1) Jam Values

Jam values are an important part of disclosure, used to set limits on some derived measures such as calculated medians, aggregates, and averages, when those measures contain statistical outliers. A jam value is a hard-coded value displayed instead of a derived measure because that measure falls within a certain range and cannot be displayed because of disclosure, causing some categories to be top or bottom-coded, for example a value that is 110 or greater, represented by 110+. Jam values also act as informational symbols that explain the absence of data, represented by a symbol in the data display, such as ".". All ACS tables should contain these values when estimates are missing due to certain special conditions.
The following symbols are special data values and their definitions, used to symbolize the absence of certain estimates

Missing Value = ".": A dot ., indicates that the estimate is unavailable.

Filtered Value or Geographic Restriction = "": A blank indicates that a value is filtered or is not shown due to a geographic restriction.

Margin of Error (MOE) or Standard Error (SE) = 0: Estimate is controlled, which indicates the estimates are controlled and a statistical test is not appropriate.

Following are more examples of the special data symbols and their definitions:

Table ID	Low Value	Jam Value Definition	High Value	Jam Value Definition
B01002	0	1-	116	115+
B06011	2499	2,500-	100001	100,000+
B19013	2499	2,500-	200001	200,000+
B25035	1939	1940-	2005	2005
B25076	9999	10,000-	1000001	1,000,000

2) Rounding Rules Margin of Errors (MOE) and Standard Errors (SE)
A. The B00XX1/B00XX2 and B99XXX/B99XXXX tables are not estimates and do not have MOEs and SEs associated with them. The MOE and SE are set to -1 for these tables.

B. These are a few special rules on how certain margin of error and standard error are determined for ACS estimates. The accuracy of the estimate (decimal place) in the base tables to determine how many digits the standard error should be rounded.
The only exceptions to these rules are tables B19082: SHARES OF AGGREGATE HOUSEHOLD INCOME BY QUINTILE and B19083: GINI INDEX OF INCOME INEQUALITY. On American FactFinder (AFF), the estimates display to 2 decimal places and the MOE to 3.
3) Display of Estimates

Display whole numbers in the summary files:The estimates in the summary files should always be stored as whole numbers instead of in scientific notation. The largest estimate in the 2006 Summary File contains 14 digits because since SAS was used to generate the 2006 Summary File, SAS automatically stores and displays values with more than 12 digits in standard notation, but these values should be stored in the summary files as whole numbers.
4) Base Table Changes and Restrictions

Changes to the 2006 data products which contain complete lists of new, removed, and modified tables can be referenced on the ACS Website on: http://www.census.gov/acs/www/Downloads/Table_Changes2006.xls
ACS 2006 base tables that were subject to geographic restrictions are located on: http://www.census.gov/acs/www/Products/users_guide/ACS2006_Data_Product_Geographic_Restriction.xls
5) Special Table ID Sequence Rules
The data from the American Community Survey contains 49 detailed tables that have different table layouts for the Puerto Rico than their United States counterparts. The United States version of the tables are limited to geographic areas in the US (50 states plus the District of Columbia) while the Puerto Rico versions are limited to areas in Puerto Rico. In the summary file the Puerto Rico version of the tables are in sequence file 138 through 144. Therefore, for all geographic areas within the US sequence files 138 through 144 will be blank. In addition, sequence files 20, 24-28, and 136 will be blank for Puerto Rico as they contain the US version of the tables. The decision to separate the files was made because the comparable tables between the US and Puerto Rico may have a different number of cells (such as C06001 with 45 cells and C06001PR with 37 cells) which would significantly alter the metadata between the US and Puerto Rico. To identify what tables are available for Puerto Rico only look for the "PR" suffix in the table id (e.g. C06001PR).
For more information on PR only sequences that are counterparts to states-only sequences, please reference the following table.

PR Only Sequences	State-only Sequences
138	20
139	24
140	25
141	26
142	27
143	28
144	136

6) Correction to Sequence 0139
The ACS 2006 Summary File was re-run and re-released because it was discovered that C06010PR was missing from the summary file and the accompanying documentation because all PR tables should have a non-PR counterpart. Table C06010 did not exist in the ACS table package, so C06010PR was not assigned a sequence number in the 2005 or 2006 summary file. Table C06010PR was re-run and placed in sequence 0139, which breaks the ACS summary file rule of forcing collapsed tables to be in the same sequence as the base table (B06010PR is in 0140). The reason that this was done is that there wasn't enough space in sequence 0140 to include C06010PR and still maintain our cell width limit, so we placed it in the nearest logical sequence that fits all of our other rules so as to minimize the total affect on the files that have already been published.

7) Removal of Detailed Table B23003 from Sequence 0103
Due to problems with the PAOC variable, it was decided to remove detailed table B23003, which uses this variable, from the summary files. Please refer to errata note #40: ACS 2006 - PUMS variable PAOC (URL: http://www.census.gov/acs/www/UseData/Errata.htm) on the ACS Website for more information. A new sequence, 0103, was produced and contains the same data as the original sequence 0103, but with B23003 excluded. Table B23003 now contains blank values for all cells for all geographies in summary file sequence 0103. This change does not affect any of the summary file technical documentation.