Two neural network paradigms-multilayer perceptron and learning vector quan
tization-were used to study voluntary employee turnover with a sample of 57
7 hospital employees. The objectives of the study were twofold. The Ist was
to asses whether neural computing techniques offered greater predictive ac
curacy than did conventional turnover methodologies. The 2nd was to explore
whether computer models of turnover based on neural network technologies o
ffered new insights into turnover processes. When compared with logistic re
gression analysis, both neural network paradigms provided considerably more
accurate predictions of turnover behavior, particularly with respect to th
e correct classification of leavers. In addition, these neural network para
digms captured nonlinear relationships that are relevant for theory develop
ment. Results are discussed in terms of their implications for future resea
rch.