For a gentle but thorough introduction, maybe try Crawley's 'Statistics: an introduction using R' (2005) - it's aimed at non-mathematicians and non-statisticians, and includes a chapter on contrasts. Or scan its 'further reading' list for the more advanced references... http://www.amazon.co.uk/Statistics-Introduction- Michael-J-Crawley/dp/0470022981 Katherine > From: Gordon Smyth <smyth at="" wehi.edu.au=""> > Subject: [BioC] Reference on linear models [was LIMMA makeContrasts > question] > To: "J.delasHeras at ed.ac.uk" <j.delasheras at="" ed.ac.uk=""> > > The introduction by Glantz and Slinker: > http://www.amazon.com/Primer-Applied-Regression-Analysis- Variance/dp/0070234078 > > might be useful. Although the emphasis is on regression with > continuous covariates, it includes material on linear models with categorical predictors.