Saturday, April 20, 2019

New England Voting by Town

The map comes to me via Barry McMillion, who deserves the credit. Bsking sent it along.


To those outside of New England, this might be interesting only to note that there are strongly red sections even in very blue states. Even I, who am a geography nut, have to strain at some of these sections to figure out which towns are intriguing exceptions in their areas. The yellows tend to be wealthy towns. It also pays to remember that "rural" in New England does not automatically suggest farming - it sometimes just means "almost empty." So I wouldn't work too hard to get your head around what is going on in the minds of those pale and dark purple towns in Maine.  They are 50-50 red/blue to begin with, so small movements in the electorate can flip them. They have few people, and include much rural poverty, folks depending on five side businesses and food stamps to get by. You can also see why we in NH consider the towns along the Connecticut River to be practically part of Vermont, and why I call them the West Coast of NH. No surfing or beaches though.

Many of the darker blues and reds have more population, and illustrate the continuity of voting for the Democrat even if they hated Hillary (as many in Vermont do after her treatment of Bernie) or voting for the Republican even if they hated Trump.  Donald captured an interesting group that other candidates might not have, but most of his votes still came from people whose families have voted Republican since Harding.

As for those who switched, I am intrigued by that dark purple stretch that covers Eastern Connecticut (fancy prep schools there), Western Rhode Island, and Central Massachusetts. It is not broken up by blues that much or yellows at all, but by red, even creeping over the border into NH.  I don't have a good sense for who those people are, but they look like a consistent demographic a candidate might cultivate. (For those who like such things, there is a similar cluster at the borders of Illinois, Iowa, Minnesota, and Wisconsin that was usually blue but not deeply so.  If a candidate could figure out what they have in common and win them over, s/he could make deep inroads in four states. Trump sort of did that, and it was key to 2016.  Holding rallies in that 100-mile radius wouldn't hurt him going forward, and I have to think some Democratic strategist (they have some very good ones) has figured this out for later, though it's pretty useless in the primaries.)

Interesting patterns you might notice are welcome.

5 comments:

Christopher B said...

...there is a similar cluster at the borders of Illinois, Iowa, Minnesota, and Wisconsin that was usually blue but not deeply so.

Those tend to be the strongly Irish Catholic working-class industrial towns in an area otherwise populated by German and Scandinavian Protestant farmers.

RichardJohnson said...

Hillary's percent of the CT vote was about 4% less than what Obama won in 2012. (Most of the dropoff in Hillary's vote went not to Trump, but to Independents). There were, however, some towns where Hillary won a higher percentage of the vote than Obama.

Consider the five biggest outliers, where Hiilary won a higher percentage of the vote than Obama. For example, Hillary won 52.9% of the vote in Darien compared to Obama's 34.4%.

Darien 18.5%
New Canaan 17.9%
Weston 13.5%
Wilton 13.1%
Greenwich 13.1%

These are very wealthy towns. Fairfield County. Gold Coast.
Darien (6/6) $105,846
New Canaan (3/3) $105,846
Weston (2/2) $92,794
Wilton (3/3) $78,131
Greenwich (12/12) $90,087

Rank, Per Capita Income
2 Darien (6/6)
3 New Canaan (3/3)
5 Weston (2/2)
7 Wilton (3/3)
6 Greenwich (12/12)
These are for 2010, when CT's per capita income was $36,775.

Consider the 5 towns that had the biggest drop in Hillary versus Obama

Chaplin -16.6%
Killingly -17.7%
Voluntown -18.2%
Sterling -18.4%
Plainfield -18.6%

Consider their per capita income.
Chaplin $32,188
Killingly $25,215
Voluntown $32,760
Sterling $25,557
Plainfield $24,825

If you run a correlation of per capita income with increase in % Democrat vote from Obama to Hillary, you get .758, a rather high correlation.

JMSmith said...

Most of my colleagues in the geography department put their faith in spatial statistics. I prefer squinting at the map. Squinting at this map, I see three zones. Greater Boston, the Old Industrial Fringe, and the Woods. The Green Mountains and Berkshires are the heart of the Woods, and here the population is low enough to be tipped by various sorts of expatriates from Greater Boston. These places relate to Boston like Santa Fe, N.M. relates to Los Angeles. The Old Industrial Fringe is outside the commuting range of Greater Boston, but is not yet sufficiently hilly and wooded to attract the expatriates.

It would be interesting to compare this map with a map of perceived desirability of residence. My guess is that most New Englanders would prefer Greater Boston or the Woods, and so this is where we find them. Not exclusively, of course, but in sufficient numbers to tip the vote.

RichardJohnson said...

Massachusetts Stereotoypes. Dragons switched.

Assistant Village Idiot said...

@ Richard - I had forgotten that. The answer was nearby, but I was looking elsewhere. A great reminder! I remembered how the towns were different, but still could quite put my finger on the voting. 25% of my ancestry comes from Fitchburg/Leominster/Shirley, but that grandmother died the year before I was born, so I have little sense of the place. I have been through many times.

@ JM Smith. My brother lives in Hippie Farmers and works in Hippie Students. He has fully embraced this. When contemplating The Woods, while we think of retired preppies hiking in the Berkshires, the concentration of colleges - two of the Seven Sisters, two Little Ivies, plus large UMass-Amherst and a bunch of other schools and some fancy prep schools, drives the culture enormously. It may be more liberal than Boston, even though the mill towns of Springfield, Chicopee, and Holyoke are there.