Missouri Census Data Center

What's New at the MCDC in 2014

Go to most recent posting(s).

New 2008-2012 ACS Data Accessible Via MCDC's ACS Profile Extract Assistant

Posted 1-11-2014

You can now access the latest 5-year ACS period estimates data via Dexter and the ( ACS Profile extract assistant preprocessor app (the second item in our Quick Links menu box). The application makes it relatively simple to specify the geography and profile tables you want to extract.

Of course, what makes this all possible is that we have created the required datbase datasets which the application can access. It is the same application as before, just working with a whole new set of 2012 vintage 5-year period estimates.

New 2008-2012 ACS Data Accessible Via Uexplore/Dexter

Posted 1-11-2014

Similar to prior years we have created an extensive collection of datasets based on the Census Bureau's 2008-2012 5-year period estimates data. In addition to the standard MCDC profile datasets (most of which are accessible using the previously cited Extract Assistant application), we also have a new set of detailed base tables which can be accessed. Access the smaller mcdcpofiles datasets using this URL: http://mcdc.missouri.edu/cgi-bin/uexplore?/pub/data/acs2012 (or, better yet, use the Datasets.html page within this directory. The new detailed base tables are such a huge collection that they merit their own special subdirectory. Access it at the following URL: http://mcdc.missouri.edu/cgi-bin/uexplore?/pub/data/acs2012/btabs5yr. Note the presence of a Varlabs directory that can be used to search for and identify the tables of interest. The TableTopicCodes.txt file can help you to find the table-group of interest. (Use it to find out, for example, that the Poverty topic code is 17, so you can see the outlines of all available tables dealing with poverty in the file Varlabs/17_20labels.txt).

New 2008-2012 ACS Data Available for Circular Areas

Posted 1-16-2014

A version of the MCDC's CAPS web application for generating reports for circular areas (i.e. based on user-specified point and radius parameters) is now avaiable for use with the new 2008-2012 ACS data. To access it, use the trial version page at http://mcdc.missouri.edu/websas/caps10acs2012.html . This will change on Jan. 27 when we plan to replace the old version (which worked with 2007-2011 data) with this new one (at which time you can use the usual (current) link to the application - http://mcdc.missouri.edu/websas/caps10acs.html .

New Readme File for the acs2012/btabs5yr Directory

Posted 1-21-2014

We have created a new Readme.shtml file for our collection of custom 5-year period estimate base tables. We mostly copied the corresponding file from last year (in acs2011/btabs5yr) but did note some changes for this year. Such as going back to our standard table-variable naming convention using "i" before the cell number, instead of "e" (for estimate) as we did with the 2011 vintage tables.

New 2012 SAIPE Data Downloaded and Added to Archive

Posted 1-23-2014

The SAIPE (Small Area Income and Poverty Estimates) data (Census Bureau) was downloaded and converted for the data archive. These data were released in December of 2013. Conversion and data set naming conventions have not changed for several years now. We simply added another set of data to the time series. The new data sets are usstcnty2012 and usschldst2012 (new data for the entire US just for 2012 and at the nation-state-county and school district levels, respectively. The corresponding Missouri-only subsets are the same, just change the "us" prefix to "mo". The corresponding "20xx" data sets (i.e. usstcnty20xx and uschhldst20xx) have the new data as well as data for all other years going back to 2000. The Datasets.html index page with metadata has been updated and is the easiest way to access the data. You can also access it the old-fashioned way via uexplore on the saipe data directory. See the SAIPE web site at the Census Bureau for background information regarding these data and for access to an interactive data tool for viewing the data.

New Queries Module Extracts an ACS Base Table for All Tracts in Several States

Posted 1-23-2014

Inspired by a data user who was needing such data we coded a Dexter sample query showing how you can extract a specified ACS base table for all the census tracts in 3 selected states. Of course, you could just as easily do it for just one state, or for all states; or for a different level of geography such as a county, ZCTA or PUMA; or for multiple tables, if needed. The sample query is stored as a web page in the Queries subdirectory of the btabs5yr of the ;latest ACS data directory (/pub/data/acs2012). The module is mnemonically named OneTable4StatesAlltracts.html . A key feature of this example is that it illustrates how you can select tables and the corresponding MOE variables by using the "3rd way" to select variables in Section III of the Dexter Query Form. This entails entering the names of all the varibles you want in a text box and taking advantage of the variable interval notation. This turns out to be much easier than finding and selecting all 346 (173 estimates and 173 MOEs) variables (base table B23001) off a monster select list.

MABLE/Geocorr12 Updated To Fix Missing PUMA 2000 Codes

Posted 1-27-2014

A problem was reported related to the old PUMA 2000 codes. There were a few blocks where no such code had been assigned. We were able to trace the problem back to a step where we built a bridge file linking 2000 and 2010 census blocks. A small number of 2010 blocks were not linked to any 2000 block using our algorithm (which was based on data from the blkrel10 block relationship files). We tweaked that code, so that any block not matched would be assigned the same puma2k code as the preceding block if it were in the same county. The data base was updated today 1-27-2014, circa 7:30 AM CST.

New ACS PUMS Data for 2010-2012, Missouri and Entire U.S.

Posted 2-13-2014

We have downloaded the two 3-year ACS PUMS SAS data sets for Missouri into the acspums data directory (data sets mohrecs123yr and moprecs123yr). Note also the ACS2010-2012_PUMS_README.pdf file which contains news regarding the way PUMA codes are being handled on these data sets. This is the first multi-year ACS PUMS file released since the Bureau started using the revised 2010 PUMA codes (aka "2012" because they were not readily available or used in the ACS data until vintage year 2012). Basically, they give us 2/3 old 2000 codes and 1/3 new 2010 PUMA codes. Which means you really can't do much at the PUMA level with these data sets. This sets the precedent so that we should expect the same thing to happen when the 2008-2012 5-year ACS PUMS data are released next month.

We have also downloaded the collection of 8 data sets containing the latest 3-year (2010-2012) ACS PUMS data for the entire country. Because of the size of this data collection the Bureau distributes it in 4 state-interval subsets. These are identified with a single alpha suffix code a thru d, where "a" sets contain data for the states of Alabame thru Hawaii, "b" meaning Idaho thru Mississippi, "c" Missouri thru Oregon, and "d" indicating data for Pennsylvania thru Wyoming.

The datasets as distributed have 2 PUMA variables called PUMA00 and PUMA10, only one of which contains a value for any observation depending on the year of the survey response. We were able to create our own new PUMA variable which has the same value as PUMA10 for 2012-vintage responses; for earlier responses we attempted to impute a value of PUMA from the value of PUMA00 by doing a table lookup. Turns out that about 37% of all 2000 PUMAs have a 1-to-1 correspondence to a 2010 PUMA. Our table lookup identifies those cases and makes the assignment of the equivalent 2010 PUMA.

Addendum posted 2-19-2014 .

(Some additional detail in response to questions we got regarding what we talked about in the previous paragraph.)
We used our MABLE/Geocorr12 geographic web application to generate an equivalency file showing the relationships between 2000 and 2010 PUMAs for the U.S. We saved the results as a
tab-delimited file as well as the usual SAS data set that can be accessed via Dexter. We also created a special lookup table in the form of a SAS format code, $pma2k12, which is what we used to assign values to PUMA10 based on the values of PUMA00 in some (about 37% as mentioned above) observations on these new 3yr acspums data sets. The entries of this lookup table represent PUMA00/PUMA10 combinations where at least 98% of the 2010 population of PUMA00 was contained in PUMA10, and conversely at least 98% of the 2010 population of the 2010 PUMA was contained in the 2000 PUMA. So, for example, there is this entry:

'2901701'='01801'
.

This says that PUMA Code 01701 in state 29 (Missouri) corresponds to PUMA 2010 code 01801. It turns out this is one of eight entries in a row that have equivalents across the decades. These 8 PUMAs are all subareas of St. Louis County, MO; so if you are someone doing research on that county you are in luck - you can do analysis using the new 3yr acspums data set and have PUMA10 filled in for all the observations, not just the (approximately 1 in 3) observations based on 2012 surveys.

New Thresholds in caps10c Application for Choosing Geographic Units

Posted 2-17-2014

We ran into a fairly bizarre case of running our caps10c (Circular Area Profiling System, 2010 Census version) where we were able to generate data for a .5 mile radius of a point, but not for a 4.4 mile radius of that same point. We discovered the curious case that we had dozens of census block centroids that were in the circle but NO block group centroids. To fix the problem we changed our thresholds in the caps10c application: we now use 6 miles as the threshold for using blocks; and, while we were at it we also raised the threshold to start using tracts instead of BGs from 20 to 30.

New IRS Tax Returns Summary Data by ZIP and County

Posted 2-19-2014
Updated 3-3-2014

We have created a new filetype, irstaxes, containing data from the IRS summarizing tax returns for 2011. We have created four national files with ZIP code and county level summaries of returns, with and without gross income category breakdowns. These data sets are basically as received, with variable names corresponding to those used in the provided layout file, and with no derived variables such as means or percentages.

We have now (circa 2-26) created a set of four enhanced data sets based on the original four created by reading the data as delivered. These "plus" data sets contain all the variables from the original files, but most have been assigned new, more mnemonic names. For example, the variables N00200 and A00200 contain the number of returns reporting Salaries and Wages and the total amount of salaries and wages, respectively. On the "plus" data set these variables are now Nwages and AmtWages. In addition, we have added three new variables to help describe the item:

  1. PctWithWages: indicates the percentage of returns that reported Salary & Wage income.

  2. Avgwages: indicates the average amount of salaries and wages reported, using Nwages as the denominator.

  3. AvgTRwages: indicates the average amount of salaries and wages reported using TotRtrns (total returns) as the denominator. Note that this figure is in dollars, while the AmtWages variable is in thousands of dollars.

This same pattern of renaming and assigning the percentage of all returns and the two average figures are assigned to each return line item.

New Government Finance Data

Posted 3-3-2014

The Census Bureau publishes data annually regarding aggregate revenues and expenditures at the state and nation levels. Data are also reported for the aggregate of all local governments with the nation or state. We have downloaded these data for 2011 (latest available) and 2006 (to allow viewing 5-year trends) and have created a new filetype, govfin . These datasets have a very simple structure, each observation representing a line item by state/nation and and level type (state, local, or state+local). We have added an ItemDesc variable (Item Description) that contains the meaning of the ItemCode variable.

We are doing this on an experimental basis. Let us know if you find these data useful and/or would like to see more years of data.

New Default Version of ACS CAPS

Posted 3-4-2014

We have copied the current default version of the CAPS application for ACS data to caps10acs2011.html . It has been edited to reference the fact that it is not the latest version of the data and has a link to the current (latest data) version. The latest version is still caps10acs2012.html and it has been edited to provide a link back to the previous version. Most importantly, we have deleted the caps10acs.html file and replaced it with a soft link which now points to caps10acs2012.html. So when a user goes to the default caps10c version and sees the link to the ACS version that link now takes the user to the latest version, per the soft link.

Sample Queries for New irstaxes Data

Posted 3-11-2014

We have created a new Queries library for the irstaxes filetype. There are three sample queries, one of which actually invokes Rankster directly instead of Dexter. It produces a report (which we saved and you can access without having to run the query) showing the 40 wealthiest ZIP codes in every state (based on averaged Adjusted Gross Income values).

The transpose_states sample query shows how to use the transpose capabilities of Dexter to create an output report/file with the variables as rows and geographic areas (5 selected states in this example) serving as the variables.

Enhanced Readme.html Page for 2010 Census

Posted 3-17-2014

See the revised overview at http://mcdc.missouri.edu/census2010/ (or follow the link labeled 2010 Census in the Navy Blue Navigation Box (at left of most mcdc pages). We have added descriptions and links for our SF1-based profile applications.

Cycle Update for the Missouri Social & Economic Profile Report

Posted 3-17-2014

We have modified the Missouri Social and Economic Profile Report beta test version to reflect recent data additions and especially to upgrade the links pointing to the latest available data from the American Community Survey (ACS). We have not yet created a corresponding set of narrative profile versions of the new data. We plan to do so in the summer, after the new round of population estimates for 2013 are made available.

As the title implies, these reports are for Missouri counties only. Easiest way to access them is old home page - www.oseda.missouri.edu/countypage/county_seir.shtml - click on any of the "Data Tables" links (which takes you to an old 2009 version for now - we plan to change that soon) and then follow the link at the top of that report page to the"new beta test version". You can also access the profiles via our Applinks application, which for Missouri counties is most easily done using the Missouri County Data Map, accessible via our Quick Links navigation box (on right of this page). You just click on your county, and when the application links menu page for that county appears, look for "MO Social and Economic Profile" in the Applicaiton/Link column. When the report appears click on the "here" link at top to get the new beta version. Note that data for the state (totals) are included for comparison on each report.

Small Area Health Insurance Estimates (SAHIE) Data for 2012

Posted 3-18-2014

The recently released data estimates, now based on ACS data, for 2012 have been downloaded and converted. These are the same kind of data and in the same format the Bureau has been using since 2008. We store these data together with the SAIPE (Small Area Income and Poverty Estaimtes) data) in the saipe data directory. The raw downloaded data are in the sahie12all.txt file, while the data for the entire US are stored in the sahie12all and sahie12totals data sets, with mosahie12 stored as the Missouri subset of sahie12all. Easiest access is via the index page, Datasets.html. Learn more about these data sets and view highlights (maps and pdf files) at the Bureau's SAHIE data web site.

MCDC Home Page Features Article re new IRS Taxes Data

Posted 3-18-2014

We posted a rather extensive article describing this important new economic data source. Complete with links and snapshots of outputs from some of our several sample queries. Plans to do thematic maps based on these data are underway.

New All About 2010 PUMAs

Posted 3-22-2014

See the new puma2010.shtml page with lots of information regarding the revised Public Use Microdata Areas, 2010 update. Companion piece to our earlier page that dealt with the 2000 edition of this important geographic level.

MCDC Home Page Features Article re new IRS Taxes Data Completed

Posted 3-24-2014

As promised in the article as posted earlier this month, we have completed creating some thematic maps and the corresponding Queries modules that were used to generate the data used for those maps. We also added a brief testimonial regarding the data resource from the Maryland State Data Center. [On 4-3-14 this article was moved to this oldhome page.]

Census Bureau Releases New County Estimates With Components of Change

Posted 3/27/2014

This is the first in a series of three major sub-state estimate products to be released in the first half of this year. The key item in this set is the 2013 estimated total population for each state and county. (The state figures have actually already been published back around New Years but they are repeated here.) The data are also reported for CBSAs (Metropolitan Statistical Areas and Micropolitan SA's). These figures contain annual estimates (starting with July 1,2010 and going through July 1, 2013) and report components of change (births, deaths, migration) for each year since 2010 plus the 3-month period April-June, 2010. These data are stored in a series of data sets within the popests data directory, and the Datasets.html file for navigating though those sets has been updated to reflect the 5 new data sets (mocom13, mocom13t, uscom13, uscomcbsas13, and mocomregns13). Follow the metadata links in Datasets.html for details. They are all the same as last year but now with data for (or through) 2013 rather than 2012.

We have updated the poptrends1 web application, best accessed using applinks. We have a link to the Applinks master menu in our Quick Links box. For those interested just in Missouri we recommend using the MO County Data Map link, also in Quick Links. Would you like to see these reports and charts for every county in the state? Just replace the value of the &co parm with the word all instead of a county code. For example the URL
http://mcdc.missouri.edu/cgi-bin/broker?_PROGRAM=websas.poptrend1.sas&_SERVICE=sasapp&st=29&co=all
results in a rather long report showing basic pop trends (no component detail, just total pops) for each county in Missouri going back to 2000. Change the value of the &st parm if you want it for a different state (e.g. &st=06 will generate a report for California counties.)

We have updated our Population Estimates page to reflect what's new. That page contains a link to the Reports library (bullet item 3 under Our Best Estimates), with new reports for Missouri and the nation at the county and CBSA (metro area) levels.

Access the data as released by the Census Bureau at their Population Estimates main page, under Latest Releases.

Updated Social and Economic Profiles for Missouri Counties

Posted 3/31/2014

These popular reports have been updated to reflect the latest population estimates with components of change which we recently added to our data archive. We also created a new directory with updated narrative profiles for each county. And, finally, we created a new master menu page which can be accessed at http://www.oseda.missouri.edu/countypage/county_seir14.shtml . We have made this revised version of the "cntypage" web application the default version (it was previously the 2010 version with data no later than 2009). If you choose this applicaiton from an applinks menu page it will now take you directly to the latest version with the most up-to-date data.

New MCDC Home Page Features Article Re New 2013 Pop Estimates

Posted 4/3/2014

The article features a report produced by our poptrends1 web application showing a trend report for the state for the last 13 years. It also produces a 9-bullet list of interesting trends related to Missouri's population as extracted from the new mocom13 and uscom13 datasets (and one item using mocomregns13). The previous home page article related to the IRS Tax summary data files was moved to the oldhome directory.

Bug Fixed in irstaxes Enhanced Data Sets

Posted 4/10/2014

A problem was found and fixed in our "plus" data sets in the irstaxes data directory (see earlier entry here). The IRS reported 0 values as .0001 (see footnote 2 in the tech doc). So variables that appear to have a value of 0 (for both the N returns and A amount reported values). This caused the Avg values on our "plus" data sets to have a value of $1,000 which makes no sense. The Avg value for an N of 0 is missing. We have rerun our conversions for the plus data sets and the Avg variables are now correct. We also replaced all of the .0001 pseuo-values to zero. The original data sets still contain the .0001 values as distributed by the IRS. We also updated the metdata modules to alert users to this situation.

IRS Releases 2010-2011 Migration Data

Posted 4/23/2014

The IRS Statistics of Income division recently posted this notice on their web site:

State and county migration data for 20102011 are now available on Tax Stats. Migration data for the United States are based on year-to-year address changes reported on individual income tax returns filed with the IRS. The data present migration patterns by State or by county for the entire United States and are available for inflows (the number of new residents who moved to a county or State and where they migrated from) and outflows (the number of residents leaving a county or State and where they went).

The Missouri Census Data Center has downloaded and converted these files, storing the results in the irsmig data directory. The data sets are a little different from earlier editions. Variables TotalHHIncome and AvgHHIncome are now called Aggr_AGI and AvgAGI, for example. Median and Per Captita income variables have been dropped.

The web application for generating IRS Migration Profile reports has been upgraded to recognize the new year of data. Access this application at the IRS Migration Profile menu page (for single counties, anywhere in the U.S.).

The two state-level irsmig data sets, usstateinmig20xx and usstateoutmig in the irsmig directory have also been updated, with the new year of data (2010-2011 based) added. The variable names have also changed consistent with the county names (to aggr_agi and AvgAGI). The web app to display state-to-state migration profiles has also been updated.

Go to the top of this page.


See the whatsnew page for last year.

This file last modified Wednesday April 23, 2014, 10:17:51


Site Map    |    Our URL    |   
The Missouri Census Data Center is a sponsored program of the Missouri State Library within the office of the Missouri Secretary of State. The MCDC has been a partner in the U.S. Census Bureau's State Data Center program since 1979.

Questions/Comments regarding this page or this web site are strongly encouraged and can be sent to