External Research Resources Support and Dissemination Core

Core D将支持我们的附属机构、科学界、老年人权益倡导者、政策制定者和

Core D will support our affiliates, the scientific community, advocates for the aging, policy makers, and others who are interested in, or may benefit from, research and data resources generated by our affiliates. Specifically, we propose to continue and expand current activities funded by this core in order to support secondary users of the major NIA-supported data resources; survey researchers and secondary users who are creating significant value-added resources in health and aging; and those who can employ and apply our affiliate's research results for the good of the aging population and of society more generally. We propose to continue the intensive type of support that our Data Analyst/Archivist, Janet Eisenhauer Smith, and our Data Librarian, Chiu-Chuang (Lu) Chou, now provide to secondary users of the major NIA-funded survey projects. We will continue to match secondary data users with data resources that support their research objectives, and to encourage and train them to use web-based search engines and extraction tools as an aid to identifying and sub-setting variables and cases of interest, and in downloading data in convenient formats. When these tools are not available, we will create custom abstracts for users who cannot do it themselves. We will also continue to help affiliates whose research is innovative and promising, but who do not have the requisite experience with the major statistical packages, to program their own analyses. We also propose to create value-added data resources for aging-related data of particular interest to our affiliates or the wider research community and post these to our on-line public-use data archive which we call BADGIR (Better Access to Data for Global Interdisciplinary Research). This on-line archive is powered by the Nesstar software which has unique capabilities, so we request funds to pay for our annual license. Studies currently disseminated on BADGIR include SABE (7 countries), NSFH (3 waves), PREHCO (2 waves), and the National Health Measurement Study (NHMS). With the exception of SABE, all these studies were either substantially upgraded or newly posted this past year. Specifically, we plan to geocode the third wave of NSFHS and to create a cross-wave data file for this important study. As discussed in Core E, the geocoded wave 3 data will initially be made available only through the geo-merge procedure initially developed by NSFH for waves 1 and 2, for which we inherited responsibility for when NSFH's closed its doors approximately two years ago. At least initially, the cross-wave file is intended for internal use (see Core E), but we may eventually disseminate it on BADGIR, depending on how much progress the Minnesota Population Center has made with its own version of the cross-wave file. We also plan to support a cross-country data file for SABE that is being created by Mary McEniry, a CDE post-doctoral research affiliate. This kind of collaboration with a secondary user creating a unique and valuable set of data from one of the major NIA-funded surveys, is exactly the kind of opportunity that was envisaged by CDHA's Director, Robert M. Hauser, when he proposed to include the two data cores in his initial P30 Center Grant proposal. Because Dr. McEniry will not complete her work for a couple of years, we propose to begin our own efforts with NSFH. We have secured a commitment from Larry Bumpass, one of the principal investigators on this project, to advise us in this endeavor. When creating comparative data, identifying changes in coding and measurement of key concepts over time (or space), and calculating new variables that are measured and coded in the same way across waves are critical steps in the process. The real value in this sort of exercise is generated by creating variables that permit valid comparison of characteristics, opinions, preferences, and behavior of the same individual over time and of members of the same family. The key is to ensure that these new variables summarize similarities and differences (and the magnitude of these differences) in intuitively useful ways. In light of the questions and comments we now receive from NSFH and SABE users, we anticipate that the value-added geocoded and comparative data that we are proposing to create will be well received by secondary users of these important NIA-funded resources. Creating comparative data of value is not a trivial task and could not be accomplished at current funding levels, so we have requested an increase in funding for this core to support more of the Analyst/Archivist's time in this activity. Creating cross-wave data for NSFH and cross-country data for SABE is an integral part of a new initiative in which we propose to develop a new module for the DDI, a content and formatting standard for codebooks (metadata) in social science survey research. After developing a preliminary draft of the proposal, we will solicit comments from interested parties including the creators of BLAISE and CASES, the two most popular software programs for conducting computer aided telephone and personal interviews (called CATI and CAPI software, respectively). We also aim to solicit comments from Nesstar developers and from Jeremy Iverson, a programmer who is creating a software program that harvests detailed information from the CASES setup files including skip patterns, associated variable universes, and other important features of CATI/CAPI survey instruments. We also plan to approach Marcus Petersen at the University of Minnesota, who is developing a database application to help users navigate codebooks for longitudinal studies. We plan to submit this proposal to the DDI Alliance for inclusion in the next major revision of the metadata standard. Because Alliance resources are already fully occupied with other important issues, a committee was not formed to address this issue, despite the tremendous difficulties secondary users have in using existing longitudinal studies to their full advantage. Because of her professional experience in survey research before joining CDHA and her experience with the DDI since joining the center, the Analyst/Archivist will take the lead on this activity. The DDI Alliance is a membership-based organization, and we propose that our annual dues continue to be funded by this core. The increase in the percentage of the Analyst/Archivist's time in this core is, in part, accounted for by this activity. Because the metadata proposal to be submitted to the DDI will have to include changes and additions to the XML specification, we are also requesting funds to support a part-time XML programmer. We have also budgeted for travel for working trips to Statistics Netherlands, Minnesota, and Norway in support of this objective. Although we will continue to disseminate public-use data on BADGIR, we would also like to use it to publicize our restricted data holdings. Clearly it will not be possible to disseminate these restricted data files on BADGIR, but we do think it very important to create and publish standards-compliant informative codebooks on the internet as proposed and budgeted under Core E. Now that so much public-use data is available on-line, it has become second nature for researchers to begin a search for relevant data on the Internet. Because of this, we firmly believe that the important and useful (restricted) data from aging-related studies stored on our secure data server and available to researchers through licensing or via our remote computing system are likely to be overlooked and, as a result, underutilized if detailed information about these studies and protocols for access to them is not available online. As discussed in the opening paragraph of this proposal, we have three different audiences in mind when developing our proposed work plan, i.e., secondary users of the major NIA-funded survey projects, data producers who are funded to produce these surveys, and other groups and individuals with an interest in aging. For our survey researchers and secondary users, who contribute significantly to the pool of survey and administrative data about health and aging, the Analyst/Archivist will continue to advise and assist in developing data sharing plans to publicize and disseminate these data as widely as possible. Funding from this core will be used to create standards-compliant codebooks for these studies, and to serve codebooks and data up on our online data archive, BADGIR. We will provide access to restricted data through one of two mechanisms: By license in our secure Cold Rooms, where qualified (licensed) researchers may analyze these restricted confidential data themselves, and via our remote computing service, which provides access to restricted data for users who do satisfy licensing criteria. See Core E for a description of these services. The final activity that we are proposing under this core is to continue to maintain and develop CDHA's website as a source of information for CDHA members, other researchers, and the public at large. The website hosts a searchable directory of affiliates, a searchable bibliography containing citations for aging-related publications produced by our affiliates since 1999 when the center was first funded, and a searchable projects database describing aging-related research in which are affiliates have been engaged over the past ten years. The website also serves as an archive for CAAR reports and provides information on our weekly seminars and on special workshops and conferences hosted by CDHA. The databases that populate our website also populate our annual progress reports: without these it would be impossible for us to create the summary appendices which are so useful to NIA in justifying their budget requests to Congress and to the public who benefit from the research spearheaded by NIA. Although web programming does not absorb very much of our Special Librarian's time, collecting and entering data for the annual report is a very time-consuming activity and, on an annual basis, absorbs the majority of her time. We are proposing to take better advantage of our Librarian's expertise and are requesting funds that would allow us to hire students and a PA in Library Science who will assist her in collecting and entering the required information into our databases. In all of these activities, the External Research Resources Support Core of CDHA complements and extends services of the print library for the Center for Demography and Ecology (CDE), and the Data and Information Services Center (DISC), the administrative unit which currently houses services provided by CDHA and by the CDE Data Library and the Data and Program Library Service. CDHA's data services do not overlap with those provided by either the CDE print library or the other data service units which comprise DISC. In fact, CDHA provides no print library services, so CDHA members rely entirely on the CDE print library for those. CDHA's data Core provides support for CDHA researchers that is more intensive and user-friendly than that provided to CDE affiliates and to the rest of the social scientists on campus, and it acquires, manages, and disseminates data resources that are of unique interest to researchers in the demography of health and aging.