Augusta (mostly), ME census block/ward/house district boundary issues

[jimrtex]

In Hudson, I have discovered that the address ranges in the 2010 edges shapefile may be wrong. That is, the 2010 shapefiles might not indicate which houses were enumerated in the 2010 census.

The 2015 edges shapefile used for the BBSP do not have address ranges in them. Instead there is a separate DBF file with address ranges.

The Census Bureau has a utility where you can enter an address and determine where the census bureau estimates where it is. There is a batch mode/

Link to Census Geocoder

[Kevinstat]

The internet is a wondrous thing. Google has indexed the census block file maps (PDF with readable text). That is how I found Nellman Pond. By the way sometimes map companies will put errors in their maps. The underlying data (where a street is, etc.) is not copyrightable, but the presentation is. If the pond is actually Wellman, and someone copies "Nellman" you have evidence that they copied the presentation.

If you select the edge of Nellman pond, you will see that the CBBFLG = 9 which means it is not eligible for use as a block boundary, or at least it is not intended to be used as one by the Census Bureau. If you compare the 2020 prototype blocks to those used in 2010, you can see that Nellman Pond is not slated as being a block boundary.

Directory of Prototype block files. Maine is state 23, Kennebec is county 11. The zip file is for the entire state, but I haven't found a shortcut to get county specific files.

Other changes in Kennebec include eliminate some blocks around the airport and interstate interchanges.

In the V2 files, the house boundary looks OK - District 80 is Windsor, and then comes into Augusta from the northeast. District 85 includes the Augusta portion  of the pond. I don't know how Maine communicated their district boundaries to the census bureau - it might not be by the statute text. Someone might have corrected it as part of transmission.

A curiosity is that the block numbers are quite similar across the Windsor-Augusta border.

[Kevinstat]

By the way sometimes map companies will put errors in their maps. The underlying data (where a street is, etc.) is not copyrightable, but the presentation is. If the pond is actually Wellman, and someone copies "Nellman" you have evidence that they copied the presentation.

So, might the Census Bureau be trying to catch people copying their presentation?  Or might the Census Bureau have copied someone else's presentation without permission?  (Any permission might have come with a note to correct certain errors they had put in to catch pirates.)

Other changes in Kennebec include eliminate some blocks around the airport and interstate interchanges.

The airport changes, if they hold through this process, would make the problem of the extension of North Street not being a block boundary worse than in 2010, although not as bad as in 2000 or (even worse) 1990.  It's quite likely that no one in 2010's CT 103, Block 1000 lives west of the extension of North Street (in Ward 1 rather than Ward 3).  One couldn't say that about the blocks including the Boothby Street neighborhood in 1990, 2000 or in the Census Bureau's prototype 2020 census blocks.

In the V2 files, the house boundary looks OK - District 80 is Windsor, and then comes into Augusta from the northeast. District 85 includes the Augusta portion  of the pond. I don't know how Maine communicated their district boundaries to the census bureau - it might not be by the statute text. Someone might have corrected it as part of transmission.

Yeah, maybe so.  Thanks for that information, and also about the lakeshore not being slated as a block boundary for 2020.  Lake and pond boundaries generally have been in Maine, but that may be being changed as part of the Census Bureau's efforts to reduce the number of census blocks.  In the Nellman/Wellman Pond case at least, I have no problem with that.  Hopefully lines connecting streams flowing in and out of ponds are being added block boundaries where the Census Bureau is eliminating the lakeshore.  (It's too late for me to do anything about that now probably.)

A curiosity is that the block numbers are quite similar across the Windsor-Augusta border.

Yeah, particularly around Nellman/Wellman Pond.  I think I might have noticed that once a year ago or more but had forgotten about it.

[Kevinstat]

Probably too late to make a difference, but...

You must be logged in to read this quote.

I had written "athough" instead of "although".  Funny the things I notice later on.

You must be logged in to read this quote.

I haven't heard back yet.  My guess is I'll receive a reply on Tuesday or Wednesday, on the day of or after the deadline for her to make submissions to the Census Bureau for this round of the project (the later round of state submissions to the Census Bureau as part of the Block Boundary Suggestion Project being to verify the Census Bureau's implementation of the state's suggestions, and seemingly not for making new ones).

[Kevinstat]

You must be logged in to read this quote.

You must be logged in to read this quote.

[Kevinstat]

Can anyone verify (or refute) that a redistricting liaison can make additional changes (not just clarifications to something the liaison had made some note about but perhaps not in the right way, but something where nothing had been done in the initial phase) in the verification phase?  With her comments on the Bridge Street area, I'm not sure if Amanda meant that she hadn't even added the extension of North Street (a current ward boundary) as a must-hold line, or if she just hadn't been able to add the additional line I was suggesting - I hadn't communicated a decent place for that line to end on the end other than Bond Brook until just last week which I pretty much knew was too late for this phase of the project.

Also, can the voting districts in the Voting District Project phase of the Redistricting Data Program split the blocks that result from the Block Boundary Suggestion Project phase?  I had assumed they couldn't.  But Amanda's latest message to me suggests otherwise.  Perhaps the Census Bureau would just divide a block into two or more blocks in that case.

I don't think there's much chance the Augusta City Council will want to move the Ward 1-Ward 2 boundary west of Sewell Street, so I found the ward boundary justification for holding the existing boundaries along "Waddington Street" (while getting rid of that street as a street) a little odd.  It might not have mattered, though, since the boundary between the "southwest Sewall blocks" and the "big honking Howard Hill block" as they existed in 2010 is the current (legal) boundary between House Districts 85 and 86.  Amanda's must-hold line (lines probably) there might be helpful actually as they will draw attention to that area and when someone sees a must-hold line going through a condo (the Murdock condo), there could be some follow-up that could result in a better line.

I'm glad the outside of Gardencrest will all be in one block, although I'm not sure the golf course boundary was necessary.  I hadn't made it clear, in re-reading my e-mail where I first mentioned Gardencrest, that the three homes in the other block from the rest of the outside (not in the two loops) of the neighborhood were indeed on the other side of the power lines from the rest of the neighborhood.  Those power lines (if that was indeed what the Census Bureau thought they were following as a block boundary) were pretty badly misaligned further north, although it didn't look like any homes in Manchester were impacted.  In my reply, I did mention that her golf course boundary might result in Gardencrest (now all of the outside instead of missing those three homes) remaining in a separate block from homes further west where it might make sense for them to be in a separate census block.  But that depends on if you just go around the golf course or somehow connect with the prototype block line following (roughly) the power lines.

I was surprised at all of Maine's township lines becoming block lines.  That will result in a decent increase in the number of blocks (3 more apiece for most 4-corner township intersections probably).  I wonder if the fact that the person who seemed most upset at how the boundary between House Districts 150 and 151 was interpreted was a former longtime Speaker of the House (still quite influential, I believe) had anything to do with that?  I didn't think to ask Amanda if that included the Penobscot Indian Island Reservation, which includes islands in the Penobscot River from Old Town to Medway.  Department of Transportation and DeLorme maps (when DeLorme existed) have that area parceled off among the various towns on either side of the river.  While it probably would have worked numbers-wise to parcel off the part of the Penobscot Reservation north of what the DOT and DeLorme considered Old Town into other districts (several of which crossed the river and were thus arguably not contiguous the way the lines were drawn), as has been done in the past, the entire area of the reservation was kept in one district, which made me wonder if some federal law were at play that people finally noticed in 2013.

[jimrtex]

You must be logged in to read this quote.

You must be logged in to read this quote.

Does the power line cut a street?  You might want to check the .addr.dbf file.  In Hudson, some of the TO-FROM addresses were completely messed up in the 2010 Census edges file, but had been corrected in the partnership .addr file.

I don't know if this had been messed up during the 2010 census, or it had been caught at that time 2010, but not found its way into the census maps.

[jimrtex]

Can anyone verify (or refute) that a redistricting liaison can make additional changes (not just clarifications to something the liaison had made some note about but perhaps not in the right way, but something where nothing had been done in the initial phase) in the verification phase?  With her comments on the Bridge Street area, I'm not sure if Amanda meant that she hadn't even added the extension of North Street (a current ward boundary) as a must-hold line, or if she just hadn't been able to add the additional line I was suggesting - I hadn't communicated a decent place for that line to end on the end other than Bond Brook until just last week which I pretty much knew was too late for this phase of the project.

Also, can the voting districts in the Voting District Project phase of the Redistricting Data Program split the blocks that result from the Block Boundary Suggestion Project phase?  I had assumed they couldn't.  But Amanda's latest message to me suggests otherwise.  Perhaps the Census Bureau would just divide a block into two or more blocks in that case.

I don't think there's much chance the Augusta City Council will want to move the Ward 1-Ward 2 boundary west of Sewell Street, so I found the ward boundary justification for holding the existing boundaries along "Waddington Street" (while getting rid of that street as a street) a little odd.  It might not have mattered, though, since the boundary between the "southwest Sewall blocks" and the "big honking Howard Hill block" as they existed in 2010 is the current (legal) boundary between House Districts 85 and 86.  Amanda's must-hold line (lines probably) there might be helpful actually as they will draw attention to that area and when someone sees a must-hold line going through a condo (the Murdock condo), there could be some follow-up that could result in a better line.

I'm glad the outside of Gardencrest will all be in one block, although I'm not sure the golf course boundary was necessary.  I hadn't made it clear, in re-reading my e-mail where I first mentioned Gardencrest, that the three homes in the other block from the rest of the outside (not in the two loops) of the neighborhood were indeed on the other side of the power lines from the rest of the neighborhood.  Those power lines (if that was indeed what the Census Bureau thought they were following as a block boundary) were pretty badly misaligned further north, although it didn't look like any homes in Manchester were impacted.  In my reply, I did mention that her golf course boundary might result in Gardencrest (now all of the outside instead of missing those three homes) remaining in a separate block from homes further west where it might make sense for them to be in a separate census block.  But that depends on if you just go around the golf course or somehow connect with the prototype block line following (roughly) the power lines.

I was surprised at all of Maine's township lines becoming block lines.  That will result in a decent increase in the number of blocks (3 more apiece for most 4-corner township intersections probably).  I wonder if the fact that the person who seemed most upset at how the boundary between House Districts 150 and 151 was interpreted was a former longtime Speaker of the House (still quite influential, I believe) had anything to do with that?  I didn't think to ask Amanda if that included the Penobscot Indian Island Reservation, which includes islands in the Penobscot River from Old Town to Medway.  Department of Transportation and DeLorme maps (when DeLorme existed) have that area parceled off among the various towns on either side of the river.  While it probably would have worked numbers-wise to parcel off the part of the Penobscot Reservation north of what the DOT and DeLorme considered Old Town into other districts (several of which crossed the river and were thus arguably not contiguous the way the lines were drawn), as has been done in the past, the entire area of the reservation was kept in one district, which made me wonder if some federal law were at play that people finally noticed in 2013.

In previous censuses, the block boundary suggestion project and the VTD project were conducted at the same time. The states specifically requested that they be split.

2010 was the first time that VTD boundaries could split block boundaries - that the VTD boundaries could force splits of blocks on artificial lines. My understanding would be that you could identify non-traditional block boundaries such as streams, power lines, ridge lines that could be used as block boundaries during the first phase; but totally artificial lines such as are used in Hudson would have to wait to the second phase.

This presentation 2020 Census Redistricting Data Program (PDF), by the Census Bureau to the NCSL suggests that block changes can continue to be suggested throughout the process. Note that 2016 is pretty early to be finalizing geography for the 2020 census.

[Kevinstat]

You must be logged in to read this quote.

You must be logged in to read this quote.

Does the power line cut a street?  You might want to check the .addr.dbf file.  In Hudson, some of the TO-FROM addresses were completely messed up in the 2010 Census edges file, but had been corrected in the partnership .addr file.

I don't know if this had been messed up during the 2010 census, or it had been caught at that time 2010, but not found its way into the census maps.

If you zoom in really closely on the Google areal map you can see the trio of power lines.  It crosses US 202 (called "Route 202" in Maine if not by some street name, like Western Ave. in Augusta west of the west side rotary) between the two entrances to Fairway Overnite & Extended Stay, goes just west of the various buildings in the lot (I think the one in the back may be the owner's family home) and crosses over the corner in of the dead-end portion of Country Club Road (what Google Maps still has as both Country Club Rd and Joel Ave which used to be the name of that dead-end portion).  I've stood at that corner around noon and the power lines (at least the middle one) doesn't quite touch the center line even if you assume the center line doesn't curve where the inside of that corner curves (Google's satellites have it off a bit in the other direction from the Census Bureau), but it is closer to that center line than to the outside corner of the pavement.  If the power lines were aligned properly, there would probably have been an additional 2010 and 2020 prototype block in Gardencrest as Joel Ave from that corner to the intersection with Apple Tree Lane (it's never Apple Tree Drive or Appletree Drive) and the beginning of Woodridge Drive would divide the outside of Gardencrest (or the portion east of the power lines) on either side of the two interior blocks.  Both interior blocks would border both exterior blocks (east of the power lines) in that case, allowing for a snaking legislative district boundary (a very unlikely one but it would be possible).

Going north and west, the Census Bureau's line starts to go a bit east of the power lines, but then cuts hard to the west until it ends on the Manchester-Winthrop town line.  Another line begins just north of where the power lines cross into Wintrhrop, seemingly an attempt to follow that same set of power lines (it has the symbol for a pipeline or power line on 2010 census block maps; see http://www2.census.gov/geo/maps/dc10map/GUBlock/st23_me/county/c23011_kennebec/DC10BLK_C23011_014.pdf ).  But there isn't an approximation of the power lines between Case Road and (almost) Carlton Pond Road.  That may be because the Census Bureau's ... whatever they have that makes the lines acted as if the power lines stopped at Case Road and that a new set of power lines started a bit further west (see that same link above).  But given that the Census Bureau is trying to cut down on the number of census blocks, I would have thought those power lines (that weren't a block boundary for the 2000 census, at least around Gardencrest) would have made sense not to be block boundaries, as from reading what you have written is the norm (power lines not being block boundaries) elsewhere in the country.

I'm not sure how I'd go about checking the .dbf file (or even what to open it with).  You might have covered that in your recent tutorials (I haven't done anything since Part 7).  By .addr.dbf did you mean PVS_15_v2_addr_23011.dbf ?  I've been more focused on trying to get the lines right than where the Census Bureau counts people (improving the first ought to make the second easier and more intuitive, like if the eastern boundary of the Howard Hill property was a block line instead of "Waddington Street" to borrow from another of my areas of focus).

[Kevinstat]

Can anyone verify (or refute) that a redistricting liaison can make additional changes (not just clarifications to something the liaison had made some note about but perhaps not in the right way, but something where nothing had been done in the initial phase) in the verification phase?  With her comments on the Bridge Street area, I'm not sure if Amanda meant that she hadn't even added the extension of North Street (a current ward boundary) as a must-hold line, or if she just hadn't been able to add the additional line I was suggesting - I hadn't communicated a decent place for that line to end on the end other than Bond Brook until just last week which I pretty much knew was too late for this phase of the project.

Also, can the voting districts in the Voting District Project phase of the Redistricting Data Program split the blocks that result from the Block Boundary Suggestion Project phase?  I had assumed they couldn't.  But Amanda's latest message to me suggests otherwise.  Perhaps the Census Bureau would just divide a block into two or more blocks in that case.

I don't think there's much chance the Augusta City Council will want to move the Ward 1-Ward 2 boundary west of Sewell Street, so I found the ward boundary justification for holding the existing boundaries along "Waddington Street" (while getting rid of that street as a street) a little odd.  It might not have mattered, though, since the boundary between the "southwest Sewall blocks" and the "big honking Howard Hill block" as they existed in 2010 is the current (legal) boundary between House Districts 85 and 86.  Amanda's must-hold line (lines probably) there might be helpful actually as they will draw attention to that area and when someone sees a must-hold line going through a condo (the Murdock condo), there could be some follow-up that could result in a better line.

I'm glad the outside of Gardencrest will all be in one block, although I'm not sure the golf course boundary was necessary.  I hadn't made it clear, in re-reading my e-mail where I first mentioned Gardencrest, that the three homes in the other block from the rest of the outside (not in the two loops) of the neighborhood were indeed on the other side of the power lines from the rest of the neighborhood.  Those power lines (if that was indeed what the Census Bureau thought they were following as a block boundary) were pretty badly misaligned further north, although it didn't look like any homes in Manchester were impacted.  In my reply, I did mention that her golf course boundary might result in Gardencrest (now all of the outside instead of missing those three homes) remaining in a separate block from homes further west where it might make sense for them to be in a separate census block.  But that depends on if you just go around the golf course or somehow connect with the prototype block line following (roughly) the power lines.

I was surprised at all of Maine's township lines becoming block lines.  That will result in a decent increase in the number of blocks (3 more apiece for most 4-corner township intersections probably).  I wonder if the fact that the person who seemed most upset at how the boundary between House Districts 150 and 151 was interpreted was a former longtime Speaker of the House (still quite influential, I believe) had anything to do with that?  I didn't think to ask Amanda if that included the Penobscot Indian Island Reservation, which includes islands in the Penobscot River from Old Town to Medway.  Department of Transportation and DeLorme maps (when DeLorme existed) have that area parceled off among the various towns on either side of the river.  While it probably would have worked numbers-wise to parcel off the part of the Penobscot Reservation north of what the DOT and DeLorme considered Old Town into other districts (several of which crossed the river and were thus arguably not contiguous the way the lines were drawn), as has been done in the past, the entire area of the reservation was kept in one district, which made me wonder if some federal law were at play that people finally noticed in 2013.

In previous censuses, the block boundary suggestion project and the VTD project were conducted at the same time. The states specifically requested that they be split.

2010 was the first time that VTD boundaries could split block boundaries - that the VTD boundaries could force splits of blocks on artificial lines. My understanding would be that you could identify non-traditional block boundaries such as streams, power lines, ridge lines that could be used as block boundaries during the first phase; but totally artificial lines such as are used in Hudson would have to wait to the second phase.

This presentation 2020 Census Redistricting Data Program (PDF), by the Census Bureau to the NCSL suggests that block changes can continue to be suggested throughout the process. Note that 2016 is pretty early to be finalizing geography for the 2020 census.

Good point.  I've wondered in the past how changes like new roads and such late in the decade are factored in for the coming census, or even if they are.  I'm sure some things still happen too late to be factored in.  I mean, there's always something changing somewhere.

So perhaps the extension of North Street in Augusta (a current ward boundary) could be added during the Voting District Project as a line between voting districts and thus become a block boundary.  My idea for an additional line in that area probably wouldn't be allowable in that phase, though, but it still might be addable during the verification phase (or sub-phase) of the current Block Boundary Suggestion Project phase of the Redistricting Data Program.

[jimrtex]

You must be logged in to read this quote.

You must be logged in to read this quote.

Does the power line cut a street?  You might want to check the .addr.dbf file.  In Hudson, some of the TO-FROM addresses were completely messed up in the 2010 Census edges file, but had been corrected in the partnership .addr file.

I don't know if this had been messed up during the 2010 census, or it had been caught at that time 2010, but not found its way into the census maps.

If you zoom in really closely on the Google areal map you can see the trio of power lines.  It crosses US 202 (called "Route 202" in Maine if not by some street name, like Western Ave. in Augusta west of the west side rotary) between the two entrances to Fairway Overnite & Extended Stay, goes just west of the various buildings in the lot (I think the one in the back may be the owner's family home) and crosses over the corner in of the dead-end portion of Country Club Road (what Google Maps still has as both Country Club Rd and Joel Ave which used to be the name of that dead-end portion).  I've stood at that corner around noon and the power lines (at least the middle one) doesn't quite touch the center line even if you assume the center line doesn't curve where the inside of that corner curves (Google's satellites have it off a bit in the other direction from the Census Bureau), but it is closer to that center line than to the outside corner of the pavement.  If the power lines were aligned properly, there would probably have been an additional 2010 and 2020 prototype block in Gardencrest as Joel Ave from that corner to the intersection with Apple Tree Lane (it's never Apple Tree Drive or Appletree Drive) and the beginning of Woodridge Drive would divide the outside of Gardencrest (or the portion east of the power lines) on either side of the two interior blocks.  Both interior blocks would border both exterior blocks (east of the power lines) in that case, allowing for a snaking legislative district boundary (a very unlikely one but it would be possible).

Going north and west, the Census Bureau's line starts to go a bit east of the power lines, but then cuts hard to the west until it ends on the Manchester-Winthrop town line.  Another line begins just north of where the power lines cross into Wintrhrop, seemingly an attempt to follow that same set of power lines (it has the symbol for a pipeline or power line on 2010 census block maps; see http://www2.census.gov/geo/maps/dc10map/GUBlock/st23_me/county/c23011_kennebec/DC10BLK_C23011_014.pdf ).  But there isn't an approximation of the power lines between Case Road and (almost) Carlton Pond Road.  That may be because the Census Bureau's ... whatever they have that makes the lines acted as if the power lines stopped at Case Road and that a new set of power lines started a bit further west (see that same link above).  But given that the Census Bureau is trying to cut down on the number of census blocks, I would have thought those power lines (that weren't a block boundary for the 2000 census, at least around Gardencrest) would have made sense not to be block boundaries, as from reading what you have written is the norm (power lines not being block boundaries) elsewhere in the country.

I'm not sure how I'd go about checking the .dbf file (or even what to open it with).  You might have covered that in your recent tutorials (I haven't done anything since Part 7).  By .addr.dbf did you mean PVS_15_v2_addr_23011.dbf ?  I've been more focused on trying to get the lines right than where the Census Bureau counts people (improving the first ought to make the second easier and more intuitive, like if the eastern boundary of the Howard Hill property was a block line instead of "Waddington Street" to borrow from another of my areas of focus).

I'm having a hard time following you.

This is the area of concern?



I am displaying the 2010 census blocks and Partnership edges file. Is the problem that block 1035 (the green one) wraps around the other two blocks. Where are the power lines?  I see what looks like a power line cut through the woods on the south side of Western Avenue a bit to the east in block 2010.

[Kevinstat]

I'm having a hard time following you.

This is the area of concern?



I am displaying the 2010 census blocks and Partnership edges file. Is the problem that block 1035 (the green one) wraps around the other two blocks.

No, the problem is that three houses on Country Club Road (two on the "Joel Avenue" portion) are in a different block from the rest of the exterior of the neighborhood when the only feature separating those homes from the rest of the exterior are the power lines.  And the power lines do run between those three homes and all the others, but it's not a very obvious line and I can't be confident people would be counted in the right block.  The tax maps for Manchester don't have address numbers on them, and I'm not sure when Joel Avenue officially changed to part of Country Club Road, and I wasn't able to use the Census Geocoder to locate those homes (the lat/long figures didn't check out) and thus couldn't verify what census block they were considered part of.  Anyway, the three homes I'm talking about are the ones on lot 7A (handwritten, not circled) on U-13 and 6 and 7 (again, handwritten, not circled) on U-14 in the Manchester tax maps.  Those tax maps show the right of way for the power lines.

Where are the power lines?  I see what looks like a power line cut through the woods on the south side of Western Avenue a bit to the east in block 2010.

Zoom into where 2010 block 1037 almost touches block 1036.  Then unselect all your layers except for Google Satellite.  Then go down to the corner in the dead-end road.  Then zoom way in.  You should be able to see a set of three power lines going over that corner, although the Google Satellite view shows them just a dite to the right (east) of where they actually are vis a vis the road, although it's closer than the Census Bureau's satellite apparently.

I just realized that the Google error is because the satellite or whatever is viewing that area from the east, and so the height of the power line poles causes the power lines to show further east than stuff at ground level.  Below that corner you can see where the shadow of a power line pole is and where the bottom of that shadow hits the bottom of the actual pole.  The next pole to the north is similar, but without the same shadow.  But if you drew a line between the bottoms of those poles you'd have a decent idea of where the power lines crossed.  You could also use a street view in Google maps or some other map site that has a street view of that area.

I see what looks like a power line cut through the woods on the south side of Western Avenue a bit to the east in block 2010.

The boundary line between blocks 2005, 2010 and 2012 (that small blue block west of block 2010) on the one side, and 2017, 2024 and 2013 on the other is almost certainly supposed to be those power lines.  Going roughly north-northwest, they turn west about where the block boundary turns, and the same when it turns again to go north across Route 202.

Compare block maps for that area in 1990, 2000 and 2010.  I think the 2000 setup is the best for Gardencrest (the inner loops might be useful to keep in separate blocks for enumeration purposes, even though they couldn't be drawn in a different district than the block that surrounds them).

[jimrtex]

You can display the direction of the lines in QGIS:

Right Click on Layer, Select Properties, Select Style.

In the second box you should see:

> ___ Line.

Click on the > to expand:

> ___ Line.

    ___ Simple Line

Click on the Green +

> ___ Line
   ___ Simple Line
   ___ Simple Line

Select one of the simple lines and select Marker Line (upper right)

Select Simple Marker, and then choose the > symbol

Apply.

To get back to the normal simple line, select the marker line and the red "-".

Select the lines corresponding to the three streets (Country Club Rd, Joel Ave, and Apple Tree Ln).

Make note of the TLID.

You can drag the addr.dbf file into the canvas. Make it the current layer. Select by Expression:

"TLID" = '75481176'  for example is Country Club Rd

Vermont and bring the selected features to the top. There should be four for Country Club Road. See the fields SIDE, FROMHN, and TOHN

Side is either L or R, for the left or right side of the road.

Right (east) is 53-79
Left (west) is 8-48, 56-60, 37-51

I don't see any addresses on the tax maps.

I suspect the three houses are enumerated based on the street edge rather than the location of the building.

What is the population and housing county for Block 1037?

A better boundary might well be the Augusta Country Club.

[Kevinstat]

What is the population and housing [count] for Block 1037?

2000 Census counts:
Block 1022 (split into blocks 1035, 1037 and 1039 for 2010): 210 people; 84 housing units
Block 1026 (became Block 1036 in 2010): 29 people; 10 housing units
Block 1027 (became Block 1040 in 2010): 61 people; 22 housing units

2010 Census counts:
Block 1035: 214 people; 84 housing units
Block 1036: 27 people; 9 housing units
Block 1037: 0 people; 0 housing units
Block 1039: 0 people; 0 housing units
Block 1040: 42 people; 20 housing units

So yeah, I imagine you're right about what block those three homes in Gardencrest west of the power lines and the people in them were counted in.  The zero population and housing block count for the two blocks in Manchester west of the power lines and north of Route 202 suggests that no line is needed separating that area from Block 1035.  1035's a sprawling block, kind of like Block 2009 in Tract 104 in Augusta, but the separation of area west of the power lines doesn't seem to have helped matters.

The population loss (and the loss of two housing units) in the longer and narrower of the two interior blocks (1027 in '00; 1040 in '10) is surprising.  One or both of the 2000 or 2010 counts was probably inaccurate is what I'm thinking.

[jimrtex]

What is the population and housing [count] for Block 1037?

2000 Census counts:
Block 1022 (split into blocks 1035, 1037 and 1039 for 2010): 210 people; 84 housing units
Block 1026 (became Block 1036 in 2010): 29 people; 10 housing units
Block 1027 (became Block 1040 in 2010): 61 people; 22 housing units

2010 Census counts:
Block 1035: 214 people; 84 housing units
Block 1036: 27 people; 9 housing units
Block 1037: 0 people; 0 housing units
Block 1039: 0 people; 0 housing units
Block 1040: 42 people; 20 housing units

So yeah, I imagine you're right about what block those three homes in Gardencrest west of the power lines and the people in them were counted in.  The zero population and housing block count for the two blocks in Manchester west of the power lines and north of Route 202 suggests that no line is needed separating that area from Block 1035.  1035's a sprawling block, kind of like Block 2009 in Tract 104 in Augusta, but the separation of area west of the power lines doesn't seem to have helped matters.

The population loss (and the loss of two housing units) in the longer and narrower of the two interior blocks (1027 in '00; 1040 in '10) is surprising.  One or both of the 2000 or 2010 counts was probably inaccurate is what I'm thinking.

The household size for 2000 is kind of high. What was the age breakdown? You can have a sharp decline about 25-30 years after an area is developed, as the children move out.

I count 22 houses on the block, but maybe some are new/replacement?

[Kevinstat]

The household size for 2000 is kind of high. What was the age breakdown? You can have a sharp decline about 25-30 years after an area is developed, as the children move out.

I count 22 houses on the block, but maybe some are new/replacement?

Block 1027, 2000 Census:
Total:         61
Under 5 years      0
5 to 9 years      3
10 to 14 years      15
15 to 17 years      2
18 and 19 years      0
20 years         0
21 years         0
22 to 24 years      0
25 to 29 years      2
30 to 34 years      1
35 to 39 years      5
40 to 44 years      8
45 to 49 years      6
50 to 54 years      8
55 to 59 years      3
60 and 61 years      1
62 to 64 years      1
65 and 66 years      0
67 to 69 years      0
70 to 74 years      1
75 to 79 years      4
80 to 84 years      0
85 years and over   1

Block 1040, 2010 Census:
Total:         42
Under 5 years      1
5 to 9 years      2
10 to 14 years      0
15 to 17 years      1
18 and 19 years      0
20 years         0
21 years         0
22 to 24 years      2
25 to 29 years      0
30 to 34 years      0
35 to 39 years      2
40 to 44 years      3
45 to 49 years      2
50 to 54 years      6
55 to 59 years      7
60 and 61 years      1
62 to 64 years      5
65 and 66 years      1
67 to 69 years      1
70 to 74 years      3
75 to 79 years      2
80 to 84 years      2
85 years and over   1

I'm not aware of any households being built or demolished in that block since my family moved to the area in 1994, and they've all been single-unit houses throughout that time as far as I could tell.  There's one place kind of back in the woods on Sylvan Way that might not have always been there, but that would be going in the opposite direction from explaining the 2-unit loss from 2000 to 2010.  The other interior block had one place built in that decade (on Sylvan Way, next to the one on the corner with Apple Tree Lane), but no losses as far as I could tell.  Again, that's the opposite direction from the difference shown in the figures.

[Kevinstat]

From a post I just made in Nhoj's "City and Town estimates" thread:

From an e-mail I sent to a friend who shares my interest in redistricting (although his interest is perhaps not quite as strong as mine):

You must be logged in to read this quote.

I later clarified my comments on Augusta with the following e-mail:

You must be logged in to read this quote.

[jimrtex]

The household size for 2000 is kind of high. What was the age breakdown? You can have a sharp decline about 25-30 years after an area is developed, as the children move out.

I count 22 houses on the block, but maybe some are new/replacement?

Block 1027, 2000 Census:
Total:         61
Under 5 years      0
5 to 9 years      3
10 to 14 years      15
15 to 17 years      2
18 and 19 years      0
20 years         0
21 years         0
22 to 24 years      0
25 to 29 years      2
30 to 34 years      1
35 to 39 years      5
40 to 44 years      8
45 to 49 years      6
50 to 54 years      8
55 to 59 years      3
60 and 61 years      1
62 to 64 years      1
65 and 66 years      0
67 to 69 years      0
70 to 74 years      1
75 to 79 years      4
80 to 84 years      0
85 years and over   1

Block 1040, 2010 Census:
Total:         42
Under 5 years      1
5 to 9 years      2
10 to 14 years      0
15 to 17 years      1
18 and 19 years      0
20 years         0
21 years         0
22 to 24 years      2
25 to 29 years      0
30 to 34 years      0
35 to 39 years      2
40 to 44 years      3
45 to 49 years      2
50 to 54 years      6
55 to 59 years      7
60 and 61 years      1
62 to 64 years      5
65 and 66 years      1
67 to 69 years      1
70 to 74 years      3
75 to 79 years      2
80 to 84 years      2
85 years and over   1

I'm not aware of any households being built or demolished in that block since my family moved to the area in 1994, and they've all been single-unit houses throughout that time as far as I could tell.  There's one place kind of back in the woods on Sylvan Way that might not have always been there, but that would be going in the opposite direction from explaining the 2-unit loss from 2000 to 2010.  The other interior block had one place built in that decade (on Sylvan Way, next to the one on the corner with Apple Tree Lane), but no losses as far as I could tell.  Again, that's the opposite direction from the difference shown in the figures.

If you group by 5-year cohorts and offset by 10 years between censuses you see:

-     1
-     2
0    0
3    1
15  2
2    0
0    0
2    2
1    3
5    2
8    6
6    7
8    6
3    2
2    3
1    2
1    2
4    1
0    -
1    -

In 2000, you had 15 children between 10 and 14. In 2010, they would be between 20 and 24. The Census Bureau counts college students based on their place of residence on April 1. So some of the middle school gangs of 2000 might still have an attachment to their parents home in 2010, but would not be counted there.

In 2000, 22 of the 40 persons over 30 were between 40 and 54 (they were the parents of those tweens and early teens).

In 2010, 19 of the 36 persons over 30 were between 50 and 64. The same people still lived in the same houses they lived 10 years earlier.

There are a bit less than two adults per house, but that can be explained by widows and divorced. As time goes on, these people will gradually move out. They don't need such a large house except when the grandkids visit. And simply maintaining the house and yard becomes a burden when people enter their 60s and 70s.

You could see the same phenomena on a large scale in the 1980 Census for Nassau County, New York. There was a massive population boom to Long Island after WWII, with people moving out of overcrowded New York City and had boomer-sized families.

1950 +65.4%
1960 +93.3%
1970 +9.8%
1980 -7.5%

People who bought a house at age 23 in 1955 (they needed a larger house as their second-born was on the way), were 48 in 1980. Their 4 children who had swelled the 1960 census, were now 22 to 29. That was a 67% decline for that one house.

Only gradually over time has the population increased above its 1960 level. Since families (and households) are smaller, there has been an infill of apartments.

[jimrtex]

Massachusetts requires cities and towns to redistrict prior to legislative redistricting. This is true even for the many towns and cities that have a town meeting or representative town meeting form of government, and have no reason for equal-sized wards. And larger cities really don't need to have equal-sized precincts within a ward. But Massachusetts requires it. If a town or city fails to redistrict, the Commonwealth will do it for them.

The maximum size for a precinct is 4,000, which is 9.5% of the quota for a house seat.

When you are assembling house districts from precincts of this size, you can always be within 5% of the target without busting. If you are anywhere from about 88% to 96% of the quota, you can add another precinct to get you within range.

Massachusetts does not split any precincts in its house or senate districts. If it divides a town, it will be based on precinct boundaries.

They do split the minimum number of precincts for congressional districts, but that is because of their misguided belief that congressional districts have to have identical populations.

If Maine were to follow the same principles, it would need to pre-define electoral areas of less than 900 persons, which is likely too small to be of electoral use, but would at least permit towns to define the division lines which follow natural communities.

Even better would be to switch to weighted voting. Augusta could be split into two districts at the river, and never have to bother with changing districts. The voting weights would be updated each census.

[jimrtex]

From a post I just made in Nhoj's "City and Town estimates" thread:

From an e-mail I sent to a friend who shares my interest in redistricting (although his interest is perhaps not quite as strong as mine):

You must be logged in to read this quote.

I later clarified my comments on Augusta with the following e-mail:

You must be logged in to read this quote.

The latest messages in Hudson QGIS explain how to present your table graphically.

[Kevinstat]

Dear Ms. Rector,

The 2018 prototype blocks came out sometime within the last week and I noticed that and looked at those in a few counties (including Kennebec of course) yesterday.  I was pleased to see that there is indeed now a line going from what could be considered the end of North Street in the Augusta Public Works property to Bond Brook, separating homes in the Boothby Street neighborhood, the east side of North Street, North Street Place and parts of Bridge and State Streets in that same block (all of which are currently in Ward 3) from those on the west side of North Street (north of Bennett Street), various parts of Bennett, Prospect, High and Granite and Winthrop Streets and homes as far west as the east side of Fuller Road behind Northeast Bank (all of which are currently in Ward 1).  While the new line (which connects from an existing node which seems to correspond with one of the intersections of roads within Augusta Public Works to the intersection of a set of power lines and Bond Brook) doesn't quite match the ward line on the city's maps, the affected area is fairly small (much smaller than even in the 2010 census blocks which itself was smaller than in previous censuses) and it's obvious that there are no residences in that territory, and will not be any residences in that territory by April 2020.  Therefore people won't have to guess at what the 2020 census populations of Augusta's wards are (at least not because of that issue, and I haven't identified any others and I think I would have if there had been any), or of what the populations of proposed new wards 1 and 3 are if the North Street/North Street extension boundary is retained (which seems likely from the 2014 experience).

Thank you also for getting the odd jog in Capitol Street that showed up in last year's prototype blocks fixed.  And for the earlier fix in Gardencrest in Manchester (my parents live in that neighborhood and I used to take misdelivered mail to a home on the other side of some power lines that the fix brought into the same block as the rest of the outside of the neighborhood) and getting township lines in Aroostook County (most importantly along the Route 11 corridor) added as block boundaries.

I know your time as the "Governor's liaison" to the U.S. Census Bureau, and as Maine's State Economist) may (or may not) be coming to a close.  I wanted to thank you for your attention to and action on my suggestions and wish you all the best in the future.  Thanks also to the Augusta City Manager and the Director of Development Services (both of whom I've Cc:ed in this e-mail) for any helpful information they provided in your getting a block line added that approximates the current 1-3 ward boundary.

Sincerely,

Kevin Lamoreau
Augusta

Thanks for the note, Kevin – I’m happy to have been of assistance.
 
Best,
Amanda
 
 
Amanda Rector
Maine State Economist
Department of Administrative and Financial Services
78 State House Station
Augusta, Maine 04333-0078
[State e-mail and phone number]

[Kevinstat]

Hello again, Amanda.

The partnership shapefiles for the Voting District Project Verification Phase (2018_V2) were added online (at https://www.census.gov/geo/maps-data/data/partnership.html ) sometime since this past weekend (the Census Bureau's web site doesn't say on the page describing the Redistricting Data Program for the 2020 census ( https://www.census.gov/programs-surveys/decennial-census/about/rdo/program-management.html ) that those files or the 2018 prototype blocks are there, but I've been clicking the links under "VTD Geographic Files" when I check for updates).

I added the layer PVS_18_v2_vtd_23011.shp to a project I have on my QGIS (which GUPS is apparently based off), and overall it looked very good (although I can't vouch for the Waterville ward boundaries and haven't bothered to try and verify that, except that I know Waterville has 7 wards rather than 8 and it looks like a (likely unpopulated) eastern section of one of them is cut off from the rest of that ward by the area between some railroad tracks and Colby Street near the intersection of Front Street and College Ave.).  I've noticed one error in Augusta, though.  The boundary between Wards 1 and 3 which I checked first looks good (following that added line connecting the roads around Augusta Public Works with Bond Brook).  But along the boundary between Wards 1 and 2, the area bounded by (clockwise starting from the north) Hitchborn, State, Capitol and Chamberlain streets is shown as being in Ward 2 while it's actually in Ward 1 (see http://www.augustamaine.gov/2014-03-24%202010%20Census%20Ward%20Boundaries%20-%20Approved.pdf ).  The most recognizable landmark there would be the Blaine House.  That area was in Ward 2 before the 2014 redraw*, but so was the area bounded by Hitchborn, Chamberlain, Memorial Circle and State (the Preti Flaherty/Verrill Dana building) and it seems the "triangle" formed by Memorial Circle and the entrance/exit thereto/therefrom from/to State Street, and those areas are both correctly shown as being in Ward 1.  The block boundaries in that area are good, so if necessary one could subtract the population of the two misplaced blocks (one on either side of Higgins Street) from that of the Voting District corresponding otherwise to Ward 2 and add it to the population of the Voting district otherwise corresponding to Ward 1, but it seems like this error is one that could be fixed in the VTD Verification Phase (being the kind of thing that phase was designed for).

I've Cc:ed the Clerks of Augusta and Waterville on this e-mail as they could perhaps verify what I've said (or correct me if I'm wrong), plus James Whitehorne as I pointed out that the availability of the latest prototype block and partnership shapefile data isn't yet indicated on one key page.  Feel free, everyone to forward this message to whomever you think it ought to be forwarded to.

Thanks again for all your efforts,

Kevin

*Potentially fun fact: In 2012 the Blaine House was in the "southeastern Augusta" ward (Ward 2) but in the main "western Augusta" House District (then House District 57).  By the time of the November 2014 General Election (and still today), the Blaine House was in the "southwestern Augusta" ward (Ward 1) but the main "eastern Augusta" House District (House District 85).

The Blaine House is the Governor's mansion in Maine.  In the 2010 census, it was in Block 1019 of Kennebec County Census Tract 104, which had a 2010 Census population of 2 (probably John and Karen Baldacci, although there's a building just north of the Blaine House that I'm not sure is on the premises (but it might be, like some kind of museam) and the Wrapped Up sandwich shop is in that same census block and appears to have a residence above it, but it could be more recent than 2010 or been vacant at that time).  On the other side of Higgins Street is Block 1029 which, while fairly small in area, has at least 11 buildings in it (one of them has been the Maine Republican Party headquarters since sometime in the late 2000s (decade)) and a 2010 Census population of 38.  Blocks 1019 and 1029 together comprise the territory which has been placed in the voting district corresponding to the wrong ward in the "2018_V2" Partnership Shapefiles.  The PretiFlaherty building was in Block 1020 and the triangle on the south side of Memorial Circle was Block 1022 (both had a 2010 Census Population of 0).  The four census blocks I just mentioned (all in Tract 104) comprise the territory that was moved from Ward 2 to Ward 1 in 2014.

What I had thought might have been an error in Waterville now appears to just be the case of an odd ward boundary (see http://www.waterville-me.gov/wardmap.pdf and note the narrow southern extension of the Ward 1 (in purple), which is between a loop over College Ave. that is part of a connector between Front Street and Main Street... it's between that loop and a set of railroad tracks, cutting off a likely unpopulated eastern sestion of Ward 2 (in blue) from the rest of that ward).  I sent Amanda Rector a follow-up e-mail last Sunday (mentioning that things seemed okay with Waterville but restating what I had noticed in Augusta), but haven't heard back yet.

[Kevinstat]

Dear Ms. Rector and Ms. Packard,

Maps that tell a story:

2000 census block maps for Farmington, Maine: https://www2.census.gov/geo/maps/blk2000/st23_Maine/CountySubdivision/2324775_Farmington/CBS2324775_001.pdf (lower right-hand corner) and https://www2.census.gov/geo/maps/blk2000/st23_Maine/CountySubdivision/2324775_Farmington/CBS2324775_003.pdf (upper right-hand corner)

2010 census block maps for Farmington: https://www2.census.gov/geo/maps/dc10map/GUBlock/st23_me/cousub/cs2300724775_farmington/DC10BLK_CS2300724775_001.pdf (lower right-hand corner) and https://www2.census.gov/geo/maps/dc10map/GUBlock/st23_me/cousub/cs2300724775_farmington/DC10BLK_CS2300724775_003.pdf (upper right-hand corner)

2010's Block 2040 (of Census Tract 9712) straddles the Sandy River (it looks like a dam or something resulted in no visible water just south of a reservoir or something).  In 2000, what became Block 2040 in 2010 was contained in Blocks 4043 (west of the Sandy River) 4030 (east of the Sandy River, north of "Beaver Brook" and south of Sunny Hill Drive and it's extension), 4042 (east of the Sandy River and south of "Beaver Brook" along Route 4) and most of Block 4025 (north of Sunny Hill Drive, also including what in 2010 became Block 2029 north of Frederick Lane).  The reservoir (or whatever it is) was its own block in both 2010 and 2010, as was a cul de sac at the end of Sunny Hill Drive.

I noticed this while beginning to try and verify population figures for proposed Franklin County Commissioner districts under LD 940, which would increase the number of Franklin County Commissioners to 5 and divide Farmington (which has more than 1/5 of Franklin County's population) along the Sandy River.  I've Cc:ed the sponsor and co-sponsors of LD 940 on this e-mail.  I'm e-mailing you, Ms. Packard, as you testified on the bill (NFNA) on behalf of the Department of the Secretary of State.  I've downloaded edge and face shapefiles in my QGIS and unfortunately there isn't even a face boundary separating the two sides (of the Sandy River) of 2010's Block 2040.  A couple new blocks are carved out of 2010's Block 2040 in the southeast, but there are still seeming residences in the remaining block on both sides of the river.  The reservoir and the cul de sac are proposed to be part of that same block in 2020 (which makes perfect sense for the cul de sac which probably has a population of 0), but they are still individual faces.  And a recognized edge boundary along the river going upstream from the bridge along Routes 2 and 4 comes pretty close to the reservoir.  So if the edge following the river upstream from the bridge along Routes 2 and 4 were just extended along the river bed to the reservoir, which could become its own block again, you could separate the part of 2010's Block 2040 west of the river from the remainder of that block (or from the remainder of that portion that hasn't been carved out already).

I don't know what LD 940's fate will be, but given that there is clearly some interest in expanding the size of the Franklin County Board of Commissioners to 5 which would likely result in a split of Farmington with the Sandy River being a decent candidate for part or all of the in-town boundary, it would make sense to get rid of the one instance where a census block crosses the river.

Thanks,

Kevin Lamoreau
Augusta, Maine

You can view information on LD 940 at http://legislature.maine.gov/LawMakerWeb/summary.asp?ID=280071999 .

[Kevinstat]

Boredom-prompted bump.

[Kevinstat]

Hello again, Ms. Rector,

I just thought I'd bump this e-mail again since we're in the last month of the second-to-last review period of the 2020 Census Redistricting Data Program.  I know, on December 16 I sent you an e-mail thanking you for all you had done and perhaps suggesting I might be done making requests, and then three days later when the latest batch of Partnership Shapefiles came out I started bugging you again.  I'm Cc:ing Melissa [O'Neal (I had misspelled her name as O'Neil)], Director of Boards and Commissions with the Governor's Office, just to make sure there hasn't been some change or vacancy (hopefully not the latter) in the position of Maine's liaison to the Census Bureau's Redistricting Data Program that I wasn't aware of.  I'll be bumping (and forwarding to Melissa [O'Neal]) my e-mail about the census block straddling the Sandy River in Farmington once I send this e-mail.

Thanks again for all you have done with this process, including getting census block boundaries to align with township boundaries in the unorganized territories (most importantly along the Route 11 corridor in Aroostook County) and the Ward 1-3 boundary along North Street and it's extension in Augusta.

Sincerely,

Kevin Lamoreau
[Address, phone number and e-mail]

Hi Kevin,
 
We’re working with the Maine Office of GIS on verification this time around – your suggestions are in the queue, we just haven’t gotten to those counties yet. I’ll follow up with you once I’ve heard from MEGIS.
 
Best,
Amanda

Amanda Rector
Maine State Economist
Department of Administrative and Financial Services
78 State House Station
Augusta, Maine 04333-0078
[State e-mail and phone number]

Hi Amanda,

Okay, cool, thanks.  That's neat that you're working with the GIS Office on the census geography verification.  They'll be well positioned to catch things that no one else might notice because of all the data they already have.

I look forward to hearing from you.

Best,

Kevin