|
Northwest Florida Map » |
|
Consumer Expenditures methodology
The consumer expenditure database consists of a multi-level hierarchical classification of household expenditures that covers the majority of annual household expenditures. It is derived from an extensive modeling effort using the 1992, 1993, 1994, and 1995 Consumer Expenditure Survey data from the Bureau of Labor Statistics (BLS). The BLS survey is a comprehensive survey that averages over 5,000 households four times a year using a rotating sampling frame. The use of several consecutive years of data provides a rich base of expenditure data from which to build expenditure models based on household demographics.
The survey includes a wide range of demographic attributes related to "consuming
units" (generally households) that have been modeled separately for each
discrete expenditure category. The 1992 1993, and 1994 surveys were first inflated
to 1995 price levels using the detailed consumer price index series. For each
individual expenditure category in the survey, summary statistics were calculated
for each separate element in the list below. In several cases, it was possible
to utilize cross tabulation data (e.g. income by age of head of household).
These variables are listed below:
• geographic region (Northeast, South, Midwest, West)
• metropolitan status (metropolitan, non-metropolitan)
• housing tenure (owner or renter)
• age of head of household (<25 years, 25-34 years, 35-44 years, 45-54
years, 55-64 years, 65-74 years, and 75+ years)
• size of household (1 person, 2 persons, 3 persons, 4 persons, 5 persons,
6+ persons)
• household income (<5000, 5-10000, 10-15000, 15-20000, 20-30000, 30-40000,
40-50000, 50-70000, 70000+)
• race (White, Black, American Indian, Asian)
• number of vehicles (none, 1, 2+ vehicles per household)
The total sample was utilized to obtain an average expenditure for each item.
For each expenditure item, a series of adjustment factors were derived for each
unique demographic attribute. These adjustment factors were then applied to
the block group level using the same demographic variables in order to create
estimates at the local level that are consistent with local characteristics.
Consistency checks were undertaken in order to ensure that the results at the
block group level were consistent in the aggregate with overall income levels
and published expenditures. Finally, the 1994 estimates were inflated using
detailed consumer price indexes to 1999 levels.
|