*****  To join INSNA, visit http://www.sfu.ca/~insna/  *****

> Hi all,
>
> Some time ago I posted this query. Here are the responses I received.
> Thanks to all for your help. Claire
>
> Dear all,
> >
> >I am currently designing my research instrument- a questionnaire to be
> >distributed on-line- for my research. The questionnaire is essentially
> >complete and is also quiet long! However, as I want to ensure that I
> don't
> >forget anything before I send it out, could someone please advise me as
> to
> >the following:
> >
> >I am focusing on social capital and career outcomes (intrinsic e.g. job
> >satisfaction and extrinsic e.g. salary). My sample is Human Resource
> >Development professionals in Ireland. I wish to measure their social
> capital
> >(weak tie and structural hole) by asking the respondent questions such
> as:
> >
> >*       Name (initials) 3 people who have acted to help your career
> >*       For each of these 3 contacts I ask for:
> >*       gender
> >*       age
> >*       education
> >*       relationship to respondent
> >*       closeness to respondent
> >*       frequency of contact
> >*       time known to respondent
> >
> >        1) Will these questions provide me with enough information to
> >measure weak/strong ties and structural holes?
> >        2) Given that I am going to put this questionnaire on-line and
> the
> >package I am using puts the responses to the questionnaires into a
> database,
> >which database do you recommend I use to store the responses to the
> social
> >capital measures until I analyse them? e.g. excel
> >
> >        If there is further information I can give you to make this
> question
> >clearer then please don't hesitate to ask. There seem to be very few
> people
> >in my University here knowledgeable in this area so I am relying strongly
> on
> >this group.
> >
> >        Any and all advic or guidance on this query greatly appreciated.
> >
> >        Regards
> >        Claire
>
> Responses
>
> *     Granovetter's original discussion distinguished between ties in
> which
> the two people know a lot of people in common (strong, e.g., fellow
> employees), and those where they don't (weak, e.g., the postman).  This
> is different from how well you know someone.  Because of the words
> themselves (strong and weak) I think these concepts have been confused.
> Both are important, of course, but it doesn't seem to me that the
> questionnaire you describe will capture the former, but is geared to
> defining the latter.
>
> *     The short answer is not really
> You are collecting ego centred data. For each ego you ask for three
> alters. Why three, there can be no reason for this. I would let them
> free list as many as they want but you may want to limit them by a
> time period or amount of influence. Eg in the last 5 years or have
> had a strong influence.Far more importantly for this to be a network study
> you must ask
> about the connections between the alters (what you call
> respondents) do they know each other. For software you should look at
> egonet, it is listed on the INSNA website.
>
> *     If you are going to study structural holes, you'll need to get some
> information on the contacts between the three people that ego nominates.
> Else, my only advice is to keep the survey as short as possible!
>
> *     regarding your question on strong/weak ties I would guess that the
> information you demand is sufficient. In my thesis I proposed a social
> capital measure based on the idea of role equivalence using triadic role
> positions (see the references in the attached article on Hummell and
> Sodeur,
> 1987 or Burt, 1990, "Detecting Role Equivalence"). I applied the
> instrument
> to the data on the Newcomb student fraternity (part of UCINET III.) by
> making use only of the first three nominations of each student. Symmetric
> ties are then corresponding to strong ties whereas asymmetric relations
> account for weak ties. I attach a paper which contains an upshot of my
> work (forthcoming in the
> next issue of Connections).
>
> *     Why not ask them what each person did for you:
> write letter of recommendation
> pass on job information
> provide mentorship
> work together etc.
>
> *     The best work on measuring social capital I have found is:
>
> Van Der Gaag, Snijders T. A. B.  2003a, May.  ICS: University of Gronigen,
>
> Vrije Universiteit Amsterdam, Unpublished  Report.
>
> Van Der Gaag, Snijders T. A. B.  2003b, November. A comparison of measures
>
> for individual social capital.  ICS: University of Gronigen, Vrije
> Universiteit Amsterdam, Unpublished  Report.
>
> *     I used the same measures of closeness as yours in my investigation
> of
> information behaviour of Mongolians.  In addition I used residential
> proximity, but I don't think that would apply in your case.  I also
> followed Nan Lin's methodology for determining social capital.  If you
> haven't heard of it try these references:
>
> Lin, N. & Dumin, M. (1986).  Access to occupations through social ties.
> Social Networks 8, 365-385.
>
> Lin, N., Cook, K. & Burt, R. S. (Eds.) (2001).  Social capital: Theory
> and Research.  New York: Aldine de Gruyter.  (See especially his chapter
> on the position generator, pp. 57-81).
>
> Lin, N. (2001).  Social Capital: A theory of social structure and
> action.  Cambridge, UK: Cambridge University Press.
>
> He might also suggest you get the occupation of the career helper in
> order to determine status.
>
> *     some questions and considerations about the research design:
>
> 1.  Does your sample include the all (members of) Human Resource
> Development
> professionals in Ireland or just a subset working in a particular
> organization?
> If there is a membership roster, you can format it so members can rate
> their
> relationships with each other.
>
> 2.  Is there any substantive reason for focusing on 3 alters?. Depending
> on the
> number of nodes, 3 alters may not provide enough variability to capture
> weak
> ties/structural holes.  If there is no justification for limiting the
> alters,
> you could use your knowledge of the organization to chose a maximum number
> of
> alters. Of course, you may want to consider the relative advantages and
> disadvantages of using closed or open alter lists - for details see
> Wasserman &
> Faust, Chapter 2)
>
> 3. To reduce response burden, you would want to balance the above two
> considerations (sample, and number of alters to rate) with the length of
> the
> questionnaire. Usually, response burden is higher if there are a lot of
> alters
> to rate on many relations.
>
> 4. Re databases for storage, Excel could do, but a lot depends on the
> native
> database used in storing the online responses, and how various fields are
> defined. What would you be using - mysql, dbase, oracle or access, etc,
> and then
> you may need to write a few sql-statements to re-arrange the data in a
> format
> suitable for export to Excel, Spss or UciNet.
>
>
> Claire Gubbins
> Department of Personnel & Employment Relations
> Kemmy Business School
> University of Limerick
> Limerick
> Ireland
> Phone: 00353-61-202666
> E-mail: [log in to unmask]
>

_____________________________________________________________________
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.sfu.ca/~insna/). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.