Correspondence Analysis: Theory, Practice and New Strategies

Chapter 12Computing and Correspondence Analysis

12.1 Introduction

Since the early development of correspondence analysis, undertaking the necessary computations has been an ever-important aspect. Naturally, the evolution of programs and code during that time for performing correspondence analysis (and other related techniques) started in France, and it started small. Much of the early contributions ran on DOS and required, by today's standard, a very tiny amount of memory to be run. Initially the availability of the packages, programs and code was only through direct contact with the author of the software (or his/her colleague) and was, again by today's standard, often very slow. The impact of the World Wide Web during the 1990s helped bring the world closer together and kept people informed about issues on a global level more quickly than ever before –correspondence analysis was no exception. The development of packages and code to perform correspondence analysis also quickly grew with the advent of the Web. As a result, the past three decades have seen an increasing number of articles that describe the need for, and resources available for, performing correspondence analysis –see, for example, the brief discussions of Beh (2004a) and Beh and Lombardo (2012). In this chapter, we shall explore this important aspect of correspondence analysis more fully.

12.2 A Look Through Time

12.2.1 Pre-1990

Prior to the World Wide Web, the availability of packages was generally only ...