Key Concepts About Identifying Correct Sample Weights and File Locations
Most of the environmental chemicals were measured in randomly selected subsamples of examined person within specific age groups. Subsample weights were constructed for each two-year survey cycle to account for the complex survey design (including oversampling), survey non-response, and post-stratification.
|
|
When analyzing environmental chemical data, it is important to select the correct subsample weight variable. Several subsample weights are associated with the NHANES environmental chemical data, and different environmental chemicals measurements may have accompanying subsample weights or share the same subsample weights depending on the chemical group of interest. Generally, when environmental chemicals were measured in randomly selected subsample of the survey participants, the data files contain specific subsample weight variables. The 1999–2000 and 2001–2002 data files and documentation includes two-year and four-year subsample weights. The four-year weights are used when the 2001-2002 data are combined with 1999–2000 data to produce a combined estimate for 1999–2002.
The following is an information note from the Continuous NHANES Web Tutorial:
Continuous NHANES Web Tutorial - Weighting module
|
|
Beginning in 2003, the survey content for each two-year period is
held as constant as possible to be consistent with the data
release cycle. In the first four years of the continuous survey
(1999–2002), this was not always the case, and special data
release and access procedures had to be developed for selected
survey content collected in "other than two-year" intervals. (For
more details see the
- Below is the list of variables of subsample weights related to environmental chemicals available from NHANES between 1999–2004. Beginning in 1999, most environmental chemical measurements were obtained after dividing the entire survey sample into three randomly assigned subsamples (i.e. subsample A, B, and C). This required that three sets of subsample weights be assigned (i.e. for subsample A, B, and C). Use of any one of these sets of subsample weights will result in national weighted estimates for the chemical in that subsample.
Variables for Subsample Weights
| Chemical Group | Specimens | 1999-2000 Subsample Weight Variable Names | 2001-2002 Subsample Weight Variable Names | 2003-2004 Subsample Weight Variable Names |
|---|---|---|---|---|
| Metals | ||||
| Heavy Metals | Urine | WTSHM2YR, WTSHM4YR | WTSHM2YR, WTSHM4YR | WTSA2YR |
| Total Arsenic and Speciated Arsenics |
Urine |
na | na | WTSA2YR |
| Persistent organic pollutants | ||||
| Dioxins, Furans, and Coplanar Polychlorinated Biphenyls (PCBs) | Serum | WTSPO2YR, WTSPO4YR | WTSPO2YR, WTSPO4YR | WTSC2YR |
| Non-dioxin-like Polychlorinated Biphenyls (PCBs) | Serum | WTSPO2YR, WTSPO4YR | WTSPO2YR, WTSPO4YR | WTSC2YR |
| Organochlorine Pesticides | Serum | WTSPO2YR, WTSPO4YR | WTSPO2YR, WTSPO4YR | WTSB2YR |
| Polybrominated Diphenyl Ethers and Polybrominated Biphenyl | Serum | na | na | WTSB2YR |
|
Polyfluoroalkyl Compounds |
Serum |
na | na | WTSA2YR |
|
Polycyclic Aromatic Hydrocarbons |
Urine |
WTSPH2YR, WTSPH4YR | WTSPH2YR, WTSPH4YR | WTSB2YR |
|
Phthalates |
Urine |
WTSPH2YR, WTSPH4YR | WTSPH2YR, WTSPH4YR | WTSB2YR |
|
Phytoestrogens |
Urine |
WTSPH2YR, WTSPH4YR | WTSPH2YR, WTSPH4YR | WTSB2YR |
|
Pesticides |
Urine |
WTSPP2YR, WTSPP4YR | WTSPP2YR, WTSPP4YR | WTSC2YR |
|
Environmental Phenols |
Urine |
na | na | WTSC2YR |
|
Perchlorate |
Urine |
na | na | WTSC2YR |
|
Volatile Organic Compounds |
Whole Blood |
WTSVOC2Y, WTSVOC4Y | WTSVOC2Y WTSVOC4Y | WTSVOC2Y |
