Data Suppression Question

dlehman 7 months ago

I am looking at table B19013 for Median Household Income in a particular CBG (Block Group 4, Census Tract 42, Tulare County, California) and the 2021 5 year estimate is missing. It was present in the 2019 5 year estimate table, however. The guide to data suppression says that filtering due to large MOE is done on 1 year and supplemental estimates but not on the 5 year estimates. Does that mean the 2021 5 year estimate I am referring to is not due to large MOE? It is a small CBG but I don't think that is triggering the missing data. What are the possible reasons for this data to be missing in the 2021 data but not the 2019 data?

Thanks.

David Dorer 7 months ago

If you download the table using the API you will find numeric codes in the cells that are suppressed. The codes are here.

https://www.census.gov/data/developers/data-sets/acs-1year/notes-on-acs-estimate-and-annotation-values.html

There are other rules : (from ACSO Users support:)

Good afternoon David,

The reason you are not finding Census Tract 51.02 for San Diego, CA is that many times at the lower levels of geography, like Census Tracts, the estimates are suppressed due to the Margin of Error (MOE) being greater than the estimate. ACS has suppression rules that we must follow in order to make sure the data that is published meets quality standards and/or protects confidentiality. In this case, it is probably more so due to the quality and the MOE being large in comparison to the estimate itself.

Best,
Charles

As a statistician I don't agree with this MoE > Estimate rule because you can still interpret a count even though the lower end of the interval defined by Estimate and MoE may be below zero. In that case the lower bound is something near zero. The bound given by Estimate +MoE still makes sense.

Good afternoon David,

Best,
Charles

Estimate and Annotation Values

Annotation values are character representations of estimates and have values when non-integer information needs to be represented. See the table below for a list of common Estimate/Margin of Error (E/M) values and their corresponding Annotation (EA/MA) values. Please note that ACS data may return the following in place of data.

Estimate Value	Annotation Value	Meaning
-666666666	-	The estimate could not be computed because there were an insufficient number of sample observations. For a ratio of medians estimate, one or both of the median estimates falls in the lowest interval or highest interval of an open-ended distribution. The estimate could not be computed because there were an insufficient number of sample observations. For a ratio of medians estimate, one or both of the median estimates falls in the lowest interval or highest interval of an open-ended distribution. For a 5-year median estimate, the margin of error associated with a median was larger than the median itself.
-999999999	N	The estimate or margin of error cannot be displayed because there were an insufficient number of sample cases in the selected geographic area.
-888888888	(X)	The estimate or margin of error is not applicable or not available.
Varies	median-	The median falls in the lowest interval of an open-ended distribution (for example "2,500-")
Varies	median+	The median falls in the highest interval of an open-ended distribution (for example "250,000+").
-222222222	**	The margin of error could not be computed because there were an insufficient number of sample observations.
-333333333	***	The margin of error could not be computed because the median falls in the lowest interval or highest interval of an open-ended distribution.
-555555555	*****	A margin of error is not appropriate because the corresponding estimate is controlled to an independent population or housing estimate. Effectively, the corresponding estimate has no sampling error and the margin of error may be treated as zero.
*	N/A	An * indicates that the estimate is significantly different (at a 90% confidence level) than the estimate from the most current year. A "c" indicates the estimates for that year and the current year are both controlled; a statistical test is not appropriate.
null	null	A null value in the estimate means there is no data available for the requested geography.