The Antiplanner

Stuck in the 1960s

2nd November 2009

Stuck in the 1960s

posted in Useful Data |

Each year, about one out of 40 households receive a letter from the Census Bureau demanding that they fill out an American Community Survey asking such nosy questions as how much money each person in the household earns each year, how they heat their house, and whether they have a flush toilet. Though some have suggested that people should boycott the decennial census as too “intrusive,” the Antiplanner is a voracious consumer of census data, and so I was proud to receive and fill out the 28-page survey form form this year.

This survey is an annual extension of the Census Bureau’s so-called long form, which has been given to one out of six households each decennial census since at least 1960 (including the Antiplanner’s in 2000). As I filled out the 2009 form, it occurred to me that some of the questions have not been significantly updated since 1960.

The survey asked whether we have a refrigerator, but didn’t ask whether we have a personal computer. It asked whether we have a sink with a faucet, but didn’t ask whether we have dial-up or high-speed Internet connections. It asked whether we lived in a house or apartments, but didn’t ask if we lived in a mixed-use development. It asked if we have telephone service — including cell phones — but it might have been useful to ask if households had only cell phones and no land line. It asked there was a store or medical office on the property, but didn’t ask if we have a home office.

It so happens the Antiplanner once lived in a house with no refrigerator, but such households are rare — and ones with no sink and faucet rarer still. The things that divide people by wealth and lifestyle today are very different than in 1960, and the surveys should be updated to reflect those differences.

The Census Bureau has added a few new questions lately having to do with health insurance and marital status. I suppose these are important to someone, but I’d like more information about other topics.

The survey asks how people “usually” get to work, but doesn’t allow for people who drive or bicycle partway and take transit the rest of the way–or for people who take a bus and transfer to a train. Moreover, since commuting is now only about a fifth of all travel, it would be nice to know how people got around for the other four-fifths.

I realize there is only so much the Census Bureau can ask without getting resistance or unreliable results. But it asks whether people completed nursery school but not kindergarten, or kindergarten but not first grade. It asks if people live in a building with 2, 3 to 4, 5 to 9, 10 to 19, 20 to 49, or 50 or more apartments. It asks how much money people spent on electricity, gas, water, sewer, coal/oil/kerosene/wood heat, property taxes, fire insurance, and rent or mortgage. Why couldn’t it be just a little more specific about transportation, like how many miles did you put on your car(s) and how many times did you take transit or ride your bicycle last week?

You can’t always get what you want, and I am sure there is no end to the list of questions someone would like to see added to the form. I’ll be happy to use whatever data I can get from the surveys as they are posted by the hardworking Census Bureau.

Speaking of data, when I downloaded the 2008 National Transit Database last week, it occurred to me that in past years the Federal Highway Administration has always posted its annual Highway Statistics first, followed by the Federal Transit Administration’s transit database. For 2008, the transit data came first. Moreover, several tables are still missing from the Highway Statistics for 2006 and 2007. Is this an indicator of the changing fortunes of the two agencies?

This entry was posted on Monday, November 2nd, 2009 at 12:00 am and is filed under Useful Data. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

There are currently 16 responses to “Stuck in the 1960s”

Please feel free to submit comments. Constructive debate is welcome. Ad hominem attacks and name-calling will reveal the shallowness of the author. Foul language may be deleted. Your first comment will require moderation; after that, all comments will be approved unless they contain too many links in which case the software may hold them for approval as potential spam.

  1. 1 On November 2nd, 2009, C. P. Zilliacus said:

    The Antiplanner and some other readers might recall the phony “controversy” manufactured by certain members of the Republic Party in the U.S. House of Representatives and U.S. Senate during the 2000 Census. They claimed that the long form questions were somehow “intrusive,” which was one reason that the ACS was expanded and extended.

    And I agree with that the ACS data are useful, but in one aspect the “long form” was much better – its sample size was massive, and it covered all of these United States, as well as its territories and overseas possessions. Now with a good sampling plan (which is something that the Census Bureau should be good at), the ACS is almost as good as the long form. Of course, I would prefer to have data from both of them!

    You are correct – the questions are indeed somewhat dated, but one possible reason for that is continuity of the time-series they represent, so trends can be tracked over time.

    posted on November 2nd, 2009 at 5:59 am

  2. 2 On November 2nd, 2009, Dan said:

    They claimed that the long form questions were somehow “intrusive,” which was one reason that the ACS was expanded and extended.

    We have some nutters around here who claim that they will be targeted and won’t be filling out the census. Sigh. Whaddya gonna do?

    Nonetheless, I agree with CPZ that the questions need to have continuity to track trends. Surely they can figger out how to word a question about multi-modality, but surely that number is small and won’t start growing in this census, likely the next one will see mode changes as gas prices go up.

    DS

    posted on November 2nd, 2009 at 8:15 am

  3. 3 On November 2nd, 2009, Neal Meyer said:

    Antiplanner,

    I can still remember when I was one of those picked to respond to the 2000 Census (or ACS?) survey. The thing that got me mad was that on the postage, there was a statement that if I did not fill out the entire form, it was a federal offense punishable, by if I can remember correctly, a $300 fine. Forgive me if I can’t remember all the details, but it was nine years ago.

    The Census should ask only one question: How many people live in your household?

    Instead, what we have here is the federal government sticking a gun in your face, cocking the gun, and telling you, if you don’t hand over lots of information about yourself, you will be found in are contempt and we will legally rob you.

    posted on November 2nd, 2009 at 10:17 am

  4. 4 On November 2nd, 2009, Mike said:

    I, too, expect that most of the “dated” questions are likely longitudinal and should probably be phased out as more currently-applicable questions phase in. After all, when they ask how many horses and buggies your home has, that number has to have dropped to something statistically near zero for effectively everybody, no matter that a few stragglers might still be out there. And the trend ain’t reversing. At that point, that particular longitudinal analysis is effectively over.

    Part of my expectation of the government protecting my individual rights includes the expectation that I may be required to provide extremely rudimentary demographic data to it. To the extent that this provides government with the information necessary to run the military, police, courts, and pan-border functions (epidemiology, land title recording, vital records) it’s fine. To the extent that it’s used to further the execution of wealth-expropriatory programs such as Social Security, Medicaid, subsidized High-Speed Rail, etc, it is not fine. As a normal citizen will have no way of granulating answers so as to support one function over the other, the citizen’s recourse remains to push for government to discontinue functions that are not part of its legitimate purpose of safeguarding individual rights — not to attempt a backdoor solution by resisting a census that does support that legitimate purpose, as well as other non-legitimate purposes.

    posted on November 2nd, 2009 at 11:46 am

  5. 5 On November 2nd, 2009, t g said:

    Mike, I would like to extend that prohibition on the government’s use of statistical measures to include the Consumer Price Index. It’d be interesting to see how the market on it’s own, starting from scratch, merely responding to demand, would create a measure of inflation. I’m not aware of a private sector alternative to the CPI. In another category, we have the OFHEO home price index with it’s private sector alternative of the Case-Shiller. But no private alternative for the CPI – which seems like something the financial sector would be willing to pay for.

    The CPI was modified in the 90′s in order to reduce the COLAs (by the purportedly liberal Clinton Administration). There’s criticism from both sides that it under and over estimates inflation. One could argue from Friedman that you could adjust your price data for inflation by using the money supply. But it seems like something that is so subject to political whim (change this basket to reflect this) that it has lost at least some of its accuracy.

    You’re the lawyer right? What’s the constitutionality of the CPI?

    posted on November 2nd, 2009 at 12:26 pm

  6. 6 On November 2nd, 2009, Dan said:

    Let’s also extend that prohibition to GDP as a measure of well-being. And ag output/ac as a measure of farm health. And economic growth as an indicator of a country’s well-being. And and and…

    IOW: meaningful indicators, please.

    DS

    posted on November 2nd, 2009 at 1:56 pm

  7. 7 On November 2nd, 2009, Borealis said:

    Very rough measurements, like GDP and output/acre have their problems, but adjusting them have their problems too. Enviros can’t even agree on an enviro GDP, and all the options are full of value-laden adjustments. Might as well have a PETA-approved GDP as many of those enviro GDPs.

    posted on November 2nd, 2009 at 2:15 pm

  8. 8 On November 2nd, 2009, Dan said:

    …adjusting them [has] problems too…Enviros can’t even agree on an enviro GDP, and all the options are full of value-laden adjustments. Might as well have a PETA-approved GDP as many of those enviro GDPs.

    Of course, there are numerous well-being indices that capture well-being far better than GDP. And why would one want to tie GDP to the environment, except to track exploitation or degradation? How does one meaningfully track overall ecosystem health with one metric? Ridiculous.

    DS

    posted on November 2nd, 2009 at 3:04 pm

  9. 9 On November 2nd, 2009, bennett said:

    Mike and Neal,

    I think that the Census and the Demographic Health Survey are the American Government at it’s best. Collecting Data and reporting it. Without advocacy and providing facts. Yes this data can be used by governments (Federal/State/Local) to make a case for High Speed Rail or the likes, but that is not the fault of the census itself. Maybe government planners use the data to forward an agenda, but the antiplanner uses the same date to refute that agenda. Not everything that is intrinsically good is always used for good. That doesn’t mean it should go away. Penicillin has been used to save the lives of murders. Does that mean that we shouldn’t have penicillin?

    posted on November 2nd, 2009 at 4:09 pm

  10. 10 On November 2nd, 2009, bennett said:

    This is why data like this is important…

    http://www.ted.com/talks/hans_rosling_at_state.html

    posted on November 2nd, 2009 at 4:10 pm

  11. 11 On November 2nd, 2009, Dan said:

    I loves me some Ted!

    DS

    posted on November 2nd, 2009 at 5:36 pm

  12. 12 On November 2nd, 2009, Frank said:

    Let’s also extend that prohibition to GDP as a measure of well-being.

    About 70% of GDP is consumption, so while it might show short-term well-being, which is debatable, it can also point to an inevitable crash of our non-sustainable lifestyle.

    posted on November 2nd, 2009 at 5:52 pm

  13. 13 On November 2nd, 2009, Borealis said:

    Let’s just stipulate that GDP is a measure of economic trading, not a measure of well-being or sophistication as a nation. I would rather we keep the measurements clean and argue about their meaning than design value-laden measurements and argue about the measurements.

    posted on November 2nd, 2009 at 8:46 pm

  14. 14 On November 3rd, 2009, Dan said:

    About 70% of GDP is consumption, so while it might show short-term well-being, which is debatable, it can also point to an inevitable crash of our non-sustainable lifestyle.

    Good point.

    DS

    posted on November 3rd, 2009 at 6:19 am

  15. 15 On November 3rd, 2009, Mike said:

    t g,

    I’m not presently a practicing attorney (a license is not required for the work I do) but I do have a law degree, yes. The CPI itself is really just a mathematical exercise, but there may be constitutionality issues depending on what that data is used to accomplish by other government instruments. The problem, as you allude, isn’t that someone is aggregating that data and making a declaration of what it means, but that vastly powerful economic entities and governmental modules are taking that declaration for holy truth (and may be statutorily required to do so, in the case of social welfare programs). The private market, lazy and seeking maximum efficiency, essentially gets to be a free rider on that information. In the absence of governmental tabulation of the CPI, private industry would have to depend on independently-generated projections. This is not unprecedented: FICO scores, which are projected estimates of creditworthiness generated by a proprietary process developed and owned by the Fair Isaac Co., are perfect examples of what might emerge.

    bennett,

    I don’t think we disagree here. In my post I mentioned that, rather than using the census issue to mount a backdoor attack on the programs one does not wish to fund, the better approach is to challenge those programs directly and let the census remain its own issue. The census data is necessary for legitimate functions like police protection, epidemiology, and vital records, even though that data is also used to administer illegitimate social welfare programs. Indeed, saying the census causes bloated government is like saying video cameras cause pornography.

    posted on November 3rd, 2009 at 11:00 am

  16. 16 On November 3rd, 2009, John Thacker said:

    The Antiplanner and some other readers might recall the phony “controversy” manufactured by certain members of the Republic Party in the U.S. House of Representatives and U.S. Senate during the 2000 Census. They claimed that the long form questions were somehow “intrusive,” which was one reason that the ACS was expanded and extended.

    I agree that the idea is extremely unlikely, but then again so is the idea that the U.S. Government would actually be reading everyone’s library records and emails and international phone calls (instead of selectively looking at those that were already targets of interest), and so is the idea that the no fly list targeted people for reasons of political advocacy.

    Yet somehow a lot of people who thought that the Census thing was a “phony controversy” believed in the others.

    posted on November 3rd, 2009 at 4:26 pm

Leave a Reply

You must be logged in to post a comment.