I'm looking at table B19067, "AGGREGATE PUBLIC ASSISTANCE INCOME IN THE PAST 12 MONTHS (IN 2015 INFLATION-ADJUSTED DOLLARS) FOR HOUSEHOLDS". Looking at block groups, last year, for the 2010-2014 ACS, there were 131,952 block groups with data (the rest being suppressed). This year, 2011-2015 ACS, only 36,781 block groups have data. Almost a hundred thousand block groups that had data last year are suppressed this year.
ACS documentation this year mentions that dollar amount medians and quintile limits are being suppressed where the margin of error is greater than the estimate: http://www.census.gov/programs-surveys/acs/technical-documentation/user-notes/2016-01.html. However, this document specifically states that aggregate values are still available. I'm specifically seeing block groups with margins of error greater than their estimates (see Philadelphia County, Pennsylvania, Block Group 2, Census Tract 3; estimate 564,300, MOE +/-603,472; http://factfinder.census.gov/bkmk/table/1.0/en/ACS/15_5YR/B19067/0500000US42101.15000).
I see a similar pattern with census tracts.
Anyone know what's going on here?
Hello Bernie, I have consulted with some folks here at Census and because of data quality, B19067 is not available for small geographies such as block groups. If you have any further questions related to this please feel free to contact us directly via email at < acso.data.user@census.gov >. Have a good one, Justin