. 2020 Sep 17;8:466. doi: 10.3389/fpubh.2020.00466

Table 3.

Results of the logistic regression analysis predicting preferences for the profiles of PE classes among students (N = 1,340).

Predictors of profiles of PE classes	Univariate model							Multivariate model
	B	*SE B*	OR	95% CI		Wald's	p	B	*SE B*	OR	95% CI		Wald's	p
				LL	UL	χ²					LL	UL	χ²
1. Fun
Gender (male)	−0.52	0.11	0.59	0.47	0.74	20.72	0.00001	−0.53	0.11	0.59	0.47	0.74	21.06	0.00000
School level (high)	−0.38	0.11	0.69	0.55	0.86	11.03	0.001	−0.39	0.11	0.68	0.54	0.85	11.39	0.001
2. Exercise
Gender (male)	0.45	0.12	1.56	1.24	1.98	14.19	0.0002	0.45	0.12	1.57	1.24	1.98	14.34	0.0002
School level (high)	0.27	0.12	1.31	1.04	1.66	5.31	0.02	0.28	0.12	1.32	1.05	1.67	5.45	0.02
3. Control
Gender (male)	0.86	0.37	2.36	1.15	4.83	5.49	0.02	0.87	0.37	2.39	1.16	4.90	5.61	0.02
School level (high)	1.12	0.37	3.06	1.42	6.55	8.25	0.004	1.13	0.39	3.08	1.44	6.62	8.36	0.004
4. Learning
Gender (male)	0.12	0.32	1.13	0.60	2.11	0.14	0.71	0.12	0.32	1.13	0.60	2.11	0.14	0.71
School level (high)	0.09	0.32	1.10	0.58	2.08	0.08	0.77	0.09	0.32	1.10	0.58	2.08	0.09	0.77

LL, lower limit; UL, upper limit.