Add 2016 election precinct shapes for New Hampshire

[nvkelso]

Priority 4 based on population

[nvkelso]

When researching & requesting data also ask for matching election results (presidential, congressional, state house, and state senate) in addition to the precinct shape data for 2016, and for earlier election years if available.

[wboykinm]

Also dibs.

[nvkelso]

• Voting Districts, New Hampshire, 2000
  • http://geocommons.com/datasets?id=5137
  • http://www.census.gov/geo/www/cob/vt2000.html
  • https://raw.githubusercontent.com/geoiq/gc_data/master/datasets/5137.geojson

[wboykinm]

The geocommons-hosted districts suffer from the same problem the Vermont districts ( #48) had; the rural areas of NH don't vote at the county level, but rather the town. I'll see if there's anything like a polygon boundary available.

[nvkelso]

Can you clarify, please? Are votes cast at the town (or minor civil division) level only, there aren't separate precincts for reporting purposes?

[wboykinm]

Sure; votes are collected and reported at the "district" level, which is more or less analogous with "precinct". The full list of these districts is here, and election results (at least since 2012) have been reported at that level.

The complication (similar to Vermont) is that no geometric representation of the complete boundaries seems to exist. NH-GRANIT hosts a "political boundaries" layer that conforms to the upper level of the SoS reporting districts, but it's missing ward subdivisions that exist in all cities and constitute the highest level of detail the SoS reports.

Ward boundaries exist somewhere (see this example for Lebanon,NH, and the now out-of date smaller units in the above geocommons map), but neither GRANIT nor the SoS host them, making this potentially a city-by-city slog.

@nvkelso While I try to get some clarification from GRANIT, do you think it's worth importing the coarser political boundaries that are available?

[nvkelso]

Yes, we should import the coarser political boundaries and then supplement them with the more detailed wards later. There are two examples on how to do this now: - Kansas: see the Makefile and associated script - Indiana: pure Makefile Basically we can cookie cutter pieces of the more detailed wards into the more generic districts for a composite state output. If you upload the statewide districts I can take care of doing the integration work once the wards start landing (and they can land 1 by 1 as you find them with new PRs).

…

On Tue, Mar 14, 2017 at 6:55 AM, Bill Morris ***@***.***> wrote: Sure, votes are collected and reported at the "district" level, which is more or less analogous with "precinct". The full list of these districts is here <https://www.dropbox.com/s/81brnzdwwvkpdo1/TOWNS%20AND%20WARDS%20AS%20DISTRICTED%20FOR%20ELECTION%20PURPOSES2015.pdf?dl=0>, and election results (at least since 2012) have been reported at that level <http://sos.nh.gov/2012PresGen.aspx?id=28307>. The complication (similar to Vermont) is that no geometric representation of the complete boundaries seems to exist. NH-GRANIT hosts a "political boundaries" layer <http://www.granit.unh.edu/data/metadata?file=pb/nh/pb.html> that conforms to the upper level of the SoS reporting districts, but it's missing ward subdivisions that exist in all cities and constitute the highest level of detail the SoS reports. Ward boundaries exist *somewhere* (see this example for Lebanon,NH <http://lebcity.net/CityClerk/Documents/Election/wardmap.pdf>), but neither GRANIT nor the SoS host them, making this potentially a city-by-city slog. @nvkelso <https://github.com/nvkelso> While I try to get some clarification from GRANIT, do you think it's worth importing the coarser political boundaries that *are* available? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#42 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA0EOwXV_RXeMd4uXb309l4i72TFqpn1ks5rlpxMgaJpZM4MTv9h> .

[migurski]

This does suggest that our output schema should include a flag for division level: precincts and wards are a lot different from larger house districts.

[wboykinm]

I AM LEARNING SO MUCH ABOUT DEMOOCRAZY

[nvkelso]

Hahahaha. Me too :) Bill can you clarify what "district" means in your case? I assume it's something smaller than a "congressional district", but am not familiar with how New England sets things up.

…

On Tue, Mar 14, 2017 at 2:54 PM, Bill Morris ***@***.***> wrote: I AM LEARNING SO MUCH ABOUT DEMOOCRAZY — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#42 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA0EOwP8YNP9ytNMQvcCybImmMSTcRKoks5rlwydgaJpZM4MTv9h> .

[wboykinm]

"Voting District" refers to the catchment within which Granite Staters vote. One polling place per district, so it includes both towns and wards.

[wboykinm]

@nvkelso You already added a version of what we're looking for, with both towns and wards! These are just outdated since the 2012 redistricting. Did you get them from the census?

[nvkelso]

Yep, Census 2010.

…

On Mar 14, 2017, at 15:35, Bill Morris ***@***.***> wrote: @nvkelso You already added an exact version of what we're looking for! These are just outdated since the 2012 redistricting. Did you get them from the census? ― You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[wboykinm]

I suspect that means someone at the census went through what we're doing right now back in 2009, and isn't scheduled to do so again until 2019 😬

[nvkelso]

Yep :\ But you have very small states ;)

…

On Tue, Mar 14, 2017 at 5:06 PM, Bill Morris ***@***.***> wrote: I suspect that means someone at the census went through what we're doing right now back in 2009, and isn't scheduled to do so again until 2019 😬 — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#42 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA0EO70SgS-7_D5TLE0VfVgNf7qI9lfaks5rlyuRgaJpZM4MTv9h> .

[wboykinm]

Ward maps (PDF unless otherwise noted, check boxes as cities are digitized or acquired)

• Lebanon
• Manchester
• Dover (possibly-scrapable Gmaps app)
• Concord (Note the unironic use of comic sans)
• Franklin
• Keene
• Laconia (very-scrapable google mymaps page)
• Nashua
• Portsmouth
• Rochester
• Somersworth

@nvkelso This is all of the cities in NH with ward subdivisions.

[nvkelso]

These are great, thanks!

…

On Wed, Mar 15, 2017 at 4:37 PM, Bill Morris ***@***.***> wrote: Ward maps (PDF unless otherwise noted) - Lebanon <http://lebcity.net/CityClerk/Documents/Election/wardmap.pdf> - Manchester <https://www.manchesternh.gov/Portals/2/Departments/city_clerk/wards/CitywideWardMap.pdf> - Dover <https://dovernh.mapgeo.io/?latlng=43.17561%2C-70.885208&panel=themes&themes=%5B%22wards-and-polling-places%22%5D&zoom=12> (possibly-scrapable Gmaps app) - Concord <http://nh-concord.civicplus.com/DocumentCenter/View/1201> (Note the unironic use of comic sans) - Franklin <http://www.franklinnh.org/Pages/FranklinNH_Council/map.pdf> - Keene <http://www.ci.keene.nh.us/sites/default/files/WARDS_12_GRID_0.pdf> - Laconia <https://www.google.com/maps/d/u/0/viewer?mid=1oqNwYVdgUhjL5ZhPRlY-XxaEujs&ie=UTF8&hl=en&msa=0&ll=43.55004100000002%2C-71.47035599999998&spn=0.084851%2C0.137501&z=13> (very-scrapable google mymaps page) - Nashua <http://www.nashuanh.gov/DocumentCenter/View/3836> - Portsmouth <http://www.cityofportsmouth.com/cityclerk/documents/votingWards.pdf> - Rochester <http://www.rochesternh.net/sites/rochesternh/files/file/file/rochester_wards_06-2016.pdf> - Somersworth <http://www.somersworth.com/wp-content/uploads/2012/11/Somersworth-Wards.pdf> @nvkelso <https://github.com/nvkelso> This is all of the cities in NH with ward subdivisions. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#42 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA0EO5Vy6xNkcWUKYOsarFGq0wIS1d2aks5rmHZSgaJpZM4MTv9h> .

[nvkelso]

I think what's going on here is most of the state is still current in 2016 using the 2010 US Census data except the ~11 places mentioned in #42 (comment). I think the Makefile is correct, but these areas are pretty small so the state looks entirely green in the preview map – but if you look closer there are some gaps.

[wboykinm]

@nvkelso This is my next project 😀