Correspondence Analysis applied to psychological research

Abstract

Correspondence analysis is an exploratory data technique used to analyze categorical data (Benzecri, 1992). It is used in many areas such as marketing and ecology. Correspondence analysis has been used less often in psychological research, although it can be suitably applied. This article discusses the benefits of using correspondence analysis in psychological research and provides a tutorial on how to perform correspondence analysis using the Statistical Package for the Social Sciences (SPSS).Correspondence analysis (CA) has become most popular in fields such as ecology, where data is collected on the abundance of various animal species in specific sampling units/areas (terr Braak, 1985). The amount of data involved in these samples makes it difficult to get a clear sense of the data at a glance (Palmer, 1993). With the use of CA, an exploratory data technique for categorical data, ecologists have been able to transform these complicated tables into straightforward graphical displays (Hoffman & Franke, 1986). Ecological data is multidimensional making visualization of more than two dimensions difficult. Correspondence Analysis is an ideal technique to analyze this form of data because of its ability to extract the most important dimensions, allowing simplification of the data matrix (Palmer). However, this statistical technique merits further attention within the field of psychological research. In fact, the lack of awareness of this useful statistical method puts psychological researchers at a disadvantage. To demonstrate that CA can be suitably applied to psychological research, this paper will use a research question from community psychology. Additionally, due to the gap in the literature on how to perform CA using statistical software programs, the current paper will describe how to perform CA using SPSS software.

Correspondence Analysis: A New Method for Analyzing Qualitative Data in Architecture

“…CA provides useful information about the data and allows one to explore the underlying structure of categorical variables [Greenacre 2007[Greenacre , 2010Doey and Kurta 2011]. More specifically, the CA technique is a tool to transform a table of numerical information, commonly with a large number of data, into a graphical display known as a biplot (Gabriel biplot), in which each row and each column is shown as a point.…”

Minimaliojo darbo užmokesčio ir makroekonominių rodiklių sąsajos Europos Sąjungoje

“…Dažniausiai pasirenkamos dvi arba trys ašys, kurias lengviau interpretuoti grafiškai. Atitikties žemėlapis dažniausiai yra sudarytas iš dviejų ašių, atitinkančių du pagrindinius faktorius, o kiekviena suderinamumo lentelės eilutė ir stulpelis tampa taškais atitikties žemėlapyje (Doey, 2011).…”

Visualizing categorical data: An introduction to correspondence analysis for technical communication researchers

“…There are four major assumptions when using CA [6]. First, CA should be used only when all of the variables of interest (both independent and dependent) are categorical.…”

“…Even though CA is not used for hypothesis testing, it does use the chi-square statistic, or a weighted Euclidian distance, to measure the distance between points [6]. It should be noted that CA is a non-parametric statistic and has no theoretical distribution to compare observed distances [6].…”