Participation has been shown to be a protective factor for cognition in older adults, but instruments to assess it are limited. The main objective was to determine the validity of two scoring systems (direct vs transformed) for the Participation Assessment with Recombined Tools-Objective (PART-O) by applying structural equation modeling to the relationship between the subscales and the cognitive functions, in a sample of 245 people over 60 years of age. The transformed scores model showed stronger relationships and larger explained variance in overall participation (55.4% vs 37.4%), especially in the Social Relations subscale (31.4% vs 14.6% ). Participation was a direct predictor of cognitive functions in both models. Age and depression inversely influenced participation in the transformed scores model. The proposed score transformation for the PART-O provides a more appropriate measurement of the older adults’ participation. Participation has a mediating role in the relationship between cognition and both age and depression.