NCHS Research Data Center Prospective researchers must submit a research proposal that will be reviewed and approved by a committee whose judgment is based upon the availability of RDC resources, consistency with the mission of NCHS, general scientific soundness, and the feasibility of the project. It is expected that the user will develop the research proposal with the RDC staff to minimize the time required. Although researchers will sign confidentiality agreements, strict confidentiality protocols require that researchers with approved projects must complete their work using the facilities located within the RDC. Researchers can supply their own data to be merged with NCHS data sets. Completed by the RDC staff, the merged files will be only available to the originating researcher unless written permission is given to allow access to others. Two access methods are available for researchers to accomplish their tasks: Onsite and remote. Onsite researchers have the ability to use the full capabilities of the SAS system with the only caveat being a disclosure review. However, major restrictions limit the analysis capabilities of the remote access system. Totally scanned and screened, printed output using SAS procedures and functions adhere to strict minimal disclosure limits and are suppressed if the minimums are not met. Another analysis option for researchers with large, complex analytic projects may be to subcontract with the NCHS RDC to perform the tasks necessary for the research project. Regardless of the option chosen, each has an associated cost that includes capital and labor expenses involved in setting up and monitoring each project. Cost estimations both in terms of time and money, can only be as accurate as the degree of specificity of the research. Again, with the help of the staff of the RDC at an early stage, the estimates should reasonably approximate reality. Details Suggest to make a trip early in the application process to RDC Best to develop in conjunction with RDC staff Portions of doctoral proposals or grant applications may suffice Detailed summary of proposed research including why publicly available data are not sufficient Full personal and institutional affiliation required Dates of proposed visits to the RDC An estimation of the total hours expected to be working in the RDC Specific NCHS data requested including year or years and variables Details of the user-supplied data to be merged with NCHS data Software requirements including language - SAS for Windows and Stata are available Required to sign Agreement Regarding Conditions of Access to Confidential Data in the Research Data Center of the National Center for Health Statistics Required to sign the Researcher Affidavit of Confidentiality Acceptability based on availability of RDC resources, consistency with the NCHS mission, and the feasibility of the project Operations RDC staff constructs necessary data files including merged user data PC SAS, SUDAAN, STATA, FORTRAN and HLM available Other statistical packages available with sufficient lead time Output subject to disclosure review Disclosure review guidelines published in NCHS Staff Manual on Confidentiality Analyses (paper output) passing disclosure review can be taken off site Center requires staff oversight Only open during normal working hours Remote Access RDC staff constructs necessary data files including merged user data Ability to submit analytical computer programs via e-mail Output returned by e-mail to users registered address SAS programs only, certain procedures and functions not allowed PROC TABULATE and PROC IML not allowed LIST and PRINT not allowed - no listing of individual cases R_, FIRST., LAST., not allowed - no selection of individual cases No cell fewer than five observations; if found, other cells also supressed Job log scanned for conditions that spawn case listings Manual disclosure limitation review where necessary Staff Programming Subcontract with the RDC staff to perform programming tasks Specifications from researcher with input from
RDC staff Estimated for each research project Includes equipment and space rental, staff overhead and setup Knowledge of research necessary to estimate costs Specific costs: On site researcher cost $1,000 per week Remote access for National Survey of Family Growth contextual data files $500 per year Remote access for any one data set $500 per month File construction and setup $500 per day of effort Multiple workstations High speed Pentium computers All have Windows 95 All have SAS for Windows No removable media Secured printer Server for RDC private network Data National Health Interview (NHIS) - State and County level identifiers 1995 National Survey of Family Growth (NSFG) - contextual data files Third National Health and Nutrition Examination Survey (NHANES III) - County level identifiers National Hospital Discharge Survey - State identifiers Detailed Mortality, 1968-Present Other data from NCHS can be made available such as vital statistics or the Longitudinal Study on Aging with restricted data such as lower levels of geography, sample design information, etc., included Data from other HHS Agencies Assistant Secretary for Planning & Evaluation (ASPE)-sponsored data on Welfare Outcomes and NEWWS. User-Supplied Data Can be proprietary data collected and "owned" by the user Can be other publicly available information such as census data Consulting with RDC staff necessary To ensure data can be merged with NCHS data Format of the data must be consistent with NCHS data Merge is done before the arrival of the researcher Files and information used in linking that will be destroyed after the merged file is created and will not be made available to the researcher Any merged data set created belongs to the researcher Researcher can give access to others with written consent only Periodic backup files created and maintained by RDC staff Contain user-supplied data and merged files Stored in secure area Only destroyed upon written request of the user For
complete details and application form contact:
This page last reviewed March 03, 2003
|