Readme File for the acs2012/btabs5yr Data Directory

Rev. Jan. 21, 2014

Vintage 2012 American Community Survey summary ("base") table files containing 5-year (2008-2012) period estimates.

Type of data: Summary tables covering a broad range of subjects including

  • basic demographics such as age, race, sex, ethnicity

  • socioeconomic indicators such as income, poverty, education, employment, occupation, household/family types and marital status

  • housing characteristics such as tenure, units in structure, rent, value of home, mortgage status and housing costs.
The data are very similar to what you once would have found in a decennial census summary file. They are sample estimates based on a roughly 2% yearly household sample, plus a separate group quarters survey component.

Geographic coverage: United States, sometimes including Puerto Rico. This is a complete national collection.

Geographic summary units: A variety of geographic entities, from national summaries down to states, (selected) cities, counties, school districts, metro areas, PUMA's (still the old ones, however), census tracts, and even block groups. Geographic component (such as urban or rural portion) summaries are available for some larger areas (state and above). Since these are 5-year period estimates there are no population thresholds to worry about, and very little data suppression.

Time Period(s): The 5-year period 2008 to 2012. These data were released by the Census Bureau in December, 2013 (But not in this format, or course. Downloading and processing the files from the Bureau was a signficant challenge.)

What's New for this Year:

  1. The variable naming convention was modified last year to accomodate the combining of estimates and moe data. We have saved that concept but changed the convention slightly. The data cells for estimates contain the letter "i" prior to the cell number portion of the name, while the corresponding margin or error value contains the letter "m". For example, the first data cell in talbe B07401 would be stored in variable B07401i1, with the corresponding MOE stored in variable B07401m1. Last cycle (acs2011/btabs5yr) we used the letter "e" instead of "i" to indicate the cell containing the estimate. But we have now decided to back to our usual namning convention using the letter "i" (for "item").

  2. There are now detailed tables here dealing with the disability question. It has now been 5 years since the question changed on the survey form so we have had to wait that long to get new disability data.

  3. What is new for this cycle (as compared to last year) is that not much is new. We made lots of changes last year, and this year we are mostly being consistent with our new strategies and conventions for these data. Indeed, this Readme file is over 90% just a copy of the one we did for the 2011 cycle.

  4. We have created a Datasets.html directory page for this subcollection, which was not done for the 2011 data. This means you can invoke the Detailed metadata pages (either from the Datasets.html page, in the Detail columns, or from a link that will now appear at the top of the Dexter Query Form. It provides some very basic facts about each data set, with perhaps the most important thing being the access to the Key Variables for each set. For example, all the data sets will have State as a key variable so you can follow the link on the metadata page to get the code for the state(s) of interest, required for doing a state-based filter in Section II.

Two Geographic Universes. Each dataset is named starting with "us" or "mo". The former are national datasets with data for the entire country, while the latter are Missouri subsets of the corresponding us dataset. For example the dataset usplaces08 has data for topic 08 (journey to work) for all places in the US, and the moplaces08 dataset contains the Missouri subset of that same data. (Techies may be interested in knowing the the Missouri datasets are actually things called views that are virtual datasets that take up virtually no extra storage space; all that gets stored in a view are the instructions on how to generate the specified data when accessed.)

Table Topic Partitioning.. We continue to use the convention of breaking down the base tables into six datasets, each with tables related to a set of data topics. The TableTopicCodes.txt file can be browsed to see what the 2-digit topic codes are. For example, topic code 01 is "Age and Sex". Tables dealing with these two subjects are assigned names of the form B01xxx or C01xxx, where xxx is a 3-digit number. The dataset usstcnty00_07 will contain all the tables for topics 00, 01, ... 07 for all states and counties in the US. Dataset mobgs17_20 will contain all tables for topics 17, 18, 19 and 20 for Missouri block groups.

Finding the Table You Need. The Census Bureau's ACS web site is the ultimate source of information regarding what tables are available as well as getting detailed information regarding the underlying definitions and categories. What we have on this site is a set of codebook files that can be extremely helpful not only for finding the data you are looking for, but also for getting the variable names associated with each cell of each table. The Varlabs subdirectory (of this btabs5yr directory; it also occurs in the basetbls subdirectory) contains six txt files corresponding to the six topic groups used in creating our datasets. For example, one of the files is 17_20labels.txt, and contains information that looks like this:

/* Universe: Population for whom poverty status is determined  */

    B17001i1 ="Total:"
    B17001i2 =" Income in the past 12 months below poverty level:"
    B17001i3 ="  Male:"
    B17001i4 ="   Under 5 years"
    B17001i5 ="   5 years"
    B17001i6 ="   6 to 11 years"
    B17001i7 ="   12 to 14 years"
    B17001i8 ="   15 years"
    B17001i9 ="   16 and 17 years"
    B17001i10 ="   18 to 24 years"
    B17001i11 ="   25 to 34 years"
    B17001i12 ="   35 to 44 years"
    B17001i13 ="   45 to 54 years"
    B17001i14 ="   55 to 64 years"
    B17001i15 ="   65 to 74 years"
    B17001i16 ="   75 years and over"
    B17001i17 ="  Female:"
    B17001i18 ="   Under 5 years"
    B17001i19 ="   5 years"
    B17001i20 ="   6 to 11 years"
    B17001i21 ="   12 to 14 years"
    B17001i22 ="   15 years"
    B17001i23 ="   16 and 17 years"
    B17001i24 ="   18 to 24 years"
    B17001i25 ="   25 to 34 years"
    B17001i26 ="   35 to 44 years"
    B17001i27 ="   45 to 54 years"
    B17001i28 ="   55 to 64 years"
    B17001i29 ="   65 to 74 years"
    B17001i30 ="   75 years and over"
    B17001i31 =" Income in the past 12 months at or above poverty level:"
    B17001i32 ="  Male:"
    B17001i33 ="   Under 5 years"
    B17001i34 ="   5 years"
    B17001i35 ="   6 to 11 years"
    B17001i36 ="   12 to 14 years"
    B17001i37 ="   15 years"
    B17001i38 ="   16 and 17 years"
    B17001i39 ="   18 to 24 years"
    B17001i40 ="   25 to 34 years"
    B17001i41 ="   35 to 44 years"
    B17001i42 ="   45 to 54 years"
    B17001i43 ="   55 to 64 years"
    B17001i44 ="   65 to 74 years"
    B17001i45 ="   75 years and over"
    B17001i46 ="  Female:"
    B17001i47 ="   Under 5 years"
    B17001i48 ="   5 years"
    B17001i49 ="   6 to 11 years"
    B17001i50 ="   12 to 14 years"
    B17001i51 ="   15 years"
    B17001i52 ="   16 and 17 years"
    B17001i53 ="   18 to 24 years"
    B17001i54 ="   25 to 34 years"
    B17001i55 ="   35 to 44 years"
    B17001i56 ="   45 to 54 years"
    B17001i57 ="   55 to 64 years"
    B17001i58 ="   65 to 74 years"
    B17001i59 ="   75 years and over"

/* Universe: Population for whom poverty status is determined  */

    C17001i1 ="Total:"
    C17001i2 =" Income in the past 12 months below poverty level:"
    C17001i3 ="  Male:"
    C17001i4 ="   Under 18 years"
    C17001i5 ="   18 to 64 years"
    C17001i6 ="   65 years and over"
    C17001i7 ="  Female:"
    C17001i8 ="   Under 18 years"
    C17001i9 ="   18 to 64 years"
    C17001i10 ="   65 years and over"
    C17001i11 =" Income in the past 12 months at or above poverty level:"
    C17001i12 ="  Male:"
    C17001i13 ="   Under 18 years"
    C17001i14 ="   18 to 64 years"
    C17001i15 ="   65 years and over"
    C17001i16 ="  Female:"
    C17001i17 ="   Under 18 years"
    C17001i18 ="   18 to 64 years"
    C17001i19 ="   65 years and over"

When I am trying to determine if we have a table dealing with poverty ratios I just go to the 17_20labels.txt file in Varlabs and use my browser's find command to see if the string "poverty ratio" is found. When that fails (as it always does) I try finding just "ratio" and that turns out to be just what I needed: the tables with this info are labeled "Ratio of Income to Poverty Level". Very low tech but very high value.

Finding the Table You Need at the Block Group Level. Unfortunately we do not have all these tables availalbe to us when looking at block group level data. Because they tend to be quite small the Bureau made the decision to only let us have selected tables at the BG level. Our Varlabs modules do not currently have any annotations indicating which tables are available for block groups. But for the 2011 vintage data we can tell you - here are the tables available at the block group level:

B00001 B00002 B01001 B01002 B01002A B01002B B01002C B01002D B01002F B01002G B01002H B01002I B01003 B02001 B02008 B02009 B02010 B02011 B02012 B02013 C02003 B03002 B03003 B07201 B07202 B07203 B08007 B08008 B08009 B08016 B08017 B08018 B08134 B08135 B08136 B08301 B08302 B08303 B09002 B09018 B09019 B09020 B11001 B11001A B11001B B11001C B11001D B11001E B11001F B11001G B11001H B11001I B11002 B11002A B11002B B11002C B11002D B11002E B11002F B11002G B11002H B11002I B11003 B11004 B11005 B11006 B11007 B11012 B11015 B11016 B12001 B14002 B14005 B14007 B14007A B14007B B14007C B14007D B14007E B14007F B14007G B14007H B14007I B15002 B15003 B16002 B16004 C17002 B17010 B17011 B17017 B17021 B19001 B19013 B19013A B19013B B19013C B19013D B19013G B19013H B19013I B19025 B19025A B19025B B19025G B19025H B19037 B19049 B19050 B19051 B19052 B19053 B19054 B19055 B19056 B19057 B19059 B19060 B19061 B19062 B19063 B19064 B19065 B19066 B19067 B19069 B19070 B19101 B19113 B19113A B19113B B19113C B19113D B19113G B19113H B19113I B19127 B19201 B19202 B19202A B19202B B19202H B19214 B19301 B19301A B19301B B19301C B19301D B19301F B19301G B19301H B19313 B19313A B19313B B19313C B19313D B19313F B19313G B19313H B20001 B20002 B20003 B21001 B21002 C21007 B22010 B23003 B23007 B23008 B23009 B23022 B23024 B23025 C23023 C24010 C24010A C24010B C24010C C24010D C24010E C24010F C24010G C24010H C24010I C24020 C24030 B24080 B25001 B25002 B25003 B25003A B25003B B25003C B25003D B25003E B25003F B25003G B25003H B25003I B25004 B25006 B25007 B25008 B25009 B25010 B25014 B25015 B25016 B25017 B25018 B25019 B25020 B25021 B25022 B25024 B25032 B25033 B25034 B25035 B25036 B25037 B25038 B25039 B25040 B25041 B25042 B25043 B25044 B25045 B25046 B25047 B25049 B25050 B25051 B25053 B25054 B25055 B25056 B25057 B25058 B25059 B25060 B25061 B25063 B25064 B25065 B25066 B25067 B25068 B25069 B25070 B25071 B25072 B25073 B25074 B25075 B25076 B25077 B25078 B25079 B25080 B25081 B25082 B25083 B25085 B25087 B25088 B25089 B25091 B25092 B25093 C25095 .

What if I Want Single-year or 3-Year Data? You are in the wrong directory. The "5yr" in the btabs5yr directory name indicates that we have decided to segregate our 5-year period estimates data (at least the base tables). Assuming you are interesed in detailed (base) tables you should be looking in the basetbls directory - a "sibling" of this one.

Is There any Single Dataset With All the Base Tables? A user contacted us expressing disappointment when he found out that the motracts dataset in the parent directory (/pub/data/acs2012) did not contain all the base tables for Missouri tracts. The answer to the question is No, we currently only have partioned-by-topics base tables datasets. We have those stored in two subdirectories: basetbls (with the 1 and 3-year data) and this one (btabs5yr) with just 5-year estimates. The motracts and all the other datasets in the parent directory do not contain base tables data; they are standard extract datasets. Those standard extracts are much smaller and were created by manipulating the data stored in the base tables data sets. You should try to answer your data question using the data in those extract datasets before you resort to accessing these monsters.

Key Bookmarks:

Go/return to Uexplore the acs2012 directory.