Skip to main content
. Author manuscript; available in PMC: 2014 Oct 1.
Published in final edited form as: Soc Networks. 2013 Oct;35(4):10.1016/j.socnet.2013.09.003. doi: 10.1016/j.socnet.2013.09.003

Table 2.

Maximum percent missing to retain target correlation with true score

Network Target Correlation In-Degree Out-Degree Total Degree Bonacich Power Closeness Betweenness
Interlock 0.90 -- -- 23 20 8 12
0.80 -- -- 40 36 19 25
Prison 0.90 27 17 19 15 10 21
0.80 45 31 35 29 21 24
Sorority 0.90 20 23 18 13 16 5
0.80 37 40 34 26 31 13
6th Graders 0.90 43 38 38 29 46 22
0.80 64 58 58 48 60 38
Coauthor 0.90 -- -- 58 54 3 24
0.80 -- -- * * 12 47
Prosper 0.90 31 26 25 22 18 10
0.80 51 45 43 39 33 21
Co-Citation 0.90 -- -- * * 12 33
0.80 -- -- * * 36 60
Elites 0.90 * 23 * 45 15 25
0.80 * 40 * 67 30 45
HS 13 0.90 43 25 34 30 21 17
0.80 64 44 54 49 36 30
BioTech 0.90 -- -- 58 47 0 44
0.80 -- -- * 70 8 62
HS 24 0.90 39 27 32 30 25 17
0.80 60 46 52 48 39 31
CSprings 0.90 -- -- * 47 0 38
0.80 -- -- * * 7 62
Mean (Std Dev) 0.90 39 (16.1) 25.6 (6.4) 42.9 (20.9) 35.2 (17.4) 14.5 (12.7) 22.3 (11.5)
0.80 55.9 (11.9) 43.4 (8.2) 55.5 (14.6) 51.8 (17.0) 27.7 (15.2) 38.2 (16.9)
*

Cases where percent missing is above 70, our observed maximum. In these cases, 70 is used to calculate overall means. The maximum percent missing was calculated based on a quadratic fit to the data.