I usually refer to these types of data collection methods as census roster and census nomination. For a brief comparison of the 2 methods see pages 47-48 of Valente, T. W. (2010). Social Networks and Health: Models, Methods, and Applications. Oxford University Press.
From pages 47-48:
“The advantages of the nomination method are (1) it is an unaided recall; (2) may be less demanding for the respondent because he or she does not have to read every name on the roster; (3) the rank order of the nomination can be preserved and used as a proxy for tie strength; (4) the number of alters allowed for each respondent can be varied (i.e., selecting only the first five or four nominations); (5) the data entry and management are easier (i.e., the five ID numbers of those named can be stored easily with the attribute data); and (6) it is more likely to identify individuals outside the imposed boundary of the network. The advantages to the roster method are (1) weak ties are measured as well as strong ones and (2) there is no ambiguity regarding the boundary of the network, everyone on the list is in the network.
The advantages to the roster method are (1) weak ties are measured as well as strong ones and (2) there is no ambiguity regarding the boundary of the network, everyone on the list is in the network. The roster method will usually record more links since respondents scan the list and indicate all of those network contacts, not just the top five or seven that occur to them. The strength or frequency of communication and contact can be measured with a roster, and this information used to reduce the number of links. For example, next to each name the researcher can ask the respondent to circle the number that represents frequency of contact and in the analysis decide to retain only those links in which the frequency is greater than some value. Often the best approach is to use the nomination format but provide a roster for the respondents to refer to. The roster can include names and ID numbers with the respondents instructed to simply write the ID numbers of his or her network nominations.”
From: Social Networks Discussion Forum [mailto:[log in to unmask]]
On Behalf Of amit gal
Sent: Friday, December 30, 2016 5:59 AM
To: [log in to unmask]
Subject: "nomination networks"
***** To join INSNA, visit http://www.insna.org *****
Typically we try to collect data on a full network (e.g. everybody rates everybody else on some measures), but sometimes from practical reasons it is only feasible to ask participants to name (e.g. nominate) a few (typically 3 or 5) friends and rate only them. I call these "nomination networks", and obviously the out degree of every node is constant (up to some small variance due to non-response)
I guess these kind of networks and data collection methods have been studied and discussed before, but I cannot even find the common term for these ("nomination networks" is my own name). Is there any common term for this? is there a literature that discusses this collection methods and its limitations compared to analyzing full network data?
Any references would be helpful.
Happy new year,
_____________________________________________________________________ SOCNET is a service of INSNA, the professional association for social network researchers (http://www.insna.org). To unsubscribe, send an email message to [log in to unmask] containing the line UNSUBSCRIBE SOCNET in the body of the message.