ABSTRACT

THE methods of analysis to be discussed all apply to data which can be set out in a table with a number of columns, say m, and a number of rows, say N. Each column refers to a psychological test and each row to a person: the entry or element in the Nth row and m th column is the Nth person’s score on the mth test. Any such N × m table can be represented geometrically by a dispersion or scatter of N points in an m-dimensional space. To examine this geometrical model let us begin with one person’s score on two tests.