Print

Print


*****  To join INSNA, visit http://www.insna.org  *****

YL:

> Dear all,
>     I am having a hard time understanding what really is a qualitative
> social network or qualitative social network analysis. Can you show me
> an example illustrating the differences between quantitative and
> qualitative networks?
>    Much appreciated!

Ryan:

I agree wit Jeffrey Broadbent's note preceding this one.  Something that follows
may offer some other insight.

Wasserstein & Faust (p. 29)assert that there are two types of variables that can
be in network data, structural and composition.  Compositions are measurements
of vertice (they say "actor") attributes.  Structural variables measure ties
between pairs of actors.

You might decide whether SNA is as "strongly typed" as this assertion suggests
from your own perspective or you might take their assertion as an appropriate
defining standard.
If you largely accept the W&F characterization of SNA variables, then
"qualitative" would probably be limited to a form of measurement used in SNA
that is not quantitative, but that is used in very similar ways.  By strongly
typed, I mean that it follows strict category conventions.

As you probably know, qualitative is often taken to be that which is not
quantitative or statistical (e.g. in Strauss and Corbin 1998).  That is
probably a reasonable convention to apply here as well.

The exact semantics are open to contention.  There are various levels of
acceptance for conventions in communities of various sorts, both formal and
informal.  There are now communities of primarily qualitative researchers who
are part of association meetings, etc.  So the legitimacy for inclusion in
"SNA" is perhaps greater than it might have once been.

For example, you might envision any number of networks that would measure
qualitatively, say, a politician's dislike of other politicians on a scale that
is descriptive (hate, dislike, love, adore) rather than digital scale of measure
(scale of 1-10). That would be the most straightforward sort of example of a
difference in measurement.  But such an example is relatively trivial of
course.

As a matter of epistemology rather than SNA convention, some might argue that
dichotomies like Quant and Qual create issues related to how categories are
defined and who gets to define them.  Such arguments can become emotionally
charged and power can be applied to edit opinions or to coerce discipline or to
force a discipline open.
Consequently, you would get an argument from some that such terms (Quant and
Qual) potentially obscure more than
they illuminate because they are perspective-based or, to some, "political."
Others would accept them as necessary conventions for practical action and
exercise of tools.  Still others might seem them as essential defining truths
established as universal scientific categories rather than practical
conventions. This last group would probably be relatively rare in today's
academy in most fields including SNA.

Again, you must decide how valuable conventions and norms are to you or whether
you are coerced by a discipline (or openly choose) to be a disciple--one who
follows the standards set of a discipline.  If you are to be a disciple (an
admittedly contentious word), or if standards and norms are quite valuable to
your research for other reasons--e.g. membership in certain research
communities of practice, approval of committees, etc., you will want to
adopt general standards for your research like Wasserstein and Faust unless less
conventional approaches are
specifically allowed.

Because quantitative measures allow set mathematics to be applied to graphs,
they can yield results that may be easier to interpret or to summarize than
more qualitative studies.  This may aid in the usability of results (e.g.
measures of centeredness) or in the discussion of such results.

Qualitative networks may offer less succinct summaries and, thus, perhaps less
actionable or interpretible results.  On the plus side, their descriptions,
while inherently descriptive, can offer a rich portrait of a complex topic that
might enlighten certain complex or heuristic judgments about the same topic or a
similar one referenced comparatively at some point.

A book like Aihwa Ong's Flexible Citizenship is, to my mind, an example of a
qualitative social network work that is at the fringe of what might be accepted
as "SNA" by some within the SNA interest community. But that work is clearly not
SNA in any form that would be conventional to the history of the topic, nor to
most of its experts, nor based on SNA's defining texts (e.g. W&F).  A classic
hybrid work with perhaps more acceptance might be Claude Fischer's early work
on communities around San Francisco.

Whether the field is open to a broader definition of what might be reasonably
discussed as "SNA" is of course a question for whatever may constitute the
field.  There are advantages to tight definitional limits, and advantages to
broader inclusion.

For now, there is a set of standard research conventions that emphasizes
primarily quantitative networks and graph theory, and there are those
researchers with primarily qualitative interests who may stand with less
legitimacy given their tenure in the research area and the less summary nature
of their results.  However, they are

Hope something here helps.

Ryan Lanham

_____________________________________________________________________
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.