Table 4.

Relationship of individual network variables to MSRS-R

Variable	β	Standard error	Adjusted P-valuea	FDRb
Network structure
Size	−0.025	0.013	0.052	0.197
Density	0.007	0.365	0.985	0.985
Constraint	0.004	0.007	0.537	0.729
Effective size	−0.035	0.05	0.487	0.712
Maximum degree	–0.041	0.044	0.347	0.564
Mean degree	0.003	0.052	0.958	0.985
Network Composition–General
Percent kin	0.001	0.004	0.769	0.876
Percent who are supportive	−0.005	0.004	0.198	0.47
Standard deviation of age	−0.006	0.017	0.701	0.876
Diversity of sex	−0.332	0.359	0.356	0.564
Diversity of race	0.686	0.423	0.105	0.333
Percent contacted weekly or less often	−0.009	0.004	0.023	0.147
Percent who have been known for less than 6 years	0.001	0.004	0.784	0.876
Percent who live more than 15 miles away	−0.003	0.004	0.346	0.564
Network Composition–Health
Percent who smoke	0.006	0.003	0.045	0.197
Percent who do not exercise	−0.003	0.003	0.296	0.564
Percent who do not take medications regularly	0.018	0.012	0.123	0.334
Percent who do not go to doctor’s appointments	0.045	0.015	0.002	0.023
Percent who have a negative influence on health	0.017	0.005	0.001	0.016

^aAdjusted for potential confounders, sex, age, marital status, and years of education via linear regression as described in the Methods

^bFDR is false discovery rate, controlling for multiple testing