Copyright © 2017 by John Wiley & Sons, Inc. All rights reserved.
Published by John Wiley & Sons, Inc., Hoboken, New Jersey.
Published simultaneously in Canada.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permission.
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com.
Library of Congress Cataloging-in-Publication Data
Names: Upton, Graham J. G., author.
Title: Categorical data analysis by example / Graham J.G. Upton.
Description: Hoboken, New Jersey : John Wiley & Sons, 2016. | Includes index.
Identifiers: LCCN 2016031847 (print) | LCCN 2016045176 (ebook) | ISBN 9781119307860 (cloth) | ISBN 9781119307914 (pdf) | ISBN 9781119307938 (epub)
Subjects: LCSH: Multivariate analysis. | Log-linear models.
Classification: LCC QA278 .U68 2016 (print) | LCC QA278 (ebook) | DDC 519.5/35–dc23
LC record available at https://lccn.loc.gov/2016031847
This book is aimed at all those who wish to discover how to analyze categorical data without getting immersed in complicated mathematics and without needing to wade through a large amount of prose. It is aimed at researchers with their own data ready to be analyzed and at students who would like an approachable alternative view of the subject. The few starred sections provide background details for interested readers, but can be omitted by readers who are more concerned with the “How” than the “Why.”
As the title suggests, each new topic is illustrated with an example. Since the examples were as new to the writer as they will be to the reader, in many cases I have suggested preliminary visualizations of the data or informal analyses prior to the formal analysis. Any model provides, at best, a convenient simplification of a mass of data into a few summary figures. For a proper analysis of any set of data, it is essential to understand the background to the data and to have available information on all the relevant variables. Examples in textbooks cannot be expected to provide detailed insights into the data analyzed: those insights should be provided by the users of the book in the context of their own sets of data.
In many cases (particularly in the later chapters), R code is given and excerpts from the resulting output are presented. R was chosen simply because it is free! The thrust of the book is about the methods of analysis, rather than any particular programming language. Users of other languages (SAS, STATA, ...) would obtain equivalent output from their analyses; it would simply be presented in a slightly different format. The author does not claim to be an expert R programmer, so the example code can doubtless be improved. However, it should work adequately as it stands.
In the context of log-linear models for cross-tabulations, two “specialties of the house” have been included: the use of cobweb diagrams to get visual information concerning significant interactions, and a procedure for detecting outlier category combinations. The R code used for these is available and may be freely adapted.
GRAHAM J. G. UPTON
Wivenhoe, Essex
March, 2016
A first thanks go to generations of students who have sat through lectures related to this material without complaining too loudly!
I have gleaned data from a variety of sources and particular thanks are due to Mieke van Hemelrijck and Sabine Rohrmann for making the NHANES III data available. The data on the hands of blues guitarists have been taken from the Journal of Statistical Education, which has an excellent online data resource. Most European and British data were abstracted from the UK Data Archive, which is situated at the University of Essex; I am grateful for their assistance and their permission to use the data. Those interested in election data should find the website of the British Election Study helpful. The US crime data were obtained from the website provided by the FBI. On behalf of researchers everywhere, I would like to thank these entities for making their data so easy to re-analyze.
GRAHAM J. G. UPTON