Applied Statistics with SPSS

Eelko Huizingh

• Publisher: SAGE Publications Ltd
• Year: 2007
• Print ISBN: 9781412919319, 9781412919302
• eBook ISBN: 9781446223000
• Edition: 1
• Topic: Social Sciences
• Subtopic: Social Science Research & Methodology
• Index: Social Sciences

Accessibly written and easy to use, Applied Statistics Using SPSS is an all-in-one self-study guide to SPSS and do-it-yourself guide to statistics. Based around the needs of undergraduate students embarking on their own research project, the text?s self-help style is designed to boost the skills and confidence of those that will need to use SPSS in the course of doing their research project.

The book is pedagogically well developed and contains many screen dumps and exercises, glossary terms and worked examples. Divided into two parts, Applied Statistics Using SPSS covers :

1. A self-study guide for learning how to use SPSS.

2. A reference guide for selecting the appropriate statistical technique and a stepwise do-it-yourself guide for analysing data and interpreting the results.

3. Readers of the book can download the SPSS data file that is used for most of the examples throughout the book.

Geared explicitly for undergraduate needs, this is an easy to follow SPSS book that should provide a step-by-step guide to research design and data analysis using SPSS.

Part I

Learning to work with SPSS

The first part of this book consists of three introductory and three do-it-yourself chapters. The first three chapters explain the background and applications of SPSS, while the other three, in the form of tutorials, show the way SPSS works.

Chapter 1 presents the history and features of the SPSS data analysis system. Chapter 2 discusses the question of how you can use SPSS in your own research. Seven stages are distinguished in the research process and we explain how SPSS and the module Data Entry can be of help in each of these stages. In each case the discussion includes an overview of the relevant SPSS commands.

Before any statistical analysis can be performed, a database needs to be made available. This usually involves ‘translating’ the measurements or survey questions into variables, and, after coding the data, entering the data into a computer file. This process is summarised in the form of a code book, which is the subject of Chapter 3.

The last three chapters of Part I are of a practical nature. Each consists of a session which takes a simple example to show all essential topics that one encounters in working with SPSS. The three sessions have been written in a way to enable you to study them while using your computer, that is, all mouse clicks and keystrokes are explicitly noted, with further explanations where necessary. In these sessions we simulate a small research project. During this guided tour you will perform the same tasks that you have to do in your own research.

What is the purpose of the sessions?

For experienced SPSS users, the sessions are a means of quickly getting acquainted with the most recent version of the software. For those who have no prior knowledge of the program (in SPSS parlance they are new friends, as distinguished from the experienced old friends), the sessions provide an overview of the structure, the operational aspects and the features of SPSS.

1	Background to SPSS for Windows

In a world in which the pace of change is constantly accelerating, SPSS is one of the oldest, yet still widely used, statistical software packages. The first version of SPSS appeared on the market way back in 1968! Section 1.1 summarises the development that the package has seen since that time. The SPSS software consists of a base module needed for all applications, plus a number of add-on modules for additional purposes that are briefly discussed in Section 1.2.

1.1 SPSS: history and development

Many statistical computations involve sequences of simple arithmetic operations repeated many times over. Consider, for example, the computation of a common statistic such as the standard deviation. This is done as follows:

• determine the number of cases;
• compute the sum of all observed values;
• divide this sum by the number of cases: this gives the mean;
• for each case, compute the observed value minus the mean;
• compute the squares of these differences;
• determine the sum of all squares;
• divide this sum by the number of observations minus 1;
• find the root of the division result.

Although the detailed recipe thus involves quite a number of steps, clearly the knowledge required to execute each individual step barely exceeds the level of primary education. It is a useful exercise to perform this computation by hand once or twice, to get some feel for statistics. But doing the arithmetic in situations involving a hundred cases would be a formidable task, not to mention the likely errors. It is not surprising, therefore, that from the very beginning of the computer age statisticians have been eagerly using electronic technology as an aid to arithmetic. Equally understandably, it was the advent of computers that marked the true take-off of applied statistics.

Ever since 1968 when SPSS was introduced as a statistical analysis program for the social sciences, the package has been part of this development. The original meaning of the abbreviation SPSS is Statistical Package for the Social Sciences, but in its present form SPSS is much more than that. It has grown into a comprehensive modular package that handles data, data processing and format. The target group has long included users from disciplines other than the social sciences. SPSS can be used wherever data are collected, analysed and presented in the form of tables and charts.

One of its strong points is that SPSS is capable of performing almost all common types of analysis. This makes the software particularly suitable for analysing data obtained by questionnaires. Such research projects usually include variables at different measurement levels, that is, ratio or interval data occur together with nominal or ordinal variables. The broad applicability of the program is also evidenced by the number of users, which SPSS claims to exceed 3 million worldwide.

1.1.1 Evolutions in statistics and technology

Over the many years since 1968 the SPSS package has gone through many adaptations to match the ongoing developments, both in statistics and in technology. Many of the statistical methods that have been developed or improved over the last three decades have been included. Additionally, SPSS has seen the development from computing centres dedicated to operating large mainframes, to notebook computers the size of a writing pad. In the past, users had to prepare sets of instructions and deliver them, at first in the form of stacks of punch cards at the desk of a computing centre, and in later years as electronic files via a computer terminal. At some later time the SPSS user would then pick up the printed output from the service desk. Subsequent program versions were increasingly user-friendly, a development that took place by many modest steps in addition to two major advances. The first major step was the introduction, in 1983, of an SPSS version for personal computers (SPSS/PC). The second major step came nine years later, with the introduction of SPSS for Windows in 1992. This version was still easier to use, as it was the first that did not require knowledge of the special SPSS command language. Since then, SPSS has continued to become ever easier to learn and use.

1.1.2 The users

Thus, over time, the SPSS software has gone through a far-reaching evolution process. In addition, the user profile has significantly changed as well. The ‘first’ SPSS user, towards the end of the 1960s, was used to carrying out many of the techniques by hand. As a result, this user was thoroughly familiar with the assumptions, the computation methods and the interpretation of the outcomes. Many present-day SPSS users are much less familiar with these fundamentals, which carries the risk of incorrectly performing or interpreting analyses. The increased ease of use enhances that risk: with SPSS, any layperson nowadays can perform advanced statistical analyses, but a full grasp of what this entails is another matter. SPSS offers little protection against such risks and therefore a good basic understanding of statistics remains a prerequisite for properly applying these techniques.

1.2 The SPSS data analysis system

SPSS offers a very broad range of statistical methods. As such, this has the potential disadvantage that users have options at their disposal that many will never use. That is why SPSS has been organised into modules that together form the SPSS data analysis system. Users can decide which of these special modules they need in addition to the base module; these modules can be bought separately.

This book deals only with the various features of the base module and the module Data Entry. All applications of SPSS require the base module. It contains commands for creating and transforming data files, plus the most widely used methods of analysis. Examples of these include frequency tables, crosstables, m...

Table of contents

• Cover Page
• Title
• Copyright
• Contents
• Preface
• Part I Learning to work with SPSS
• Part II Working with SPSS and Data Entry
• Index