The proper first place to explore is the graphics
manual [G], although the main trick I shall
recommend involves recourse to [R] separate.
This sounds like wanting to draw a map of response
in terms of latitude and longitude. I am not clear
whether you have data from a area large in comparison
with the world as a whole, in which case you need
to worry about the curvature of the Earth, i.e. choose
a suitable map projection, or small,
in which case you can take the Earth to be (locally)
flat.
As your response is (0,1), you could do something
like
separate latitude, by(response) short
This gives two new variables which you can
then distinguish by different point symbols:
scatter latitude? longitude, ms(oh dh)
where the wildcard -latitude?- will pick up
the new variables created by -separate-.
A limitation of -scatter-, and -twoway- more
generally, is the lack of a handle for enforcing
equal scales on y and x axes (in this case degrees of lat or long
per cm on map). This may or may not
matter to you, especially for exploration of data.
Nick
[email protected]
Aslihan Arslan
> I have plot level data on crop varieties, and latitude and
> longitude of each
> plot and I am trying to plot the variety data (0,1) on latitude and
> longitude axes to see whether there are some niche
> environments dominated by
> one type of crop.
> I have been searching in stata graphics web pages but
> everything I found was
> on plotting one variable against another (twoway)...
>
> I hope somebody can help me to figure out how to plot three
> variables as I
> described above, Thank you very much!
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
