go to the School of Psychology home page Go to the UNE home page
Chapter 7 - Analysing the Data Part IV - Analysis of Variance Chapter 1 - Behavioural Science and research Chapter 2 - Research Design Chapter 3 - Collecting the Data Chapter 4 - Analysing the Data Part I - Descriptive Statistics Chapter 5 - Analysing the Data Part II - Inferential Statistics Chapter 6 - Analysing the Data Part III - Common Statistical Tests The Scientific Approach The Research Process Variables Confounds The Logic of Cause and Effect Research Hypothesis and Predicition Readings and Links

Chapter 1
Behavioural Science and Research



Confounds are special types of variables that we would rather not know about! They are nuisance variables that interfere with our attempts to explain a relationship between our variables of interest. A confounding variable is one that provides an alternative explanation for the thing we are trying to explain with our independent variable. A simple example, which we deal with later in more detail, is a study in which we want to find an improved way of teaching children the alphabet. We get two groups of kids and teach one of them a new "sing the alphabet" method and the other group boringly and repetitively pronounce each letter of the alphabet. Lo and behold we find that the children learn much better using the new singing method! So, here we want to explain the improved learning (DV = number of letters correct) as being caused by the improved singing method (the IV). We should encourage all primary teachers to get their children to sing the alphabet every morning! But what if the children in the group that got the singing alphabet method were all 4th grade kids and the kids in the standard boring group were only 2nd grade kids? Might not the improved learning be due to the older age of the kids in that group? Age here is a confounding variable. It competes with our IV of interest in trying to explain the differences we found with the DV. Similarly if the singing group were all females and the pronouncing group were all males (gender is the confounding variable), or the singing group were all top class students and the pronouncing group were all bottom class students (IQ is the confound). Motivation might also be a confound.




© Copyright 2000 University of New England, Armidale, NSW, 2351. All rights reserved

UNE homepage Maintained by Dr Ian Price
Email: iprice@turing.une.edu.au