General Information about NHANES Documentation Files
The data documentation, codebook, and frequencies for each Demographic, Dietary, Exam, Lab, and Questionnaire component of the survey have been combined into a single master document, accessible from the single Docs link corresponding to that particular survey component.
This combined documentation file is one of several important references to use in analyzing the data. It should be used in conjunction with the specific Exam/Lab/Questionnaire procedures manual and/or questionnaire instrument, the analytic guidelines, and other information available on the NHANES website.
The data documentation section of the master document contains detailed information about each particular survey component including the component description, the eligible sample, protocol, and other analytic notes and references. The codebook section of the master document provides detailed information about each data item that is available in the public release data file. At the top of a codebook is the name of the section or component represented in the codebook and the type of records on the file. The first item listed in each codebook is the sequence number or ‘SEQN’. This data field is used to merge different data files together and uniquely identifies each survey participant (sample person).
The following metadata (information about a data item) is incorporated into the NHANES codebooks:
|Item (or Variable) Name||The actual eight character maximum alphanumeric name of the data item.||SEQN, VIQ180||Short eight character names have been chosen to allow the SAS transport files to be compatible with the SAS Statistical Package and other appropriate packages.|
|Target||Age group and gender of survey participants eligible for this item.||Target Gender: Both males and females
Target Age: 0 YEARS - 150 YEARS
represents both genders and all ages are eligible.
Target Gender: Females only
Target Age: 20 YEARS - 44 YEARS
represents females 20-44 years of age are eligible.
|The Target Gender is listed as Males only, Females only, or Both males and females. For Target Age, the number 0 indicates all children less than 1 year of age and the number 150 represents the maximum age eligible in the survey.|
|Hard Edits||Edits automatically performed by a machine or the survey instrument, or enforced by other means at the time of data collection.||20 to 200||In some cases the instrument forces the examiner or interviewer to re-measure or re-ask a question since the original value is not within the accepted range. Some hard edits are built into the bio-medical machines while others are programmed into the data collection software. If hard edits are not applicable to an item, they are left blank on the codebook.|
|SAS Label||A brief 40-character description of the data item.||Eye surgery for near sightedness?||The SAS label is a 40-character label so that the SAS transport data files can be read by multiple versions of SAS, SPSS, and STATA, which may not allow for long data labels.|
|English Text||For questions in the examination or household/sample person questionnaire this is the actual question text. For examination and laboratory measurements this is a description of the measurement.||Have you ever had eye surgery to treat or prevent nearsightedness or myopia?||We also include the English text for the item. For questionnaires, this is the text of the question in the questionnaire. For examination and laboratory items, this is the description of the item seen by the technician on the computer screen in the MEC.|
|English Instructions||Specialized instructions for the interviewer or examiner conducting the interview or examination.||The interest is in laser or radial keratotomy (RK) surgeries used to improve nearsightedness. Some other terms to describe this surgery include: PRK, RK, Lasik, Eximer, or Refractive Surgery.||The English Instructions are instructions to the interviewer or examiner ONLY and are not read as part of the question to the respondent/examinee.|
|Codes||Domain of possible values for a data item. Includes code, description, frequency of occurrence (count), and Skip to Items, if applicable.||
|For data items that are continuous in nature, the codes are blank. Code descriptions, counts of occurrence, and Skip to Items, if applicable, are included to the right of the codes. Some of the codes will have repeating digits or fill values. For example a ‘Refused’ for a data item that is five digits in length would appear in the data as 77777.|
In some questionnaire sections, ‘check items’ providing logic about item eligibility are included to better understand the flow thru a section. These ‘check items’ usually incorporate information about answers given in previous sections or question items. A user must also look at the item targets to understand who is eligible for a particular question. Please note however that the release data does not necessarily include all of the questions asked in a section. Data items may have been removed due to confidentiality, quality, or other considerations. Thus, it’s possible that a dataset does not completely match all of the questions asked in a questionnaire section.
Each data file has been edited to include only those sample persons eligible for that particular section or component. A blank record for any sample person indicates that the person did not participate in that particular questionnaire section or examination/laboratory component, although they were eligible. Distinct responses of refused (7-fill) and don’t know (9-fill) are included in the questionnaire data. Reasons for missing data in the exam and laboratory components can be found in the completion status and the comment codes for that section. Individual items for any section or component may be missing due to different targets within a section or component, or as a result of previous results or responses. To completely understand each section or component, the entire survey section/component content should be reviewed.