What's New at MCDC in 2006

Quick Links ACS Profile Reports (2006) SF3 Profiles (2000) Circular Area Profiles SF3 Trend Reports Detailed SF3 Profiles IRS Migration Profile Missouri County Data Map Applinks Master Menu MABLE/Geocorr 2K RxA Cure for the Common Codes American FactFinder 2006  January || February || March || May || June || July || August || September || November || December December 2006 2000 Census Data For PUMAs -- These new data are available in the data archive as well as in the format of our standard 2000 census demographic profile reports (dp3_2k series). To access these go to the main menu page for U.S. geographies at mcdc2.missouri.edu/websas/dp3_2kmenus/us/ and choose PUMA's (or, take this shortcut directly to the main menu page for PUMAs). Follow the linked menu pages to a state and then to a PUMA within that state. If you are not familiar with PUMA geography we suggest you see the discussion of these entities as Item #10 on our Ten Things to Know About the American Community Survey page as well as our topics page devoted to PUMAs. Suffice to say here that there are maps, reports and cross-walk files available that can help you get a handle on what and where these PUMAs are. After displaying a profile (which is based on the standard extract data only) you can access the data in the complete SF3 tables using the drill-down "parent table" links that appear in the "SF3 Table" columns of the report. Instructions for how to code a URL to access any (1 or more) complete SF3 tables using the sf3tabgen utility module can be found at the very end of the Usage Notes page for the dp3_2k application. (For example, here is a link that will display tables P50 and P51 for Missouri PUMA 01602.) These data (PUMA-level SF3 tables) are not published by the Census Bureau (and are not available via American FactFinder). The MCDC had to aggregate/allocate small area (block group) level data from Summary File 3 to create data at the Public Use Micodata Area (PUMA) level for the entire United States. These are the PUMAs as used in the 5% PUMS files for 2000 (as opposed to the larger "Super PUMAs" which are used as well and are supersets of these smaller PUMAs). Nationwide, there are 2071 of these PUMAs for 2000. The American Community Survey publishes data for these areas each year (with 2005 data already available), since the ACS threshold population is 65,000 (to get data each year) and all PUMAs are required to have a minimum population of 100,000. We created the uspumasph dataset (containing all P and H tables) within the sf32000 data directory in our data archive (we have not created any PCT or HCT tables yet). This dataset can be accessed via our Uexplore/Dexter software using this link. The corresponding standard SF3 extract (about 400 variables rather than the 5500+ on the full tables dataset) has been created as dataset sf32000x.uspumas. This Dexter-accessing-the-source-dataset link also appears at the bottom of the dp3_2k profile report pages. Ten Things to Know About Urban Vs. Rural -- This web page, the second in our Ten Things to Know About ... series, looks at the current definition of the urban/rural twins as formally defined and used in the census. It includes some fascinating facts (such as what percent of the U.S. and Missouri populations live on what percent of the land), tips on where to find data related to the urban/rural portions of the population, and concludes with a picture of rural population in Missouri. November 2006 Population Estimates by Age - This new dynamic web application lets you generate custom extracts of state and county level population estimates for either pre-defined or user-specified age intervals. The data used in this application are the special "bridged" race estimates from the National Center for Health Statistics (and the Census Bureau) with breakdowns by single years of age, race (just the 4 special bridged categories), hispanic origin and sex. You can choose your geography (by state and level), your time periods (1990 to 2005), your own custom age groupings (under 11, over 27, 18 to 64, etc.), and demographic items of interest (race, hispanic and gender detail). Population Estimates Page Revamped - If you have not recently (or ever) visited the MCDC's Population Estimates page (7th item down in the navy blue navigation box to the left on this and most other recent MCDC web pages), you might want to check it out. It has been thoroughly revised and enhanced. It contains links to the best stuff we know about related to estimates. Only our more current stuff is out there now; a big part of the upgrade was getting rid of stuff that was becoming out of date. The page also contains a link to a Trivia Challenge page and an invitation to express interest in a training session. September 2006 New Web Page Regarding the 2005 ACS -- Ten Things to Know about the American Community Survey (2005 Edition) is our attempt to make sure users are aware of some of the more important potential "gotcha"s that go with this important new data collection. American Community Survey Data for 2005 (Wave 2) The Census Bureau has released "Wave 2" of the American Community Survey data based on surveys collected in calender year 2005. These new data can most easily be accessed via the American FactFinder web application. The new release includes an extensive collection of base tables and profiles related to economic data -- income, poverty, employment, etc. The Bureau also released ACS Public Use MicroSample (PUMS) data files for people wanting to create their own tables from sample microdata. These can be downloaded from the ACS section of American FactFinder, or can be accessed via uexplore/dexter from our acspums data directory (filetype). The MCDC is still working on a document that will provide a good overview of this very important -- but in some ways rather limited -- new data source. Users should be aware that the Census Bureau is strongly cautioning users from doing economic trend analysis with these data because the income being measured in the ACS is not exactly comparable to what was measured in the 2000 census. All of the (national) economic trends that the Bureau cited last week in their press release (relating to things such as poverty rates and median incomes) were based on data from the most recent Current Population Survey rather than from the ACS. (Unfortunately, because of its relatively small sample size the CPS cannot be used as a source for most sub-national geography, including most states.) The MCDC has been downloaded the new ACS raw data files and created versions of the data in our data archive. You can see these data by accessing the acs2005 data directory via uexplore/dexter. Note particularly the issues paper prepared by the New York SDC summarizing comparability of decennial and ACS 2005 data. One of the more important resources we have available is in the profiles subdirectory where we created datasets with the dp1 (Demographic), dp2 (Social) and dp3 (Economic) summary profiles for all geographic areas. We intend to create enhanced versions of these profile datasets with more mnemonic variable names and with cumstom percentages added, but for now users can access these important datasets in their original form. They are (just) small enough (i.e. have few enough variables/columns) to fit in an Excel spreadsheet. For those wanting more detail we have also created a basetabs subdirectory with individual datasets for each base table (Waves 1 and 2 only, thus far). These are the detailed "B" and "C" tables that you see from American FactFinder. For example, for base table B04007 there is a dataset named b04007. It contains 7016 rows, the same number as for all the base table datasets, correponding to the 7016 geographic entities summarized in the 2005 ACS. (To see just what geography is available for the entire U.S. see the Geography.txt file in the acs2005 directory.) The rows are comprised of 8 geographic identifiers (the same on all table datasets) and 20 numeric data items; 7 basic table cells, with 7 corresponding margin-of-error variables and 6 percentage variables (there is no pct variable for the first data cell; for all the other data cells the corresponding pct variable contains that cell value as a percentage of the first cell.) To see what geographic areas in Missouri are summarized on these datasets see MoGeography.html in the acs2005 directory. (To see a comparable report for any other state, access the geography dataset in acs2005 using Dexter and filter using the State variable and specify that you want an output report in html or pdf format.) Finally, we have attempted to combine (merge) the base tables to create datasets (in the main acs2005 directory) that resemble a decenial Summary File type of dataset. We created a series of such datasets based on just the Wave 1 tables. When we tried to do the same thing for Wave 2 we ran into problems because of too many variables in the resulting datasets. We shall continue to examine our options for creating such datasets. If you are anxious to have access to such large collections please use the feedback button (Questions/Comments, below) to let us know. New Report Shows Household Vs. Group Quarters Population Trends A new population estimates report, Trends in Population: Household Vs. Group Quarters, 2000 Census to July, 2005 Estimates shows household and group quarters populations for 2000 and 2005 for all states and counties in the U.S. This pdf file is 119 pages but has a state index making it easy to view the data for an individual state. This report can be useful to users of the new ACS data for 2005, which provides summaries for only the household population, since it gives you some idea of what you may be missing for a given area. You can access the report, along with the corresponding dataset used to create it, from our Population Estimates Reports page. August 2006 American Community Survey Data for 2005 -- The long-awaited release by the Census Bureau of the first large-scale set of data tables based on the American Community Survey took place early today (Aug. 15). This is actually the first of several "waves" of tables to be released based on the data collected in 2005. Per the ACS American FactFinder page: Additional data from the 2005 American Community Survey (ACS) are scheduled to be released late August through November. This is followed by a 2005 ACS release schedule link. That schedule is not obvious on the page, but if you choose a topic from the Base Tables Shells - by topic table on the page it gives you a list of tables related to the topic and each table's release date. The next important release date appears to be Aug. 29, when tables related to income and poverty (among other things) are scheduled to become available. The best way to access these data is via FactFinder, which not coincidentally features a major new release today geared to delivering the ACS data. See our previous note on this subject which appeared here in July. To view complete sets of tables and profiles for geographic entities in Missouri as Excel spreadsheet files (each containing 3 worksheets, 2 of them smaller profiles and the other the complete set of base tables) access the special web directory http://www2.census.gov/acs2005/Tables_and_Profiles/Missouri/. It works for other states as well - just change Missouri to Illinois to access profiles for that state. Or, just drop the state name from the URL (http://www2.census.gov/acs2005/Tables_and_Profiles) and get a directory that lets you choose a state OR a summary level and get access to all tables/profiles for the category. The MCDC has created SAS dataset versions of the base tables for this first wave which can be accessed in our public archive via uexplore/Dexter in the new acs2005 data directory ("filetype"). The actual data tables are stored in the basetabs subdirectory. Each base table is represented by its own data set (333 of them), each containing exactly 7016 rows/observations. The latter is the number of geographic entities nationwide for which there is data published this year, meaning they have met the 65,000 population threshold. The easiest way to see what areas have data is to access the Geography.txt file in the acs2005 directory. Levels of geography for which there are data include states, counties, places (cities), metro areas (CBSA's - metropolitan and micropolitan statistical areas), 109th Congressional Districts, PUMAs (2000 5%), and school districts. The big giant footnote to remember when accessing these data is that they summarize only persons living in households - group quarters was not covered in the 2005 survey. They are being covered in 2006 so the data we get at this time next year should include the entire population. Whether the ACS will continue to include the GQ population in future years (2007 and beyond) depends on congressional funding.   Census Bureau Releases Estimates by Age, Race, Sex and Hispanic Origin -- Detailed population estimates for states and counties for the years 2000 through 2005 were released by the Bureau on Friday, August 4. The MCDC has a compete national collection of these data. See Population Estimates by Age, Race, Sex and Hispanic 2000 - 2005 for access to reports for Missouri, Illinois and Kansas. Or, you can access our web application that takes you to comparable reports for any state in the U.S. If you are interested in accessing these in more detail you can do so using our Uexplore/Dexter software. Go to the popests data directory and look for datasets such as mocasrh05. We have 1 such file per state with county level data and one file, usstatescasrh05, with state level data for the entire country. For those interested in Missouri data by selected age categories (such as Under 5, 18 and over, 65 and over, etc.) and gender, you can access the moagesex05.csv file in the popests directory. This file contains a subset of the data contained in the SAS dataset of the same name (moagesex05.sas7bdat) that can be accessed via Dexter. Note that none of these datasets/reports reflect the City of St. Louis' successful challenge of their total population figure for 2005 (from 344,362 to 352,572). Detailed characteristics numbers that sum to the revised estimate will not be available from the Census Bureau until next year -- just as the figures on this year's datasets for July 1, 2004 are the first to reflect the City's successful challenge to their 2004 estimate. New Senior Report Web Site and Data -- You have probably read or heard about the new Missouri Senior Report, with its in-depth analysis and county level ranking of the state of the state's elderly population. OSEDA (the Office of Social and Economic Data Analaysis at the University of Missouri Columbia, an MCDC core agency) was reponsible for the data and web site development on this project. The data used in the web site reports are stored as part of the MCDC's data archive. You can access these data in the mosenior data directory. (See links under Data and Maps/Download Data Files on the MO Sr Report web site). Population Trend for City of St. Louis Was Incorrectly Reported -- In a highlights article that appeared here in June and early July we reported that the city of St. Louis was "by far the fastest-shrinking Missouri city" over the 2004-2005 estimate period. That figure was based on the original estimate for the city as published by the Census Bureau. It has since been pointed out to us that the City has successfully challenged that number and the new, official estimate for the City for 2005 is 352,572, an increase of more that 6,000 over the original estimate (344,362). Using the new figure the city actually grew by almost 2,000 persons over the year and ranked as the 4th fastest growing city during the period. The MCDC has adjusted the figures in the subcounty estimates datasets mosc05 and ussc05 to reflect the new after-challenge numbers. However, other estimates files such as mocom05 and the newly-released mocasrh05 datasets still do not reflect the challenges. The Census Bureau does not publish new components of change or detailed demographic characteristics that reflect the successful challenges until the following year. In fact, the only place where the Bureau even mentions the fact that there was a successful challenge is on a rather obscure page on their estimates site - see http://www.census.gov/popest/archives/challenges.html. If you follow the links to the Challenge results pages you will see that the City of St. Louis has had a successful challenge in each of the last 3 estimate years. The MCDC apologizes for the original incorrect report. July 2006 Census Bureau To Release 2005 American Community Survey Data in August -- The American Community Survey is about to about to pass a major milestone on its path to becoming America's new primary source of detailed demographic data. In August, the Census Bureau will publish data based upon the ACS survey responses for calendar year 2005. Tables will be published for geographic areas with a population of at least 65,000 persons. For Missouri, this means that we should be getting tables for eight cities (Columbia, Independence, Kansas City, Lee's Summit, O'Fallon, St. Joseph, St. Louis and Springfield) and 15 counties (Boone, Buchanan, Cape Girardeau, Cass, Christian, Clay, Cole, Franklin, Greene, Jackson, Jasper, Jefferson, Platte, St. Charles, and St. Louis county). Data should also be available for metro areas, congressional districts and Public Use Microsample Areas ("PUMAs"). The data will be released in a series of topical "waves" starting some time in August. We do not have a specific set of release days yet, except we have seen one note that indicating the final wave would be released on the last day of the month (Aug. 31). The Census Bureau is working on a major upgrade to its American FactFinder web tool, with most of the changes specifically targeted towards making it easier to find and extract from these new tabulations. They have announced August 15 as the target date for this new FactFinder release. (A powerpoint presentation from the Census Bureau is available summarizing what to expect from the new AFF release.) Since the 2005 survey did not include any group quarters, the data will provide summaries only of the population living in households. (Data to be released in August of 2007, based upon data collected in 2006, will include the group quarters population.) The next major step in the ACS data release schedule is slated for 2008 when we will get 3-year moving average data for geographic areas of at least 20,000 population. This assumes continued funding from Congress, which is not a sure thing. In each of the past 3 years there have been threats to dramatically cut funding for the Census Burea, and hence the ACS. There is currently a proposal in Congress to significantly reduce the Bureau's FY07 budget. One of the results of such a cutback would be to once again eliminate group quarters from the ACS. See the Bureau's impact statement detailing the program reductions that would result from the proposed budget cut. To learn more about the ACS see the MCDC's ACS data directory which currently has very little data but a series of documents describing the survey. Included there are a detailed Readme file, a Bureau-prepared FAQ page, and an in-depth article (24 page pdf file) from the Population Reference Bureau describing the survey. You can also visit the ACS Home page at the Census Bureau's web site. One of the things you can do from that site is sign up for ACS Alerts, which are e-mailed mini newsletters regarding the ACS. These are also saved as html documents that can be accessed from the Bureau's web site. New ZIP Code (ZCTA) Master Geographic Reference Dataset -- One of the most frequently asked question categories for the MCDC involves users wanting to know how ZIP codes can be related to other geographic areas. People frequently have access to address-level data containing ZIP codes and they want to be able to determine what county(s) or metro area(s) or place(s) or ... (etc) is/are associated with the ZIP code. We have created a ZIP Code Resources Page that has extensive information regarding the nature of ZIPs and how to relate them to other geographies. This page has recently been updated to document a new dataset that we have created, which we call the "ZCTA Master" dataset. ZCTA is the Census Bureau's approximation of ZIP codes and is the geography used in all census data products as a proxy for true ZIPs. Details on what this dataset is about and how to access it (via Uexplore/Dexter) is contained in the updated resources page. CAPS Reports Get More Recent Estimates and Projections We have obtained more recent estimates and projections for plugging into the reports generated by the CAPS (Circular Area Profiling System) application. Estimates are now for 1-1-2005 and projections for 1-1-2010. June 2006 Census Bureau Releases Place Level Population Estimates -- The Census Bureau released new population estimates for places (cities) and other subcounty governmental units today (6-21). The estimates are for each year starting with July 1, 2000 through July 1, 2005. There are no demographic characteristics or components of change with these estimates - just total population figures. Users can access the data for Missouri in an excel spreadsheet in the MCDC data archive. Or, you can access the datasets via Dexter for either Missouri or for the entire United States. You can also access the data along with a press release at the population estimates web page at the Census Bureau (http://www.census.gov/popest/estimates.php). See the Population Estimates Reports page which has links to two new reports based on these new estimates. The reports show the 50 fastest growing and shrinking cities over the latest year and over the five-year post-2000 time period. There is one set of reports (pdf file) just for Missouri and another for the United States. Some highlights: Wentzville, O'Fallon, Columbia, Lee's Summit, Nixa, Blue Springs and Raymore were the fastest growing Missouri cities over the 2004-2005 time period (listed in rank order). These were the only Mo cities to gain at least 1000 persons over the period. St. Louis was by far the fastest-shrinking Missouri city over the same period with a net loss of over 6000 people. This reversed an earlier post-2000 trend which had seen St. Louis finally ending its decades-long downward spiral. The city went from an increase of 2,666 in the 2003-2004 year to this dramatic loss of over 6000. (These numbers are not really new; St. Louis City is also a county equivalent and the county numbers were released in March.) SIC - The city of St. Louis successfully challenged its estimate for 2005. The new official estimate for the City is 352,572 which is an increase of almost 2000 over the 2004 estimate (which was also the result of a successful challenge.) So instead of being the fastest-shrinking city in the state, it was actually the 4th fastest growing city in terms of absolute growth for the latest year. The only other cities with a loss of 400 or more were Independence (-758), Florissant (-481), Raytown (-411) and Springfield (-403). Looking at growth over the entire period from the 2000 census through 7-1-2005, O'Fallon is by far the fastest growing with 20,878 new residents, almost twice the number for second-place Wentzville, which grew by 10,964. Lee's Summit, Columbia, Ozark, Blue Springs, Raymore and Nixa ranked 3 through 8 and were the only other cities with more than 4,000 new residents. The list of fastest-shrinking cities over the 5-year period was basically the same as the list for the most recent year. At the national level there were 17 cities that added 10,000 or more residents during the 2004-2005 period. Phoenix was by far the biggest gainer with 44,456 new people, with two Texas cities, San Antonio and Fort Worth, in a virtual tie for second, each with about 21,000 net growth. Of these 17 cities, 4 were in California, 4 in Arizona, 2 each in Florida, Nevada, North Carolina and Texas. The 4 fast growers in California were not the ones you might expect: Bakersfield, Moreno Valley, Elk Grove and Rancho Cucamonga. The Census Bureau press release has the headline Elk Grove, Calif., Named Fastest-Growing City This is qualified (in a much smaller font) in the first sentence which says "Elk Grove, Calif., had the nation’s fastest growth rate among large cities (100,000 or more population)..." [italics ours]. They measured growth in terms of percent increase instead of actual new people, and they created the somewhat arbitrary category of "large" cities being those with at least 100,000 population. (so that Surprise, AZ with a 22.8% increase was not eligible because it had only 60,591 people in 2004 and thus wasn't "large" enough.) Using percent change as the measure of growth instead of numeric growth almost always results in the winner being the one with the smallest base number and a decent increase (hence Nevada always gets the headline as the fastest-growing state, not because it has the most growth, but because it has the smallest denominator to use when figuring growth as a percent of something.) Elk Grove, CA was actually the 13th fastest growing city (based on numeric growth), but it was the one whose base (2004) population was the lowest figure over the 100,000 cutoff (it was 100,686). Los Angeles is still the fastest-growing city (measured by numeric increase) for the post-2000 period, edging out Phoenix 150,345 to 139,948. But since Phoenix out-gained LA by about 37,000 during the latest year, it is a pretty good guess that Phoenix has by now (2006) actually passed Los Angelses as the fastest grower but we'll have to wait a year for official confirmation of that. There were 12 cities that had a net loss of over 5000 people in the latest year. New York City led the way with a net loss of 21,509. St. Louis had the 7th largest loss but of the 12 biggest losers it had the second highest percentage loss, -1.8% vs. -2.8% loss for Norfolk, VA. (We think measuring population loss as a percentage of population base makes more sense than measuring population gain that way because the more people who live in a place, the more potential there is for them to leave.) May 2006 Domestic Net Migration in the United States: 2000 to 2004 — This report describes recent patterns of population redistribution within the country. Analysis focuses on net migration for different geographic areas, including regions, divisions, states, metropolitan and micropolitan statistical areas and counties. The report focuses on the post-Census 2000 period but also includes data from 1990 to 1999. Updates on Domestic Migration Trends for States - The new P-25 series Census Bureau report (see above) uses population estimates through 2004, as released in 2005 by the Census Bureau. Because of the time required to prepare their report for publication, the report is a year behind their own data release (the July, 2005 estimates were released last month, March, 2006). The MCDC accessed the latest data at the state level and generated a report that shows state migration trends for the most recent year, along with data for both the previous year (2003-04) and for the 4 and 5-year post-2000 annual averages. See the report at //mcdc.missouri.edu/pub/data/popests/Reports/dmig_report_states.html. Some of the more interesting trends shown in these reports include: Missouri's domestic migration in the most recent year (2004 to 2005) jumped to 8,283 as compared to 4,236 the previous year and to an annual average of 4,752 over the first 4 years of the decade. This is still below the 10,762 annual average increase the state experienced during the decade of the 1990's. Missouri ranked 16th in terms of domestic migration growth in 2004-2005, up from a rank of 20th in 2003-2004. Florida continues to be the fastest growing state in terms of net domestic migration, adding another 262,511 persons in '04-'05, up from the 254,029 increase of the previous year. While New York continues to be the state with the dubious distinction of having the highest average annual outmigration since 2000, it did not lead the country in out-migration for the most recent year. California, with a net out-migration of 239,417 persons during the 2004-2005 period ranked just ahead of New York's 232,638 net loss. This net outflow of persons from California is a significant increase over the annual average of just under 100,000 during the 2000-2004 period. The loss represents a return to the outmigration trend of the 1990's when California averaged a net loss of 220,871 persons per year. We looked at domestic migration trends for the states for the two most recent years (2003-04 and 2004-05) and looked at the change in these two values. The states with the largest increases in net migration in the latest year were Arizona Oregon North Carolina Tennessee Texas Utah Colorado Indiana Florida Alabama. Besides California and New York, states that experienced significantly less growth (or higher losses) in 04-05 than they did in 03-04 included: Alaska, going from a gain of 1,462 to a loss of 2,831 Maryland, going from a loss of 6,791 to a loss of 13,289. Michigan, going from a loss of 37,495 to a loss of 47,900. North Dakota, going from a gain of 744 to a loss of 2,908. Rhode Island, going from a loss of 2,052 to a loss of 10,243. For more about the latest population estimates with components of change data for 2005 see the MCDC's What's New page entry for last month. March 2006 State and County Population Estimates With Components of Change for 2005 Missouri's population grew by just under 41,000 persons (about .7%) over the period July 1, 2004 to July 1, 2005 according to figures recently released by the U.S. Census Bureau. This rate of growth is consistent with the small but steady rate reported for the state over the 5-year period since the 2000 census. During that time the state has grown by 203,627 persons, or 3.6%. The latest state population estimate is 5,800,310; this figure ranks 18th among all states, compared with the state's rank of 17 at the time of the 2000 Census. St. Charles county continues to be the fastest growing Missouri county measured in terms of persons added, with just over 46,000. There were 3 other smaller counties that had a larger percentage increase over the period than St. Charles. Christian county (south of Springfield) grew by nearly 24% over the 5-year period, from 54,285 to 67,266. Lincoln (22.6%) and Warren (17.3%) were the other two counties, both of them located on the western edge of the St. Louis MSA. The MCDC has downloaded the data for the entire U.S. and they can be accessed in our data archive using the uexplore/dexter software. The datasets are in the popests data directory, most easily accessed via the Datasets.html directory page. The four relevant datasets are mocom05, uscom05, mocomrgrns05 and uscomcbsas05. The first two consist of state and county level summaries, while the latter two are the result of our aggregating the data to get metro area and other county-based region summaries. You can access a new set of reports based on data in the popests data directory at http://mcdc.missouri.edu/pub/data/popests/Reports. This page provides links to both the reports and to the underlying datasets used to generate them. Also see [Maps] -- Powerpoint Presentation. All About ... Measures of Income in the Census We have created the first in a series of "All About..." documents that we plan to make available over the next few months. The initial document is an in depth discussion of the different ways that income and poverty are measured and reported in the census. It can be accessed at http://mcdc.missouri.edu/allabout/measures_of_income. New MCDC Filetype Provides 1980 Summary Tape File 3 Extracts The MCDC has created a new collection of standard extracts based on the complete-table data stored in the stf803 filetype directory. The best way to access these data is via uexplore/dexter. Use the link to the MCDC Data Archive and follow the link to the 1980 sub-topic of Decennial Census Data. The new filetype is stf803x2. We are still working on these datasets to make sure they have all the necessary geographic codes and to enhance the metadata. But the basic variable definitions are stable and the data are ready for public use. Census Bureau Q & A Center Points to MCDC Web Applications The Census Bureau web site has an extremely useful Question and Answer Center feature that allows users to do keyword searches or even to enter a question. The key to this searchable FAQ (frequently asked questions) system is a collection of carefully indexed FAQ module that include the Census Bureau responses to each question. The Missouri Census Data Center has two of its web applications referenced within this system, the Cure for the Common Codes geocode display system and the CAPS circular area demographic profile generator (see the Quick Links box on this page). To see how this works go to the Census Bureau web site and click on the Search button near the top of the page, leaving the FAQs radio button selected. On the next page click on the Question and Answer Center link (or you could just go directly to https://ask.census.gov .) The two FAQ entries that reference the MCDC applications are numbered 743 and 688. Try entering "Looking for a fips place code" in the Keywords or Question box and clicking the Search key. FAQ #743 is returned 3rd from this search. Curiously, the #688 entry with a reference to CAPS is also returned, even though it does not seem relevant to the search question. Entering a search string such as "circular profile" will also cause the 688 entry to be returned, this time as the first (most relevant) result. MCDC Quick Tour Updated We have completed an extensive revision to our MCDC Quick Tour web page, especially the sections related to Uexplore/Dexter. January 2006 1990 Summary Tape File 4 Upgrade We have made significant enhancements to our collection of STF4 files from the 1990 census. We have added an extensive national collection which includes the A Table file for the U.S. and a set of 12 B Table files. The B Table files provide summaries for the total population, 5 major race groups, the Hispanic population, and the 5 major race groups / non-hispanic. These are in addition to our previously available collection of File A and File B (not to be confused with Table A and Table B) datasets for the state of Missouri. In addition to adding more data, we have fixed a pair of minor bugs in the conversion routine and have added variable labels to the datasets, making it much easier to interpret Dexter extracts stored in spreadsheets. We have created dataset-level metadata in the form of a datasets dataset and corresponding Datasets.html directory page. We have also created the necessary metadata files to support access to these data at the table level in Dexter. Finally, we have created a Docs subdirectory with the complete technical documentation (as written by the Census Bureau and converted to pdf format by ICPSR), with each chapter and appendix stored as a separate pdf file referenced by the index.html page. New Population Estimates The National Center for Health Statistics commissioned the Census Bureau to do a special tabulation of their intercensal population estimates for the decade of the 1990's as well as the post-2000 period through 2004. These are referred to as the "bridged race" estimates because they use race definitions that are consistent with those used on the MARS (modified age, race and sex) files prior to the revamping of the race categories as mandated by OMB in 1997. These estimates are somewhat comparable to the casrh estimates - county age, sex, race and hispanic - that the Bureau distributed over these same time periods. The difference is that there are only 4 race categories (the race alone vs. alone or in combination catgories are gone), and the estimates are for single years of age instead of 5-year cohorts. These can be accessed via the uexplore/Dexter application in the nchsbri subdirectory of the popests filetype directory. There, you will find a collection with 4 files (datasets) per state (for all 51 states). To understand what the 4 sets are about be sure to look at the Readme file in this (nchsbri) directory. Note: this has changed since our initial posting here regarding these datasets. Inititally, we had created datasets for Missouri only and there was no subdirectory. Obviously, that has changed. We processed the latest state level estimates with components of change from the Census Bureau. These are July, 2005 estimates for the nation and its states, and provide estimates of the births, deaths, and net migration (international and domestic) over the period from April 1, 2000 to July 1. 2005. The dataset name is uscomnst05. County Business Patterns We have added new CBP data files for the entire U.S. at the nation, state and county levels for the years 2002 and 2003, using data uploaded from the latest cd-rom. Access these data via the revised Datasets.html page in the /pub/data/cbp data directory via uexplore/Dexter. Dropping Access to SAS Datasets on Old OSEDA Machine For several years, the MCDC's uexplore software for accessing our data collection, supported the ability to access data stored on two different Unix server machines: the venerable oseda machine (oseda.missouri.edu) and the current machine (mcdc2.missouri.edu, with an alias of mcdc.missouri.edu). Access to data still stored on the old machine involved using older software with a home-grown Perl-based interface to run the dynamic SAS modules. Over the last 4+ years we have moved nearly all of our key datasets off the old server onto the new one. We have now dropped support for accessing datasets on the old machine, and have moved a dozen or more filetypes over to the new machine. There are still a handful of filetypes that have not yet (and may never) be moved. Rather than just removing any reference to these filetypes, we still list most of them from our uexplore home page, but with a link to a note page that informs the user that the data are not currently available. Users encountering these cases are encouraged to send us feedback regarding their need to access these datasets, which could well lead to our moving the data collection over. Online Help for Dexter Updated We have added documentation for the Transpose Output options in Section V. (We had temporarily removed this feature from the Dexter program, but were persuaded by users to put it back.) This is the help that can be accessed by clicking on the section header on the Dexter input form. Other "What's New" by Year: 2005 | 2004 | 2003 | 2002

This file last modified Thursday February 15, 2007, 10:30:11