lation eligible for disability benefits) as well as public access to, utilization of, and satisfaction with the nation’s disability benefits programs. Although resources may dictate many of the system’s design features, the committee urges SSA to make its system one in which data are collected on a continuing basis from a valid and statistically efficient national sample of households.
SSA should consider using data from existing federal surveys in designing the monitoring system and should use these information sources to supplement data generated from the SSA system. Besides serving as a supplementary source of disability data, integration with the design features of these surveys might prove beneficial both to SSA’s monitoring system and to the other surveys. This level of design integration has already been successfully accomplished between the Medical Expenditure Panel Survey conducted by the Agency for Healthcare Research and Quality and the National Health Interview Survey conducted by the National Center for Health Statistics (Cohen, 1999).
To design and establish a high-quality SSA disability monitoring system, the committee suggests that a planning period of approximately three years would be needed. Three important tasks must be completed during this planning phase. First, SSA must recruit a sufficient cadre of qualified staff to conduct the design work and directly oversee the initial field-testing of the system largely from within SSA. The implication is that SSA should develop this system largely from within and not rely heavily on external contractors to do the work. Among those professional expertise areas that SSA would need to recruit are experts in: functional ability, cognitive measurement, survey design, and analysis.
The next planning phase would be for SSA to develop a detailed blueprint for the system. This step suggests that pilot testing an early version first in a few states or metropolitan areas before national implementation might best phase in the system. The goal here would be to produce a system design that is useful to SSA, scientifically sound, and able to withstand careful scrutiny, yet sufficiently flexible to adapt to changing information needs over time. Toward these ends, the committee suggests that SSA support a careful study of the cognitive and process effects of measuring disability in a survey context. At a minimum this research should answer the following questions: What are the effects of the mode of gathering the data (e.g., self-administered, telephone interview, face-to-face interview)? How is the portrayal of disability influenced by who provides the data (e.g., the disabled person, a proxy caregiver, a health care provider)? How do question wording, context, and format affect the picture of disability painted for a disabled person? Ultimately, the role of this research on measurement effects is to understand the effect of the survey design strategies used to measure disability