We pretty routinely ask about the "strength" of relationships; the duration; the "kind"; and how often interaction occurs.

Sam Friedman
National Development and Research Institutes
71 West 23d Street
New York, NY 10010
1 212 845 4467
Fax 1 917 438 0894
[log in to unmask]
>>> Scott Engle <[log in to unmask]> 02/23/02 10:46 AM >>>

I'm helping a colleague with network aspects of her research.  She would like to collect valued data on the strength of several types of dyadic relationships, but a roster type of instrument won't work because of the size of the organization (or at least the reasonable boundaries of the area where she is collecting data.)   We've thought of using a name-generating instrument and asking the respondent to indicate the strength of each named relationship.  In other words, we would ask a pretty standard name-generating question and then below the question provide blanks for the names.  Beside each blank there would be a Likert scale for the respondent to indicate the strength of the relationship.  I'm sure that some of you have tried something similar, but I don't recall seeing it, so I thought I'd take the idea to the list.

Obviously, the wording of the name-generating question would presuppose much about the nature of the relationship named by the respondent.  But . . . would it presuppose so much that there is no value in trying to tease out more information about the strength of the relationship?  In any event, I'd appreciate any thoughts you might have on this.


Scott L. Engle, Ph.D.
Engle Associates

5201 Southern Hills Dr.
Frisco, TX  75034
ph: 469-384-9440
fax: 360-242-2794
e-mail:  [log in to unmask]