Dear Catherine, If you have access to PsycINFO (via http://www.csa1.co.uk for instance) and search for general linear model, over a 100 papers come up, some of which look like they might have results in them. The journal Psychological Methods seems to be where the theoretical issues are discussed. Might be a useful starting point. I haven't used GLM but will probably need to in the future, so I'd be interested in what you find out. With kind regards, Brian Saxby Junior Research Associate