2010 Census: What can we look forward to, what will we miss from 2000

[Raf]

I received my 2010 Census in the mail yesterday and promptly sent it out. The hype regarding how short it was going to be is true, with a whopping 9 questions. The questions were standard:

1) how many people living in the house
2) age
3) sex
4) race
5) Hispanic origin
6) relationship status
7) do you own/rent
8) telephone number
9) do you live here

It took me all of 10 minutes to fill out for all three of us in my household. The race/Hispanic question threw me for a loop simply because it said Hispanic is not a race, so i had to put down other and obviously filled out the Hispanic section.

All this being said, what kind of demographics will we as planners be able to get with such a short form and lack of a "long form" that was sent out to various, but not all household in 2000 that gave planners a wealth of information such as income, employment, industry type, disabilities, commute times and distances, unit types, etc. Why wasn't a long form sent out this go around?

Some folks will point out that the purpose is simply a head count, nothing more nothing less, however there is great value to the data collected to planners and others to help build a snap shot of the April 1, 2010 as well as shape decisions in the future. What say you.

[Cloverhill]

All I can tell you is that we were one of the fastest growing counties on and off for the decade and our view of ourselves could (will? must?) shift after the results. We've got a pretty good idea of where we are, but is there a surprise lurking in the data?

I think that a big change is going to be on a state rep level. The shift, statewide, in population is going to really give us more influence in state government. Since we are a Dillon Rule state, that could give us, for better or worse, more control.

I am guessing that data such as income, employment, etc., that is missing in this census is available via other sources. We've gotten very good at divining data from the marketplace. I agree, though, that we will miss the 'official' version of the data and always question the 'derived' data at some level.

[WSU MUP Student]

I think that it's a good thing that the Decennial Census is getting back to its roots of being a head count for apportionment purposes.

While it's relatively new and doesn't have all the bugs worked out yet, the American Community Survey will provide the demographics that the Decennial Census will no longer provide. IMO, the fact that the information in the ACS will be released for most geographies with a lot more frequency than every 10 years (yearly for figures based on 1-year, 3-year, or 5-year averages based on an area's population) far outweighs the trade-off in margin of error.

As somebody who is constantly being asked for housing tenure statistics, per capita income, educational attainment, etc. for small communities, I would much rather be able to give numbers that are only a year or two old than something that is up to 12 years old even if I have to provide the caveat that there might be a greater margin of error. It is now 2010, it will be 2012 when all of the 2010 Census information is released. By that point, if you are still using information from the 2000 Census, how accurate can the information be anymore?

[mgk920]

I'm kind of wondering how many will check 'other' and write in 'American'.



That aside, I agree that the Census is mainly (and, Constitutionally, *ONLY*) for getting an accurate head count for apportioning USHouse seats. It is, of course, also used to apportion representative seats in the state and local governments - thus the need to get as accurate and complete count as possible. The information on names and family relationships was, IIRC, originally used to prevent double-counting and later on become of immense use to deep descendants in family history searches and related research.

The more detailed statistical stuff should be gatherable from other sources, too.

BTW, the detailed records of the 1940 USCensus will be released to the public in about two years (its 72 year confidentiality period will have expired). An interesting bit of trivia here - nearly all of the detailed records of the 1890 USCensus no longer exist - they were destroyed in a fire. http://www.archives.gov/genealogy/census/1890/1890.html

Mike

[JimPlans]

The race/Hispanic question threw me for a loop simply because it said Hispanic is not a race, so i had to put down other and obviously filled out the Hispanic section.

Oops. You are allowed to choose a race AND an ethnicity. So, you can be white and hispanic, black and hispanic, etc. Of course, many people fill out the form just like you did, as the Bureau has never been clear about how these two questions should be answered in conjunction with one another.

All this being said, what kind of demographics will we as planners be able to get with such a short form and lack of a "long form" that was sent out to various, but not all household in 2000 that gave planners a wealth of information such as income, employment, industry type, disabilities, commute times and distances, unit types, etc. Why wasn't a long form sent out this go around?

The "long form" is the American Community Survey. It replaced the old-fashioned long form in the mid-2000's to allow for continuous measurement of population and demographic changes (something that transportation planners had been asking for for years). We will get data for the block group-level and above later this year (the "five-year" data). Of course, this will represent an average of five years worth of ACS surveys, as opposed to a point-in-time measurement represented by the decennial Census, but most users shouldn't have too many problems with the change.

Unfortunately, there will be higher margins of error for small areas under the ACS, because (1) the total number of surveys sent out (about 3 million per year) represent approximately the same number of households who received the long form in 2000, but there are now more households, which reduces the sample size, and (2) large amounts of households fail to respond to the ACS (only 1,931,955 of 2,894,711 households in the sampling frame actually responded to the ACS in 2008, for example), further reducing the sample size.

MPOs are very upset because there will be a huge amount of supression for small areas in the CTPP for 2010, and there doesn't seem to be any way to fix it because of the way the ACS has been conducted.

[stroskey]

edit - message is NOT too short!

[stroskey]

edited! message is NOT too short!

[DetroitPlanner]

Hispanic can be defined as simply someone who speaks a form of spanish or comes from a country that was once predominately spanish. Hispanic is not simply a race as if you are from europe hispanics are from spain and white, if you are from florida they could be from a carrriibean island and be black, if you are from california, they are native american, and if you are from hawai they could be phillapino. Personally, I think hispanic is not a very useful term as there can be very different physical and cultural features between this varied group that is forced together.

I give race another two censuses, then it will be dropped all together. While I hate the Canadian term visible minority, it can be more reflective of reality. Though in another 20-30 years caucasian-europeans will be as much of a visible minority as everyone else.

I got the short form in the mail today. I am filling out 12 people because Michigan needs all the people it can get! (Kidding!)

[WSU MUP Student]

I got the short form in the mail today. I am filling out 12 people because Michigan needs all the people it can get!

Talk about being a Detroit booster!

(pun intended)

[Gotta Speakup]

The race/ethnicity questions are set by the federal Office of Management and Budget and are now standard throughout federal surveys. They are very confusing to the public. Most of the people who answer other or put down that they are multiple races are Hispanic. This drives the Census and other agencies nuts and they complain all the time.

The government has tried various ways to capture the concept of Hispanic/Latino. They once asked if people are spanish speakers, but his didn't work because most Latinos in the US dont speak Spanish (take that! those of you who think they are not assimilating). They tried using an algorithm that identified Spanish last names but gave up on that because of the confusion with Portuguese names. So they use the two question system now.

[Woolley]

What about religion (jedi) and how fat you are!

I always thought the census was done as a worldly unit (i.e. same year). We do ours in 2011.

[Mud Princess]

I was disappointed when I received the ridiculously short form in the mail. I understand that it's mainly a head count for apportionment purposes... but it will provide so little information for future historians and genealogists.

I've been using Census records to learn more about my grandparents, great-grandparents, etc. through 1930 (by law, individual records must be kept confidential for 72 years - the 1940 Census records will be available in 2012). What will our future descendants see? Just our name, sex, race, and age?

Of course, all of our Facebook, MySpace, and LinkedIn profiles will probably still be around to embarass ourselves in absentia.

[mgk920]

Well, the questions being asked about names, ages, birthdates and relationships of people in the residence are pretty much the same ones that have been asked since the first Census.

IIRC, everyone was sent the same forms this year.

The USA has always done its headcounts in years ending in '0'.

Mike

[Salmissra]

Got ours fills out and back in thte mail yesterday. Took hubby less than 10 minutes to complete, and I just bemoaned that we only got a short form.

[Otis]

That aside, I agree that the Census is mainly (and, Constitutionally, *ONLY*) for getting an accurate head count for apportioning USHouse seats.

How do you get that the Constitution limits the census to ONLY being an enumeration? The Constitution says that there shall be decennial enumerations, but it puts no limits on.

[mgk920]

How do you get that the Constitution limits the census to ONLY being an enumeration? The Constitution says that there shall be decennial enumerations, but it puts no limits on.

USA Constitution - Article. I. Section. 2. (Paragraph 3) says:

"(Amended sentence referring to apportioning representatives among the several states based on actual counts of people residing within those states). The actual Enumeration shall be made within three Years after the first Meeting of the Congress of the United States, and within every subsequent Term of ten Years, in such Manner as they shall by Law direct. (Sentence referring to the state-by-state apportionment of the First Congress.)"

From my read of that passage, it only refers to getting a decennial headcount in order to reapportion the US House of Representatives.

Mike

[Linda_D]

USA Constitution - Article. I. Section. 2. (Paragraph 3) says:

"(Amended sentence referring to apportioning representatives among the several states based on actual counts of people residing within those states). The actual Enumeration shall be made within three Years after the first Meeting of the Congress of the United States, and within every subsequent Term of ten Years, in such Manner as they shall by Law direct. (Sentence referring to the state-by-state apportionment of the First Congress.)"

From my read of that passage, it only refers to getting a decennial headcount in order to reapportion the US House of Representatives.

Mike

The bolded part gives Congress pretty much the right to ask what it wants in getting a head count for apportionment.

[Otis]

The bolded part gives Congress pretty much the right to ask what it wants in getting a head count for apportionment.

Exactly my point. The enumeration clause is a requirement, not a limitation.

[Veloise]

A few weeks back: The Letter. "Expect the census form soon," badee badee badee.

Last week: The Form. Multiple pages of blanks; there is one human living in my household. (I am not, however, the "head of household"; that honor is a continuing battle amongst the furry roomies.)

Yesterday: The Postcard. "We sent you a census form and you're required by law to fill it out."

Sheesh.

[chocolatechip]

A few weeks back: The Letter. "Expect the census form soon," badee badee badee.

Last week: The Form. Multiple pages of blanks; there is one human living in my household. (I am not, however, the "head of household"; that honor is a continuing battle amongst the furry roomies.)

Yesterday: The Postcard. "We sent you a census form and you're required by law to fill it out."

Sheesh.

I got the exact same things. I think they just send the third postcard automatically. That's only about $40,000,000 in postage.

[JimPlans]

I got the exact same things. I think they just send the third postcard automatically. That's only about $40,000,000 in postage.

The Census Bureau hopes that the three mailings will up the response rate. According to them, every percentage point increase in response rate saves them $75 million in costs that would be paid to enumerators.

There will also be another mailing of replacement questionnaires to unresponsive census tracts, where every house will get a second form if they responded or not (blanket mailing), and to middling-ly responsive tracts, where only non-responding hosueholds will get a new form (targeted mailing). This should happen around April 10th. Only then will the enumerators be set loose for nonresponse followup.

BTW, you can see the current response rate for your area at http://2010.census.gov/2010census/take10map/. These numbers are updated daily.

[Veloise]

Another census form arrived today.

Seems like the folks who really need reminding might get confused and fill it out again. (I almost did, thinking that perhaps they'd lost the original.)

[Masswich]

I was disappointed when I received the ridiculously short form in the mail. I understand that it's mainly a head count for apportionment purposes... but it will provide so little information for future historians and genealogists.

I've been using Census records to learn more about my grandparents, great-grandparents, etc. through 1930 (by law, individual records must be kept confidential for 72 years - the 1940 Census records will be available in 2012). What will our future descendants see? Just our name, sex, race, and age?

Of course, all of our Facebook, MySpace, and LinkedIn profiles will probably still be around to embarass ourselves in absentia.

The "Long Form" is no more- replaced by the rolling American Community Survey that is designed to provide more up-to-date data, especially in larger municipalities. The tradeoff was somewhat less information in ACS in return for updated data every 1 or 3 years, depending on the size of the community.

[boiker]

USA Constitution - Article. I. Section. 2. (Paragraph 3) says:

"(Amended sentence referring to apportioning representatives among the several states based on actual counts of people residing within those states). The actual Enumeration shall be made within three Years after the first Meeting of the Congress of the United States, and within every subsequent Term of ten Years, in such Manner as they shall by Law direct. (Sentence referring to the state-by-state apportionment of the First Congress.)"

From my read of that passage, it only refers to getting a decennial headcount in order to reapportion the US House of Representatives.

Mike

The census was also mandated for the apportionment of federal tax dollars collected via federal property tax. Since we don't have that, the US Code directs other tax money to be apportioned by population.

Lastly, the phrase "as they shall by law direct." Gave the 1790 lawmakers the latitude to decide the manner and level of thoroughness that the Census should be conducted. The debates in Congress in 1790 were nearly identical to the debates in 2010 with regards to privacy and the extent of the Census permitted by the Constitution.

[mgk920]

The census was also mandated for the apportionment of federal tax dollars collected via federal property tax. Since we don't have that, the US Code directs other tax money to be apportioned by population.

Lastly, the phrase "as they shall by law direct." Gave the 1790 lawmakers the latitude to decide the manner and level of thoroughness that the Census should be conducted. The debates in Congress in 1790 were nearly identical to the debates in 2010 with regards to privacy and the extent of the Census permitted by the Constitution.

It is my sense that the questions relating to names/ages/birthdates/sex/family relationships/etc were first used as a cross-check to prevent double counting. These were asked from the very beginning and now have to side-benefit of being a treasure-trove of useful information for those doing historical research, including family history research.

Besides, the decennial census is no match whatsoever to the income tax in intrusiveness and violations of personal privacy - and many/most of those who are totally aghast at the Census Bureau for asking their few simple questions seem to have no difficulty at all baring EVERYTHING to the gummint at income tax time.



Mike