. 2020 Sep 18;10(9):1687. doi: 10.3390/ani10091687

Table 1.

Model selection.

Model ¹	Variables ²	Likelihood Test ³	Wald Test ⁴	Coefficients β_n ⁵	Exp ⁶	Confidence Intervals ⁷
I	Tumor size		8.59; <0.0001	0.24	1.27	1.21–1.35
II	Tumor size	35.48; <0.0001	7.94; <0.0001	0.22	1.25	1.19–1.32
II	Age	35.48; <0.0001	5.88; <0.0001	0.11	1.12	1.08–1.16
III	Tumor size	0.16; 0.69	7.93; <0.0001	0.22	1.25	1.19–1.32
	Age		5.82; <0.0001	0.11	1.12	1.08–1.16
	Spayed		0.40; 0.69	0.05	1.05	0.82–1.36
IV	Tumor size	0.88; 0.35	7.96; <0.0001	0.22	1.25	1.19–1.32
	Age		5.68; <0.0001	0.11	1.11	1.07–1.16
	Spayed		0.35; 0.73	0.05	1.05	0.81–1.35
	Pure breed		−0.94; 0.35	−0.09	0.91	0.75–1.11

¹ Models are built so that the smaller models are special cases of the larger ones. Equivalently, the smaller models are obtained by sequentially setting to 0 the coefficients of the full model (IV). The general form is: log-odds = β₀ + β_1×1+ β₂X₂ + ... + β_nX_n: where β₁, β₂, ..., β_n are the coefficients of the x₁, x₂, ..., x_n independent variables (covariates) included in the model. odds is calculated according to the formula: odds = exp(β₀ + β₁X₁+ β₂X₂ + ... + β_nX_n); ² Covariates included in the model. Intercept not reported; ³ Likelihood ratio test statistic: Deviance, p value; ⁴ Wald test statistic: z and p value; ⁵ Model parameters β_n; ⁶ Exponentiated model parameters e βn; ⁷ Wald 95% confidence interval for an exponentiated model parameter.