Discrepancy in estimates between US Census and IPUMS USA

USA
1

I’m calculating unemployment rates for CA using IPUMS USA 5-YR Estimates 2014-2018. My denominator is everyone in civilian labor force (those in the armed forces are excluded) and my numerator is those in the civilian labor force who are unemployed. I applied the survey weights and arrived at the following values: 19,614,080 (civilian labor force) and 1,311,510 (unemployed)

However, when comparing the values above to table DP03 Selected Economic Characteristics (5-YR Estimates 2014-2018) for CA (from the US Census website: Explore Census Data), there is a discrepancy in the estimates for the Civilian Labor Force and the Unemployed. In table DP03, the values are 19,630,514 (civilian labor force) and 1,321,502 (unemployed).

Is it normal to see these discrepancies between the IPUMS USA data and data from the US Census website?

2

We do not generally expect to exactly replicate official US Census statistics with public use data, though we expect estimates to be within the margin of error around the official estimates (which your estimates are, though the unemployment estimate is close). Official statistics rely on more detailed versions of the data than the public use versions available through IPUMS USA. This Census Bureau FAQ and this blog post from a former colleague provide a bit more information. You may also be interested in reviewing the Census Bureau’s labor force glossary for detailed information on how they define key concepts to ensure that your calculations using the public use microdata match the definitions used in the official statistics as closely as possible.