2010 Georgia Congressional Districts

[Jfer09]

Redistricting requirements

In Georgia, under the 2011-12 Guidelines for the House Legislative and Congressional Reapportionment Committee: districts must:

1. be contiguous
2. have equal populations
3. be geographically compact
4. preserve political subdivisions and communities of interest
5. avoid pairing incumbents

Guidelines for the House Legislative and Congressional Reapportionment Committee

Algorithmic Constraints

We enforce a maximum population deviation of 0.5%.

Data Sources

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

Pre-processing Notes

No manual pre-processing decisions were necessary.

Simulation Notes

We sample 60,000 districting plans for Georgia across two independent runs of the SMC algorithm. We then thin the sample to exactly 5,000 plans.

We impose a hinge constraint on the Black Voting Age Population so that it encourages districts with BVAP above 43%, but districts with BVAP of 34% or less are not penalized as much. In addition, we impose an inverse hinge constraint on the Black Voting Age Population to penalize districts with BVAP above 61% to prevent packing.

Validation

SMC: 60,000 sampled plans of 14 districts on 2,961 units
`adapt_k_thresh`=0.985 • `seq_alpha`=0.5
`est_label_mult`=1 • `pop_temper`=0

Plan diversity 80% range: 0.63 to 0.82

R-hat values for summary statistics:
    pop_overlap       total_vap        plan_dev       comp_edge     comp_polsby       pop_white 
       1.028433        1.000465        1.006164        1.019184        1.029123        1.014853 
      pop_black        pop_hisp        pop_aian       pop_asian        pop_nhpi       pop_other 
       1.013547        1.009564        1.014845        1.037753        1.023118        1.012276 
        pop_two       vap_white       vap_black        vap_hisp        vap_aian       vap_asian 
       1.011599        1.016606        1.005109        1.009466        1.039454        1.035719 
       vap_nhpi       vap_other         vap_two  pre_16_rep_tru  pre_16_dem_cli  pre_20_rep_tru 
       1.026342        1.005372        1.020378        1.000464        1.022859        1.000818 
 pre_20_dem_bid  uss_16_rep_isa  uss_16_dem_bar  uss_20_rep_per  uss_20_dem_oss  gov_18_rep_kem 
       1.044291        1.000610        1.022661        1.000719        1.039536        1.004839 
 gov_18_dem_abr  atg_18_rep_car  atg_18_dem_bai  sos_18_rep_raf  sos_18_dem_bar sos_r18_rep_raf 
       1.032997        1.003077        1.036278        1.002255        1.008487        1.010313 
sos_r18_dem_bar          adv_16          adv_18          adv_20          arv_16          arv_18 
       1.009812        1.023670        1.016225        1.041143        1.001537        1.001523 
         arv_20   county_splits     muni_splits             ndv             nrv         ndshare 
       1.000756        1.007849        1.002823        1.024799        1.001580        1.000859 
          e_dvs          pr_dem           e_dem           pbias            egap 
       1.001079        1.000464        1.032985        1.039055        1.029460 

Sampling diagnostics for SMC run 1 of 2 (30,000 samples)
         Eff. samples (%) Acc. rate Log wgt. sd   Max. unique Est. k 
Split 1    23,635 (78.8%)     20.2%        0.25 19,004 (100%)      9 
Split 2    23,300 (77.7%)     23.2%        0.55 17,328 ( 91%)      7 
Split 3    22,972 (76.6%)     29.1%        0.63 17,302 ( 91%)      5 
Split 4    22,490 (75.0%)     26.3%        0.67 17,098 ( 90%)      5 
Split 5    21,777 (72.6%)     23.7%        0.71 16,991 ( 90%)      5 
Split 6    20,955 (69.8%)     26.0%        0.75 16,841 ( 89%)      4 
Split 7    20,352 (67.8%)     30.0%        0.79 16,686 ( 88%)      3 
Split 8    19,843 (66.1%)     27.2%        0.80 16,569 ( 87%)      3 
Split 9    19,332 (64.4%)     32.0%        0.82 16,565 ( 87%)      2 
Split 10   18,722 (62.4%)     28.6%        0.83 16,417 ( 87%)      2 
Split 11   19,393 (64.6%)     18.1%        0.84 15,792 ( 83%)      3 
Split 12   19,673 (65.6%)     19.4%        0.82 15,465 ( 82%)      2 
Split 13   15,689 (52.3%)      7.3%        0.86 13,678 ( 72%)      2 
Resample    3,553 (11.8%)       NA%        1.70 12,654 ( 67%)     NA 

Sampling diagnostics for SMC run 2 of 2 (30,000 samples)
         Eff. samples (%) Acc. rate Log wgt. sd   Max. unique Est. k 
Split 1    23,614 (78.7%)     20.3%        0.25 18,977 (100%)      9 
Split 2    23,294 (77.6%)     27.0%        0.55 17,338 ( 91%)      6 
Split 3    23,044 (76.8%)     29.2%        0.63 17,220 ( 91%)      5 
Split 4    22,387 (74.6%)     31.8%        0.67 17,139 ( 90%)      4 
Split 5    21,876 (72.9%)     36.2%        0.71 16,993 ( 90%)      3 
Split 6    21,236 (70.8%)     42.8%        0.75 16,944 ( 89%)      2 
Split 7    20,290 (67.6%)     38.7%        0.78 16,834 ( 89%)      2 
Split 8    19,804 (66.0%)     27.0%        0.81 16,466 ( 87%)      3 
Split 9    19,485 (64.9%)     24.1%        0.81 16,483 ( 87%)      3 
Split 10   19,044 (63.5%)     21.2%        0.81 16,356 ( 86%)      3 
Split 11   19,615 (65.4%)     13.7%        0.83 16,123 ( 85%)      4 
Split 12   19,621 (65.4%)     13.9%        0.81 15,561 ( 82%)      3 
Split 13   14,736 (49.1%)      6.8%        0.78 13,983 ( 74%)      2 
Resample    3,323 (11.1%)       NA%        1.70 11,805 ( 62%)     NA 

   bvap_rank       dem
       <dbl>     <dbl>
 1         1 0        
 2         2 0.0000833
 3         3 0.0693   
 4         4 0.426    
 5         5 0.625    
 6         6 0.551    
 7         7 0.128    
 8         8 0.0620   
 9         9 0.0282   
10        10 0.0828   
11        11 0.247    
12        12 0.947    
13        13 1.00     
14        14 1        

  n_black_perf     n
1            2   778
2            3 40877
3            4 17330
4            5  1005
5            6    10

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

delete this line and all the tags except the reviewers you need @CoryMcCartan @christopherkenny

[CoryMcCartan]

I'll let Chris weigh in on VRA stuff, but this looks very solid to me!

Can we also edit the top of the documentation?

• remove mention of the VRA as a constraint
• copy the link from the bottom in-text, like this: [Guidelines for the House Legislative and Congressional Reapportionment Committee](https://www.dropbox.com/s/i8zqyivtr8iozs8/GeorgiaSenateCommitteeGuidelines2011-12.pdf)

[Jfer09]

changed!