U.S. CitiesCity-Data Forum Index
Go Back   City-Data Forum > U.S. Forums > General U.S.
 [Register]
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
View detailed profile (Advanced) or search
site with Google Custom Search

Search Forums  (Advanced)
 
Old 05-05-2021, 11:46 PM
 
452 posts, read 302,230 times
Reputation: 750

Advertisements

Since the estimated release was a bust and we'll have to wait until at least August before the official numbers are released at the county level, and knowing that some areas are sure to be undercounted and some appear that they may be overcounted based off the official 2020 state releases, figured I would start a thread to discuss which areas may be in one of the two categories.

Regardless of whether it ends up having merit or not, I would like the predictions to be based on some sort of tangible numbers. For example, all the numbers I will cite come from the US Census Bureau, Bureau of Labor Statistics, Bureau of Economic Analysis and Ohio Department of Education.

For right now, I'm interested in the three big Ohio counties, so that's what I've looked into. But I'm sure there are other numbers people out there who are looking into other areas and not sure where to put it. Let's make this the post to discuss those until the official numbers come out and see where this goes. Personally, I'm interesting in reading some of that data, even if it's not from an area I will put time into looking into.

For the three "Cs" ... Cleveland, Columbus, Cincinnati, from what I've compiled leads me to believe that Cuyahoga could be a surprise when the numbers are released (we already know Ohio is roughly 110,000 undercounted, though that doesn't mean that certain counties can collectively surpass that number if others were overcounted).

If I had to take a somewhat educated guess, on a county level, I think Cuyahoga (Cleveland) will see the biggest positive change. Hamilton (Cincinnati) looks like it is undercounted as well. Franklin (Columbus) is tough, may be about accurate, but I'm not seeing how there is going to be a big upward correction and wouldn't be surprised to see it go a little down.

On the metro level, however, while I haven't ran outlying county numbers like the three core ones, I would say that Cincinnati, and SW Ohio in general, is going to come out way ahead. In Columbus, which is no doubt still growing, any additional gains will be due to surrounding counties (Delaware/Licking/Fairfield). In Cleveland, the gains in Cuyahoga may be offset by overcounts in outlying counties (though Lake County may also be in the undercounted category, but don't have the data to put that out there).

I'll try to make this as easy to read as possible, looking at different numbers.

Here are the most recent estimated populations (and while the recently released numbers don't make up the difference in the official state undercount, there was some small corrections).

Change between 2010-2015 (first half of the decade):
1. Franklin (Columbus): 76,974 (6.52 percent)
2. Hamilton (Cincinnati): 9,156 (1.14 percent)
3. Cuyahoga (Cleveland): -11,315 (-.89 percent)

Changes between 2015-2020 (second half of the decade):
1. Franklin: 49,412 (3.87 percent)
2. Hamilton: 6,219 (.77 percent)
3. Cuyahoga: -26,441 (-2.11 percent)

Total change 2010-2020:
1. Franklin: 126,386 (10.39 percent)
2. Hamilton: 15,375 (1.91 percent)
3. Cuyahoga: -37,756 (-3.00 percent)

The first half look like they are reasonable, though it's hard to see, based off jobs, how Cuyahoga's decline accelerated in the second half of the decade. While county GDP only has been tracked since 2016, job gains/losses on the county level can be found.

2. Here is employment gains for each between 2010-2015:
1. Franklin: 74,154 (11.46 percent)
2. Cuyahoga: 28,747 (4.19 percent)
3. Hamilton: 22,322 (4.62 percent)

Franklin's first half gains were legit. It was head and shoulders above the other two. For the second half (2015-2019 ... 2020 county numbers are not available, so one pandemic year short)

1. Franklin: 26,870 (3.64 percent)
2. Cuyahoga: 13,070 (1.82 percent)
3. Hamilton: 11,732 (2.30 percent)

Franklin still leads and its estimated population increase did come down, it still doesn't necessarily add up, though. Population over Hamilton is estimated to have increased by roughly 3 percent over the second half of the decade over Hamilton, while job growth was only 1.3 percent higher; population gain for Franklin over Cuyahoga increased by 6 percent despite a less than 2 percent job increase.

How does that compare with GDP? As mentioned, GDP at the county level is not available for the first half of the decade, but it is for the second half. In that period, here is total GDP growth (actually this is 2016-19).

1. Cuyahoga: 4.901 billion (5.90 percent)
2. Hamilton: 3.985 billion (6.01 percent)
3. Franklin: 3.62 billion (3.86 percent)

Franklin in population, just in the second half of the decade, has still gained 75,853 in total population despite it falling behind Cuyahoga in total GDP output by 1.3 billion, and losing its grip on job gains. Same with Hamilton, which has outpaced Franklin by more than 300 million in GDP in that period, but fell short in population gain by 43,193.

This is more flimsly, but the Ohio Department of Education has databases that shows yearly public/charter school enrollment based on headcounts each October. I say flimsly because this does not include private schools, which in Ohio is mainly Catholic. And Cleveland and Cincinnati have way higher number of students attending Catholic schools in those counties (and a lot do come in from other counties, and maybe Catholic school population is decreasing more rapidly). At the same time, Columbus still does have 4 Catholic HS, and none of the four have seen student increases in terms of which level they compete against other HS's in Ohio.

Public/Charter school change 2010-2015:
1. Franklin: 15,367 (7.96 percent)
2. Hamilton: -6,369 (-5.54 percent)
3. Cuyahoga: -10,526 (-5.81 percent)

Public/Charter school change 2016-2020:
1. Hamilton: -523 (-.48 percent)
2. Cuyahoga: -9,978 (-5.90 percent)
3. Franklin: -15,789 (-7.48 percent)

You would think that even if public school enrollment isn't the best metric (and in fairness small parts of the city of Columbus are in growing districts located Delaware and Fairfield counties) the Franklin portion would still be growing. Instead, it has seen the biggest numerical and percent decrease over the past five years of the three. ... though was the only that did increase in the first half. Hamilton has seen a pretty significant increase and even Cuyahoga has seen a small increase compared to 2015.

We'll see, but at the core, doesn't seem like Columbus/Franklin is still chugging along in comparison to its two peers and it's probably a lot more close, especially the last five years, than most people realize.
Rate this post positively Reply With Quote Quick reply to this message

 
Old Yesterday, 01:15 AM
 
11,378 posts, read 11,299,251 times
Reputation: 7627
Based on NYS I’d bet the whole rust belt (plus RI and CT) got undercounted which would make me think Cleveland likely got shortchanged more than Columbus since NYS is a lot like NEO. While Sunbeltish dominate states got overestimated.

Cleveland be close to 2.1 million.

I think Buffalo and St Louis cities are going to be pretty close in population. Maybe like 289,000 and 275,000.

Providence might pass 10,000 ppsm.
Rate this post positively Reply With Quote Quick reply to this message
 
Old Yesterday, 03:42 AM
 
Location: Louisville
4,502 posts, read 4,324,713 times
Reputation: 7752
Based on the speculation that a lot of the under estimating is surrounding undocumented immigrant communities, my assumption would be that areas that typically see the largest migration patterns from these places would be affected most. More specifically places that have large Latin and Central American enclaves. Starting with the 2018 estimates pretty much all places across the board started seeing a slow down in growth, even places that hadn't experienced slowing in the last 50 years. I thought then(and now) that this was a direct reflection of the previous administrations policies. I think it's very plausible that many members of these communities started keeping lower profiles. The question I'd have regarding the OP is what areas in these states would have been affected most by this? Or do you think other factors are at play?
Rate this post positively Reply With Quote Quick reply to this message
 
Old Yesterday, 11:15 PM
 
452 posts, read 302,230 times
Reputation: 750
Quote:
Originally Posted by mjlo View Post
Based on the speculation that a lot of the under estimating is surrounding undocumented immigrant communities, my assumption would be that areas that typically see the largest migration patterns from these places would be affected most. More specifically places that have large Latin and Central American enclaves. Starting with the 2018 estimates pretty much all places across the board started seeing a slow down in growth, even places that hadn't experienced slowing in the last 50 years. I thought then(and now) that this was a direct reflection of the previous administrations policies. I think it's very plausible that many members of these communities started keeping lower profiles. The question I'd have regarding the OP is what areas in these states would have been affected most by this? Or do you think other factors are at play?
I think there is a lot of truth to that. I won't claim to be a professional demographer, but part of my job does make me have to follow different trends and one I could see without when looking at which states were under and overcounted, without delving into it, appeared to be the ones that also had the highest Central American/Mexican population growth as well. Then you add in areas that has a historical black populations on top of that, and IMO, even more recipe for being undercounted.

For Ohio, since that would be the area I have the most knowledge about, I think that black undercount will have the largest impact. Cleveland/Cuyahoga is probably the most undercounted (biggest population by number and percent, and the most transient if talking about moving between the city of Cleveland and into/or even back out of the inner suburbs). But the same is pretty much true in both Cincinnati and Columbus, which also have big black populations.

For Latinos, especially Central Americans/Mexicans, Cincinnati and Columbus may be more affected there, because both have small but growing communities, but probably not large enough to throw counts off to much. Plus, I wonder how much of an impact the hurricanes in Puerto Rico have also impacted numbers. I remember reading a story that the Spanish Club in Cleveland helped resettle something like 1,000 people from Puerto Rico after Maria. That was just one club that helps with resettlement. And since Cleveland/Lorain already have had a century long Puerto Rican community, how many more came from the island to live with family already over here without going through an agency like the Spanish American Club? How many were never taken into account for estimates?

Again, for Cleveland (Cuyahoga), it won't be one that could throw things off by a huge amount, but even if it's a couple thousand, you're talking 5 percent of the estimated losses for the decade.

In Northeast states, an undercount Puerto Ricans who came over who weren't put in the estimate model could be a significant part of their undercount.

This may be something I look into more.
Rate this post positively Reply With Quote Quick reply to this message
 
Old Today, 03:46 AM
 
Location: Louisville
4,502 posts, read 4,324,713 times
Reputation: 7752
Quote:
Originally Posted by ClevelandBrown View Post

In Northeast states, an undercount Puerto Ricans who came over who weren't put in the estimate model could be a significant part of their undercount.

This may be something I look into more.
Interesting post. Based on my understanding of immigration I don't think Puerto Rican communities would have been affected by the Immigration slowdown because they are already considered US Citizens? If they have been undercounted it would be more because the estimate models were unable to keep up with rapid number of Puerto Rican residents entering the mainland. Maybe someone else can correct me on that.
Rate this post positively Reply With Quote Quick reply to this message
 
Old Today, 09:48 AM
 
Location: Rahway, NJ
5,803 posts, read 2,128,415 times
Reputation: 4631
I think for Massachusetts, which was undercounted by about 150,000 people.. it looks more like this:

Middlesex County, MA: 1,641,000 (+31,000 from 2019est)
Worcester County, MA: 841,000 (+10,000 from 2019est)
Suffolk County, MA: 828,000 (+24,000 from 2019est)
Essex County, MA: 808,000 (+20,000 from 2019est)
Norfolk County, MA: 712,000 (+6,000 from 2019est)
Bristol County, MA: 572,000 (+7,000 from 2019est)
Plymouth County, MA: 525,000 (+4,000 from 2019est)
Hampden County, MA: 489,000 (+23,000 from 2019est) <--This county probably was very undercounted.
Barnstable County, MA: 215,000 (+3,000 from 2019est)
Hampshire County, MA: 165,000 (+5,000 from 2019est)
Berkshire County, MA: 130,000 (+5,000 from 2019est)
Franklin County, MA: 72,000 (+1,000 from 2019est)
Dukes+Nantucket, MA: 30,500 (About even, maybe 1,000 more than 2019est)

I did these estimates single handily and they come out to 7.028,000 people.. just 1k short of the 2020 census numbers. I calculated the dwelling units by average tenancy per unit.. coupled in with new construction units that were completed since February 15h 2010 (Went a little before 2010 Census numbers because people dont move immediately). I also tried accounting for Puerto Rican migration. Looks pretty spot on... im hopeful lol
Rate this post positively Reply With Quote Quick reply to this message
 
Old Today, 02:29 PM
 
452 posts, read 302,230 times
Reputation: 750
Quote:
Originally Posted by mjlo View Post
Interesting post. Based on my understanding of immigration I don't think Puerto Rican communities would have been affected by the Immigration slowdown because they are already considered US Citizens? If they have been undercounted it would be more because the estimate models were unable to keep up with rapid number of Puerto Rican residents entering the mainland. Maybe someone else can correct me on that.
Yes, I meant that there was such a massive influx in the months following Maria that it was probably virtually impossible to keep track of where people from the island were moving to, especially since so many already have family in the states and didn't need to use a resettlement agency.

In the aftermath of Maria, while nobody was able to give an exact number, estimates I read was in the following year (2017 to 2018) that went from 5,000 being relocated to Cleveland alone to as much as 13,000 (which majority would be to Cleveland, followed by Lorain and then maybe Youngstown after that).

Looking at the yearly estimates of the Puerto Rican population, it doesn't appear that was factored in.

Based on the 1-year, and then 5-year, estimates, here is the yearly change:

2010-11
Cuyahoga: 3,540 (one-year); 714 (five-year)
Lorain: 3,131; 1,598

2011-12
Cuyahoga: 3,666; 931
Lorain: -2,897; 343

2012-13
Cuyahoga: 1,939; 1,702
Lorain: 2,387; 365

2013-14:
Cuyahoga: -2,393; 1,771
Lorain: 2,084; 235

2014-15
Cuyahoga: 2,744; 1,965
Lorain: -243; 1,473

2015-16
Cuyahoga: 2,263; 1,523
Lorain: 1,086; 479

2016-17:
Cuyahoga: -2,192; 146
Lorain: 949; 954

2017-18
Cuyahoga: 2,873; 1,139
Lorain: 1,247; 1,122

2018-19:
Cuyahoga: 2,679; 970
Lorain: -2,490; 98

The one-year is more volatile and there appears to be some major "corrections" for each in certain years as to smooth the numbers. At the same time, the one-year estimates would be able to pick up quicker on an influx, and in this case it doesn't appear to have.

The five-year data is usually more reliable since it's based off of longer trends, so you don't have the drastic changes, but those show the Puerto Rican growth actually slowing the past couple years in comparision to the first half of the decade. And there is nothing there to indicate that the area did pick up anywhere between 5,000 to 10,000 following the hurricane. Granted, some may have been temporary and some families may have since moved back to the island, but they were still here for a period of time that doesn't seem to be accounted for at all.

In Cuyahoga County's case, it's a small number, but even an 5,000-10,000 undercount there takes away roughly 15 to 25 percent of its estimated population losses right there. In NY or Massachusetts, those numbers could be significantly higher, more in the 50,000 to 100,000 range and be a significant factor in why those state's populations came in much higher than estimates showed.
Rate this post positively Reply With Quote Quick reply to this message
 
Old Today, 02:43 PM
 
Location: Baltimore
11,537 posts, read 4,636,984 times
Reputation: 6170
Quote:
Originally Posted by btownboss4 View Post
Based on NYS I’d bet the whole rust belt (plus RI and CT) got undercounted which would make me think Cleveland likely got shortchanged more than Columbus since NYS is a lot like NEO. While Sunbeltish dominate states got overestimated.

Cleveland be close to 2.1 million.

I think Buffalo and St Louis cities are going to be pretty close in population. Maybe like 289,000 and 275,000.

Providence might pass 10,000 ppsm.
MA and NJ were also undercounted. MA by about 140k and NJ by 460k I think...
Rate this post positively Reply With Quote Quick reply to this message
 
Old Today, 04:33 PM
 
67,131 posts, read 93,279,043 times
Reputation: 14688
Quote:
Originally Posted by ClevelandBrown View Post
Yes, I meant that there was such a massive influx in the months following Maria that it was probably virtually impossible to keep track of where people from the island were moving to, especially since so many already have family in the states and didn't need to use a resettlement agency.

In the aftermath of Maria, while nobody was able to give an exact number, estimates I read was in the following year (2017 to 2018) that went from 5,000 being relocated to Cleveland alone to as much as 13,000 (which majority would be to Cleveland, followed by Lorain and then maybe Youngstown after that).

Looking at the yearly estimates of the Puerto Rican population, it doesn't appear that was factored in.

Based on the 1-year, and then 5-year, estimates, here is the yearly change:

2010-11
Cuyahoga: 3,540 (one-year); 714 (five-year)
Lorain: 3,131; 1,598

2011-12
Cuyahoga: 3,666; 931
Lorain: -2,897; 343

2012-13
Cuyahoga: 1,939; 1,702
Lorain: 2,387; 365

2013-14:
Cuyahoga: -2,393; 1,771
Lorain: 2,084; 235

2014-15
Cuyahoga: 2,744; 1,965
Lorain: -243; 1,473

2015-16
Cuyahoga: 2,263; 1,523
Lorain: 1,086; 479

2016-17:
Cuyahoga: -2,192; 146
Lorain: 949; 954

2017-18
Cuyahoga: 2,873; 1,139
Lorain: 1,247; 1,122

2018-19:
Cuyahoga: 2,679; 970
Lorain: -2,490; 98

The one-year is more volatile and there appears to be some major "corrections" for each in certain years as to smooth the numbers. At the same time, the one-year estimates would be able to pick up quicker on an influx, and in this case it doesn't appear to have.

The five-year data is usually more reliable since it's based off of longer trends, so you don't have the drastic changes, but those show the Puerto Rican growth actually slowing the past couple years in comparision to the first half of the decade. And there is nothing there to indicate that the area did pick up anywhere between 5,000 to 10,000 following the hurricane. Granted, some may have been temporary and some families may have since moved back to the island, but they were still here for a period of time that doesn't seem to be accounted for at all.

In Cuyahoga County's case, it's a small number, but even an 5,000-10,000 undercount there takes away roughly 15 to 25 percent of its estimated population losses right there. In NY or Massachusetts, those numbers could be significantly higher, more in the 50,000 to 100,000 range and be a significant factor in why those state's populations came in much higher than estimates showed.
I believe this should give some idea of the migration from Puerto Rico post-Maria: https://centropr.hunter.cuny.edu/sit...2-3-3-2018.pdf
Rate this post positively Reply With Quote Quick reply to this message
 
Old Today, 08:47 PM
 
452 posts, read 302,230 times
Reputation: 750
Quote:
Originally Posted by ckhthankgod View Post
I believe this should give some idea of the migration from Puerto Rico post-Maria: https://centropr.hunter.cuny.edu/sit...2-3-3-2018.pdf
There is a lot to digest there, but thanks for posting that.

I went ahead and looked at school district data for Latino students in Cuyahoga and Lorain counties just before Maria and then just after to see if there was an increase. To be clear, the state lumps all Hispanic students into the Latino category, so it doesn't necessarily mean Puerto Rican, but Puerto Ricans make up roughly 65 percent of the Latino/Hispanic population in Cuyahoga and Lorain counties.

Even though these numbers don't only reflect Puerto Ricans, I think you can draw a correlation between the uptick in Latino students in each county in the immediate aftermath of Maria, which as showed above in one of my previouis posts that yearly Puerto Rican overall population estimates did not appear to take that into account.

In the three years leading up to Maria, public/charter schools in Cuyahoga County collectively averaged an average annual gain of 308 Latino students per year. In Lorain County, the average was an increase of 177.

In the 2018-19 school year (the first full year after the Hurricane), Cuyahoga County had an increase of 621 students (202 percent increase compared to the previous three-year average). Lorain County had an increase of 312 students (176 percent increase over the three-year average).

In the 2019-20 school year, Cuyahoga went back to an increase of 233 (76 percent of what it had averaged in the three pre-Maria years). In Lorain County, it went to an increase of 167 (94 percent of what it had previously averaged).

*While students from Puerto Rico likely started coming over during the 2017-18 school year, considering the hurricane happened in late September and the Ohio Department of Education does its headcounts each October, those students likely were not counted in those numbers, and weren't reflected until the 2018-19 school year. Plus, I would love to find Catholic school data, because I know pretty much every Catholic school (especially on the westside) has really been reaching out to the Latino/Hispanic community to help boost enrollment. Considering that's a heavily Catholic population, I would imagine those schools are seeing big gains to offset the loss of the white Catholic population. I know at least one west side Catholic high school has gone from about 90-95 white 10 years ago to now only 75 percent, and the increase was mainly due to Puerto Rican/Hispanic students.

FWIW, I'm not expecting Cuyahoga to have actually gained population over the past decade, though I do suspect a decent portion of the state's undercount is there. The Maria aspect is just one small portion of that.

In general, looking at different aspects, if I had to guess right now, of the 113,000 undercount in Ohio, it would go:

1. Cincinnati metro: 40,000, probably split evenly between Hamilton County and its other Ohio metro counties (Butler/Warren/Clermont). On the Kentucky side, there also could be an increase that moves the overall metro upward of 50,000 more than estimated, pushing the overall metro toward 2.27 million.

2. Cleveland metro: 40,000, probably mostly in Cuyahoga County, though Lake County could pick up a small portion. That would put Cuyahoga about even with the 2010 numbers, and I think the city of Cleveland picks up about 15,000, also putting it back to where it was at the 2010 census. That would bump up the metro to about 2.08 million, which would be a small increase over 2010.

3. Columbus metro: 10,000: This one is tough, because I can actually see it not gaining much because Franklin has come back to the pack economically in the past 5 years, but its population estimates have just started to come down the last two estimates. Franklin averaged about 15,000 gain per year when it was clearly out-performing the rest of Ohio the first half of the decade. Economically, it started to come down in 2015, but yet, estimates still had gains 17,000 in 2016 and 21,000 in 2017, before dropping to 12,000 in 2018 and 6,000 in 2019. If Franklin does stay the same, I think Delaware, Fairfield and Licking counties will pick up 10,000. However, if Franklin does go back, then metro Columbus will probably come in right around where the estimates have it, meaning an additional 10,000 to go around.

4. Small metro/non-metro counties: 10,000: This won't be geared toward any particular county, but I can see the 50 or so rural counties having their decline a little overestimated. You're talking a couple hundred here, a couple hundred there, but over that many counties, it adds up.

5. Dayton metro: 5,000: I can see the city of Dayton gaining a couple thousand and getting back to the small positive territory from 2010 and then a small undercount in the suburban areas.

6. Rest of metro Ohio (Akron, Canton, Toledo, Youngstown): 8,000. I don't see any big increases between these. In fact, I don't expect Akron to gain at all, and actually could lose a little. I don't see big increases in the others, but collectively may pick up the remainder.

Last edited by ClevelandBrown; Today at 09:00 PM..
Rate this post positively Reply With Quote Quick reply to this message
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.

Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.


Reply

Quick Reply
Message:


Over $104,000 in prizes was already given out to active posters on our forum and additional giveaways are planned!

Go Back   City-Data Forum > U.S. Forums > General U.S.

All times are GMT -6.

© 2005-2021, Advameg, Inc. · Please obey Forum Rules · Terms of Use and Privacy Policy · Bug Bounty

City-Data.com - Contact Us - Archive 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37 - Top