2010 Wisconsin Congressional Districts

Redistricting requirements

In Wisconsin, districts must:

have equal populations

Algorithmic Constraints

We enforce a maximum population deviation of 0.5% and add a pseudo-county constraint to reduce county and municipality splits. Since Milwaukee County has a greater population than the target district population, we split Milwaukee County by municipality lines.

Data Sources

Data for Wisconsin comes from the ALARM Project's 2010 Redistricting Data Files.

Pre-processing Notes

No manual pre-processing decisions were necessary.

Simulation Notes

We sample 10,000 districting plans for Wisconsin over 2 independent runs of the SMC algorithm. No special techniques were needed to produce the sample.

Validation

SMC: 10,000 sampled plans of 8 districts on 6,290 units
`adapt_k_thresh`=0.985 • `seq_alpha`=0.5
`est_label_mult`=1 • `pop_temper`=0

Plan diversity 80% range: 0.60 to 0.87

R-hat values for summary statistics:
   pop_overlap      total_vap       plan_dev      comp_edge    comp_polsby 
      1.003023       1.000496       1.006525       1.006056       1.003808 
     pop_white      pop_black       pop_hisp       pop_aian      pop_asian 
      1.002087       1.003262       1.007569       1.000165       1.001255 
      pop_nhpi      pop_other        pop_two      vap_white      vap_black 
      1.002095       1.002484       1.002402       1.002905       1.004289 
      vap_hisp       vap_aian      vap_asian       vap_nhpi      vap_other 
      1.006855       1.000438       1.001002       1.000476       1.002564 
       vap_two pre_16_rep_tru pre_16_dem_cli pre_20_dem_bid pre_20_rep_tru 
      1.001800       1.003794       1.001837       1.002079       1.003212 
uss_16_rep_joh uss_16_dem_fei uss_18_rep_vuk uss_18_dem_bal gov_18_rep_wal 
      1.005168       1.001511       1.005151       1.001903       1.005313 
gov_18_dem_eve atg_18_rep_sch atg_18_dem_kau sos_18_rep_sch sos_18_dem_laf 
      1.000574       1.004340       1.000814       1.004094       1.001847 
        adv_16         adv_18         adv_20         arv_16         arv_18 
      1.001014       1.001186       1.002079       1.004435       1.004681 
        arv_20  county_splits    muni_splits            ndv            nrv 
      1.003212       1.006289       1.015318       1.001142       1.004263 
       ndshare          e_dvs         pr_dem          e_dem          pbias 
      1.003853       1.003965       1.001472       1.002357       1.004259 
          egap 
      1.002601 

Sampling diagnostics for SMC run 1 of 2 (5,000 samples)
         Eff. samples (%) Acc. rate Log wgt. sd  Max. unique Est. k 
Split 1     4,913 (98.3%)     10.6%        0.27 3,149 (100%)     17 
Split 2     4,862 (97.2%)     14.3%        0.34 3,125 ( 99%)     11 
Split 3     4,794 (95.9%)     23.2%        0.44 3,109 ( 98%)      6 
Split 4     4,633 (92.7%)     28.3%        0.50 3,077 ( 97%)      4 
Split 5     4,620 (92.4%)     29.7%        0.52 3,074 ( 97%)      3 
Split 6     4,521 (90.4%)     29.4%        0.56 2,951 ( 93%)      2 
Split 7     4,576 (91.5%)     10.5%        0.55 2,561 ( 81%)      2 
Resample    3,370 (67.4%)       NA%        0.56 2,888 ( 91%)     NA 

Sampling diagnostics for SMC run 2 of 2 (5,000 samples)
         Eff. samples (%) Acc. rate Log wgt. sd  Max. unique Est. k 
Split 1     4,912 (98.2%)     12.9%        0.27 3,169 (100%)     14 
Split 2     4,859 (97.2%)     17.7%        0.34 3,165 (100%)      9 
Split 3     4,793 (95.9%)     22.8%        0.44 3,140 ( 99%)      6 
Split 4     4,692 (93.8%)     28.3%        0.51 3,078 ( 97%)      4 
Split 5     4,616 (92.3%)     30.0%        0.53 3,052 ( 97%)      3 
Split 6     4,581 (91.6%)     29.1%        0.54 2,926 ( 93%)      2 
Split 7     4,680 (93.6%)     10.2%        0.48 2,619 ( 83%)      2 
Resample    3,760 (75.2%)       NA%        0.49 2,949 ( 93%)     NA 

•  Watch out for low effective samples, very low acceptance rates (less than
1%), large std. devs. of the log weights (more than 3 or so), and low numbers
of unique plans. R-hat values for summary statistics should be between 1 and
1.05.

Checklist

[X] I have followed the instructions
[X] I have updated the tracker
[X] All TODO lines from the template code have been removed
[X] I have merged in the master branch and then recalculated summary statistics
[X] I have run enforce_style() to format my code
[X] The documentation copied above is up-to-date
[X] There are no data files in this pull request
[X] None of the file output paths (for the redist_map and redist_plans objects, and summary statistics) have been edited

@tylersimko

alarm-redist / fifty-states