Skip to main content
. Author manuscript; available in PMC: 2015 Aug 1.
Published in final edited form as: Sociol Methodol. 2014 Mar 6;44(1):273–321. doi: 10.1177/0081175013520159

Table 3.

Edge Portion of Models 1 to 4 Ranked by BIC Score

Edge Model
Model 1 SE Model 2 SE Model 3 SE Model 4 SE
BIC 8,166.2395* 7,929.2997* 7,699.5743* 7,689.1466*
Intercept −2.3162* (0.0359) −4.3081* (0.0362)
Mixing regular (R) 1.1091* (0.0430) 1.1389* (0.0434)
Mixing ¬ R ↔ R 0.5560* (0.0737) 0.5595* (0.0742)
Mixing ¬ R — — — —
Inline graphicIndiv 06 −0.6563* (0.0384)
Inline graphicIndiv 17 −0.9519* (0.0397)
Inline graphicIndiv 16 −0.4602* (0.0473)
Inline graphicIndiv 37 −0.4161* (0.0459)
Inline graphicIndiv 39 −0.5880* (0.0638)
Inline graphicIndiv 46 −0.7159* (0.0502)
Inline graphicIndiv 05 0.5852* (0.0485)
Inline graphicIndiv 07 −0.5901* (0.0475)
Inline graphicIndiv 20 −0.8542* (0.0497)
Inline graphicIndiv 40 0.7993* (0.0552)
Inline graphicIndiv 15 −2.8539* (0.0485)
Inline graphicIndiv 19 0.3049* (0.0580)
Inline graphicIndiv 02 −0.8645* (0.0624)
Inline graphicIndiv 26 −0.7884* (0.0580)
Inline graphicIndiv 51 0.1373* (0.0712)
Inline graphicIndiv 54 0.3120* (0.0645)
Inline graphicIndiv 24 1.1369* (0.0580)
Inline graphicIndiv 28 0.2357* (0.0504)
Inline graphicIndiv 33 −0.1863* (0.0623)
Inline graphicIndiv 42 −1.1086* (0.0532)
Inline graphicIndiv 44 1.4785* (0.0594)
Inline graphicIndiv 50 −0.6624* (0.0801)
Inline graphicIndiv 08 0.2206* (0.0569)
log (nt) 0.6394* (0.0118) 0.3884* (0.0120) 4.0946* (0.0121)
Yt−1 0.8946* (0.1019) 0.3120* (0.1042) 0.2808* (0.1052)
log (9-cyclet−1 + 1) 0.0880* (0.0095) 0.1077* (0.0095)
Inline graphicMonday −5.9929* (0.1768) −12.2986* (0.1776)
Inline graphicTuesday −4.1357* (0.1665) −9.5061* (0.1643)
Inline graphicWednesday −3.7315* (0.1225) −10.6229* (0.1248)
Inline graphicThursday −4.0714* (0.1057) −10.6006* (0.1108)
Inline graphicFriday −4.7021* (0.1305) −11.6135* (0.1308)
Inline graphicSaturday −4.2353* (0.0564) −11.4279* (0.0569)
Inline graphicSunday −4.9328* (0.0837) −12.5474* (0.0840)

Note: The lag term is t − 1. The days of the week stand in for the intercept term in models 3 and 4; indiv k indicates the density effect for individual k (as indexed in the data set); log (nt) indicates the contagious propensity effect and is the log of the network size at time t; again, Yt−1 represents the lag effect and log (9-cyclet−1 + 1) is the embeddedness cycle statistic; and Inline graphic indicates that a given variable is represented via a dummy variable.

*

p <.05.