The Effect of Physical Activity on Anxiety: The Mediating Role of Subjective Well-Being and the Moderating Role of Gender

Figures & data

Table 1 Descriptive Data for Main Variables

Variables	Total Sample (N = 1153)
Age, M (SD)	20.03 (0.98)
Gender, n (%)
Male	834 (72.3%)
Female	319 (27.7%)
PA (MET-min/week)	3214.16 (2121.95)
PA level, n (%) (M/F)
Low-intensity	34 (4.1%) / 13 (4.1%)
Medium-intensity	102 (12.2%) / 55 (17.2%)
High-intensity	698 (83.7%) / 251 (78.7%)
Subjective well-being, M (SD)	79.08 (14.28)
Anxiety, M(SD)	6.56 (5.13)

Table 2 Demographic Difference Analysis

Variables	Groups	Physical Activity	Subjective Well-Being	Anxiety
Nationality	Han	3193.12±2062.83	79.21±14.38	6.50±5.10
	Other	3404.04±2598.48	77.89±13.35	7.05± 5.46
	t	−0.842	0.943	−1.094
	p	0.402	0.346	0.274
Registered residence	Urban	3197.53±2068.74	79.45±14.62	6.57±5.21
	Rural	3293.88±2364.53	77.29±12.43	6.49±4.78
	t	−0.534	2.166	0.200
	p	0.594	0.031*	0.842
Age	18 years	3547.33±2864.72	77.64±14.37	7.33±5.12
	19 years	3170.22±1919.96	78.72±13.94	6.74±5.04
	20 years	3254.28±2068.81	79.69±14.25	6.23±5.11
	21 years and older	3149.56±2271.51	78.68±14.67	6.76±5.24
	F	0.522	0.574	1.233
	p	0.667	0.632	0.296

Note: *P<0.05.

Table 3 Correlation Analysis Between Variables

Variables	Physical Activity	Subjective Well Being	Anxiety
Physical activity	1
Subjective well being	0.130**	1
Anxiety	−0.080**	−0.718**	1

Note: **P<0.01.

Table 4 Mediating Model Test of Subjective Well-Being

Regression Equation (N=1153)		Fitting Indicator			Coefficient
Outcome Variable	Predictive Variable	R	R^2	F	B	t	p
Anxiety		0.0839	0.0070	4.0714
	RR				−0.0041	−0.1407	0.8881
	PA				−0.0837	−2.8472	0.0045**
SWB		0.1362	0.0185	10.8659
	RR				−0.0595	−2.0354	0.0420*
	PA				0.1235	4.2282	0.0000**
Anxiety		0.7120	0.5070	393.8270
	RR				−0.0466	−2.2443	0.0250*
	PA				0.0045	0.2153	0.8296
	SWB				−0.7137	−34.1335	0.0000**

Note: *P<0.05, **P<0.01.

Abbreviations: RR, Registered residence; PA, Physical activity; SWB, Subjective well-being.

Figure 1 Meditation effect model diagram.

Note: **P<0.01.

Table 5 Fit Index Metrics for Multi-Group Analysis of Gender Differences

Model	χ^2	df	CFI	TLI	RMSEA
M1	0	0	1	1	0
M2	8.592	3	0.993	0.987	0.04
M3	9.455	4	0.993	0.99	0.034

Notes: M1: Unconstrained model, M2: Structural weights model, M3: Structural covariances model.

Figure 2 Multi-group structural equation model plot of gender difference.

Notes: **P<0.01; (A): male, (B): female.