FAQ: Population Estimates

  1. What is the latest population estimate for <whatever-area-I'm-interested-in>?
    Use the Population Finder feature on the Census Bureau home page. This is on the right side of the page, just below the "population clocks" (which provide a continuously updated estimate of the population of the world and the U.S., if those are of interest). Type the name of the area (city, town, county of ZIP code) and select a state from the drop-down; then click on the GO button. You may get back a menu of multiple entities that match (try typing "washington" without specifying a state and you'll get back dozens of choices and it will take over a minute to do the search). Clicking on the "correct" result displays not only the latest estimate, but also the population counts from the last two censuses. (If you enter a ZIP code all you will get is the population for that area from the most recent (2000) census, assuming it is a residential ZIP that existed at the time of that census. Neither the Census Bureau or any other government agency provides post-2000 population estimates for ZIP codes.) You also get links that will take you to reports showing data for all entities of the same type in the same state; e.g. if you get data for Madison County, Illinois there will be a link to a report for all counties in Illinois (alphabetical by name or ranked-by-population sorts).

    Users interested only in Missouri and only in numbers for cities and counties and are Excel users might find it more convenient to check out our curmoests.xls Excel workbook file. It gets updated each year in the spring and summer as the estimates are released by the Census Bureau.

  2. How many hispanics have moved into my [state / county / city] since [2000 or other year] and what are the latest estimates for hispanics?
    These data are not available at the city level, but are for state and county. We tried to find a way to get answers to such questions from the Census Bureau web site but were not successful. But we have a solution on our MCDC web site. If you go to the MCDC home page (mcdc.missouri.edu) and follow the navigation link (in the navy blue box at the left) to Population Estimates and then on that page to bullet item 4 (this item number could change -- look for the string "by age, race, sex and Hispanic origin"), the link at the end of this item to MCDC reports for any state in the U.S.. This takes you to our

  3. What is the trend in the elderly population of our area?
    Assuming your "area" is a state, a county (or group of counties) or a metropolitan/micropolitan area then the answer to this question is almost the same as for the previous one regarding hispanic population. If you follow the links provided there, you'll note that the reports that broke down the population estimates by hispanic/non-hispanic also contain numbers for age categories. The actual age cohort detail varies with the two sets of reports. If you choose from the first set of reports on the intermediate menu screen ("Estimates for Each of the 4 Demographic Dimensions Separately") you get estimates for 5-year age cohorts, ending with "85 and over". That may be just the kind of detail you want. On the other hand choosing a report from the second set ("Estimates for Major Race (Alone) and Hispanic by Age and Sex") gets you data for the 65-and-over cohort (with no further detail), but gives you this cohort broken down by race, hispanic origin and sex.

    Alert readers who have been following the links to the demographic-detail estimates reports in this and the previous question will note that these reports also take care of common questions regarding trends for race and gender as well as for Hispanic and age.

  4. How can I get data about my state and its counties that shows me how our population has changed since 2000 including a breakdown of the components of that change (births, deaths and net migration.)?
    These data are available from a number of sources. We'll point you to what we feel may be the easiest for most users to access.(The "best" way to access such data depends on whether you just want to see the data or whether you want to do some custom processing of the data.)

    Start at the Census Bureau's Population Estimates page (http://www.census.gov/popest/estimates.php). This page has a very slick user interface that some users don't recognize. Most of the functionality comes from using the brown menu bar across the top. You move the cursor over the "Estimates Data" entry and a drop-down menu appears. There are actually multiple ways to get to the same thing; you can start by choosing your geography type or you can select what they call "by event", to get data for births, deaths and migration. (This is how it should work, but we discovered on a recent visit that if we started with choosing the "births and deaths" event, it would not allow us to get data for the latest year by county. This is probably a temporary oversight by the Bureau; they do have a menu entry that lets you get the data by county for the latest year when you start by selecting the "counties" entry as we shall see next.) So, choose "counties" from the "By geography" section. A page appears with background information on what kind of data are available at the county level. To get to the data (assuming you don't want to "download entire data set" yet - that is for very serious users) click on the "--choose a popular table--" drop down and then choose the item "Cumulative Components of Population Change, 2000 to 200X" (where X is the latest year; it was 2005 at the time this was written.) After making your choice you have to click the GO button. This leads to one of those 3 x 17 hyperlink matrices with one entry per state. You can access the data in Excel or CSV format. Depending on how your browser is configured it may not make any difference (most IE users who have Excel on their PCs will have their browsers configured to automatically invoke Excel and import a CSV filetype when you click on such a file.) So click on the Excel file for your state (or, you can always right click and save/download the file and then access the local copy using Excel outside your browser, giving you more complete Excel functionality). This spreadsheet does a nice job of giving you just the components of change data, but it is too bad they do not include a column with the base and current total population estimate. That data is readily available elsewhere (from other choices on the "Popular Tables" menu, for example), but it would be nice to include it here.

  5. How many people aged 7 to 12 [for example] live in my county?
    This is not a specific request but a class of requests. It involves somebody wanting estimates for a very specific age range. In this example it is 7 to 12, but it could also be 6 and under, 72 and older, or 16 to 23 (for example). The geography does not have to be a single county; it can be multiple counties or states. The Census Bureau does generate estimates by single years of age at the county level, but they do not advertise this fact and we have never found anything regarding such estimates on their web site. However, they did a special estimates tabulation for the National Center for Health Statistics (NCHS) that are county level and by single years of age. You can download these data from the NCHS web site (http://www.cdc.gov/nchs/about/major/dvs/popbridge/popbridge.htm) but this requires custom downloading and processing. The MCDC has put these data up in our data archive and data can be extracted using a custom web application titled Populations Estimates by Age. It actually does more than just age; it also handles the 3 other demographic characteristics - race, hispanic origin and sex. All of these can be cross-tabulated with age (but not with each other, i.e. you can get age by race and age by sex, but you cannot get race by sex with this application).
    As far as getting it for a specific 3 counties, this application does not at this time allow selecting counties. But you can select the state and get output for all counties in that state, and then simply ignore / throw away what you don't want.

  6. Can I get a breakdown of my community by age for the hispanic and non-hispanic population?
    See the previous question/answer. The web application that lets you extract data for specific age ranges, also lets you do so with detail by sex, race and hispanic origin. The only limitation is that your community must be comprised of counties. There are no such detailed estimates available for smaller geographic units such as places (cities) or ZIP codes. (At least not in the public domain; there are plenty of data vendors who will be more than happy to sell you their estimates.)

  7. How many persons in my county aged 5 to 17 are below 185% of the poverty threshold (and thus eligible for the free and reduced lunch program)?
    This is an example of a question that we do not consider to be part of population estimates. This question deals with much more detailed data, the kind that need to be collected in a special survey, such as a decennial census, the current population survey or the Census Bureau's American Community Survey. This particular question may be answered (depending on the size of the county) using the latest American Community Survey datasets. This is an example of the kind of question that is frequently asked and rarely answered. (In case you really care: You need to go to the American FactFinder web site and access the latest ACS data set. Then see if your geographic area is available. Choose the Detailed Tables option and search for a relevant table using keywords such as age and poverty ratio. The answer is B17024 (that is the table that comes close to answering your question - you'll have to settle for age 6-17.)

  8. What is the latest estimate of the population of (these) ZIP codes?
    The Census Bureau does not publish any estimates at the ZIP code level. (If they did, they would be for something called "ZCTA"s -- ZIP Census Tract Areas -- which are the Bureau's geographic estimates of true ZIP codes.) There are dozens, if not hundreds, of private sector firms who will be more than happy to sell you all kinds of estimates at the ZIP level. Some even offer free online access to current population estimates for ZIP codes as a teaser to get you to purchase more data. To see what we mean try using your favorite web Search engine and look for something such as "Demographic updates ZIP".
    (Update, 2010): City-Data.com is a good site for getting ZIP data that includes a recent population estimate, among many other things. We visit it quite a bit, and have not yet been asked to pay a fee.

  9. Can I get the latest population estimate for a circular area, such as a 10-mile radius of my house?
    Yes, almost. See our CAPS web application. Most of the data produced by this application is 2000 census data, but it does include a more recent total population estimate for the circular area(s) summarized (as well as a 5-year projection). Details regarding the source of these updates are available in the online Usage Notes page linked to from the CAPS application.

  10. How reliable are the Census Bureau's estimates?
    See OMB's formal review of the Bureau's intercensal estimates program. The bottom line: basically okay, but they have had problems with the international migration component, which has meant (in the lates 1990's, at least) some significant underestimating of the hispanic population in some areas.

    Population estimates for places (cities) employ a methodology that assumes a constant vacancy rate and persons per unit. This leads to consistent overestimating of small places in rural areas which typically are losing population because of an aging-in-place population that leads to an increase in vacant units and fewer persons per household.

    Watch out for successfully challenged estimates. The Bureau has a challenge process that allows jurisdictions to submit evidence showing why the Bureau's estimate may be wrong. Usually these challenges get resolved in the spring or summer of the year in which the estimates are published. But no publications issued in that calendar year reflect the successful challenge; they all show the original, supposedly incorrect, figures. To learn more about challenges and see a list of successful challenges go to the Bureau's estimates page and choose Estimates Topics / challenges off the menu bar / drop down to get general information regarding the challenge process; choose Estimates Topics / revised estimates to see successful challenges for the current year.

  11. Where can I go to get an estimate of what my county's population may be in 2020?
    In the spring of 2008 the Missouri Office of Administration (where Matt Hesser, the state demographer works) released a set of projections for Missouri counties by age and sex for every 5 years out to 2030. These were the first such official projections released in over 10 years. See the bullet item related to projections on the MCDC's estimates page.

    By the way: technically, such a number (your county's population in 2020 as seen from 2008) would not be called (by demographers, at least) an "estimate", but instead a "projection" since it deals with projecting the future. Projections are somewhat difficult; it is generally considered much easier and safer to do estimates, which usually deal with the recent past and can be based on hard information that allows us to have some confidence in their general validity. Projections require assumptions, and often involve complex mathematical models that try to identify trends and project them into the future. The Census Bureau does not do projections except at the state and national level (see the Bureau's projections page, which is linked to from the top of their Estimates page.)

  12. Why is the estimate for the total population for my county in the 2006 ACS data identical to number I see on the FSCPE official estimates files, but the estimate for my city is more than 5% different from what it is in the official estimates?
    Because of the way the Bureau adjusts the weights used in tabulating the ACS data. It basically "rakes" the ACS tables so that the total population (as well as some age/sex cohorts of that population) at the county level matches the figures in the latest official estimates. This is not done at the place (city) level. The result is differences (sometimes rather significant) in the total population figure reported in the ACS vs. the estimates program.

    If you live in a city of less than 65,000 population then you do not have any ACS data to compare with the estimates. In 2008 we'll be getting estimated data for cities (and all other geographic areas) of at least 20,000 and these numbers will be statistically flawed in the same way the larger place numbers are now. It gets worse as the geographic areas get smaller. The numbers that shall come out in 2010 for areas of any size (including all census tracts, ZIP codes, and very small cities) will all be distorted by the Bureau's attempt at making the county level numbers match. (This is an opinion rather than a fact, but a lot of people share it.)