[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: SPMAP: interpreting class breaks / adding state outlines
Your main question is answered in the help:
clbreaks(numlist) is required when option clmethod(custom) is specified. It defines the custom class breaks to be
used for classifying variable attribute. numlist should be specified so that the first element is the minimum
value of variable attribute to be considered; the second to kth elements are the class breaks; and the last
element is the maximum value of variable attribute to be considered. For example, suppose we want to group
the values of variable attribute into the following four classes: [10,15], (15,20], (20,25] and (25,50]; for
this we must specify clbreaks(10 15 20 25 50).
Look carefully and you will see a standard notation for closed and
half-open intervals that is evidently new to you. See for
example http://mathworld.wolfram.com/ClosedInterval.html and similar.
When [ is used it denotes that values are >=
the lower limit. When ( it denotes that values are > the lower
limit. Similarly ] and ) imply <= and < the upper limit.
Thus you want [0,0], (0,1], (1,2], (2,3], (3, larger than three]
which is what your breaks will achieve.
Julie Susan Darnell
> With enormous thanks to Scott Merryman's suggested syntax, I
> have successfully created a map of core-based statistical
> areas using spmap.
> First, I wish to confirm that I understand how the data are
> classified when using the clmethod(custom) option.
> # delimit;
> spmap numberoffreeclinics if conterminous using "uscoordinates2",
> id(id) ndfc(white) fc(Blues) ocolor(gs12 ..) clmethod(custom)
> clbreaks (0 1 2 3 4 32)
> This produces a legend with:
> Should the first group be interpreted to include or exclude 1?
> My goal is to produce five groups, with the first group
> containing only 0s, the second group containing only 1s, etc.
> In other words, I want to distinguish CBSAs with 0, 1, 2, 3,
> or 4+ [free clinics].
* For searches and help try: