Key Concepts About Weighting in NHANES II

Weights are created in NHANES II to account for the complex survey design (including oversampling), survey non-response, and post-stratification.  When a sample is weighted in NHANES II it is representative of the U.S. civilian non-institutionalized Census population.  

How weights are created in NHANES II

Each sample person in the NHANES II dataset is assigned a sample weight. This sample weight is created in three steps: 

  1. the base weight is calculated;
  2. adjustments for non-response are made; and
  3. post-stratification adjustments are made using independent estimates provided by the US Bureau of the Census of the US population.

 

(1) Calculating the base weight

In general a sample person is assigned a weight that is equivalent to the reciprocal of his/her probability of selection.  In other words:

formula sample person's weight equals one divided by the probability of selection

However, calculating the base weight for a sample person in NHANES II is much more complicated due to the survey's complex, multistage design. In NHANES II, the following equation, which takes into account the survey design, is used to determine the base weight for a sample person:

formula base weight equals one divided by the final probability

where

formula final probabiltiy equals probability of PSU being selected * probability of a segment of the PSU being selected * probability of a household being selected * probability of an individual being selected

 

(2) Adjusting for non-response in NHANES II

Adjustment to the interview or exam

The base weights were adjusted for non-response to the in-home interview when creating interview weights and further adjusted for non-response to the MEC exam when creating exam weights.

In NHANES II, an individual can be classified as a non-respondent to the interview portion of the survey and/or the exam portion. An individual is considered a non-respondent to the interview if he/she was selected to be in the sample, but did not complete the medical history questionnaire in the in-home interview. Similarly, an individual who agreed to complete the interview but did not agree to, or come in for, the MEC portion of the survey is considered a non-respondent to the exam. Adjustments made for survey non-response account only for sample person interview or exam non-response, but not for component/item non-response (i.e., a sample person declined to have their blood pressure measured in the examination component but completed all other examination components).

As with continuous NHANES, separate weights are provided to account for unit non-response to the various portions of the survey (interview, MEC exam, and subsamples). 

In general for NHANES II, 27,801 persons were selected into the sample, 25,286 completed the interview, and 20,322 were MEC-examined. For detailed response rates by age and other selected demographic characteristics, please click the link below to see the Table of Response Rates for NHANES II.

For more information on component/item non-response adjustment and re-weighting the data for analyses, see

1. Lohr, Sharon L. Sampling: Design and Analysis, pp.265-272. Duxbury Press, 1999; and

Examples of papers with re-weighted NHANES data

2. Gregg E, Sorlie P, Paulose-Ram R, Gu Q, Wolz M, Eberhardt MS, Burt VL, Engelgau MM, and Geiss LS. Prevalence of lower extremity disease among persons 40 years and older in the US with and without diabetes. Diabetes Care. 2004 Jul;27(7):1591-7.

3. Ostchega Y, Dillon CF, Lindle R, Carroll M, Hurley BF. Isokinetic leg muscle strength in older americans and its relationship to a standardized walk test: data from the national health and nutrition examination survey 1999-2000. J Am Geriatr Soc. 2004 Jun;52(6):977-82.

 

Adjustment for NHANES II subsample components

NHANES II respondents were asked to participate in a variety of survey components that are statistically defined (or random) subsamples of the NHANES MEC-examined sample. These included a variety of lab components (Please see the respective survey protocol/documentation for more specific information.)  For example, some, but not all, participants were selected to give a fasting blood sample on the morning of their MEC exam. The subsamples selected for these components were chosen at random with a specified sampling fraction (for example, 1/2 of the total examined group) according to the protocol for that component. Each component subsample has its own designated weight, which accounts for the additional probability of selection into the subsample component, as well as the additional non-response.

Subsample weights are not designed to be combined. In fact, many subsamples are mutually exclusive. If it is necessary to combine two or more subsamples for your analyses, appropriate weights would need to be recalculated. However, details on how to recalculate weights when combining subsamples go well beyond the scope of this tutorial. Therefore, it is strongly advised that you do not attempt to combine subsamples in any analysis of NHANES II data.

In NHANES II there are six sample weights for special laboratory components:

  1. The oral glucose tolerance test or “fasting” subsample (approximately one-half of all participants 20-74 years);
  2. The lead subsample (all participants 6 months -6 years and one-half of all participants 7-74 years);
  3. The carboxyhemoglobin subsample (one-half of all participants 3-74 years);
  4. The bile acids subsample (subsample of adults 35-74 years);
  5. The “anemia” subsample or the folate, vitamin B12, and differential leukocyte counts subsample (random ten percent sample of examined participants 3-74 years plus persons who had a predetermined “high” or “low” value for one or more of the following: hemoglobin, hematocrit, red blood cell count, white blood cell count, and mean corpuscular volume); and
  6. The serum ferritin subsample (subsample of participants 3-74 years).

Each of these samples has its own designated weight, which accounts for the specific probability of selection into that subsample, as well as the appropriate non-response rate associated with it.  Variable names of these weights are provided in Task 1.

As mentioned previously, these sample weights are not designed to be combined. Therefore, it is strongly advised that you do not attempt to combine these subsamples in NHANES II.

 

 

(3) NHANES II post-stratification adjustment to match independent estimates provided by the US Bureau of the Census of the US population

In addition to accounting for sample person non-response, weights are also post-stratified to match the population control totals for each sampling subdomain. This additional adjustment makes the weighted counts the same as an independent count of the US population as provided by the U.S. Bureau of the Census.

Post-stratification for NHANES II was performed to match the U.S. population counts. The estimates of the number of examined persons were ratio adjusted within each of 75 age-sex-race cells to independent estimates provided by the U.S. Bureau of the Census of the population as of March 1, 1978, the approximate midpoint of the survey.  The ratio adjustment used a multiplication factor in which the numerator was the U.S. population and the denominator was the sum of the weights adjusted for non-response for the specified population subgroup. This ratio estimation process brings the population estimates into close agreement with the U.S. Bureau of the Census estimates of the civilian, non-institutionalized population of the U.S.

 

Summary

In summary, it is important to utilize the weights in analyses to account for the complex survey design (including oversampling), survey non-response, and post-stratification in order to ensure that calculated estimates are truly representative of the U.S. civilian non-institutionalized population. 

 

close window icon Close Window