Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

Re: st: spmap myseriously dropping a category

From   Scott Merryman <>
Subject   Re: st: spmap myseriously dropping a category
Date   Fri, 26 Feb 2010 11:57:03 -0600

I suspect it is due to the distribution of the var categories within
year 1995.  The default method is quantile so I would guess the var
was only broken into two groups.  Is you only have 3 categories
wouldn't it be easier to use -clm(u)-?

Using one of the example data sets from -spmap- compare:

 use "Italy-RegionsData.dta", clear
 keep if inrange(id, 8,13)
 spmap zone using "Italy-RegionsCoordinates.dta"  , id(id)
 spmap zone using "Italy-RegionsCoordinates.dta"  , id(id) clm(u)
 replace zone = 1 in 2
 replace zone = 3 in 5
 spmap zone using "Italy-RegionsCoordinates.dta"  , id(id)


On Fri, Feb 26, 2010 at 11:08 AM, Dorothy Bridges <> wrote:
> Dear Statalist,
> Allow me to preface this inquiry with an apology: I'm sorry that I am
> not able to replicate my problem with publicly available datasets, and
> that therefore this question may well be annoying and unanswerable.  I
> simply don't know where else to turn.
> I am using spmap to graph a ternary variable onto a map of the world.
> My variable is time-series, and I would like to create one map for
> each year.  When I create the map for 2005, or for 2000, or for 1990,
> it appears just as I imagined it.  When I create the map for 1995,
> though, Stata treats the first two categories as a single category.  I
> cannot fathom why this is the case (yes, I'm sure that all three
> levels of my variable occur in 1995).  The only difference in the
> command line is an "if" clause.  In other words:
> spmap var using file if year==2005 works beautifully, displaying three
> colors corresponding to the three levels of var, but
> spmap var using file if year==1995 displays only two colors, the first
> corresponding to levels one and two, and the second corresponding to
> level three.
> There is nothing fancy about the variable -- it takes the values 1, 2, and 3.
> Thanks so much in advance for your thoughts.
> Dorothy
> *
> *   For searches and help try:
> *
> *
> *

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index