Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Re: Simultaneously accounting for clustering at two different levels with vce(cluster) option


From   Lily Yor <lilyyor1@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: Simultaneously accounting for clustering at two different levels with vce(cluster) option
Date   Mon, 29 Mar 2010 01:03:16 -0700

Thank you so much, Misha and Padmakumar, for your suggestions to look
into Cameron et al's method of dealing with multi-way clustering and
Caskey's implementation of Cameron et al's method into Stata
programming.  But as Mark has pointed out (thank you so much for your
note on this, Mark), it will not work for my problem because my data
is (1) nested (classrooms nested in schools), and (2) number of
schools in my sample is too small (n=8).  After much research and
consideration, I have decided that the best way to deal with my
problem is to adjust the standard errors using the vce option to
account for clusteing at the classroom level (I have >1000 classrooms)
and just include 7 school dummies in the same models.  If something
else strikes you as a better solution, I would love to hear from you.
Thanks so much again, Lily

On Tue, Mar 9, 2010 at 12:40 PM, Lily Yor <lilyyor1@gmail.com> wrote:
>
> Hi, I have data that is clustered at two different levels -- i.e., I have data at classroom level and at school level.  I need to account for these two sets of intraclass clustering, but the clusters at each level are too few to run a formal multi-level model (for example, there are only 8 schools in my data).  Thus, I would like to run a logistic regression model (my dependent variable is binary -- whether a student has passed a certain test or not) and adjust the standard errors by using vce(cluster) option, but there doesn't seem to be a way to simultaneously, in a single model, account for clustering at both levels (i.e., classroom level and school level).
>
> The way I have approached this problem thusfar is to run the logistic regression model with vce(cluster) option applied to account for classroom clustering, and then include school dummies in the same model, but this does not seem to be a satisfactory solution:
>
> logit [dep var] [ind_var1] [ind_var2] [ind_var3] [school2] [school3] [school4] [school5] [school6] [school7] [school8], vce(cluster classroom_id)
>
> If you have any tips and suggestions on how I could use the vce(cluster) option to account for both classroom and school clustering, I would very much appreciate your help.
>
> Thank you so much.

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index