A GRAPHICAL TECHNIQUE FOR DISPLAYING CORRELATION-MATRICES

A recent multilaboratory study to investigate phenotypic and genotypic markers for alcoholism is described. A preliminary investigation of t he reliability of electroencephalographic data amassed at six laborato ries was undertaken; data from each laboratory could be usefully summa rized with sample correlation matrices. A graphical method for depicti ng these correlation matrices is presented. The Larntz-Perlman procedu re for assessing the equality of correlation matrices can immediately be incorporated into this graphical technique. The experimental data f rom the preliminary investigation are illustrated with the graphical m ethod, which readily conveys large amounts of correlational informatio n and reveals meaningful patterns in the electroencephalographic respo nses.