Abstract

The present paper describes a package of two computer programs that provide a very convenient method of analysing data that are generated by the difference between two scores. Our focus has been on the difference between a target score and a performance score; and such error scores have been gathered in a variety of experiments involving serial, coincident and single aiming tasks with the measures being in the dimensions of time or space. Four main kinds of error are calculated: constant error (CE), absolute error (AE), variable error (VE) and total error (E); and these may be submitted to a variety of proportional transformations. The results are provided for each individual subject, for a single group or for a combination of groups; and may be formatted for input for standard analysis of variance packages. The plotting program allows considerable flexibility in constructing graphs; and the interactive features help to provide a tool that is easily learned, powerful and efficient. Verf.-Referat