The Korea National Health and Nutrition Examination Survey as a Primary Data Source

article should be submitted as soon as possible. Otherwise someone else may publish a similar study from the same dataset before you do. Second, investigators must be familiar with the huge and complex data structure and study design of the KNHANES. Nonresponse and multi-stage probability sampling should be taken into account by using survey statistics to estimate the data of the whole Korean population. Despite this, lots of studies submitted to KJFM using the KNHANES do not apply sampling weight in their analysis. In such a case the estimated association might be biased and revision is required. Third, the KNHANES is a secondary data source like the Korea Youth Risk Behavior Web-based Survey and the Community Health Survey, and has the same limitations that they have. The main purpose of the KNHANES is to produce national statistics, not to answer a specific research question. Therefore, the particular information the investigator wants may not be collected. For these reasons, the dataset should be examined carefully to confirm that it includes the necessary data. There also exists a potential for errors or mistakes in the data, even with its rigorous documents control system. Missing data is also a problem in the KNHANES.