st: clarification on asclogit command

[Carena J Vanriper <cvanripe@uvm.edu>]

Hello everyone:

I am new to the STATA listserve so please forgive my former, fairly  
verbose message -- I'm still learning the ropes! I am hoping that  
someone will be able clarify what the differences are between the  
clogit command and the asclogit command. I used the following clogit  
command to analyze a stated choice dataset (6 attributes, each with  
three levels, main effects only):

clogit choice SOC_ontrail_some SOC_ontrail_many SOC_offtrail_some  
SOC_offtrail_many ECO_ontrail_some ECO_ontrail_many ECO_offtrail_some  
ECO_offtrail_many MGMT_ontrail_some MGMT _ontrail_many  
MGMT_offtrail_some MGMT_offtrail_many, group(id)

If anyone has insight into whether asclogit treats data as panel data,  
I would GREATLY appreciate a response!

Most humbly,
-Carena


----- Forwarded message from cvanripe@uvm.edu -----
    Date: Tue, 13 Jan 2009 12:54:58 -0500
    From: Carena J Vanriper <cvanripe@uvm.edu>
Reply-To: Carena J Vanriper <cvanripe@uvm.edu>
 Subject: analyzing paired choice data with asclogit!
      To: statalist@hsphsun2.harvard.edu


Hello fellow stat-A-listers:

I am seeking advice on an analysis of paired choice data measured via
a public survey. A fractional factorial design was used to generate
the paired choice experiment - 6 attributes, each with 3 levels; main
effects only design. The 18 choice set design was blocked into two
survey versions so that each respondent was asked to evaluate 9 paired
choice sets. Survey versions were administered randomly.

The form of the data are as in the matrix below, with:
ID = respondent ID
QS = paired choice set ID
Scenario = 1 for option A, 2 for option B in choice set
Attribute 1 = value (level) of attribute 1 in corresponding option
Attribute 2 = value (level) of attribute 2 in corresponding option
.
.
.
Attribute 6 = value (level) of attribute 6 in corresponding option

Thus, for each respondent, there are 18 rows of data ? one row for
each option of each of the 9 paired choice sets. Because I have
repeated measures (i.e., multiple choice set observations from each
respondent), I estimated a conditional logit (panel data estimator)
using the clogit command. However, I am seeking clarification and
guidance as to whether the clogit or the asclogit is the more
appropriate analysis procedure for my data, given that they include
multiple observations from individual respondents and are formatted
such that each choice set observation is entered as two rows of data
that should be processed together.

ID QS  Scenario  Choice  Attribute 1 Att 2 Att 3 Att 4 Att 5 Att 6
1  1   1         1       1           1     1     1     1     1
1  1   2         1       2           1     3     1     2     1
1  2   1         0       1           1     1     3     2     2
1  2   2         1       2           2     2     2     1     3
1  3   1         0       1           2     1     1     1     1
1  3   2         1       1           1     3     1     2     2
1  4   1         1       3           3     1     2     1     1
1  4   2         0       1           1     2     2     1     2
1  5   1         1       1           2     1     3     3     3
1  5   2         1       3           3     3     1     2     2
1  6   1         0       2           1     1     3     2     1
1  6   2         0       1           3     3     1     1     3
1  7   1         1       1           3     2     2     1     3
1  7   2         1       2           2     1     3     3     2
1  8   1         0       3           2     3     2     3     3
1  8   2         0       1           1     2     1     2     1
1  9   1         1       2           3     1     1     1     3
1  9   2         1       1           2     1     3     2     1


Any suggestions would be greatly appreciated!

Thanks,
-Carena






----- End forwarded message -----



*
*   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/