Using a novel genetic algorithm to assess peer influence on willingness to use pre-exposure prophylaxis in networks of Black men who have sex with men

Applied Network Science

Table 20 Mean self-weight, mean weight, proportion, and number by network, adjacency matrix and leadership for self-efficacy

Network	Adjacency	Leader	Self-Weight	Weight	Proportion	Number
1 (\(N=12\))	Build	Leader	0.76 (0.18)	0.05 (0.07)	0.57	8
	Build	Non-leader	0.97 (0.05)	0.07 (0.15)	0.50	4
	Remove	Leader	0.50 (0.23)	0.04 (0.05)	0.58	21
	Remove	Non-leader	0.55 (0.39)	0.05 (0.12)	0.54	43
2 (\(N=11\))	Build	Leader	0.60 (0.53)	0.20 (0.31)	0.90	9
	Build	Non-leader	0.72 (0.32)	0.13 (0.20)	0.88	7
	Remove	Leader	0.35 (0.43)	0.04 (0.06)	0.57	13
	Remove	Non-leader	0.32 (0.29)	0.09 (0.16)	0.70	51
3 (\(N=8\))	Build	Leader	0.96 (NA)	0.05 (0.08)	0.33	1
	Build	Non-leader	0.90 (0.11)	0.02 (0.03)	0.71	5
	Remove	Leader	0.91 (NA)	0.02 (0.01)	0.67	2
	Remove	Non-leader	0.56 (0.29)	0.07 (0.10)	0.78	35
4 (\(N=5\))	Build	Leader	1.00 (NA)	0.12 (0.02)	1.0	2
	Build	Non-leader	0.47 (0.54)	0.22 (0.44)	0.25	1
	Remove	Leader	1.00 (NA)	0.30 (0.42)	0.75	3
	Remove	Non-leader	0.61 (0.41)	0.03 (0.05)	0.36	5
5 (\(N=4\))	Build	Leader	0.80 (0.12)	0.03 (0.06)	0.25	1
	Build	Non-leader	1.00 (0.00)	0.14 (0.21)	0.5	1
	Remove	Leader	0.80 (0.12)	0.02 (0.04)	0.17	1
	Remove	Non-leader	1.00 (0.00)	0.05 (0.12)	0.33	2

Mean self-weight, mean weight from other agents, and proportion and number of possible weights from other agents that exceed 0.005 for self-efficacy with standard deviations where applicable