Attribute data for Snowballs

[Kathryn]

Hi Cai - did you ever evaluate whether it would be possible to collect
attribute data for snowball surveys when they nominate a name? The
challenge for me would be consistency, so you would want in some cases
to prepopulate the options (e.g. locations: x, y, z, other).

I'm move more to snowballs then bounded networks just because of the
sheer size of the networks I'm looking at these days.

Let me know. Cheers, ke

[Cai Kjaer]

Kathryn,

We are still thinking about that one.

For other people reading this here is a short description of the issue/
challenge. I think we all run 'snow-ball' surveys every now and then,
ie where we allow respondents to nominate people who are not on the
original list of respondents. We can capture attributes about the
respondents, but we not for those additional people who aren't
surveyed, and we therefore miss out on some really important
information.

I have run projects where we have asked the respondents to provide us
this information, but often get conflicting information back from
multiple respondents, so this isn't a bullet proof method. I have also
tried to use selectively capture attribute data for key nodes
subsequently, but this isn't easy either. The company phone book
rarely includes all the attribute information we are after.

So, if anyone has a simple idea about how we can capture this data I
am all ears.


Rgs

Cai Kjaer
Partner, ONA Surveys.