Abstract
In complex human-machine systems, multidimensional behavior is required of the operator. Consequently, there is no commonly-accepted, single measure of operator performance which can be utilized to determine the efficiency of the human-machine interaction. Because the behavior is multidimensional, multivariate statistics must be used to analyze the multiple measures gathered during system evaluation. While multivariate analogues to analysis of variance (ANOVA) exist, there are also a number of candidate multivariate analogues to the post-ANOVA simultaneous comparison tests. This paper describes a newly developed multivariate, simultaneous comparison test–Finite Intersection Test (FIT)–and provides an example of FIT's application to the analysis of multivariate data.
Get full access to this article
View all access options for this article.