2010 Virginia Congressional Districts

[kevpwang]

Redistricting requirements

In Virginia, districts must, under Commitee Resolution No. 1 adopted by the Senate and House Committees on Privileges and Elections in 2001:

1. be contiguous
2. have equal populations
3. be geographically compact
4. preserve county and municipality boundaries as much as possible
5. preserve communities of interest, as defined by criteria that "may include, among others, economic factors, social factors, cultural factors, geographic features, governmental jurisdictions and service delivery areas, political beliefs, voting trends, and incumbency considerations"

Algorithmic Constraints

We enforce a maximum population deviation of 0.5%. We use a pseudo-county constraint described below which attempts to mimic the norms in Virginia of generally preserving county, city, and township boundaries.

Data Sources

Data for Virginia comes from the ALARM Project's 2020 Redistricting Data Files.

Pre-processing Notes

No manual pre-processing decisions were necessary.

Simulation Notes

We sample 10,000 districting plans for Virginia across two independent runs of the SMC algorithm, and then thin the sample down to 5,000 plans. To balance county and municipality splits, we create pseudocounties for use in the county constraint, which leads to fewer municipality splits than using a county constraint. Note that Fairfax County must be split due to its large population, although within the county, we avoid splitting any municipality.

Validation

SMC: 5,000 sampled plans of 11 districts on 2,373 units
`adapt_k_thresh`=0.985 • `seq_alpha`=0.5
`est_label_mult`=1 • `pop_temper`=0  
ℹ Running simulations for VA_cd_2010
Plan diversity 80% range: 0.61 to 0.85
ℹ Running simulations for VA_cd_2010
R-hat values for summary statistics:
   pop_overlap      total_vap       plan_dev      comp_edge    comp_polsby      pop_white      pop_black 
      1.003868       1.000114       1.005894       1.003689       1.001729       1.002286       1.001260 
      pop_hisp       pop_aian      pop_asian       pop_nhpi      pop_other        pop_two      vap_white 
      1.000052       1.003187       1.000871       1.004234       1.000297       1.007810       1.001300 
     vap_black       vap_hisp       vap_aian      vap_asian       vap_nhpi      vap_other        vap_two 
      1.000727       1.000004       1.002956       1.002090       1.002937       1.002055       1.002543 
pre_16_dem_cli pre_16_rep_tru pre_20_dem_bid pre_20_rep_tru uss_18_dem_kai uss_18_rep_ste uss_20_dem_war 
      1.003154       1.005587       1.005808       1.004005       1.005674       1.005871       1.004615 
uss_20_rep_gad         adv_16         adv_18         adv_20         arv_16         arv_18         arv_20 
      1.003080       1.003154       1.005674       1.005205       1.005587       1.005871       1.003531 
 county_splits    muni_splits            ndv            nrv        ndshare          e_dvs         pr_dem 
      1.000645       1.004895       1.004043       1.004709       1.005044       1.004990       1.006348 
         e_dem          pbias           egap 
      1.012382       1.000935       1.010968 

Sampling diagnostics for SMC run 1 of 2
         Eff. samples (%) Acc. rate Log wgt. sd  Max. unique Est. k 
Split 1    4,854 (194.2%)     15.0%        0.35 3,156 (200%)      7 
Split 2    4,776 (191.0%)     19.8%        0.43 3,151 (199%)      5 
Split 3    4,652 (186.1%)     22.6%        0.52 3,061 (194%)      4 
Split 4    4,570 (182.8%)     26.9%        0.57 3,076 (195%)      3 
Split 5    4,467 (178.7%)     12.8%        0.62 3,080 (195%)      6 
Split 6    4,458 (178.3%)     14.0%        0.62 2,994 (189%)      5 
Split 7    4,459 (178.4%)     20.1%        0.61 3,005 (190%)      3 
Split 8    4,451 (178.0%)     13.2%        0.61 2,970 (188%)      4 
Split 9    4,397 (175.9%)     13.2%        0.65 2,826 (179%)      3 
Split 10   4,472 (178.9%)      6.4%        0.62 2,570 (163%)      2 
Resample   3,033 (121.3%)       NA%        0.62 2,844 (180%)     NA 

Sampling diagnostics for SMC run 2 of 2
         Eff. samples (%) Acc. rate Log wgt. sd  Max. unique Est. k 
Split 1    4,857 (194.3%)     11.5%        0.35 3,144 (199%)      9 
Split 2    4,757 (190.3%)     16.7%        0.44 3,082 (195%)      6 
Split 3    4,581 (183.2%)     22.3%        0.54 3,116 (197%)      4 
Split 4    4,535 (181.4%)     26.9%        0.59 3,088 (195%)      3 
Split 5    4,504 (180.1%)     24.3%        0.62 3,059 (194%)      3 
Split 6    4,484 (179.4%)     30.2%        0.60 3,042 (192%)      2 
Split 7    4,497 (179.9%)     27.0%        0.58 3,003 (190%)      2 
Split 8    4,472 (178.9%)     16.9%        0.60 2,975 (188%)      3 
Split 9    4,462 (178.5%)     19.4%        0.62 2,850 (180%)      2 
Split 10   4,546 (181.8%)      6.5%        0.59 2,594 (164%)      2 
Resample   3,330 (133.2%)       NA%        0.59 2,855 (181%)     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.
ℹ Running simulations for VA_cd_2010

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

@christopherkenny

[christopherkenny]

Thanks Kevin, this looks great. One small thing: can you link to the rules in the documentation, so that people with interest in "Commitee Resolution No. 1 adopted by the Senate and House Committees on Privileges and Elections in 2001" can locate it easily.

[kevpwang]

Thanks Chris—have made the update.