Open AccessAbsent ties in social networks, their treatments, and blockmodeling outcomes
Author(s) -
Anja Žnidaršič,
Patrick Doreian,
Anuška Ferligoj
Publication year - 2012
Publication title -
metodološki zvezki
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.127
H-Index - 7
eISSN - 1854-0031
pISSN - 1854-0023
DOI - 10.51936/redp2838
Subject(s) - interpersonal ties , reciprocity (cultural anthropology) , strong ties , equivalence (formal languages) , computer science , imputation (statistics) , social network (sociolinguistics) , network structure , social network analysis , set (abstract data type) , data mining , theoretical computer science , mathematics , psychology , social psychology , combinatorics , machine learning , discrete mathematics , missing data , world wide web , social media , programming language
An absent tie is one for which we have no information regarding its nature. Absent ties for a network is a set of such ties. This lack of information can be present anywhere in network data and has the potential to compromise the results of all network analytic tools. To assess this impact, we used real networks and based simulations on them by introducing varying amounts of absent ties. They were treated with four treatments of absent ties. Blockmodeling, using structural equivalence, was applied to the known networks and then to every treated network. The results were compared. The amount of absent ties, their treatments, the block structure of a network, and the level of reciprocity all have an effect of the adequacy of the results of blockmodeling. Reconstruction combined with imputation based on modal values was the best overall treatment. However, treatments of absent ties can work for some networks but not others and we recommend treatments of absent ties based on the form of networks.