Table 4.

Testing for mediating effect of perceived sufficiency and quality of family communication^a

	Predictor	Outcome	Association
			Crude β [95% CI]	p	Adjusted β [95% CI]b	p
Step 1	SAS-SV score (range: 10–60)	Perceived family communication sufficiency (range: 1–5)	−0.016 [−0.021, −0.011]	<.001	−0.016 [−0.020, −0.011]	<.001
	SAS-SV score	Perceived family communication quality (range: 0–10)	−0.016 [−0.024, −0.008]	<.001	−0.016 [−0.024, −0.008]	<.001
Step 2	SAS-SV score	Family well-being (range: 0–10)	−0.010 [−0.017, −0.003]	.007	−0.011 [−0.018, −0.004]	.002
Step 3	Perceived family communication sufficiency	Family well-being	0.504 [0.450, 0.558]	<.001	0.489 [0.436, 0.542]	<.001
	Perceived family communication quality	Family well-being	0.606 [0.571, 0.641]	<.001	0.591 [0.558, 0.625]	<.001
Step 4	SAS-SV score + perceived family communication sufficiency	Family well-being	−0.002 [−0.009, 0.005]	.527	−0.003 [−0.010, 0.003]	.312
	SAS-SV score + perceived family communication quality	Family well-being	−0.001 [−0.006, 0.005]	.808	−0.001 [−0.007, 0.004]	.589

Note. SAS-SV: Smartphone Addiction Scale – Short Version; CI: confidence interval.

^aWeighted by sex, age, and educational attainment distribution of the Hong Kong general population. ^bAdjusted for sex, age, educational attainment, marital status, employment status, and monthly household income.