technical difference between B17024 and S1702 and B17012

Good morning. We are trying to find the best measure for looking at family poverty at the locality-level. We use B17024 for child poverty. S1702 and B17012 were suggested to me as measures for looking at family poverty. A couple of questions:

1. What is the technical difference between child poverty (B17024) and family poverty (S1702 and B17012) for the ACS tables? 

2. Ideally we would like to look at family poverty at 200% FPL. It looks like S1702 does provide different levels of the FPL and B17012 is only 100% FPL. Would S1702 be the best table to look at family poverty? 

3. Is the only way to look at children below 200% FPL by race and ethnicity by creating PUMS groups?

Thanks so much!


  • Hi Beth,

    The difference between Table B17024 and Tables S1702 and B17012 is the unit of analysis. Table B17024 has children as the unit of analysis while Tables S1702 and B17012 have families as the unit of analysis. Which one you use depends on your goal.

    S1702 is a summary table, which means it provides more limited information. Using this table, you know the total number of families in each category—the denominator. For example, there are 77,000,000 families and there are 37,000,000 families with related children under 18. Of those 37,000,000 families with related children, 18% are living below the poverty level. The challenge with summary tables is that you do not know the numerator or the number of families that are living below the poverty level.

    Table B17012 provides all the information needed to calculate the denominator, numerator, and percentage for families with related children, all families, and families without children.

    Using S1702 for calculating the number of families below 200% of poverty could be problematic depending on your goal. The number reported in the table is the number of ALL families below 200% of poverty. This includes families with and families WITHOUT children. If you want the number of families with children that are below 200% of poverty, you would need to use PUMS.

    To estimate the number of children living below 200% of poverty by race and ethnicity at the local level you would need to use the PUMS data and use PUMAs as the geographic unit.

    If you decide to use the PUMS to obtain the estimates that are not available via American FactFinder (AFF), just a reminder that the PUMS files are a sample of the entire ACS data. As a result, PUMS estimates will slightly differ from estimates obtained from AFF.