The Research Archive on Disability in the U.S. (RADIUS)

The heart of the RADIUS archive is a collection of 19 studies comprised of over 25,000 variables and over 1.5 million (unweighted) cases that address the topic of disability. These data sets permit analyses on topics such as: the incidence and prevalence of specific diseases, disorders, and impairments, including deficits of cognition, emotion, physiology, and anatomical structure; functional limitations across a variety of specific organ systems; disabilities in relation to major life roles and activities, such as work, parenting, education, and recreation; societal limitations including physical, attitudinal, and economical barriers that restrict full participation in society; psychosocial and interpersonal factors such as coping with stress, sexuality, feelings of control and productivity, quality of life, and family relations and support; health care and rehabilitation issues such as medical costs, coverage, service utilization, use of orthotic, prosthetic, assistive devices, effectiveness of rehabilitation; as well as a variety of basic demographic factors on respondents such as age, race, sex, income, occupation, marital status, family size, and living arrangements. The purpose of the project is to facilitate access to the best data sets on the prevalence, incidence, correlates, and consequences of disability in the U.S.

Studies and data sets for the archive were selected with the help of a National Advisory Panel of experts using scientific criteria of technical quality, substantive utility, policy relevance, and potential for secondary data analysis. Because more high quality data sets were identified than could feasibly be archived, the National Advisory Panel chose not to select several high quality data sets that were readily available from government sources. A full list of data sets considered by the Panel appears as an appendix in the RADIUS Comprehensive User's Manual. RADIUS was established with the guidance of a National Advisory Panel of experts and funded by the National Center for Medical Rehabilitation Research (NCMRR) within the National Institute for Child Health and Human Development (NICHD).

The RADIUS archive includes 19 studies-with complete data files, SAS and SPSS setup files, documentation, and search and retrieval software. Click on any study title (below) to view a study abstract and ordering information.

RADIUS Studies include:

• 1991 National Maternal and Infant Health Follow-up Survey [National Center for Health Statistics]
• National Pediatric Trauma Registry, 1988-1994; (Phase II: 1988-1992) [Carla DiScala and Bruce M. Gans]
• 1986 ICD Survey: Employing Disabled Americans [The International Center for the Disabled].
• 1986 Survey of Disabled Americans: Bringing Disabled Americans into the Mainstream[International Center for the Disabled].
• 1994 NOD Survey of Americans with Disabilities [The National Organization on Disability].
• 1991 NOD Survey of Public Attitudes Toward People with Disabilities [The National Organization on Disability].
• National Longitudinal Transition Study of Special Education Students, 1987-1991[Stanford Research Institute: Mary Wagner, Kathy Valdes, and Associates].
• School-to-Work Transition Survey of Deaf Youth, 1986-1992 [Thomas E. Allen, Brenda W. Rawlings, Arthur N. Schildroth, and Kay H. Lam]
• Annual Survey of Deaf and Hard-of-Hearing Children and Youth, 1975-1994 [Arthur N. Schildroth, Thomas E. Allen, Sue A. Hotto, Kay H. Lam, and John K.C. Woo]
• The Next Generation of Upper-Limb Prostheses, 1994 [William H. Donovan, Diane J. Atins, Denise C.Y. Heard].
• 1988 National Maternal and Infant Health Survey [National Center for Health Statistics].
• Rehabilitation Services Administration-Social Security Administration Data Link, 1980-1988 [Rehabilitation Services Administration, and Social Security Administration].
• 1994 National Survey of Sexuality Issues among Women with Physical Disabilities [Margaret A. Nosek].
• Survey of Income and Program Participation Core and Disability Modules, 1992/1993[Bureau of Census, Jack McNeil].
• 1991 Study of Outpatient Physical Therapy Practices [Alan Jette].
• Medical Outcomes Study, 1986-1992 [Alvin R. Tarlov John E. Ware, Jr. Kenneth B. Wells Anita L. Stewart Cathy D. Sherbourne Ron D. Hays William H. Rogers Sheldon Greenfield Sandra H. Berry] .
• 1994-1995 National Health Interview Survey on Disability, Complete Phase I [National Center for Health Statistics].
• 1994-1995 National Health Interview Survey on Disability, Complete Phase II [National Center for Health Statistics].
• 1998 NOD/Harris Survey of Americans with Disabilities [The National Organization on Disability].

The comprehensive RADIUS archive can be ordered on a set of 6 CD-ROMs which includes complete data, instruments, documentation, and search & retrieval/extract software for all 19 studies as well as machine readable User's Guides for each study and a Comprehensive User's Manual.

Go to: Standard Data Archive Product Features.

Download the README file for the above data sets.

Search & Retrieval Product Features. RADIUS was the first Sociometrics archive to include the following four search & retrieval innovations.

1. Although all other data archives at Sociometrics have included search & retrieval capabilities based on an assigned Topic and Type code for each archived variable, RADIUS includes primary and secondary Topic codes (in addition to a Type code) for variables that require them. This innovation enhances the ability of RADIUS users to search for and retrieve variables related to their particular interests.
2. RADIUS allows the user to highlight a variable of interest and then to immediately display its unweighted univariate statistics (including total valid N, mean, mode, SD, Min & Max values, and total N missing) and, for variables with fewer than 20 unique values, the unweighted frequencies of each value, including designated SPSS missing value frequencies. This will be useful to archive users who want to quickly determine whether variables of substantive/topical utility may also meet criteria for statistical utility for desired analyses, e.g., having sufficient number of valid cases (total or within strata of interest), meeting assumptions of variance, and other distributional characteristics. Thus, users will be able to efficiently make more informed choices about variables to include or exclude from SPSS and SAS program extracts.
3. RADIUS includes machine-readable User Guide files (in addition the printed Guides) as part of the search & retrieval software. This allows users to quickly search for study summaries (overviews of methodology, etc.) and similar documentation for any study in RADIUS as well as to perform full-text keyword searches within or across User's Guides in the archive.
4. The RADIUS extract software allows users to create SPSS/Windows command extract files.

General Criteria for Ranking Prospective Data Sets. The following criteria were used by the National Advisory Panel to assess the priority of prospective data sets for RADIUS. Since initial discussions by the Panel indicated that too many data sets would be ranked highly than could feasibly be archived, an additional selection criterion was considered - current public availability from reliable sources. Consequently, several data sets were not given "high priority", even though they were acknowledged as having high technical and substantive qualities, because they were viewed as readily available from government sources, e.g., several data sets distributed by the National Center for Health Statistics. These and other meritorious data sets are listed with brief descriptions in an appendix to the RADIUS Comprehensive User's Manual so that archive users will be aware of highly regarded disability data sources that were not archived by RADIUS.

1. Technical quality - The factors considered were high response rates, low attrition rates, use of reliable and valid measures, and sound sampling and design elements;
2. Substantive importance to field - Factors included the potential to address contemporary issues, to break new ground, and to replicate or confirm important findings;
3. Potential for secondary analysis, including:

(national) scope of sample - The broader or more diverse the scope of the sample, the greater the potential of the data for generalization;

size of sample - Sample size is always an important consideration. This is even more true for data intended for secondary analysis: sample sizes adequate to support the originally intended analysis may be too small to support other analyses, especially if the new analyses focus on data cells that have a very low proportion of cases;

breadth of variables and constructs covered - The potential for secondary analysis is directly related to the breadth of variables measured in the data set. The more numerous and diverse the set of variables, the more possibilities there are for new or expanded analyses; and

4. Disciplinary balance - An archive should attempt to be representative of the entire field of research. Variations in state of the art exist between different subareas within any discipline. Thus a somewhat flexible standard (as measured by the other criteria above) was used to ensure that all major areas of the discipline were represented in the archive. Thus, a range of epidemiological, sociological, psychological, and clinical studies - encompassing both basic research and rehabilitation research perspectives - are represented in the collection.

Substantive Criteria for Inclusion. RADIUS includes data sets that permit analyses on topics such as: the incidence and prevalence of specific diseases, disorders, and impairments, including deficits of cognition, emotion, physiology, and anatomical structure; functional limitations across a variety of specific organ systems; disabilities in relation to major life roles and activities, such as work, parenting, education, and recreation; societal limitations including physical, attitudinal, and economical barriers that restrict full participation in society; psychosocial and interpersonal factors such as coping with stress, sexuality, feelings of control and productivity, quality of life, and family relations and support; health care and rehabilitation issues such as medical costs, coverage, service utilization, use of orthotic, prosthetic, assistive devices, effectiveness of rehabilitation; as well as a variety of basic demographic factors on respondents such as age, race, sex, income, occupation, marital status, family size, and living arrangements.

Substantive Criteria for Exclusion. To provide further substantive focus that is consistent with the goals of the RADIUS funding agency and overall limitations in archiving resources, data sets were excluded if they focused primarily on the following topics: emotional conditions (depression, paranoia, schizophrenia, head aches, etc.), cognitive developmental disabilities (mental retardation, autism and epilepsy), or such primary medical conditions as diabetes, arthritis, cardiac conditions, pulmonary problems, cancer, inborn errors of metabolism, and others not generally considered as the province of medical rehabilitation research. It is hoped that RADIUS can be extended in the future to include excellent data on these topics.

Topic codes for RADIUS were developed to maximize consistency on three dimensions: (1) consistency with the National Center for Medical Rehabilitation Research (NCMRR) model and definitions of pathophysiology, impairment, functional limitation, disability, and societal limitation, (2) consistency with the International Classification of Impairments, Disabilities, and Handicaps (ICIDH), and (3) consistency with the Topic codes employed by the Data Archive of Social Research on Aging (DASRA) and other Sociometrics data archives. The most important modification of the DASRA Topics for use with RADIUS was the reformulation and addition of several Topic codes (based on ICIDH categories of impairments and disabilities) that reflect the NCMRR model of impairments (RADIUS Topic codes IA - II & IZ) and functional limitations (RADIUS Topic codes JA - JI & JZ).

Because disabilities and societal limitations can be associated with a wide variety of roles and social activities, a single code for disability: superordinate/multiple (DS) and a single code for societal limitation: superordinate/multiple (SL) were created with the expectation that each would often be paired with a secondary Topic code that represents a specific role, social activity, or life domain. For example, an item that focused on a job related disability would have been coded for occupation (OC as the primary Topic) and disability (DS as the secondary Topic). An item that focused on a barrier to education would have been coded for education (ED as the primary Topic) and societal limitation (SL as the secondary Topic). The relationships among these codes are shown below.

To Receive RADIUS Announcements: mail, fax or e-mail your name and mailing address to: