Problem with spmap - Border Lines Disappearing when Variable to be Plotted is 0

[Maxence Morlet]

Hi all,

My data are at the end.

When I run the following code, it produces a map where there are no visible border lines between contiguous states with a value of X of 0. Please could someone help me produce a map with all state borders visible?

Also, is there a colour palette for which states with a value of X of 0 could be just white, instead of very lightly red shaded?

Thanks in advance!

shp2dta using "$data2/cb_2015_us_state_500k", data("us_data")  coor("us_coordinates") genid(id) gencentroids(c) replace
use us_data, clear
quietly destring STATEFP, generate(fips)
drop if fips > 56
save usstates,replace

//Determine IDs for Hawaii and Alaska
use usstates,clear
l fips id  if fips == 2 | fips == 15

// Alaska is reduced in size by 60% and then both Alaska and Hawaii are re-positioned under the southwestern states.
//Drop northwest islands of Hawaii and western part of Aleutian Islands 
use us_coordinates,clear

gen order = _n
drop if _X < -165 & _X != . &  _ID == 50
replace _X = _X  + 55  if  _X != .  &  _ID == 50
replace _Y = _Y  + 4  if _Y != .  &  _ID == 50

replace _X = _X*.4  -55 if  _X !=  . &  _ID == 15
replace _Y = _Y*.4  + 1 if _Y != .  & _ID == 15
drop if _X > -10 & _X != . & _ID == 15
sort order 
sort _ID
drop order

save us_coordinates2, replace

use us_coordinates2, clear
bys _X _Y: keep if _N==_n 
save "$data2/us_coordinates2", replace

use usstates,clear
spmap using us_coordinates2  , id(id)

rename STUSPS state
merge 1:m state using "$data2/file_with_data"
rename id _ID

keep if _merge==3
drop _merge

bys state: keep if _N==_n
spmap X using us_coordinates2 , id(_ID) clb(0 0.01 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 2 3 4 5 10 15 20 25 30 35 40 60 80 100) clm(custom)  legorder(lohi) fcolor(Reds2) ocolor(none) clnumber(5)

input byte _ID float(x_c y_c) str10 state float X
15 -152.22098 64.217476 "AK"  3.012048
25  -86.82837 32.789825 "AL"  19.62733
 7  -92.43925  34.89974 "AR"         0
22  -111.6644 34.293102 "AZ" .10857763
54 -119.61073 37.246075 "CA"  .4787962
10  -105.5478  38.99854 "CO"  1.645692
45 -72.725716  41.62027 "CT"       100
41  -75.50023  38.99154 "DE"        10
38  -82.49748 28.628437 "FL"  13.51883
 6  -83.44633  32.64922 "GA" 1.4010508
50 -156.36305 20.254757 "HI" 11.305242
20  -93.50006  42.07462 "IA" 1.1914544
36 -114.65937  44.38908 "ID"         0
17  -89.19839  40.06503 "IL"  .8257279
14  -86.27565  39.90814 "IN" 1.2314225
39  -98.38021   38.4847 "KS" 1.5873016
 5  -85.29046 37.526665 "KY"  5.178571
33  -91.98564  31.05839 "LA"  .8701472
35  -71.79853  42.25332 "MA"  2.857143
48  -76.77244   39.0369 "MD" 1.8120046
49  -69.22665   45.3714 "ME"  54.38596
34  -85.43715  44.35067 "MI" 13.861868
43  -94.30901  46.31633 "MN"         0
28  -92.47739   38.3676 "MO" .16798656
 9  -89.66521  32.75086 "MS" 35.737705
19 -109.64512  47.03353 "MT"         0
47  -100.4693  47.44633 "ND" .24390244
 1  -99.81079  41.52714 "NE"  .2967359
40   -71.5776  43.68563 "NH"  42.57907
46 -74.660904  40.18424 "NJ" 1.8258427
 3 -106.10837 34.421368 "NM"  5.263158
16 -116.65522  39.35623 "NV" .14925373
55  -75.50312  42.94038 "NY"         0
24  -82.79011  40.29176 "OH" 2.0053475
30  -97.50829  35.58348 "OK" 14.251781
27  -120.5554  43.93665 "OR"         0
 8  -77.79955  40.87383 "PA"  2.656609
53  -71.55364  41.67609 "RI"         0
21  -80.89611  33.90795 "SC"  39.30754
 4  -100.2305  44.43616 "SD"         0
12  -86.34338  35.84298 "TN" 2.1995287
42  -99.35069 31.484453 "TX" 1.8089565
11  -111.6782  39.32379 "UT" 1.4285715
31  -78.81226  37.51502 "VA"  43.19527
18  -72.66266  44.07519 "VT" 4.0816326
 2 -120.44686  47.38096 "WA"         5
26   -90.0117  44.63802 "WI"       100
32  -80.61371  38.64259 "WV"   29.8633
13 -107.55145  42.99963 "WY"         0

[Andrew Musau]

Also, is there a colour palette for which states with a value of X of 0 could be just white, instead of very lightly red shaded?

You can specify your own colors or create your own palette, including taking an existing palette and adding white as one of the colors. But then, how do you distinguish states with zero values and states with no observations? Your example is really not reproducible without some work. The borders are controlled by the outline color, which you specified as -ocolor(none)- in your code.

copy http://www2.census.gov/geo/tiger/GENZ2015/shp/cb_2015_us_state_500k.zip cb_2015_us_state_500k.zip
unzipfile cb_2015_us_state_500k.zip, replace
shp2dta using  cb_2015_us_state_500k, data("us_data")  coor("us_coordinates") ///
genid(id) gencentroids(c) replace
use us_data, clear
quietly destring STATEFP, generate(fips)
drop if fips > 56
save usstates,replace

//Determine IDs for Hawaii and Alaska
use usstates,clear
l fips id  if fips == 2 | fips == 15

// Alaska is reduced in size by 60% and then both Alaska and Hawaii are re-positioned under the southwestern states.
//Drop northwest islands of Hawaii and western part of Aleutian Islands
use us_coordinates,clear

gen order = _n
drop if _X < -165 & _X != . &  _ID == 50
replace _X = _X  + 55  if  _X != .  &  _ID == 50
replace _Y = _Y  + 4  if _Y != .  &  _ID == 50

replace _X = _X*.4  -55 if  _X !=  . &  _ID == 15
replace _Y = _Y*.4  + 1 if _Y != .  & _ID == 15
drop if _X > -10 & _X != . & _ID == 15
sort order
sort _ID
drop order

save us_coordinates2,replace

clear
input byte _ID float(x_c y_c) str10 state float X
15 -152.22098 64.217476 "AK"  3.012048
25  -86.82837 32.789825 "AL"  19.62733
 7  -92.43925  34.89974 "AR"         0
22  -111.6644 34.293102 "AZ" .10857763
54 -119.61073 37.246075 "CA"  .4787962
10  -105.5478  38.99854 "CO"  1.645692
45 -72.725716  41.62027 "CT"       100
41  -75.50023  38.99154 "DE"        10
38  -82.49748 28.628437 "FL"  13.51883
 6  -83.44633  32.64922 "GA" 1.4010508
50 -156.36305 20.254757 "HI" 11.305242
20  -93.50006  42.07462 "IA" 1.1914544
36 -114.65937  44.38908 "ID"         0
17  -89.19839  40.06503 "IL"  .8257279
14  -86.27565  39.90814 "IN" 1.2314225
39  -98.38021   38.4847 "KS" 1.5873016
 5  -85.29046 37.526665 "KY"  5.178571
33  -91.98564  31.05839 "LA"  .8701472
35  -71.79853  42.25332 "MA"  2.857143
48  -76.77244   39.0369 "MD" 1.8120046
49  -69.22665   45.3714 "ME"  54.38596
34  -85.43715  44.35067 "MI" 13.861868
43  -94.30901  46.31633 "MN"         0
28  -92.47739   38.3676 "MO" .16798656
 9  -89.66521  32.75086 "MS" 35.737705
19 -109.64512  47.03353 "MT"         0
47  -100.4693  47.44633 "ND" .24390244
 1  -99.81079  41.52714 "NE"  .2967359
40   -71.5776  43.68563 "NH"  42.57907
46 -74.660904  40.18424 "NJ" 1.8258427
 3 -106.10837 34.421368 "NM"  5.263158
16 -116.65522  39.35623 "NV" .14925373
55  -75.50312  42.94038 "NY"         0
24  -82.79011  40.29176 "OH" 2.0053475
30  -97.50829  35.58348 "OK" 14.251781
27  -120.5554  43.93665 "OR"         0
 8  -77.79955  40.87383 "PA"  2.656609
53  -71.55364  41.67609 "RI"         0
21  -80.89611  33.90795 "SC"  39.30754
 4  -100.2305  44.43616 "SD"         0
12  -86.34338  35.84298 "TN" 2.1995287
42  -99.35069 31.484453 "TX" 1.8089565
11  -111.6782  39.32379 "UT" 1.4285715
31  -78.81226  37.51502 "VA"  43.19527
18  -72.66266  44.07519 "VT" 4.0816326
 2 -120.44686  47.38096 "WA"         5
26   -90.0117  44.63802 "WI"       100
32  -80.61371  38.64259 "WV"   29.8633
13 -107.55145  42.99963 "WY"         0
end

tempfile data
save `data'


use usstates,clear
rename STUSPS state
merge 1:m state using `data'
rename _ID _ID2
rename id _ID

keep if _merge==3
drop _merge

bys state: keep if _N==_n
spmap X using us_coordinates2 , id(_ID) ///
clb(0 0.01 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 2 3 4 5 10 15 20 25 30 35 40 60 80 100) ///
clm(custom)  legorder(lohi) fcolor(none red*0.5 red*1.0 red*1.5 red*2.0 red*2.5 red*3.0 red*3.5) ocolor(gray ..) ///
clnumber(5) legend(order(1 "No data" 2 3 4 5 6 7 8 9) pos(5))

[Maxence Morlet]

Thanks Andrew! I ran your code and it worked!