The respondent universe for the 1997 NHSDA
was the civilian, noninstitutionalized population aged 12 years or older
within the United States, including residents of noninstitutional group
quarters (e.g., shelters, rooming houses, dormitories), as well as residents
of civilian housing on military bases. Persons excluded from the universe
included those with no fixed address, residents of institutional quarters
(such as jails and hospitals), and active military personnel. The survey
used basically the same multistage area probability sample design that
has been employed since the 1988 NHSDA. This design uses a composite size
measure methodology and a specially designed within-dwelling selection
procedure to ensure that desired sample sizes would be achieved for subpopulations
defined by age and race/ethnicity. Oversampling was used to meet specified
precision constraints for these subpopulations. As in the 1988 through
1996 NHSDAs, the 1997 NHSDA oversampled Hispanics in areas of high Hispanic
concentration to reduce survey costs. Unlike the 1996 NHSDA, which used
virtually the same segments surveyed in the 1995 NHSDA, the 1997 NHSDA
basically surveyed a new segment sample. Approximately 95% of the 1997
sample, or 1,844 segments, consisted of the unused members of the pairwise
segment sample selected at the same time the 1995 and 1996 segment samples
were drawn. The remaining 5%, or 96 segments, overlapped with the 1996
survey year. A combination of the pairwise sampling feature and the overlap
feature helped reduce the sampling variance, as well as minimized field
costs.
In addition to national estimation provided
by the 1997 NHSDA, at a later stage of designing the survey, State-level
estimation was needed for Arizona and California. Hence, a supplementary
sample was selected for quarters 2 through 4 from each State large enough
to yield defensible State estimates.
Another feature of the 1997 design was
the inclusion of two field tests. The first field test was designed to
determine the operational feasibility of using an electronic questionnaire
version and to identify the issues associated with the use of an electronic
version. The second field test, conducted in quarter 4, used contingent
questioning (skips) to make data collection more efficient, reduce respondent
burden, and reduce inconsistencies in reporting.
As in previous NHSDAs, the basic sample
design of the 1997 NHSDA involved five selection stages:
-
the selection of primary sampling units (PSUs,
e.g., counties),
-
the selection of subareas (blocks or block
groups) within PSUs,
-
the selection of listing units (housing units
[HUs] or individual dwelling units [DUs] within noninstitutionalized group
quarters occupied by one or more civilians) within these subareas,
-
the selection of age domains (age groups 12
to 17, 18 to 25, 26 to 34, 35 to 49, and 50 or older) within sampled listing
units, and
-
the selection of eligible individuals within
the sampled age domains.
