The objective of this study was to determine the internal validity and separation reliability of the Inventory of Reading Occupations - Adult (IRO-Adult), an assessment of functional literacy participation for adults. Participants included 69 adults with typical reading abilities and difficulties with reading. The fit of the IRO-Adult items was analyzed with the Rasch model of unidimensionality using goodness-of-fit analysis and analysis of standardized residuals. Also analyzed was the tool's ability to reliably separate distinct levels of reading participation and the optimal functioning of its rating scales. The IRO-Adult items fit the Rasch model of unidimensionality. Three test items can be further refined to strengthen internal validity. Items are able to reliably detect multiple levels of reading participation and change in functional literacy participation over time. Several categories in the tool's rating scale can be combined for better clarity and to optimize rating scale functioning. The study provides preliminary evidence of validity and reliability of the IRO-Adult as a useful tool to assess functional literacy participation of adults. Further studies on other psychometric properties and clinical utility are warranted.