
his confidence in Sennoske by intrusting to him the military
command of that region, as well as of the country along the Tokaido
from the Hakone Mountains to the province of Kuwana. The
contingent from the latter province, which was to return for the
purpose, was to be under his orders, together with such other troops
as he might think it advisable to raise in the districts assigned.
Nominally, it was necessary to have this appointment acknowledged
by the daimios in whose dominions the new commander was to act;
but this was merely a matter of form entailing a short delay. No one
would have been willing to displease the victorious Nitta, or to incur
the suspicion of being a Hōjō sympathizer, by in any way impeding
the measures deemed necessary to bring the insurgents to justice.
 SCENE AT THE STORMING OF KAMAKURA.

The delay which would ensue before he could take active

command, Sennoske determined to take advantage of by returning
home; and leaving the Kuwana soldiers to follow to the place which
he was to make the base of his operations, he started early in the
morning on the day succeeding his appointment, accompanied by
Yamagawa, both mounted on good horses and attended only by two
running footmen. He was in good spirits. The honors he had gained;
the charm of the important command intrusted to him; the thought of
meeting his father and the Duke, with the great expectations that
they entertained of him more than fulfilled in so short a time; and
last, but not least, however he might try to disguise it to himself, the
joy of again seeing O Tetsu,—all these were certainly good and
sufficient reasons for his elation.

SENNOSKE ON HIS RETURN.

 Pushing on as fast as possible, on the evening of the same day he
reached Odawara, a strong castle town at the northern foot of the
Hakone Mountain-range. The fame of his exploits had preceded him,
and the lord of the castle received him in a way which could not but
be flattering to the young man, pressing him to remain at least a day
or two longer. No inducement, however, could detain Sennoske; and
leaving their horses, which would only have impeded their way over
the mountain-passes, he started on foot the next day at dawn with
his faithful henchman.

PILGRIMS TO FUJIYAMA.

The road, as it was at that time laid out, wound between the
Hakone and the Ashigara Mountain-ranges; and here the young man
made perhaps less haste than might have been expected. It was late
in autumn, the weather was fine, the air bracing; and in spite of his
anxiety to reach his home he could not but loiter a little to view the
beautiful scenery that presented itself. The mountains and hill-sides
were clothed in their rich autumnal garb of gold and red and purple
and brown of every shade and depth of coloring; while as if to supply
the need of a stronger contrast, clusters of evergreen trees here and
there reared their crowns of unchanging hue high into the air.

ROADSIDE TEA-HOUSE.
 FUJIYAMA, WITH ANGEL HOVERING OVER IT.

The whole effect was greatly heightened by occasional glimpses

of Fuji, which presented itself under the most varying forms and
expressions. Sometimes just a glimpse of its top was visible over an
intervening mountain-range, while again a portion of the huge mass
would reveal itself through a narrow gorge; sometimes one of its
bleak sides, within range of the eye, but up to this moment obscured
by clouds which now suddenly parted, would astonish the traveller
who had just gazed at what he thought empty space, while the
shifting curtain, hiding the view the next moment, would make him
doubt whether he had seen aright. Two or three times the whole
mountain with its perfect outline stood in majesty like a giant sentry
watching over the land. With the varying beauty of scenery which
each successive step thus unclosed, it was impossible for a lover of
Nature like Sennoske not to stop for a moment now and then to
admire it. Yet it was only for a moment, as the thought of the
welcome awaiting him prevented him from lingering as he otherwise
would have done. Many a good-looking young girl at the roadside
tea-houses followed with strained eyes the figure of the dashing
young cavalier, whose pleasant, gentle ways so well set off his
unmistakably martial air. Old Yamagawa also came in for his share
of attention; and now that they neared home, honored and
successful, the old servitor indulged occasionally in what was
probably his only fault,—a love of wine,—to a greater extent than
was good for him; but his young master had not the heart to chide or
even to restrain him.

COURT LADY’S HEAD-DRESS AND STATE FAN.

 CHAPTER IX.
It was late at night on the second day before Sennoske arrived at
Mishima, at the foot of the other side of the mountains. The town
was full of people; for besides the ordinary contingent of travellers,
who always muster here in force, there were a great many samurai
—some who had started too late to join the army, others, like
Sennoske, already returning home—and a considerable number of
Hōjō adherents fleeing in various disguises to find an asylum before
pursuit was fairly organized against them. The inns and hostelries
were over-crowded; but the Kuwana nobility had always enjoyed a
good reputation on the road, and the charm of the young man’s
manner moreover failed not in its usual effect; so at the inn at which
he determined to stay, two parties of merchants, already closely
packed, were forced to content themselves with even less space in
order to clear a room for the new arrival. Sennoske, who was weary,
at once repaired to the bath, accompanied by Yamagawa, carrying
the Muramasa sword, which was never out of sight or reach of either
master or servant.
After his master had retired, the old servitor also indulged in the
luxury of the bath, the daily use of which in Japan is considered
almost a necessity. Already half overcome by the fumes of the wine-
cup, the hot bath and the close air of the room still further increased
his stupefaction; and in returning to his apartment he staggered
along the corridor in a confused, aimless manner, until, mistaking
one of the passages, he stopped at the end of it before what of
course was the wrong room.
The sound of unknown voices from within reached his ears; but
thinking they must belong to visitors of his master, he unhesitatingly
opened the paper sliding-doors, and falling on his hands and knees
with his head bowed down, in the fashion of Japanese servants, he
pushed the sword forward as far as he could, uttering the customary
phrase, “Have you any further orders for me?” Receiving no answer,
and fearing that Sennoske had noticed his drunken condition and
was displeased at his showing it before visitors,—something which
heretofore had never happened,—he remained a minute or two in
his prostrate position, bowing his head still lower, and muttering
apologies. At last a voice which he did not recognize spoke out: “We
will excuse you because you insist on being excused; but tell us first
what it is that you require, and what brings you hither.”
Yamagawa looked up, and the sight that presented itself,
unexpected as it was, partly sobered him. There were six or seven
samurai, all unknown to him, seated on the mats; and the foremost
among them, who also was the last speaker, had taken up the sword
pushed towards him, and, regarding it with the eye of a connoisseur,
was evidently surprised at the rich ornamentation and workmanship.
“Give me back my sword,” Yamagawa cried; “it belongs to my
master, and I am responsible for it. Give it back to me, oh! please
give it back to me at once,” he repeated, raising his voice and
eagerly holding out his hands.
“Gently, my man! Of course I shall give it back to you,” said the
other; “but do not be so importunate, and do not speak so loudly
before your superiors, or you will have to be taught better manners.
Who is your master?”
“His name is Sennoske Mutto, from Kuwana, and he is just
returning home from the war,” was the reply. “This is a Muramasa
sword, which he values highly, and he will begin to be anxious about
it; so please give it back to me at once,” pleaded the old man. He
was still partly under the influence of wine; and this fact, together
with the mortification of his position and the anxiety and irritation
under which he labored, had made him forget discretion, common
sense, and his dignity as a samurai, all other considerations being
merged and lost in his impatient eagerness to regain possession of
the cherished weapon.
“A Muramasa sword, and your master intrusts it to a drunken fool
like you!” said another of the party, who until now had been sitting in
the rear of his companions, but who now moved forward to where
the light fell fully upon him. “Does he attach so little value to it, in
spite of what you say, as to leave it like a toy in the hands of one
who in his cups would not mind trading it off for a measure of wine?
His opinion differs from mine as to the care which he ought to
bestow upon it. Be off! and tell your master that if he wants this
sword he must come for it himself.”
What was it which, at the sight of the new speaker and at the first
word uttered by him, made the old man look as if petrified with horror
and stupefaction, clammy perspiration bathing his face, while his hair
literally stood on end? Only his eyes spoke; and although fastened
with almost stony fixedness upon the man who had just addressed
him, they yet had an expression of such intense, deadly hatred that
the other, as he fairly caught sight of it, involuntarily recoiled a step
and laid his hand on the sword he carried at his side.
 YAMAGAWA DISCOVERING HIS MISTAKE.

Gradually, by a series of sharp, convulsive movements,

Yamagawa’s form and features relaxed from their rigid cast, and it
was evident that a full consciousness of his position and
surroundings was slowly taking hold of him, and that as it did so he
was making powerful efforts to regain his self-control. Every trace of
drunkenness had disappeared, and it was only extreme mental and
nervous excitement which made his voice tremble when he again
spoke:—
“Give me back the sword, Taka Suke; and after returning it to my
young master, I will try to refrain from telling him or his father that I
have again met you. Perhaps I am selfish enough not to wish to
stand disgraced before one in the service of whose family I have
become old and gray, whose welfare has occupied every thought of
my mind, to whom every fibre of my body is devoted, and for whose
sake I would willingly shed my blood drop by drop. Perhaps I have
become too feeble in mind as well as in body fully to realize the
duties of a samurai, for I would fain save my young master from
facing one who has been the curse of his house, instead of doing my
utmost to bring about a meeting, so that he could either kill you or fall
as a samurai should. As for me, I know I shall not long be able to
endure the thought of having met you and not only failed myself to
take vengeance, but even shielded you against punishment by those
whose right and duty it is to inflict it. For twelve years there has
never been a day when you have been absent from my thoughts,
and I could never think of you without cursing you. Yet all the
sufferings I have so far endured are as nothing compared with the
humiliating misery of this hour, when I find myself compelled, through
my own fault, to plead to one who has caused blight and ruin to fall
upon those I love and hold dear.”
“I will give you still further reason to curse me, then,” said the
other. “The gods seem kind to me; for although our cause has failed,
I am still able to strike at the only one whom, from some foolish
whim, I spared, leaving him life and liberty when I had the power to
take both. So Numa is still alive, and you now serve his son. Well,
the brat has his father’s courage, as he showed in the war; and I
have no doubt he has also the same conceit. Curse you all, with your
pretended fine feelings and virtues, which you constantly flaunt
before our eyes! They will not impose upon me, however. Go, and
repeat to your master what has been told to you here; go, and tell
him that if he knows a samurai’s duties, of which his father always
prated so much, he must be aware that a drunkard is not fit to be
intrusted with a sword, least of all with one like this. Go, and tell him
that if he cares for it more than you seem to have done, he must
come and humbly beg for it himself.”
There was somewhat of truth in this taunt, which the devilish
cunning of the other had formulated so as to make it more cutting
than the keenest-edged weapon. Seeing from the tenor of it that all
further parley would be useless, Yamagawa without another word or
look made his way back to his master’s room. As he entered,
Sennoske started in amazement and affright at the change which
half an hour had effected in the appearance of his old servant. The
signs of acute physical pain, as well as of mental anguish, were
graven in deep lines on his features, and spoke with equal emphasis
out of his hollow, sunken eyes; shuffling along as he did on his right
side, with his right hand convulsively clasping his left bosom, it is
probable that the terrible ordeal through which he had passed, had
brought on a partial heart-stroke. Yet overlying all these
manifestations of suffering, which became almost tangible and
somatic as it were by comparison, there was a look of utter, hopeless
despair,—such a look as is seen in certain types of incurable
madness; such a look as hunters see in some animals hunted down
to the last stages of exhaustion, with the dogs fastening on them,
and no hope of escape.
In a sufficiently coherent way to make himself understood,
Yamagawa explained the circumstances attending the loss of the
sword, without, however, hinting at the identity of the samurai whom
he had recognized among his despoilers. Sennoske, although
greatly annoyed and angry at the impudence of the demand
transmitted to him, showed no outward signs of perturbation, and in
reality a sincere pity for the poor old man before him mastered every
other emotion. From the version given to him, he could of course
scarcely understand why this accident, although implying a serious
neglect of duty on the part of Yamagawa according to the code then
prevailing, should yet have affected the old servitor in such a terrible
way; and in spite of the latter’s urgent solicitation to go at once about
the recovery of his weapon, he stayed to console and cheer him.
 ROADSIDE HOTEL.

“Sorrow is bad for old age,” he said; “it withers up the tree of life
quicker and surer than the cold north winds wither the blooming
chrysanthemum. Cheer up, and do not let me see you so downcast
at this trumpery business, especially now that we are nearing home.
As soon as I have regained my sword, I shall feel like teaching these
gentry a lesson at which you may have to assist me; so—”
“Yes, as soon as you have regained possession of your sword,”
said the other, interrupting him. “Oh, pray, my dear master, go at
once! this suspense is horrible.”
Sennoske at these words started up, filled with dark forebodings.
He felt that there must be something more in this affair than he yet
apprehended, and he hesitated no longer. Calling a servant, he had
himself conducted to the room which Yamagawa had mistakenly
entered; and announcing his name, made a fair apology for what had
occurred, and courteously but firmly asked for the return of his
sword. Taka Suke, who had sent him the message to come, and who
was evidently the leader of the party, replied to him in what was
plainly a prepared speech:—
“I have heard of you, Sennoske, and of the renown which you
have gained on the field of battle; but it seems to me that, in spite of
this, you are greatly deficient in the duties and obligations of a
samurai. Courage and bravery and prowess in battle are common
enough in our country; but a sword like this of yours is rarely found,
and its possession probably more than any quality of your own has
helped you to achieve success and renown. You know the old
saying, ‘The sword is the soul of the samurai.’ It owes its origin to a
feeling hallowed by the custom of centuries,—a feeling which has
been outraged by the careless, negligent way in which this treasure
has been handled. Your youth and inexperience may plead in
extenuation of yourself, but the fault of your retainer, who is a
samurai, and old enough to know his duty, cannot be condoned; and
we will listen to no demand for the restoration of this blade unless it
is accompanied by the head of that drunken brute through whose
culpable carelessness it might easily have been lost or spoiled.”
“But this is preposterous,” Sennoske rejoined; “this is horrible! You
surely cannot mean what you say—and yet the life of a faithful old
retainer is not a fit subject for sport. I recollect him from the day
when consciousness first dawned upon me; memory recalls him as
watching over my childhood, guiding and instructing my early youth,
in the most disinterested, self-sacrificing way. I would willingly risk
my life for him at any moment; and sword was never yet forged, nor
ever will be, which could weigh equally in the balance with such
faithfulness and such devotion as he has always shown. I respect
your feelings; but this man has suffered more than enough already