Quantitative Analysis of Questionnaires

Techniques to Explore Structures and Relationships

Steve Humble

• Publisher: Routledge
• Year: 2020
• Print ISBN: 9780367022839
• eBook ISBN: 9780429682735
• Topic: Education
• Subtopic: Education General

Bringing together the techniques required to understand, interpret and quantify the processes involved when exploring structures and relationships in questionnaire data, Quantitative Analysis of Questionnaires provides the knowledge and capability for a greater understanding of choice decisions. The ideal companion for non-mathematical students with no prior knowledge of quantitative methods, it highlights how to uncover and explore what lies within data that cannot be achieved through descriptive statistics. This book introduces significance testing, contingency tables, correlations, factor analysis (exploratory and confirmatory), regression (linear and logistic), discrete choice theory and item response theory.

Using simple and clear methodology, and rich examples from a range of settings, this book:

• provides hands-on analysis with data sets from both SPSS and Stata packages;
• explores how to articulate the calculations and theory around statistical techniques;
• offers workable examples in each chapter with concepts, applications and proofs to help produce a higher quality of research outputs;
• discusses the use of formulas in the appendix for those who wish to explore a greater mathematical understanding of the concepts.

Quantitative Analysis of Questionnaires is the ideal introductory textbook for any student looking to begin and or improve statistical learning as well as interpretation.

Chapter 1

Introduction

In today’s world, choice has become part of our daily routine. From which coffee to order, which mode of transport to use to get to work or which university to attend the multitude of options makes for exciting decision-making processes. Buying online has brought about the emergence of rating providers, sellers and products. This informs future would-be market participants around quality, reliability and appropriateness. Therefore completing rating scales and questionnaires has become commonplace.

As a researcher, having the knowledge and capability to understand, interpret and quantify why and how answers are given when there are choices to be made allows for a greater understanding of the structures around choice. When collecting data the questionnaire is one protocol that can provide rich data relatively easily and simply.

This book works for all types of researchers from those with no prior knowledge of quantitative methods to those who wish to expand their range of statistical techniques. It provides statistical approaches that are appropriate for in-depth analysis of questionnaires. These include questionnaires that can be simple dichotomous, Likert scale, or stated and revealed preference choice.

A range of techniques highlights how to explore and uncover what lies within the data to reveal structures and relationships that are not discovered through descriptive statistics. This accessible book will both illustrate and expand the reader’s understanding of the statistical techniques used in this area.

Each of the chapters in this book introduces a specific statistical concept required to evaluate questionnaire data. All the chapters contain illustrative data sets for the reader to explore while learning. These sections provide students undertaking research dissertations at masters or undergraduate level or those studying for higher degrees such as a doctorate in the humanities and social sciences the basic principles, techniques and applications of questionnaire statistical analysis. The text guides the reader through hands-on analysis with data sets and support for using both SPSS and Stata packages for these statistical concepts. The book hopes to bridge a gap between theoretical understanding and easy to use statistical packages. The chapters all detail how to use statistical package calculations and give reasons why the statistical techniques are explored and articulated through a range of relevant data sets for humanities and social science students.

Chapter 2 examines statistical significance and the use of contingency tables. The chapter begins by providing a brief introduction to statistical significance and p-values. Data on coffee-drinking habits is explored to illustrate how to use contingency tables, Chi-square test, Cramer’s V test and how to calculate odds ratios. The concluding part of the chapter looks at how to report and calculate contingency tables when using SPSS and Stata.

Chapter 3 considers exploratory factor analysis and why this is an important tool when exploring structures and relationships in Likert scale data. This chapter sets out the principles and procedures that are involved when carrying out exploratory factor analysis describing and justifying the different possible methods of selection that can be performed when using factor analysis. The first part of this chapter discusses essentials that need to be considered in order to decide whether factor analysis is suited to your data. The chapter explains how communality, Kaiser-Meyer-Olkin test, Bartlett test of Sphericity and Scree plots can all be used to help inform the factor model. Example data sets are used to illustrate the difference between principal factor analysis and principal component analysis. The chapter concludes by discussing when to use a particular latent factor score extraction method. The three methods considered are the Regression method, Bartlett method, and the Anderson-Rubin method.

The statistical procedures correlation and regression are examined in Chapter 4. The chapter first explores how we can visually understand correlation through scatter diagrams and moves on to appreciate why more rigorous statistical techniques are required to prove correlations in data. The chapter then explains how to carry out Pearson Correlation, Spearman’s Rank Correlation and Kendall’s Tau Correlation illustrating these techniques through a parental motivation study. The second part of the chapter explores linear and multivariate regression. The chapter goes on to describe how the technique of purposeful selection can be used to aid the researcher when deciding on which variables are to be included in the model.

Confirmatory factor analysis is explored in Chapter 5. This chapter takes forward the concepts from Chapters 3 and 4 to show how we can confirm latent structures in a questionnaire. Firstly the chapter explains how a model can be constructed. The happiness study is used to illustrate how to carryout confirmatory factor analysis using Structural Equation Modeling (SEM). The chapter then shows how to assess hypothesized fit structures of items in grouped constructs using fit and comparison indices. Following on from this example two further studies are explored to illustrate the depth of complex themed structures that can be uncovered when confirmatory factor analysis is applied to questionnaire data. The chapter ends with details on how to build, calculate, assess and report confirmatory factor analysis.

Chapter 6 explores the use of logistic regression for discrete dependent variables having two or more possible values. This chapter acts a precursor to Chapter 7 on discrete choice theory where the techniques learnt in Chapter 6 will be explored in greater depth. The chapter starts by looking at how to perform simple logistic regression when the independent variable is either dichotomous, polychotomous, or continuo...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• List of illustrations
• About the author
• About the book
• Chapter 1. Introduction
• Chapter 2. Statistical significance and contingency tables
• Chapter 3. Factor analysis: Exploratory
• Chapter 4. Correlation and linear regression
• Chapter 5. Factor analysis: Confirmatory
• Chapter 6. Regression: Logistic
• Chapter 7. Making choices: Discrete choice theory
• Chapter 8. Item response theory
• Appendix
• References
• Index