2010 Indiana Congressional Districts

philipwosull commented 1 year ago

Redistricting requirements

In Indiana, districts must:

be contiguous
have equal populations

Algorithmic Constraints

We enforce a maximum population deviation of 0.05%.

Data Sources

Data for Indiana comes from All About Redistricting and the ALARM Project's Redistricting Data Files.

Pre-processing Notes

No manual pre-processing decisions were necessary.

Simulation Notes

We sample 5,000 districting plans for Indiana via two independent runs of 2,500 each. No special techniques were needed to produce the sample.

Validation

validation_20230124_1001

SMC: 5,000 sampled plans of 9 districts on 5,321 units
`adapt_k_thresh`=0.985 • `seq_alpha`=0.5
`est_label_mult`=1 • `pop_temper`=0
ℹ Preparing IN shapefile
Plan diversity 80% range: 0.61 to 0.84
ℹ Preparing IN shapefile
R-hat values for summary statistics:
   pop_overlap      total_vap       plan_dev      comp_edge    comp_polsby      pop_white      pop_black       pop_hisp       pop_aian 
     1.0056079      1.0157829      1.0023495      1.0056936      0.9998995      1.0017240      1.0054397      1.0076329      1.0085208 
     pop_asian       pop_nhpi      pop_other        pop_two      vap_white      vap_black       vap_hisp       vap_aian      vap_asian 
     1.0069785      1.0062443      1.0123559      1.0078574      1.0002113      1.0070624      1.0060313      1.0116287      1.0095481 
      vap_nhpi      vap_other        vap_two pre_16_rep_tru pre_16_dem_cli pre_20_rep_tru pre_20_dem_bid uss_16_rep_you uss_16_dem_bay 
     1.0047945      1.0158224      1.0077122      1.0013002      1.0034142      1.0019680      1.0041439      1.0022212      1.0081546 
uss_18_rep_bra uss_18_dem_don gov_16_rep_hol gov_16_dem_gre gov_20_rep_hol gov_20_dem_mye atg_16_rep_hil atg_16_dem_arr atg_20_rep_rok 
     1.0022046      1.0072408      1.0027240      1.0133391      1.0012153      1.0068348      1.0023948      1.0083989      1.0018739 
atg_20_dem_wei sos_18_rep_law sos_18_dem_har         adv_16         adv_18         adv_20         arv_16         arv_18         arv_20 
     1.0036051      1.0029485      1.0081045      1.0093922      1.0067329      1.0063869      1.0022252      1.0018852      1.0017001 
 county_splits    muni_splits            ndv            nrv        ndshare        e_dvs.x       pr_dem.x        e_dem.x        pbias.x 
     1.0069484      1.0066482      1.0072644      1.0021762      1.0175232      1.0173780      1.0027929      1.0036146      1.0035849 
        egap.x        e_dvs.y       pr_dem.y        e_dem.y        pbias.y         egap.y 
     1.0042604      1.0173780      1.0027929      1.0036146      1.0035849      1.0042604 

Sampling diagnostics for SMC run 1 of 2 (2,500 samples)
         Eff. samples (%) Acc. rate Log wgt. sd  Max. unique Est. k 
Split 1     2,361 (94.4%)      9.1%        0.49 1,600 (101%)     19 
Split 2     2,333 (93.3%)     15.6%        0.55 1,550 ( 98%)     11 
Split 3     2,289 (91.6%)     21.7%        0.63 1,528 ( 97%)      7 
Split 4     2,245 (89.8%)     26.1%        0.64 1,520 ( 96%)      5 
Split 5     2,233 (89.3%)     22.6%        0.63 1,534 ( 97%)      5 
Split 6     2,257 (90.3%)     22.5%        0.60 1,481 ( 94%)      4 
Split 7     2,245 (89.8%)     17.0%        0.61 1,441 ( 91%)      4 
Split 8     2,183 (87.3%)      7.6%        0.69 1,302 ( 82%)      3 
Resample    1,364 (54.6%)       NA%        0.68 1,765 (112%)     NA 

Sampling diagnostics for SMC run 2 of 2 (2,500 samples)
         Eff. samples (%) Acc. rate Log wgt. sd  Max. unique Est. k 
Split 1     2,366 (94.6%)     16.3%        0.48 1,579 (100%)     11 
Split 2     2,323 (92.9%)     23.6%        0.57 1,558 ( 99%)      7 
Split 3     2,277 (91.1%)     28.4%        0.65 1,502 ( 95%)      5 
Split 4     2,236 (89.4%)     21.7%        0.65 1,535 ( 97%)      6 
Split 5     2,227 (89.1%)     22.1%        0.63 1,526 ( 97%)      5 
Split 6     2,249 (90.0%)     18.8%        0.60 1,503 ( 95%)      5 
Split 7     2,239 (89.6%)     21.4%        0.60 1,467 ( 93%)      3 
Split 8     2,260 (90.4%)      9.3%        0.61 1,308 ( 83%)      2 
Resample    1,672 (66.9%)       NA%        0.62 1,849 (117%)     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 Preparing IN shapefile

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 commented 1 year ago

Hi @philipwosull, can you revise the data sources to include the default sentence seen in the other states? (Citing AAR is good for the plan source, but otherwise misses the general other sources cited).

We seem to be quite a bit less compact across the board. Can you increase the rho option?

christopherkenny commented 1 year ago

Okay, I've been convinced on the compactness point that this is a feature of some very specific boundaries and that we shouldn't be too worried. Good to go by me once the documentation is up-to-date.

philipwosull commented 1 year ago

Documentation updated :)

alarm-redist / fifty-states