Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Austin Nichols <austinnichols@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Interpretation of Two-sample t test with equal variances? |
Date | Wed, 20 Mar 2013 15:28:53 -0400 |
Gwinyai Masukume <parturitions@gmail.com>: My own impression is that the first thing you should do is increase the sample size a thousandfold. Such data are easy to come by in many countries, and I would be surprised if you did not find age an important predictor, though of course you should condition on other maternal characteristics and characteristics of the place as well. See e.g. Figure 1 and Table 5 in http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1694300/pdf/amjph00539-0107.pdf A low pseudo R^2 is not a concern, however. On Wed, Mar 20, 2013 at 2:54 PM, Gwinyai Masukume <parturitions@gmail.com> wrote: > Nick - well that's an interesting question. The variables that > determine the decision as you would imagine are many. > Regards, > Gwinyai. > > On 3/20/13, Nick Cox <njcoxstata@gmail.com> wrote: >> What drives the decision on delivery mode any way? How far is it >> clinician's choice, patient's choice? Do you have data on the patient >> or clinician variables that influence or determine tha decision? If >> you don't have all the predictors -- and it would be surprising if you >> did -- there will be lots of unexplained variability. >> >> Nick >> >> On Wed, Mar 20, 2013 at 6:17 PM, Gwinyai Masukume >> <parturitions@gmail.com> wrote: >>> Hi again. Thanks, I'm learning a lot. Carlo - I'm developing a model >>> to simulate nulliparous (first time) mothers. As you note, the >>> C/section rate is about 30 percent. My model so far has few variables. >>> It's very very hard simulating reality I'm discovering. >>> With respect, >>> Gwinyai * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/