
A FACTOR ANALYSIS APPROACH TO ROUND ROBIN DESIGNS
Author(s) -
Wong George Y.
Publication year - 1981
Publication title -
ets research report series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.235
H-Index - 5
ISSN - 2330-8516
DOI - 10.1002/j.2333-8504.1981.tb01280.x
Subject(s) - inference , variance (accounting) , set (abstract data type) , factor (programming language) , data set , computer science , analysis of covariance , mathematics , linear model , statistics , artificial intelligence , accounting , business , programming language
We adopt a factor analytic approach to study round robin analysis of variance models appropriate for round robin interaction data with the actor and partner effects of any member involved linearly related. Such models are particularly important both statistically and computationally in sociometric studies on structures of social relationships involving a large number of subjects. We provide a convergent algorithm for computing the maximum likelihood estimates of the variances and covariances associated with these models. We also discuss the posterior inference of the linear effects. A detailed analysis on a set of speech data is worked out using these models.