Translate

Sunday, September 21, 2014

Reactivate yourself!

Updated with recent data and comment about mobile urban millenials and voter registration -RMF

Make sure to reactivate your status as a voter if you have moved!   In fact, it is just a good idea to log on to  myvote.wa.gov  and check your voting status now. WA state and your county are actively paring  (or making inactive) voters who have moved, passed on, not submitted change of address forms in preparation for the October/November general election period.  When I perform the PostgreSQL query below on the September 5 vs. the September 19 active voter databases in Whatcom County, I have a possibility of 3589  to 3693 voters that are no longer part of the current active voter database. If you have any doubts that the current registration process isn't a disaster for urban millennials, check out the stats below. Code for this post is here.

The active voter list has been going down then up as cleaning and registration happens simultaneously.

Active Voter  list counts 09/05 - 09/23
 127901 09/05
 126417 09/10
 124814 09/19
 124927 09/23

The numbers we see below tells that it is not just college students that become accidentally disenfranchised by forgetting to update their residence address at myvote.wa.gov .  A wide age range of urban (200 series precincts) lead the list for those voters whose status becomes 'inactive'. The best bet is that any apartment dweller or renter with a mobile life can forget to update their address at  myvote.wa.gov.

** Statistics below are for the 3480 now with statuscode = inactive on 9/23 but not on 09/05 **

Top 10 by Age All County
count | age
------+-----
  217 |  22
  206 |  21
  196 |  23
  179 |  24
  153 |  20
  131 |  25
  119 |  26
  109 |  27
   92 |  31
   87 |  19

Top 10 by Age LD42
count | age
------+-----
   88 |  22
   82 |  23
   82 |  24
   67 |  26
   65 |  25
   61 |  27
   60 |  21
   55 |  30
   55 |  34
   54 |  32

Top 10 by Precinct All County
count | precinctid
------+------------
  151 |        245
  127 |        253
   75 |        228
   72 |        252
   70 |        201
   63 |        226
   60 |        257
   60 |        247
   56 |        225
   56 |        231

Top 10 by Precinct LD42
count | precinctid
------+------------
   70 |        201
   56 |        225
   54 |        229
   53 |        227
   51 |        208
   48 |        220
   46 |        302
   44 |        206
   44 |        135
   34 |        107

/* Begin original post now - RMF */
I looked first for users that have been removed from the active voter list by querying registration numbers.

Create View UsersGone AS
SELECT lastname,firstname,middlename,registrationnumber,precinctID,statuscode
FROM voterdb090514
WHERE statuscode = 'A'
AND NOT EXISTS (
  SELECT *
  -- may want 'WHERE statuscode= 'A'' here
  FROM voterdb091914
  where voterdb090514.registrationnumber = voterdb091914.registrationnumber
  );


 Select Count(*) from UsersGone;
 count
-------
  3589
(1 row)

Registration numbers can be recycled, so a unique first,middle,last name combination can also be used to look for differences in the active voter database over time.

Create View UserArray AS  
SELECT statuscode,ARRAY[lastname,firstname,middlename] as array
FROM voterdb090514
WHERE statuscode = 'A';

Create View UserArray1 AS  
SELECT statuscode,precinctid, ARRAY[lastname,firstname,middlename] as array
FROM voterdb091914
WHERE statuscode = 'A';

drop view UsersGone;
Create View UsersGone AS
Select *  
FROM Userarray
where NOT EXISTS (
  SELECT *
  FROM Userarray1
  where UserArray.array = UserArray1.array
  );
  
Select Count(*) from UsersGone;
 count
-------
  3693
(1 row) 

I will use the high estimate from the name based tuple for a 42nd district query. When I query this year's all important 42nd district, I find that 2028 of these dropped voters come from 42nd LD precincts, many from 200 series precincts:

Create View UsersGone42nd as 
Select precinctid,Count(*) from UsersGone where precinctid in (Select precincts42 from ld42)
Group By PrecinctID Order By Count DESC;

Select * from UsersGone42nd;
 precinctid | count
------------+-------
        201 |    74
        225 |    57
        227 |    55
        229 |    55
        208 |    52
        220 |    50
        302 |    48
        135 |    46
        206 |    46
        107 |    36
        137 |    34
        204 |    34
        101 |    33
        213 |    32
        108 |    32
        203 |    31
        505 |    31
        148 |    31
        506 |    31
        211 |    30
        216 |    28
        136 |    26
        126 |    25
        218 |    25
        209 |    25
        301 |    23
        222 |    23
        215 |    21
        134 |    20
        150 |    20
        221 |    20
        133 |    19
        401 |    19
        233 |    19
        224 |    18
....

Select sum(count) from UsersGone42nd;
 sum
------
 2028

No comments: