On the Distribution of the Direction and Collinearity Factors in Discriminant Analysis

1974 ◽  
Vol 17 (2) ◽  
pp. 277-281 ◽  
Author(s):  
R. P. Gupta ◽  
R. D. Gupta

If the samples of sizes n1, n2,…, nq+1 are available from (q+1) normal populations with different mean vectors μα, α=l, 2,…, q+1 and the same covariance matrix ∑ and if x’=(x1, x2, …, xp) denotes the vector of p variates on which the measurements are made, then we obtain the following multivariate analysis of variance table.

Author(s):  
Oana Popa ◽  
Elena Iorgu ◽  
Beatrice Kelemen ◽  
Dumitru Murariu ◽  
Luis Popa

Morphometric analysis of some populations of lymnocardiid species (Mollusca: Bivalvia) from Razelm Lake Complex (Romania) In this paper we report the morphometric analysis of some populations of Lymnocardiid species from the lakes Razelm and Goloviţa. We used three measurements ratios to perform a discriminant analysis and a multivariate analysis of variance (MANOVA), in order to compare the species Hypanis colorata vs. Hypanis angusticostata in the two lakes, H. colorata in Razelm vs. Goloviţa, and H. angusticostata in Razelm vs. Goloviţa. From this analysis we concluded that the multivariate means of the morphological variables used in this study were highly significantly different (p=3.2e-05) between the two species. Concerning the geographical variability, in both species, the analysis showed no significant difference between the populations in the two investigated lakes. We also determined from a fitting curve analysis, that the growth pattern of both species shows length-height isometry and width-length and width-height negative allometry.


1972 ◽  
Vol 31 (2) ◽  
pp. 607-614 ◽  
Author(s):  
Lawrence M. Aleamoni ◽  
Makonnen Yimer ◽  
J. Maurice Mahan

The purpose of this study was to determine if (a) “folklore” about a teacher contributes to his ratings on a course evaluation questionnaire and (b) changes in students' attitudes during the course of instruction can be determined with a questionnaire on course evaluation. Multivariate analysis of variance and discriminant analysis, ideally suited for this type of research, were employed. The results indicated that there were no significant differences in attitudes toward a course in educational statistics between those who took the course in 1967–1968 and those who took it in 1968–1969. This seems to indicate that students do not build a “folklore” about a course based upon the course presented a year earlier.


Sign in / Sign up

Export Citation Format

Share Document