Skip to main content
. Author manuscript; available in PMC: 2014 Oct 20.
Published in final edited form as: Sociol Methodol. 2012 Aug 1;42(1):155–205. doi: 10.1177/0081175012455628

Table 4.

Connectivity Results for 5 Largest Add Health Networks by Sample Size

Bias a SE b
Net ID Statistic True Value 10% Sample 25% Sample 50% Sample 75% Sample 10% Sample 25% Sample 50% Sample 75% Sample
16 Component 1570 −1.042 −2.245 −1.097 −.896 16.263 9.209 5.734 3.696
17 Component 1707 5.813 .765 1.439 −2.039 21.993 16.013 8.516 5.625
18 Component 1745 −.360 .181 −.937 .060 18.768 10.245 8.319 4.723
19 Component 1894 −1.061 .772 .572 .769 14.633 9.308 6.015 4.002
20 Component 1954 −9.903 −10.846 −.193 −1.608 50.711 26.636 17.323 13.457
16 Bicomponent 1517 .769 −6.641 −4.453 −3.506 24.894 20.319 9.944 6.421
17 Bicomponent 1594 15.735 −3.531 −2.865 −1.842 32.292 27.695 11.580 8.363
18 Bicomponent 1648 8.945 7.374 5.150 3.216 38.663 20.124 11.688 6.835
19 Bicomponent 1838 −3.634 1.543 1.485 2.047 22.986 17.441 11.029 5.881
20 Bicomponent 1664 10.920 16.564 25.134 18.619 76.119 34.949 26.401 16.470
16 Distance 4.253 −.155 −.163 −.194 −.186 .094 .058 .039 .033
17 Distance 5.038 −.034 −.067 −.073 −.066 .154 .098 .076 .073
18 Distance 4.751 −.226 −.264 −.286 −.264 .142 .088 .054 .049
19 Distance 4.302 −.133 −.143 −.147 −.146 .076 .041 .034 .029
20 Distance 5.497 −.175 −.212 −.225 −.204 .201 .103 .068 .062
16 5 Step Reachability .881 .041 .043 .049 .048 .024 .016 .019 .008
17 5 Step Reachability .591 .030 .035 .036 .032 .051 .033 .026 .024
18 5 Step Reachability .720 .090 .104 .110 .104 .050 .027 .018 .015
19 5 Step Reachability .882 .032 .037 .038 .038 .019 .012 .009 .006
20 5 Step Reachability .393 .039 .047 .055 .048 .063 .033 .021 .018

Note: The values for Bias and SE are calculated over 30 independent samples, where each sample yields one estimate of the network measure. All estimates come from the ENC model.

a

Bias=E(estimates)−True Value.

b

The Standard Error is the standard deviation of the sampling distribution.