I need to find a reference that discusses appropriate response rates for network research.  I have a bounded network of 38, but only 36 were able/willing to participate.  (94.4%).  How do I discuss this response rate in terms of whether the network is accurately/adequately represented when it's missing two actors?   I have a quote that suggests that a 5% non-respondent rate in network research is acceptable, but no source.  Ideas?

Julie M. Hite
Brigham Young University
Dept. of Educational Leadership & Foundations
Provo, UT  84602
[log in to unmask]