From: brian [brian@stat.cmu.edu]
Sent: Friday, May 04, 2007 6:49 PM
To: 'Yi Jiang'
Cc: 'Zhanwu Liu'
Subject: RE: Alumni Survey final report and presentation.

Thanks

Here are a few comments on your talk.  Sorry not sooner!


Overall, a great job on the project and the talk; generated lots of useful
discussion.  

* your "outline of talk" slide listed results of statistical methods rather
than research questions of interest to Christian and Judy.  The latter makes
for better communication.

* objectives and data slides great

* EDA slides good too. would benefit from a question at the top of each
slide, the answer to which is the eda display .  E.g. "Where in the world
are they located?"  "What program were they in at CMU"?  Etc. etc.

* logistic regression stuff

   - Judy would have been helped to see the definition of "active alum"
before the
     technical details of the model

   - good display of logistic regr results

   - would be interesting to add interactions with year to the logistic regr
e.g. on slide 21

   - would be useful to see a 3-way contingency table of the 3 things you
used to define "active", to see how they are related, and to see what
dominates...

* general info on panel surveys was great.  example of crime survey was
great

* would be useful to add a couple of std references on panel surveys to your
final report (I hvaen't read report yet)

* I like the idea of reviewing other specific panel surveys to see "what
works"

* add phone as a possible survey mode...

all in all, very good job though.  have a great weekend

-BJ


-----

Yi Jiang <mailto:yijiang@stat.cmu.edu> wrote:

> Dear Christian and Judy,
> 
> Thank you so much for the opportunity to work with you on this
> very interesting project for our Statistical Practice class.  Here we
> attached the final report and the presentation ppt file for your
> review. 
> 
> As agreed in the non-disclosure agreement, we will delete all the
> alumni data used in this project on our hard drives and cd's.  As for
> the reports, presentation, and codes, please feel free to let us know
> if you think any of them should also be deleted.
> 
> Please do not hesitate to let us know if you have any questions about
> these files.
> 
> Thanks a lot again,
> 
> Zhanwu & Yi
> 
>   Yi (Jake) Jiang             INTERNET: yijiang@stat.cmu.edu
>   (412) 268 2724
>   Baker Hall 132M
>   Department of Statistics
>   Carnegie Mellon University, 15213