partnership with another survey to add a small set of measures in return for financial support of that survey.
Under this option, SSA would design an ongoing survey whose goal would be the estimation of prevalence of benefit eligibility. To be useful it would include as measures, those questions found most predictive of eligibility in the NSHA.
A decision by SSA to sponsor an independent ongoing survey to estimate the size and characteristics of the eligible population would be based on answers to several questions such as: What is the scope of SSA’s ongoing information needs that can be combined with those of disability monitoring, and what are the administrative, financial, and technical staffing burdens of designing and estimating operation for an ongoing survey? What are the results of the statistical analysis to identify a subset of measures and how detailed a measurement is required to achieve minimally acceptable sensitivity and specificity parameters? Is a separate ongoing survey needed or can the needed subset of measures be obtained from an existing continuing survey? If there are many characteristics of the population that are not now being well described in the existing surveys, then a separate SSA survey may be justified as a small fraction of the total funds allocated to fulfill its mission.
An alternative to sponsoring an ongoing survey is to add a limited set of work disability indicators to an ongoing survey sponsored by another agency. This alternative would provide continuous information about the size of the eligible population by forming a partnership with a household survey of sufficient periodicity and size. In this option, SSA would, through an interagency transfer of funds, support the testing and implementation of a short set of questions that would provide prevalence estimates as add-on or as supplement to the regular survey. This option differs from those above in that it offers SSA some control over the data used in the monitoring effort but less control in terms of content, definitions, and timing than offered by its own ongoing survey.
Survey Partners in Disability Monitoring. Several federal statistical agencies currently include some measurement of disability in one or more of their household data collection efforts; several other statistical agencies are currently developing such measures for inclusion in their studies. The candidate surveys for ongoing monitoring include the American Com-