st: Mean within- and between-cluster distances after sqom and sqclusterdat

[Jacob Day <jcday@ncsu.edu>]

Dear Statalist,

I am currently analyzing managerial career sequences with 4 potential
sequence states using the sqom (optimal matching) command and Ward's
method of clustering. I would like to compare cluster solutions like
Aisenbrey and Fasang (2010) did in their recent article in
Sociological Methods & Research. They compare mean within-cluster
distances and mean-between cluster distances for each cluster solution
and develop a rule of thumb for making a decision (p.432-433). I
cannot seem to figure out 1) how to get Stata to report the
mean-within and mean-between cluster distances or 2) the necessary
information I need and how to calculate these measures myself. Thank
you in advance for any guidance you can give me.

Sincerely,
Jake Day

The data look like this:

id  hpos_centseq3 sqorder
33	3	1
33	4	2
33	2	3
33	2	4
33	2	5
34	2	1
34	2	2
34	2	3
34	4	4
34	1	5

 The syntax looks like this:

sqset hpos_centseq3 id sqorder

sqom, subcost(hp3cost) indelcost(1.067) full standard(longer)
sqclusterdat
clustermat wards SQdist, name(wards) add
cluster generate wg=groups(2/15), name(wards) /*generates indicator
variables for 2-15 cluster solutions*/
sqclusterdat, return


Aisenbrey, Silke and Annete E. Fasang. 2010. "New Life for Old Ideas:
The 'Second Wave' of Sequence Analysis Bringing the 'Course' Back Into
the Life Course." Sociological Methods & Research 38:420-462.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/