next up previous contents
Next: Oregon Up: State-by-State Exceptions and Notes Previous: Ohio

Oklahoma

Merged at MCD Group level.

The computer merge worked well. Handmatching was done for all but 3 counties (Kay, Washington, which could be aggregated to county level) and Washita (no census data).

Oklahoma codes to go with math90ok.xls are in the two columns after the censusprec column.

00:unmatched
01:absentee vote
02:assumes 32s=320
03:match basis was wrong (check on why, see "match" variable)
04:questionable; good guess 
05:zero before "censprec" number caused no-match
06:assumes x=x+xa
07:assumes x=xa+xb
08:2 electoral precincts=1censprec (usually joined by ampersand)
09:3 electoral precincts=1cp
10:4 electoral precincts=1cp
11:5 electoral precincts=1cp
12:x=x+xa+xb (no such cases in OK)
13:dashes created no-matches
14:spelling or word similar but not exactly the same
89:2mcd=3 electoral precincts=1cp
90:3mcds=4 electoral precincts=1cp
91:3mcds=2 electoral precincts=1cp
92:4mcds=2 electoral precincts=1cp
93:2mcds=4 electoral precinct=1cp
94:3mcds=5 electoral precincts=1cp
95:3mcds=3 electoral precincts=1cp
96:2mcds=1 electoral precinct=2cps
97:2mcds=3 electoral precincts
98:countywide problem
99:2mcds=2 electoral precincts=1cp

Note some of the information in 89-99 repeats the information in 6-12 so these codes could be collapsed or changed.

Additional codes in the 3rd and following columns after "censprec" column are each of the MCD codes (from 2 to 4) that were "lost" in the merge.




Copyright © 1997-2004 [ROAD Home] Questions? Contact the ROAD webmaster.