Rk. Misra et Mdl. Easton, A note on the number of morphometric characters used in fish stock delineation studies employing a MANOVA, FISH RES, 42(1-2), 1999, pp. 191-194
In a fish stock delineation study based on MANOVA/MANCOVA of a morphometric
data set, use of measurements on many characters will result in a statisti
cal artifact generating spurious significant differences between population
s. A preliminary method is demonstrated for testing the presence of this ar
tifact. Data analyzed for stock identification purposes should employ only
a small number of morphometric characters, perhaps three or four. (C) 1999
Elsevier Science B.V. All rights reserved.