For this project, SNAP administrative data from Illinois, Mississippi, and Tennessee were linked to multiple years of data from the Current Population Survey Annual Social and Economic Supplement, which serves as the base of an annual SNAP eligibility model used to generate the denominators for estimates of SNAP participation rates at the national and State levels. The ultimate goal of this project is to improve estimates of SNAP participation rates by using the knowledge gained to enhance the simulation of eligibility, particularly among subgroups where the simulation appears most problematic. In the report, we describe our creation of the linked data, present estimates of subgroups of SNAP participants and units for which the estimated participation rates are at issue, examine selected characteristics of these subgroups, show the extent of multiple SNAP units and nonparticipating members conditional on these household characteristics, and discuss the implications of these findings for SNAP eligibility simulation.