Abstract
Abstract There is very little network data missing on respondents to the 1985 General Social Survey. Missing data on relations between discussion patterns pose the greatest problem but the unknown relations are strongly associated with relations known to be weak. The association between missing and weak relations remains strong after controls for the number, strength, and nature of a respondent's discussion relations. Moreover, the association remains strong across different kinds of respondents despite significant tendencies for certain kinds of respondents to have provided incomplete network data. The implication is that the missing network data can be replaced with quantitative data indicating a weak relation.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
