A three-wave longitudinal study on the relation between commuting strain and somatic symptoms in university students: exploring the role of learning-family conflicts

Table 4 Results of Bayesian multilevel regression analyses

	Mediator: Learning-family conflict			Outcome: Somatic symptoms
	Est	l-95% CI	u-95% CI	Est	l-95% CI	u-95% CI
Within-level
Intercept	0.02	− 0.21	0.30	0.01	− 0.06	0.07
Commuting strain¹
b	− 0.05	− 0.52	0.34	− 0.03	− 0.17	0.09
simo[1]	0.28	0.01	0.74	0.25	0.01	0.72
simo[2]	0.23	0.01	0.67	0.20	0.01	0.65
simo[3]	0.21	0.01	0.63	0.20	0.01	0.63
simo[4]	0.29	0.01	0.75	0.35	0.01	0.82
Learning-family conflict				0.01	− 0.02	0.04
Between-level
Intercept	2.14	1.16	3.12	1.18	0.95	1.41
Age	0.06	0.02	0.11	− 0.01	− 0.02	− 0.00
Sex²	− 0.53	− 0.88	− 0.18	− 0.12	− 0.20	− 0.04
Commuting distance	0.06	− 0.22	0.33	0.03	− 0.02	0.09
Commuting time	− 0.07	− 0.35	0.22	− 0.05	− 0.11	0.01
Commuting strain
b	0.85	0.21	1.49	0.13	0.02	0.24
simo[1]	0.20	0.01	0.58	0.49	0.07	0.89
simo[2]	0.20	0.01	0.65	0.27	0.01	0.72
simo[3]	0.60	0.10	0.92	0.24	0.01	0.68
Learning-family conflict				0.16	0.12	0.20
R² within³	.00	.00	.02	.01	.00	.03
R² between³	.19	.09	.30	.48	.39	.56

N_Within = 339; N_Between = 128. Est. = Estimate; l-95% CI = lower bound of a 95% credible interval; u-95% CI = upper bound of a 95% credible interval
¹The b coefficient can be interpreted as the average change in the dependent variable when the ordinal predictor commuting strain increases from one category to the next higher category (analogous to the common interpretation of regression coefficients; see Bürkner & Charpentier, 2018). ²Sex coded as 1 = female and 0 = male. ³Bayesian R² as proposed by 55 [55]

ISSN: 2050-7283