Changing Demographics of Pittsburgh - more Asians coming in... (restaurants, food)

[bradjl2009]

Originally Posted by Tiger Beer

Good information.

Yeah, consistent what you are saying, I haven't seen any other sources anywhere with numbers like this.

Also, the smaller precinct numbers. In my experience, even with zip codes, sometimes they can be very misleading when just having percentages. I mean, some zips have like 100 people, so just a couple families in there and suddenly its 15% Asian or something.

I would say the county percentages are fairly accurate but I wouldn't put much stock or try to look for a trend from a little census precincts for the reasons you said. You think they would be able to give you population considering all the other information you can get from the New York Times map. It would help explain why the county is 2% Asian but some precincts are 20% Asian.

[BrianTH]

The ACS contains a LOT more information than the Census. It is also done every year.

Because it only goes to a fraction of households, it is in fact subject to sampling error. That is why the smaller the unit, the more years they lump together before reporting--you can get annual data for the bigger units, but for the smallest units they will only give 5-year averages (this is the first year for 5-year averages because they have only been doing the full ACS that long). And even those 5-year averages are admittedly still subject to sampling error--in fact the ACS reports error bars along with the estimated value.

So the ACS isn't a substitute for the Census in terms of providing a hard count, but on the other hand, unlike the Census it covers a lot more topics and happens every year. This will be particularly useful for tracking trends in future years (there will be a new five-year average available every year going forward).

[BrianTH]

By the way, I expect in the next 50 years, Pittsburgh will become a lot more Asian and also a lot more Hispanic, but most of the Hispanics will be multi-generation U.S. and therefore relatively assimilated and intermingled with other ethnicities.

[BrianTH]

Last thought--the NYT map is only the tip of the iceberg when it comes to what you can do with this data. But I noticed the NYT map has same-sex-couple households--in Pittsburgh, the three big clusters of gay couples are in the North Side, the Regent Square area, and the Bloomfield/Lawrenceville area (note this is gay couples, not gay singles). Not a surprise, but this is the sort of thing it will be fun to check on, now that such data is available.

[XRiteMA98]

Originally Posted by BrianTH

By the way, I expect in the next 50 years, Pittsburgh will become a lot more Asian and also a lot more Hispanic, but most of the Hispanics will be multi-generation U.S. and therefore relatively assimilated and intermingled with other ethnicities.

Great! Maybe we'll finally have a Latino radio station I could listen to in the car.

[BrianTH]

Originally Posted by XRiteMA98

Great! Maybe we'll finally have a Latino radio station I could listen to in the car.

The problem is in 50 years, a Latino radio station will likely make as much sense as an Irish or Italian radio station.

[bradjl2009]

Originally Posted by BrianTH

The problem is in 50 years, a Latino radio station will likely make as much sense as an Irish or Italian radio station.

Some of them are starting to not do that well with the economy and the assimilation of many Latinos. It would take a lot of demographic change for a radio station such as that to even break even here.

[BrianTH]

Originally Posted by bradjl2009

Some of them are starting to not do that well with the economy and the assimilation of many Latinos. It would take a lot of demographic change for a radio station such as that to even break even here.

Yep. Native-born Hispanics are where most of the future Hispanics will come from, and like all prior immigrant waves, their interest in ethnic-specific cultural offerings is dropping with each generation.

It is like the German-language newspapers and such back in the day--WWI definitively killed them off, but they were dying before then due to the lack of interest among native-born U.S. people of German descent.

[Katarina Witt]

Originally Posted by BrianTH

The ACS contains a LOT more information than the Census. It is also done every year.

Because it only goes to a fraction of households, it is in fact subject to sampling error. That is why the smaller the unit, the more years they lump together before reporting--you can get annual data for the bigger units, but for the smallest units they will only give 5-year averages (this is the first year for 5-year averages because they have only been doing the full ACS that long). And even those 5-year averages are admittedly still subject to sampling error--in fact the ACS reports error bars along with the estimated value.

So the ACS isn't a substitute for the Census in terms of providing a hard count, but on the other hand, unlike the Census it covers a lot more topics and happens every year. This will be particularly useful for tracking trends in future years (there will be a new five-year average available every year going forward).

The ACS is actually a part of the Census Bureau, as the link says.

[BrianTH]

Originally Posted by Katiana

The ACS is actually a part of the Census Bureau, as the link says.

The ACS, or American Community Survey, is indeed conducted by the Census Bureau. There is an American Community Survey Office within the Census Bureau, and the Census Bureau in turn is part of the Commerce Department.

But when I wrote "the Census", I was referring to the decennial Census, also conducted by the Census Bureau. The ACS is not part of the Census, because those are two different programs, albeit both conducted by the same agency.