Looking for prognosticators in ovine anaplasmosis: discriminant analysis of clinical and haematological parameters in lambs belonging to differently susceptible breeds experimentally infected with Anaplasma ovis

Table 2 Variables and inclusion/removal parameters ^*

Variables in the analysis	Code	Tolerance	F to remove	Min D squared	Between groups
White blood cells	WBC	0.919	36.900	4.411	1 vs. 2
Mean corpuscular haemoglobin content	MCHC	0.633	25.579	3.720	1 vs. 2
Platelets	PLT	0.597	47.374	2.966	1 vs. 2
Mean corpuscular volume	MCV	0.180	17.311	4.105	1 vs. 2
Rectal temperature	T°C	0.856	34.120	3.577	1 vs. 2
Red blood cells	RBC	0.033	62.830	3.756	1 vs. 2
Packed cell volume	PCV	0.054	83.890	3.813	1 vs. 2
Variables *not* in the analysis	Code	Tolerance	F to enter	Min D squared	Between groups
Haemoglobin concentration	Hb%	0.016	3.603	4.931	1 vs. 2
Mean corpuscular haemoglobin	MCH	0.013	0.992	4.961	1 vs. 2
Neutrophils	N	0.879	1.553	4.923	1 vs. 2
Leukocytes	L	0.893	2.175	4.924	1 vs. 2

Legend: Tolerance, proportion of within-groups variance not accounted for by other variables in the analysis. F to remove, F-value of a variable included in the analysis if it was removed from the model; F to enter, F-value of a variable not included in the analysis if it was included in the model; the SPSS default thresholds for the F-to-enter and F-to-remove were adopted. Min D squared, Mahalanobis distance between the two closest groups. In our study, maximizing the minimum Mahalanobis distance was the variable selection rule while the statistical significance of the F-statistics was used as criterion to enter or remove the selected variables.
^*Only data from the last step of the inclusion/exclusion process are presented.

ISSN: 1751-0147