Question about Columbus vs. Cincinnati (Cleveland, Hamilton: lawyers, home, centers)

[The Cactus Leaguer]

Originally Posted by JoeP

Generally speaking growth is a %, because one is referring to a growth rate.

Raw population #'s don't indicate growth in a meaningful way

A "rate" can also refer to a number, as in "Our city is growing at a rate of 1,000 residents per week".

To anyone who supplies city services, raw population #'s are the growth numbers that really matter. If the electric company adds 100,000 new residents to the grid, they are figuring out how to add 100,000 new residents, not 5% or 2% or whatever (although that %age definitely matters to the shareholders )

[JoeP]

Yes within a city, I could see that. But one can't really see how a city is doing in relation to others without using % and similar statistics.

Or is NYC doomed to always be the worst city for crime because it is far larger than anywhere else?

Hey Detroit and others, you can be happy, per capita is useless - NYC by size alone will be #1 for a long long time in crime!

[OHBuckeye]

Originally Posted by The Cactus Leaguer

Actually, they look at both... and none of the examples that you linked were limited to the convenient cutoff of 1 million+. How is that "fair"?

BTW, I am not saying that was intentional on your part, since I know it would be laborious to comb through every single metro area and classify them by region.

It's really not relevant to this discussion, but the examples that I linked were for cities, not metro areas; the only purpose of my posting the links was to show that percentages are most commonly used in these type of rankings. I think JoeP's most recent post in this thread was an excellent, succinct explanation of why percentages should be and are used.

Yes, it would be quite laborious to comb through every single metro area, which is why I decided to limit my inquiry to major metro areas; I believe that, in general, the cutoff for what qualifies a metro area as 'major' is a population of 1 million+, which is why I used that particular cutoff. I certainly wasn't trying to inflate Cincy's ranking; if that had been my purpose, I would have stuck to the official Census Bureau definitions and not included Louisville, Washington and Baltimore. And yes, I have seen similar rankings in the past that only considered 'major' cities and 'major' metro areas.

With that being said, I thought of a better and simpler way to compare the Cincy metro area's growth rate to the Northeast/Midwest. Instead of limiting the comparison to other metro areas, I decided to compare it to the entire region by adding up the various year 2000 and year 2006 state totals, which can be found here: State and County QuickFacts.

The Midwest, as defined by the Census Bureau (OH, IN, IL, MI, WI, MN, IA, MO, ND, SD, NE, KS) had an overall population increase of 1,824,960, which correlates to a growth rate of 2.8%. The Northeast, as defined by the Census Bureau (PA, NJ, NY, CT, RI, MA, VT, NH, ME) had an overall population increase of 1,146,975, which correlates to a growth rate of 2.1%. Combined, the two regions had an overall population increase of 2,971,935, which correlates to a growth rate of 2.5%.

As seen on the maps at these links (Northeastern United States - Wikipedia, the free encyclopedia & Midwestern United States - Wikipedia, the free encyclopedia), some people would argue that, from a geographical standpoint, the Northeast/Midwest should also include DE, MD, DC, VA, WV and KY. If those places are included, the Northeast/Midwest had an overall population growth of 4,109,323, which correlates to a growth rate of 3.0%.

Summation of 2000-2006 Regional Growth Rates
Cincinnati Metro Area = 4.7%
State of Ohio = 1.1%
Combined States of Ohio, Kentucky and Indiana = 2.4%
Midwest (Census Bureau definition) = 2.8%
Northeast + Midwest (Census Bureau definition) = 2.5%
Northeast + Midwest (alternate definition) = 3.0%

As I've contended all along, the Cincy metro area's growth rate compares quite favorably on a regional basis. Whether you compare it to other major Northeast/Midwest metro areas, to the entire state of Ohio, to the Midwest region or to the combined Northeast/Midwest region, the Cincy metro area's growth rate is noticeably above average (more than I had thought, as it turns out). If I haven't convinced you of that by now, I guess I never will!

[Subway]

Originally Posted by The Cactus Leaguer

Percentages can be quite useful, but are also easily manipulated because (as I said before) the denominators are not equal.

You have it completely backwards! The inequality of the denominators is the precise reason why percentages should be used in cases like this! If the denominators were equal, there would be no reason to calculate the percentages for the purpose of comparing places to each other - the raw numbers would be fine under that scenario, and only under that scenario!

[The Cactus Leaguer]

Well, this thread has been hijacked by discussion of percentages and for that I apologize. Certainly the Columbus v Cincinnati discussion is interesting (and apparently has continued in a different thread). For me, if I were under 30 and enjoyed taking weekend drives/trips to other (larger) metro areas, I'd choose Columbus. Otherwise I'd pick Cincinnati. Neither is a bad choice though - Ohio in general is wildly underrated (and I've been to almost every top-30 metro area in the country for business).

Now, as far as the "debate" over percentages goes --- I think we all can agree that the usefulness of any statistic depends on what you are using it for. And I would never claim that using percentages is worthless.

My initial contention was that if you limit the sample size to metro areas over 1 million, the smaller areas are at a natural advantage if you are using percentages. This is true both from a pure math/statistics standpoint and also from a sociological standpoint. I am going to take one more shot at explaining the math of it and then move on from there:

Philadelphia added 139,601 residents between 2000 and 2006. Cincinnati added 94,545 residents over the same period. Who grew faster? The answer, of course, is that it depends on the definition. If Philly started out with fewer residents, the population gap between the two would decrease. In this case, Philly has more residents to begin with so the population gap between the two cities grows. I could state that "Philly is growing 48% faster than Cincy" and I would be correct. But when measuring it relative to the 2000 population base, the reverse is true. Cincy grew 4.7%, Philly 2.5% --- Cincy grew "almost twice as fast".

Now the reason why I'm making the case from a sociological perspective is that a larger a city gets, the more difficult it becomes to grow because you reach a tipping point where there are relatively fewer advantages (cultural amenities, airport hubs, industry clusters, etc.) and relatively more disadvantages (traffic, pollution, strain on public services, higher tax rates to support said service, etc.) to population growth. And my personal opinion is that this occurs around the 4-5 million population mark (the way our cities are constructed, requirements for sports/cultural facilities, etc.). In between 1 million to 4 million, yes there are more headaches but you also get more benefits --- maybe a sports team relocates or expands to your town, or an airline picks your city as a hub so you have more direct flights, or light rail is added, etc.

JoeP's example has nothing to do with my point at all. Of course crime would be measured as a rate because there is a more direct relationship to the population (along with the economy, number of young males, etc.). What I'm suggesting (and apparently nobody agrees with me on this) is that the starting population (denominator) does not have nearly the same direct relationship with the number of residents added, especially in today's mobile society. More residents (more likely) equals more crime; more residents (less likely) equals more growth.

As I've contended all along, the Cincy metro area's growth rate compares quite favorably on a regional basis. Whether you compare it to other major Northeast/Midwest metro areas, to the entire state of Ohio, to the Midwest region or to the combined Northeast/Midwest region, the Cincy metro area's growth rate is noticeably above average (more than I had thought, as it turns out). If I haven't convinced you of that by now, I guess I never will!

When you comparing it the way you did, yes I agree with you 100%. Actually, since the country as a whole is moving south and west, I would be shocked if Cincinnati did NOT stack up well in those comparisons. My guess is that it would not stack up nearly as well against the south, which, you could debate, would be inclusive since Cincinnati's metro area includes Kentucky.

My contention was that the "7th best" ranking could be misinterpreted. I never said that OHBuckeye was factually incorrect, but neither was I (except on that minor point about net migration when I meant domestic migration).

I'm not sure why I'm getting attacked so much on this board over this point - yes it's debatable on what measures to use (and digging deeper into the components is much more instructive), but my point is not invalid. Maybe it's because I'm an outsider, I don't know. I don't think Cincinnati has anything to hang its head about, but maybe you have a chip on your shoulder for getting lumped in with the dying rust belt(???)... Again, it's presumptive on my part.

[OHBuckeye]

Cactus Leaguer,

I thought of an additional way to see how the Cincy metro area's growth stacks up to other regional major metro areas that should satisfy your personal misgivings about the large differences in denominators. Let's use the most comprehensive definition of the combined Northeast/Midwest (which would include the "Upper South" - KY, WV, VA, MD, DC, DE), and only include metro areas that had denominators (year 2000 estimates bases) between 1 and 3 million. At the low end would be Rochester with a 2000 estimates base of 1,037,833, and at the high end would be Minneapolis with a 2000 estimates base of 2,968,817; Cincy's 2000 estimates base of 2,009,673 would essentially be smack dab in the middle of that range. Using those parameters, there are 17 metro areas to rank, and Cincy's 2000-2006 growth rate ranks 7th (6 are growing faster, 10 are growing slower).

1. Indianapolis: +9.2%
2. Richmond: +8.8%
3. Kansas City: +7.1%
4. Columbus: +7.0%
5. Minneapolis: +6.9%
6. Louisville: +5.1%
7. Cincinnati: +4.7%
8. Virginia Beach: +4.6%
9. Baltimore: +4.1%
10. St. Louis: +3.6%
11. Hartford: +3.5%
12. Providence: +1.9%
13. Milwaukee: +0.6%
14. Rochester: -0.2%
15. Cleveland: -1.6%
16. Pittsburgh: -2.5%
17. Buffalo: -2.8%

Another way of looking at it: Cincy's growth rate is 4.7% and the average growth rate for all of those metro areas is 3.5%.

[The Cactus Leaguer]

Good stuff!

So basically what that tells me is that the further south and/or west you go, the faster the growth rate --- and Columbus is actually a bit of an outlier in that regard. What do you think?

[OHBuckeye]

Originally Posted by The Cactus Leaguer

Good stuff!

So basically what that tells me is that the further south and/or west you go, the faster the growth rate --- and Columbus is actually a bit of an outlier in that regard. What do you think?

I agree. I would say that Columbus and St. Louis are the outliers. If those two traded growth rates, that statement would be especially accurate.

Then again, I guess we could call Virginia Beach-Norfolk another outlier. I was surprised when I learned that its growth rate was slightly less than Cincy's. I would have guessed that Virginia Beach-Norfolk's growth rate would have been more like 8-10%.

[The Cactus Leaguer]

Originally Posted by OHBuckeye

I agree. I would say that Columbus and St. Louis are the outliers. If those two traded growth rates, that statement would be especially accurate.

Then again, I guess we could call Virginia Beach-Norfolk another outlier. I was surprised when I learned that its growth rate was slightly less than Cincy's. I would have guessed that Virginia Beach-Norfolk's growth rate would have been more like 8-10%.

Yeah. I think Norfolk was hurt this decade because their area relies heavily on the US Navy as an employment center.

[The Cactus Leaguer]

One last thing - about the "everyone uses percentages, nobody uses numbers" argument:

http://www.census.gov/population/www/estimates/metro_general/2006/CBSA-EST2006-07.xls (broken link)

The Census Bureau looks at both - as they should.