SPSS for Psychologists, 7th Edition
Visit the link below to download the full version of this book:
https://medipdf.com/product/spss-for-psychologists-7th-edition/
Click Download Now
© Virginia Harrison, Richard Kemp, Nicola Brace and Rosemary Snelgar,
under exclusive licence to Macmillan Education Limited 2021
© Nicola Brace, Richard Kemp and Rosemary Snelgar 2000, 2003, 2006,
2009, 2012, 2016
All rights reserved. No reproduction, copy or transmission of this
publication may be made without written permission.
No portion of this publication may be reproduced, copied or transmitted
save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency,
Saffron House, 6–10 Kirby Street, London EC1N 8TS.
Any person who does any unauthorized act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The authors have asserted their rights to be identified as the authors of this
work in accordance with the Copyright, Designs and Patents Act 1988.
This edition published 2021 by
RED GLOBE PRESS
Previous editions published under the imprint PALGRAVE
Red Globe Press in the UK is an imprint of Macmillan Education Limited,
registered in England, company number 01755588, of 4 Crinan Street,
London, N1 9XW.
Red Globe Press® is a registered trademark in the United States,
the United Kingdom, Europe and other countries.
ISBN 978-1-352-00994-1 paperback
This book is printed on paper suitable for recycling and made from fully
managed and sustained forest sources. Logging, pulping and manufacturing
processes are expected to conform to the environmental regulations of the
country of origin.
A catalogue record for this book is available from the British Library.
A catalog record for this book is available from the Library of Congress.
The icons that appear throughout this book are printed courtesy of
The Noun Project, https://thenounproject.com/
To Emily Rae – my little statistical anomaly! <3
Virginia
To Penelope, Joseph, Francesca, Frankie and Wilb,
and to my wonderful colleagues Nicky and Rosie who have worked with me for so many years on
previous editions of this book.
Richard
To my lovely family, my siblings Sue, Meg and Tim, and to my nieces, nephew, great-nieces,
great-nephews and great-great-nieces – they’ve been so supportive and a continuing
delight through all the editions.
Rosemary
Contents
Prefacex
Acknowledgementsxiii
Chapter 1 Introduction 1
Section 1 Psychological research and SPSS 1
Section 2 Guide to the statistical tests covered 16
Section 3 Working with SPSS 19
Section 4 Starting SPSS 21
Section 5 How to exit from SPSS 23
Chapter 2 Data entry in SPSS 25
Section 1 The Data Editor window 25
Section 2 Defining a variable in SPSS 26
Section 3 Entering data 38
Section 4 Saving a data file 42
Section 5 Opening a data file 43
Section 6 Data entry exercises 45
Section 7 Answers to data entry exercises 48
Section 8 Checking and cleaning data files 50
Chapter 3 Exploring, cleaning and graphing data in SPSS 52
Section 1 Descriptive statistics 52
Section 2 The Descriptives command 53
Section 3 The Viewer window 57
Section 4 The Frequencies command 61
Section 5 The Explore command 64
Section 6 Using descriptive statistics to check your data 71
Section 7 Introducing graphing in SPSS 76
Section 8 Chart Builder 78
Section 9 Graphboard Template Chooser 84
Chapter 4 Data handling 88
Section 1 An introduction to data handling 88
Section 2 Sorting a file 89
Section 3 Splitting a file 91
Section 4 Selecting cases 94
Section 5 Recoding values 99
Section 6 Computing new variables 105
Section 7 Counting values 108
Section 8 Ranking cases 110
Section 9 Data transformation 113
Section 10 Data file for scales or questionnaires 120
vii
viii Contents
Chapter 5 Tests of difference for one- and two-sample designs 126
Section 1 An introduction to the t-test 126
Section 2 The one-sample t-test 127
Section 3 The independent t-test 130
Section 4 The paired t-test 137
Section 5 An introduction to nonparametric tests of difference 141
Section 6 The Mann–Whitney test 142
Section 7 The Wilcoxon test 146
Chapter 6 Tests of correlation and bivariate regression 151
Section 1 An introduction to tests of correlation 151
Section 2 Producing a scatterplot 152
Section 3 Pearson’s r : parametric test of correlation 161
Section 4 Spearman’s rs : nonparametric test of correlation 165
Section 5 Partial correlations 168
Section 6 Comparing the strength of correlation coefficients 171
Section 7 Brief introduction to regression 174
Section 8 Bivariate regression 176
Chapter 7 Tests for nominal data 186
Section 1 Nominal data and dichotomous variables 186
Section 2 Chi-square test versus the chi-square distribution 188
Section 3 The goodness of fit chi-square 188
Section 4 The multidimensional chi-square 189
Section 5 The McNemar test for repeated measures 205
Chapter 8 One-way analysis of variance 214
Section 1 An introduction to one-way analysis of variance (ANOVA) 214
Section 2 One-way between-subjects ANOVA, planned and unplanned
comparisons, and nonparametic equivalent 224
Section 3 One-way within-subjects ANOVA, planned and unplanned
comparisons, and nonparametric equivalent 243
Chapter 9 Factorial analysis of variance 259
Section 1 An introduction to factorial analysis of variance (ANOVA) 259
Section 2 Two-way between-subjects ANOVA 262
Section 3 Two-way within-subjects ANOVA 272
Section 4 Mixed ANOVA 282
Chapter 10 Multiple regression 296
Section 1 An introduction to multiple regression 296
Section 2 Standard or simultaneous method of multiple regression 305
Section 3 Sequential or hierarchical method of multiple regression 315
Section 4 Statistical methods of multiple regression 321
Contents ix
Chapter 11 Analysis of covariance and multivariate analysis of variance 326
Section 1 An introduction to analysis of covariance 326
Section 2 Performing analysis of covariance in SPSS 329
Section 3 An introduction to multivariate analysis of variance 339
Section 4 Performing multivariate analysis of variance in SPSS 343
Chapter 12 Discriminant analysis and logistic regression 351
Section 1 Discriminant analysis and logistic regression 351
Section 2 An introduction to discriminant analysis 353
Section 3 Performing discriminant analysis in SPSS 355
Section 4 An introduction to logistic regression 368
Section 5 Performing logistic regression in SPSS 369
Chapter 13 Factor analysis, and reliability and dimensionality of scales 377
Section 1 An introduction to factor analysis 377
Section 2 Performing a basic factor analysis using SPSS 388
Section 3 Other aspects of factor analysis 404
Section 4 Reliability analysis for scales and questionnaires 408
Section 5 Dimensionality of scales and questionnaires 414
Chapter 14 Using syntax and other useful features of SPSS 420
Section 1 The Syntax window 420
Section 2 Syntax examples 428
Section 3 Getting help in SPSS 431
Section 4 Option settings in SPSS 437
Section 5 Printing from SPSS 439
Section 6 Incorporating SPSS output into other documents 441
Section 7 SPSS and Excel: importing and exporting data files 443
Appendix448
Glossary463
References484
Index486
Preface
How to use this book
This book is designed to help you analyse psychological data using the software pack-
age IBM® SPSS® statistics software (SPSS). SPSS Inc. was acquired by IBM in October
2009. With the exception of the first few sections of Chapter 1, we recommend that
you read this book while sitting at a computer that is running SPSS.
Chapter 1 serves as a reminder of some issues related to research design, while
Chapter 2 shows you how to enter data into SPSS. In Chapters 3 and 4, we show you
how to explore, clean and manage your data. In Chapters 5–12, you will learn how to
undertake a variety of statistical procedures using SPSS. The order of these chapters
is designed to reflect the way many university psychology departments teach statis-
tics to undergraduates, but the chapters are largely free-standing, so you do not need
to read them in sequence. We hope that, once you are familiar with the process of data
entry, you will use this book as a reference to assist you to undertake the analysis you
need. To help you select the most appropriate statistical procedure, we have included
a brief description of each test in Chapter 1 alongside a helpful decision tree on page
18 and 19. The procedures covered in Chapters 5–8 are likely to be taught in most
undergraduate psychology research methods courses, whereas those covered in
Chapters 9–13 are slightly more advanced. We hope that the statistical procedures
described in these later chapters will be of particular help to students undertaking
their final-year research project, postgraduate students and researchers.
Within each of the analysis chapters, we briefly describe the procedure, show
you how to undertake the analysis, explain some of the options available and teach
you how to obtain suitable graphs and descriptive statistics. Critically, each section
ends with an explanation of how to interpret the output produced by SPSS, and write
it up in APA format.
Chapter 14 describes the use of syntax to control SPSS and also provides some
useful information on topics such as printing and importing and exporting files.
Finally, we have included an extensive Glossary of the statistical terms used in this
book.
The content of each chapter is now explained in a little more detail.
Chapter 1
Chapter 1 provides a brief overview of the basic concepts and terminology used in
psychological research and statistical analysis, and introduces SPSS. We describe
some basic methods of data collection and the types of data that are collected in quan-
titative research. We then consider data analysis and provide you with an introduc-
tion to the windows and buttons you will use when analysing your data with
SPSS. Finally, we show you how to start and exit SPSS.
x
Preface xi
Chapters 2–4
Chapter 2 shows you how to set up SPSS to receive your data, and how to create and
save a data file. In Chapter 3, you learn how to explore the data you have entered to
calculate some simple descriptive statistics, how to check and clean your data files
and how to use SPSS to produce useful graphical representations of your data. Chapter
4 introduces you to some useful functions you can use to manage your data set. We
show you how to do things such as split the data file or group participants together,
and how to recode existing variables or compute new ones.
Chapters 5–9
Chapter 5 covers some simple inferential statistical tests that can be used to determine:
whether a sample mean differs from a known population mean; whether there is a dif-
ference between the scores from two groups of participants; or whether there is a
difference between the scores from participants when tested under two different con-
ditions. Chapters 6 and 7 focus on statistical tests of association, while Chapters 8 and
9 describe a family of statistical procedures used to analyse data from more complex
experimental designs involving several conditions or variables.
Chapters 10–13
Chapters 10–13 look at tests that are appropriate for experiments involving more com-
plex designs or for data obtained using non-experimental methods such as surveys or
questionnaires.
In general, Chapters 5–13 cover a family of related tests, each of which is described
in its own section. We introduce each statistical test with a brief description. This
description is not intended to replace that which you will find in a statistics textbook;
rather, it is intended to act as a reminder of when and how to use the test. We also
include an example of a typical or real piece of research that might be analysed using
the test, to give you a concrete psychological example. We then give detailed, step-
by-step instructions on how to perform the analysis using SPSS. In the earlier chap-
ters, we include screenshots (pictures) and a full description of every step required to
perform the test. In later chapters, we assume that you will have become a little more
familiar with the workings of SPSS, and therefore summarise some of the simpler
operations. Each chapter includes an annotated example of the output produced by
SPSS to help you understand the results of your analyses. Finally, we include a note
on how you should report the results of your analyses. The data we use to demon-
strate the statistical tests covered in Chapters 5–7 can be found in the Appendix,
and all the data can be downloaded from the companion website (macmillanihe.com/
harrison-spss-7e).
Chapter 14
Chapter 14 introduces you to syntax and shows how using this text language to con-
trol SPSS can increase your efficiency, especially when working on larger, more
xii Preface
complex data files. In addition, we describe some useful option settings, how to print
files, how to import and export spreadsheet files, and how to incorporate SPSS output
into other documents. This chapter also describes how to access the extensive SPSS
help libraries.
Differences between versions of SPSS
This book was written using SPSS Version 26, which is quite similar to earlier versions
by IBM, and indeed to even earlier versions of SPSS (or PASW as it was briefly known).
However, the changes that have been introduced do not relate much to the tools we
introduce in this book. You may see small differences between the dialogue boxes and
the output we show and those produced by your version of SPSS. In most cases, these
are likely to be purely cosmetic, affecting only the appearance of the program (indeed,
even the same version of SPSS can look quite different on different computers, depend-
ing on the operating system and general display options). Furthermore, differences in
appearance may result from changes made to the option settings in SPSS, or as a result
of different ‘Add-Ons’ being installed. We show you how to use the option settings to
control SPSS in Chapter 14. Regardless, any appearance differences that do occur are
not likely to affect the advice and steps we present in this book.
We hope that you find our book useful and that it helps you to enjoy psychologi-
cal research.
Acknowledgements
Welcome to the seventh edition of SPSS for Psychologists. This new edition of the
book brings with it a new author, Virginia Harrison, who is very grateful to the origi-
nal authors for letting her come on board! In turn, the original authors are very glad
that Virginia agreed to join the team!
As with the previous editions, we would all like to take the opportunity to thank
the many people who have contributed to this book. In particular, we would like to
thank our colleagues, past and present, for their invaluable advice, and the various
reviewers who have provided feedback on earlier editions. These contributions have
again helped to shape the new edition. We thank you for your time and hope that you
can see where we have made amendments or additions in light of your comments. We
would also like to thank the various colleagues, supervisors, students and friends who
have contributed to our knowledge of statistics and data analysis either through for-
mal teaching, informal discussion or by presenting us with challenging data analysis
problems. Some of these people were also kind enough to allow us to utilise their data
to illustrate the use of particular statistical techniques. Richard and Gini would par-
ticularly like to thank William Hoang for his help in preparing some of the screen-
shots in this edition.
We first thought of writing a book about SPSS some 10 years before we got around
to doing it. With each new version of SPSS and each new group of students, we
revised our teaching notes and talked about what our book would look like. So, when
our first edition was published, we were delighted by the positive response it received.
Our objective was to enable students to actively engage in the discipline by undertak-
ing their own research. Reading about psychology can be interesting, but the real fun
is in doing it, and that inevitably involves the collection and analysis of data and the
use of software such as SPSS. We are therefore particularly pleased to hear from stu-
dents who tell us that our book has enabled them to complete their own research. It
is this more than anything else that keeps us motivated to update and expand our text
for each new edition. The fact that we are now into the seventh edition has amazed
and delighted us (although this demonstrates how even older we now are). Thank you
for taking the time to contact us.
Finally, we would like to thank the team at Red Globe Press for their support.
Over the seven editions of this book, we have worked with many different members
of their team. All have contributed greatly to this book, but we would particularly
like to thank Luke Block and Verity Rimmer for their help and support in producing
this edition.
Virginia Harrison
Richard Kemp
Nicola Brace
Rosemary Snelgar
xiii
CHAPTER 1
1 Introduction
In this chapter
●● Psychological research and SPSS
●● Guide to the statistical tests covered
●● Working with SPSS
●● Starting SPSS
●● How to exit from SPSS
SPSS for Psychologists online
Visit macmillanihe.com/harrison-spss-7e for data sets,
online tutorials and exercises.
Section 1: PSYCHOLOGICAL RESEARCH AND SPSS
■■ IBM SPSS statistics software (SPSS) is a widely used computer program designed
to aid the statistical analysis of data, particularly data collected in the course of
research.
■■ SPSS® has been around in various different forms for many years and has become
the ‘industry standard’ software for data analysis.
■■ SPSS is the program most widely used by university researchers, especially those
working in psychology and the social sciences. SPSS is also widely used in private
and government research organisations and many large private companies and
nongovernmental organisations.
■■ Being able to describe yourself as a competent user of SPSS will enhance your
employment prospects considerably.
■■ Although at first sight SPSS might appear rather complex, it is not difficult to learn
how to undertake a wide range of statistical analyses, and once you have mastered
these basics, they will enable you to undertake far more sophisticated research
than would be possible without the aid of such software.
1
2 SPSS for Psychologists
CHAPTER 1
ut I am studying psychology, not statistics – why
B
do I need to learn to use SPSS?
This is a common question, and the answer lies in the nature of psychology and
the type of research that many, if not most, psychologists undertake. It’s impor-
tant to remember that our current understanding of human behaviour and experi-
ence is based on the body of psychological research that has gone before us, so
knowing how this research was carried out will allow us to better understand and
evaluate the knowledge that we have; and equip us with tools to develop new
knowledge.
Much of the research that has been carried out in psychology has used the sci-
entific method, borrowing systematic research frameworks from the sciences. The
problem is that people are not the easiest of things to study because they vary both
between individuals and over time. That is, different people can react differently to
a particular situation; and how a person reacts in a situation today might be quite
different from how they react tomorrow. This means that the data collected by psy-
chologists are much more ‘noisy’ than that collected in some other sciences.
Consider, for example, a chemist investigating the properties of magnesium. The
chemist knows that, under constant conditions, every bit of magnesium will react
in the same way as every other bit of magnesium, and that how the magnesium
reacts today will be the same as how it reacts tomorrow. Thus, the chemist might
only need to do an experiment once using one piece of magnesium to draw firm
conclusions about the nature of magnesium in general, and is unlikely to need to
use statistics to help explain the results of the observations. The situation facing the
psychologist is quite different, and in order to be able to determine how, in general,
people react in a given situation, the psychologist will probably need to test a range
of different individuals and then make use of statistical techniques to determine
what trends are present in the data. So psychologists are particularly likely to need
to use statistics in their research.
In recent years, the complexity of the statistical techniques routinely used in
psychological research and taught to undergraduate students has increased consider-
ably. This routine reliance on more complex statistical analysis is made possible by the
widespread availability of sophisticated statistical analysis software such as
SPSS. Thus, in order to be able to undertake psychological research, either as a stu-
dent or a professional, you need to be able to use software such as SPSS. This book is
designed to introduce you to SPSS so that you can use the program to undertake the
statistical analyses you need for your course or research.
In the remainder of this section, we will provide a brief overview of some of the
concepts relating to research methods, data and statistical analysis that are important
to bear in mind when using SPSS to analyse your data. We refer to many of these in
later chapters. (You are probably already familiar with many of them from your lec-
tures on research methods.) Additionally, in each chapter we include an overview of
some statistical issues, but this book is not intended to be a statistics reference, and
you should also refer to a statistics text for full guidance.
Introduction 3
CHAPTER 1
Asking questions and collecting data
The aim of psychology research is to try to better understand human experience and
behaviour. This is not an easy task when you think about all of the complexities
involved in what it is to be a human being. Just think about all of the different feel-
ings, thoughts and behaviours involved in everyday things like managing your work-
load, engaging with social or mainstream media, maintaining friendships and
relationships… and that barely scratches the surface. There are a multitude of differ-
ent questions that psychologists might want to ask about human behaviour and expe-
rience… What makes someone more likely to experience depression than another?
What treatments work best for anxiety? Do people act differently on their own and in
public? How does social media affect the way we think or act? Does multitasking
reduce productivity? What factors predict criminality? Do we make good eyewit-
nesses? Why do we like some people and not others?
The different types of questions that we can ask about human behaviour are so
wide and varied that psychologists have devised a huge array of different methods
that we can use to help us better understand this vast topic. Which method we need
to use to answer a particular question depends on the type of question we are asking
and the type of data we need to answer it.
Some questions we might ask lend themselves to more qualitative methods of
research. For example, if we wanted to explore how people subjectively experience or
give meaning to certain events, we might want to collect rich qualitative data through
interviews or focus groups to answer our question. Alternatively, other questions may
be better explored using quantitative methods. Quantitative methods allow you to
measure, score or count the things that you are interested in. This allows you to col-
lect numerical data, identify patterns and make inferences about the relationships
between different variables. For example, investigating how the amount of exercise
people do relates to a measure of their mental health.
Methods commonly used in psychological research include questionnaire studies,
interviews, observation and experiments. Each of these methods can result in the col-
lection of quantitative data suitable for analysis using SPSS. What we do with that
data will depend on what questions we are trying to answer, or what hypotheses we
are trying to test. This book will focus on some of the quantitative methods commonly
used in psychological research.
Levels of measurement
SPSS is designed to aid the analysis of quantitative data, which usually consists of a
series of measurements of one or more variables. A variable is simply some quantity
that varies and can be measured; so, height, time, weight, sex (male or female) and IQ
test score are all examples of variables. A typical data set in psychological research
will consist of several variables each measured for a set of different participants or
cases.
We use a scale to make the measurements of a variable, and the characteristics of
the scale determine the characteristics of the data we collect and ultimately what
descriptive and inferential statistics we can legitimately undertake on this data. Many
statistics texts written for psychology students describe four different types of scales,
or levels of measurement: nominal, ordinal, interval and ratio. When entering your
4 SPSS for Psychologists
CHAPTER 1
data into SPSS you will need to tell SPSS what types of variable you have. SPSS uses
the following terms to classify variable types: nominal, ordinal or scale (‘scale’ refers
to either interval or ratio variables). We describe them briefly below, along with an
illustration of the icons SPSS uses to denote these different levels of measurement.
1. Nominal level
Nominal (also referred to as ‘categorical’) variables are categorised rather than
measured in the strictest sense. As SPSS needs us to enter numbers to repre-
sent our data, we have to assign a numerical value to the different groups or
categories within our variable. For example, we might decide to record the
sex of our participants and to enter these data into SPSS by adopting the
coding scheme 1 = Male, 2 = Female. Assigning 1 or 2 allows us to differen-
tiate between these two groups of participants, but the numbers themselves
are meaningless in this context; these values should not be taken to imply any
more than a label (or a name, hence the term ‘nominal’). With nominal data
just about the only thing we can do is count or measure frequency. We can
report how many men and women we have, but there is little else we can do
with these data and it would make no sense to calculate statistics such as the
mean sex of our participants.
SPSS does not automatically know about the level of measurement used
to collect your data. It is up to you to make sure that you do not ask SPSS
to undertake any inappropriate analysis, such as calculating the mean of
nominal data.
2. Ordinal level
Ordinal variables are measured using a scale that allows us to imply order or
rank. An ordinal variable tells you more than a nominal variable; for example,
it may be legitimate to say that the student who was assigned a rank of 1
performed better than the student with rank 2. However, what we can do with
these data is still limited because we do not know how much better the rank
1 student performed than the rank 2 student, and because it is unlikely that
the difference in ability indicated by ranks 1 and 2 will be the same as that
between ranks 2 and 3, or ranks 107 and 108. Thus an ordinal scale is still a
rather limited measure of performance.
3. Interval and ratio (scale) levels
Measuring at interval or ratio levels provides us with numbers that are more
number-like. If we have an interval or ratio scale, then we can be certain that
3 is more than 2 and 4 is more than 3. Furthermore, we know that the interval
between all points on the scale is the same; for example, the difference be-
tween 23 and 24 is the same as the difference between 102 and 103.
An interval scale is a scale that has an arbitrary zero so that a value of
zero doesn’t indicate that you have none of the quantity being measured. In
a ratio scale, zero means that there is none of the quantity being measured. In
Introduction 5
CHAPTER 1
practice, the difference between these two types of data is not critical and
SPSS does not distinguish between them, using the term scale to describe
both interval and ratio variables. For your research methods modules, how-
ever, you will probably need to know the difference between interval and
ratio levels of measurement.
Hypotheses
As mentioned above, psychology research follows the scientific method – formulating
ideas (or hypotheses) about the way the world works, and then testing these ideas
against reality by observing, measuring and analysing data. We are all able to hypoth-
esise or speculate about the underlying causes of people’s behaviour – we do it every
day. Whether it’s assuming the stumbling person on the street has had too much to
drink, or assuming that genetics are responsible for your statistics ability (we’ll let
you decide whether that’s a positive or a negative assumption…), we are making
inferences about the causes of behaviour. In psychological research we essentially do
the same thing, but we use statistics to test these speculations (hypotheses) in a formal
and rigorous way.
In a research context, a hypothesis is a prediction about the outcome of the
research you plan to carry out. The hypothesis, often known as a ‘research hypothe-
sis’, ‘experimental hypothesis’ or ‘alternative hypothesis’, makes a very specific pre-
diction about something that can be defined, measured and tested. Depending on
your study design, it predicts that there will be an association between variables, or
that there will be a difference between the conditions you are investigating (which
allows you to say whether there is an effect of one (or more) variable(s) on another).
The null hypothesis, by contrast, expresses the possibility that there is no effect or
association.
When analysing data using SPSS you need to be clear about whether your research
hypothesis is one-tailed or two-tailed. A one-tailed hypothesis makes predictions
about both the presence of an effect (e.g. there will be a difference in the performance
of young and old participants on a memory test) and also about the direction of this
difference or association (e.g. young participants will obtain a higher score on a mem-
ory test than elderly participants). In contrast, a two-tailed hypothesis predicts only
the presence of an effect, not its direction. With certain statistical procedures, SPSS
will ask you to specify whether your hypothesis is one- or two-tailed.
Operationalisation
To examine how a hypothesis is formulated, let’s imagine you are a psychologist inter-
ested in investigating how human memory works. This is an extremely broad research
question which could be interpreted in a number of different ways, depending on
how it is defined. For example, are you interested in recognition or recall; do you want
to know if we remember some things better than others, and if so, why; perhaps
you’re interested in how short-term memory changes according to personal character-
istics, such as age; or are you more interested in the content of memories themselves?
It is such a broad topic, it is difficult to know where you would start to carry out
research in this area. And this is where hypotheses come in.
6 SPSS for Psychologists
CHAPTER 1
Before a researcher can explore their topic of interest (e.g. how people remember
things), they first need to turn the concepts they are interested in from something
vague and abstract into something more concrete. This is a process known as opera-
tionalisation, and essentially involves identifying, narrowing down and defining spe-
cific factors or psychological constructs in a way that allows them to be directly
accessed or measured.
In this example, you would need to find a way to measure the aspects of memory
(and any other variables) that you are interested in. For example, if you were inter-
ested in how age is related to memory, you might want to look at how many words
people are able to remember from a long word list and look at how that changes as a
function of their age (measured in years).
Once your specific variables and measures have been identified and properly
defined, you can produce your specific hypothesis (or prediction). While a research
question is often exploratory, a hypothesis makes a specific prediction about the pat-
tern of results that are expected. This prediction should always be theoretically
grounded in previous research. For this example, previous research suggests that
short-term memory declines with age. As such, a suitable hypothesis for this research
might be something like: Participants will be able to recall fewer words as their age
increases.
You now have a specific statement that makes a prediction about people’s behav-
iour. This statement lends itself to being investigated in a single study and clearly
identifies the variables to be tested. Once a testable hypothesis has been identified,
you then need to think carefully about how you can design an appropriate and scien-
tifically rigorous study that can address it. The aim is to find evidence to either sup-
port or refute your prediction.
Types of study design
Broadly speaking, there are two types of study design that can be used to test hypoth-
eses: correlational designs and experimental designs. Which one you choose will
depend on your specific hypothesis and how you have operationalised your
variables.
Correlational designs
Correlational designs are used to investigate hypotheses that predict associations or
relationships between variables. That is, whether changes that happen to one variable
are related to changes that happen in another (like memory scores decreasing as age
increases). To take another example, imagine that you predict a relationship between
time spent doing statistics and general happiness scores. You could measure people’s
happiness using a scale so that each individual receives a happiness score, and then
you could also measure the amount of time they spend doing statistics each week. You
might find that, as the amount of time spent doing statistics increases, people’s happi-
ness levels decrease; in other words, as one variable changes so does the other. This is
an example of a correlational study. (Disclaimer: This is an entirely fictional example,
and is in no way suggesting that doing statistics is bad for your health. Obviously,
everybody loves statistics!)
