Certain personal and environmental factors as predictors of thermal sensation perceived by a population of students in a university setting from Timisoara, Romania: a case study

Table 6 Multinomial logistic regression models of predictors of the perceived thermal sensation (3-step scale, winter)

Effect	Model fitting criteria	Likelihood ratio tests
Effect	−2loglikelihood of reduced model	Chi square	df	Sig.(p value)
Intercept	36.933^a	.000	0
Country	79.055	42.121	14	.000
Intercept	38.986^a	.000	0
Gender	51.719	12.733	2	.002
B3-sensitivity to cold	52.895	13.909	6	.031
Intercept	49.015^a	3.509	2	.173
Humidity indoor winter	56.160	10.655	2	.005
NET—normal effective temperature	47.197	1.692	2	.429

^aThe chi square statistic is the difference in −2loglikelihoods between the final model and a reduced model. The reduced model is formed by omitting an effect from the final model. The null hypothesis is that all parameters of that effect are 0

ISSN: 1347-4715