Re: st: Sequence index plots, subpopuations

[Jörg Eulenberger <j.eulenberger@web.de>]

Dear Sonja,
try this:

sqindexplot if westgerman ==1 , saving("c:\...\Sequencindexplot1", replace)
sqindexplot if westgerman ==0  , saving("c:\...\Sequencindexplot2", replace)
grc1leg "c:\...\Sequencindexplot1" "c:\...\Sequencindexplot2" 
legendform("c:\...\Sequencindexplot2") saving("c:\...\Sequencindexplot3")

Hope this helps,
Jörg



Am 24.01.2011 11:14, schrieb Sonja Bastin:
> Hi everyone,
>
> as I have seen in other researcher's work there must be a possibilty to get
> well comparable sequence index plots for subpopulations containing differing
> numbers of observations.
> For example, I want to compare east and west Germans. My sample contains 956
> West Germans, whereas only 319 East Germans which means in the left graph
> there are 956 sequences displayed, in the right one 319. And as the vertical
> axis however is the same for both graphs the images are difficult to
> compare. I was thinking about drawing a random sample from all west germans,
> containing as many cases as there are east Germans, but with a data set in
> sequence format this is not that easy. Of course, I could draw seperate
> graphs for both subpopulations and display them in the same size, but this
> procedure seems too complicated and not very "fancy". There must be a more
> simple solution as graphs of other authors suggest.
>
> Thanks so much for your help
> Best
> Sonja, PhD Student
> *
> *   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/

*
*   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/