Researcher Supplied Dataset

You are responsible for providing the NCHS public-use dataset any non-NCHS data that will be included in the merge. Compiling the public use data set provides you the opportunity to become familiar with the data and expedite the data creation process. If you are using NHDS, NAMCS, and NHAMCS you may not need to provide a public data set: please check with your RDC Analyst.

• Non-NCHS Data includes any data collected by the researcher, another government agency, or a private institution that the researcher wishes to merge with NCHS data, often using geographic codes. Examples for policy research have included air pollution data, proximity of fast food restaurants, or location of health care providers.

When sending your public and non-NCHS data to the RDC Analyst, please explain what variables should be used to merge the various data sets. This is especially important for merges involving non-NCHS sources of data which are most commonly merged on geographic and temporal variables. Once you have provided the public and any non-NCHS data, the RDC Analyst will merge it to the restricted variables. Additionally, if geography has been used to conduct the merge and is not required for analysis, the RDC Analyst will remove the geography.

As part of the application process, you have already selected the public-use variables you will analyze, and have documented them in the required data dictionary. If you have not already done so, you must now create the public-use data files.

 

Provide the Public-Use and non-NCHS Dataset

• The public-use file that you submit can only include those variables required for your analysis. Please do not send the entire public-use files.
• Original NCHS Variables must have the name they are given in the public-use data set. If you want to rename the variables, you must include the original variable name in the variable description.
• Derived Variables must be labeled with the variables from which they were derived and any arithmetic manipulation must be explained.
• Public-Use Mortality Variables: If you are requesting access to the restricted mortality files, you cannot include any public-use mortality variables or variables derived from the public-use mortality data.
• If you have added variables that were not in the data dictionary in your approved proposal, please highlight these variables. They may require additional approval.
• Any attempt to include variables that may lead to re-identification of subjects or establishments is considered a disclosure violation and will result in the cessation of your project and possible legal actions.