On The Cusp of an Historic Day

Print This Post Print This Post

Written by: Director Robert Groves

Almost every week the Census Bureau releases statistical information to inform the society about how it’s doing. Sometimes the statistics reflect economic activity in the US; sometimes, the statistics report on the status of the American public.

Once every decade there is a unique release of statistics – the state population counts based on a just-finished decennial census. Indeed, this particular event evoked the name of our general-purpose statistical agency – the “Census” Bureau.

For this decade, the day will be December 21, 2010.

From my perch right now, I perceive this event as the culmination of millions of hours of people’s time. This data release includes the time invested by each respondent who spent about 10 minutes completing the 2010 census form. It includes over a million census staff members who visited housing units throughout the country to collect census data. It includes millions of people working as our volunteer partners, getting the word out to their communities about the value of participating in the census.

Congress has charged the Census Bureau with doing the arithmetic to compute the number of Members in the U.S. House of Representatives, and we’ll present this report on Tuesday. I encourage you to view this nifty video about how the computation of apportionment is done.

The event announcement will not contain much information; we will announce the official national population count. We will report the state resident population for all states, the District of Columbia and Puerto Rico. We report the total populations for these groups, reflecting the federal agency staffs that are assigned to non-U.S. stations for which their parent agencies have provided counts.

You can see the kind of statistics that we will release on this interactive map.

Tuesday is an important day for the country; we are grateful for this opportunity to serve it.

This entry was posted in Apportionment, Measuring America. Bookmark the permalink.

32 Responses to On The Cusp of an Historic Day

  1. Amanda says:

    Will a break down of population for each city in each state be available?

  2. Thomas DeMarco says:

    Question on the aportionment results – how does a state that is losing a
    seat(s)determine which rep(s)lose said seats and when is that effective. Is it a uniform practice or does each state do its own thing? thanks.

  3. The local area data from the 2010 Census are released on a flow basis as part of the redistricting file, beginning in February 2011.

  4. Drew says:

    Will you be releasing/hosting any sort of new data APIs (json or otherwise)?


    I never reeived the cencus form at all.


    I never reeived the cencus form at all.


    I never reeived the cencus form at all.

  8. Each state will redraw its own boundaries based on the redistricting data that the Census Bureau makes available starting in Feb. 2011.

  9. Lynn says:

    Seth @ Census Staff,
    Thanks for the reply (to Amanda) about when the breakdown of population for individuals cities will be. That’s exactly what I was going to ask!
    When you pull up that info, the latest data there is from 2009.
    Maybe someone can put a notice on that page that the new/updated information will be released beginning February 2011?
    Thanks!! :0)

  10. Cristian says:

    When will be available the data on foreign born population? And as PUMS?
    Also in February?

  11. We don’t have anything concrete to report right now in this regard.

  12. John Wekell says:

    I tried using your interactive map for Washington State and found that I could not get the arrow to respond to Pierce County, it hits on all surrounding counties such as King and Thurston county.

  13. There will not be data on the foreign-born pop. available from the 2010 census. Such data is available annually as part of the ACS. If you are referring to the 5-year ACS PUMS, it is scheduled to be released in early 2011, although an exact date has not yet been made public.

  14. Ken in Camarillo says:

    What is the algorithm for apportionment? I tried: “assume a population per district, divide that into the population of each state, round each answer to get seats. If the total seats allocated is too high or too low, adjust the population per district number until the total comes out to 435.” Using this I get 14 for NC and 1 for RI. I see your numbers are 13 and 2 respectively. Is there another test such as ratio of representation vs ratio of population to avoid anomalies? Or limiting extremes in Population per district from state to state?

  15. bob says:

    I see that North Carolina lost out on a Congressional seat by 15,000 votes. What state beat out North Carolina for that last seat?

  16. The 435th seat was awarded to Minnesota.

  17. Ken in Camarillo says:

    Outstanding method! It balances the two cases:
    If this state doesn’t get another seat, it has a high population per seat.
    If this state gets another seat, it has a low population per seat.
    The balance is that effectively it is the geometric mean of the population per seat for the two cases. Whichever state has the highest geometric mean at any point in the process gets the next seat.

  18. Bob says:

    How/when will we know the gay/lesbian portion of the census?

  19. cgseife says:

    A question and a comment.
    The question: what is the scheduled date for the release of the Census Coverage Measurement results?
    The comment: I find it interesting that we’ve stuck with the Huntington-Hill algorithm for apportionment even though its main advantage over other algorithms seems to be that that it’s guaranteed not to cause an “Alabama Paradox” (am I correct about this?) If so, since we’re not adding any seats, we’re automatically immune from the paradox… so why not choose another method that minimizes the least-squares error?
    For example, in the 2010 Census apportionment, Huntington-Hill gave +4 to TX and -1 to MO, but, in fact, +3 to TX and no change for MO is actually better from a least-squares point of view, so is arguably more “fair.” (And, as it happens, the old Vinton’s method gives exactly that apportionment.)

  20. CC Barrett says:

    Why weren’t county population totals released today, along with the state populations, as had been in the past?
    When will they be released?

  21. YESnack says:

    In the words of Mark Twain “There are three kinds of lies: lies, damned lies and statistics…” 🙂

  22. Wyatt says:

    Hey Seth,
    Will detailed data sets be available via direct file access (just like the data sets from the 2000 census are currently) beginning in February, as well? Is all the detailed data released on a flow basis?

  23. Net coverage estimates are scheduled for release by 7/31/2012. Components of error (separate estimates of undercount and overcount) are scheduled for 9/28/12.

  24. The local area data from the 2010 Census are released on a flow basis as part of the redistricting file, beginning in February 2011.

  25. Congratulations and thanks to all those who have help to gather and compile these all important numbers for our latest Census. These numbers are very important to us all for proper representation and provides much needed information for individuals, businesses, and nonprofit agencies throughout America.
    This information is crucial for grant writers who rely on this information to properly describe the challenges face within various communities so that new innovative solutions can be developed. Finally, we really like the short 2 minute video that explains the process for the computation of apportionment. Well done!

  26. Wyatt,
    The redistricting data sets will be available on AFF and for download. You may wish to look at
    On this url, you will find the release schedule for the standard Census 2010 data products. In addition, you can click on the links for the redistricting data, the demographic profile, and Summary File 1 to see the planned table shells.

  27. CC Barrett says:

    Why so late?!
    That’s ridiculous!

  28. We are working to adapt census data processing to tabulate same-sex marriages. The Census Bureau will release 2010 Census same-sex marriage data two ways – in edited and unedited forms. The edited data are still needed by the federal agencies that use the information for their own programs’ calculations.
    A much better source for measuring the number of marriages – both opposite-sex and same-sex – is the Census Bureau’s American Community Survey (ACS). The ACS is part of the official census and its questions were asked on the “long form” in 2000 and in
    earlier decades. The ACS is the nation’s largest household survey and is conducted throughout the decade. To see the latest (2009) data, go to the “Families and Living Arrangements” section of our web site (http://www.census.gov/population/www/socdemo/hh-fam.html — scroll down to “same-sex couples”).

  29. grooft says:

    I just watched the Census Apportionment video on YouTube. As a District of Columbia resident, (one of 601,723), I must take issue with the statement that apportionment happens “just like our founders intended”.
    You may not know (or care) that the 601,723 residents of DC have no apportioned representatives in our nation’s government. The residents of DC had electoral representation in the early years of our country.
    At least this census shows that there is only one state, Wyoming, that has FEWER people than the District of Columbia. Of course, Wyoming has 2 Senators and a single Representative.
    Thanks for the good work of counting. But please, inform others that DC has NO representation.

  30. Ken in Camarillo says:

    I believe you’re defining the error as the difference between the population of a district and the mean population. In other words the error for a district 10,000 people under the mean is the same as the error for a district 10,000 people over the mean.
    I think the system in use effectively defines the error in terms of the ratio between a district population and the mean population. Thus a district whose population is 66.6% of the mean has the same error as a district whose population is 150% of the mean.

  31. Ken in Camarillo says:

    The original Constitution gives representation to the States only, thus DC had no representation in Congress or Presidential electors by operation of the plain wording of the Constitution. Because this is the case, I find it hard to believe “The residents of DC had electoral representation in the early years of our country.” Of course, you may be referring to the period of time before DC was formed, when the area was still part of States. Referring to residents of the area during that earlier time as residents of DC is clearly a misrepresentation.
    In 1961 the 23rd Amendment was ratified which provides DC with Presidential electors. This shows that it is well understood that the representation status of DC is defined in the Constitution, and if that status is to be changed, a Constitutional Amendment is required.

Leave a Reply

Your email address will not be published. Required fields are marked *