We model the Social Security Administration's (SSA's) disability deter
mination process using household survey information exact matched to S
SA administrative information on disability determinations. Survey inf
ormation on health, activity limitations, demographic traits, and work
are taken from the Survey of Income and Program Participation (SIPP).
We estimate a multistage sequential legit model, reflecting the struc
ture of the determination procedure used by State Disability Determina
tion Services agencies. The findings suggest that the explanatory powe
r of particular variables can be appropriately ascertained only if the
y are introduced at the relevant stage of the determination process. H
ence, as might be expected by those familiar with the process, medical
variables and activity limitations are major factors in the early sta
ges of the process, while past work, age, and education play roles in
later stages. The highly detailed administrative information on outcom
es at each stage allows clarification of the roles of particular varia
bles. Planned future work will include policy estimates, such as the n
umber of persons in the general population eligible for the disability
programs, as well as analysis of applications behavior in a household
context.