Low and Moderate Income Data Set Frequently Asked Questions

Question: What do all the column headers mean?

Answer: All of the column headers are explained in the data element dictionary.

Question: Why are there duplicate block groups?

Answer: The census tract/block group combinations are often duplicated and this occurs when a block group splits any type of a boundary defined by the Census Bureau. Some of those boundaries include places, county subdivisions, voting districts and council districts. When you come across a set of identical tract/block groups that are within the boundaries you want, you should add the values (except one) to obtain the total for that census tract/block group. The one variable you cannot add is the low/mod percentage (LOWMODPCT). This must be calculated by dividing the LOWMOD by LOWMODUNIV.

Low/Mod data has been summarized at the block group level within each grantee's jurisdiction.

Question: Can I link the low/mod data with other Census data?

Answer: Yes! The low/mod data includes the LOGRECNO, which is the key field used to link the geography to all Census SF3 tables.

Question: Why is the LOWMODUNIV sometimes greater than POP100?

Answer: The POP100 represents the 100% population count for that geographic area. This value is taken from the Census Bureau's SF1 tables, which is the information derived from the Census "short form." But because income data is not on the SF1 tables, we use the SF3 tables, or the info derived from the Census "long form." (The POP100 data field is put on the low/mod data tables for information and reference only.) Because the long form went to, on average, 1 in 6 households, the population count on the SF3 tables is estimated, and as a result, the population counts for the same piece of geography is very often different in the Census Bureau's SF1 and SF3 tables. Additional information on this can be found on the Census Bureau's website.

Question: Why is the LOWMODUNIV sometimes significantly lower than POP100?

Answer: CPD uses the Census Bureau's definition of persons eligible, which removes persons in group housing such as college students, jails and nursing homes. There are some block groups in which a significant portion, and sometimes all, of the population is within group housing. For example, if a block group has a 100% population count (POP100) of 2,200 persons, and 1,500 of those persons are in group housing, the LOWMODUNIV would be 700.

Question: Why does the total of the LOW and VLOW values sometimes add up to more than the MOD total?

Answer: There are 6 different sets of data elements which provide counts at the moderate, the low, and the very-low income levels. These are FAM (number of family households), NFAM (number of non-family households), HH (number of households), FAMP (number of persons in family households), NFAMP (number of persons in non-family households), and P (persons).
Each of these sets of data are cumulative. For example, PVLOW contains the count of all persons below the 30% income threshold; PLOW contains the count of all persons below the 50% income level; and PMOD contains the count of all persons below the 80% income level. Therefore, if you add PMOD+PLOW+PVLOW, you would double count the persons who are low income, and triple count the persons who are very low income.

Question: When does the LOWMOD data change?

Answer:  There are two types of updates. The first change is an annual update which re-associates every one of the split block groups (almost 400,000 nationally!) to the CDBG grantees based on their participation each fiscal year. This update does not change the low/mod values.

The other type of change occurs infrequently. The values of the low/mod data are recalculated only when income values change. These are the reported income and income limits and changes must reflect the same point in time (The low/mod data is changing for FY07).

Question: Can I get LOWMOD data at the block level?

Answer: The low/mod data is calculated at the split block group level, specifically at the Census SF3 SUMLEV 090 level. The low/mod data cannot be generated at the block level because income data is not available at the block level.