In developing countries like Uganda, people with dementia are cared for by non-medically trained family members with minimal support from the formal healthcare system. The quality of care in this setting is largely unknown but significantly affects the well-being of those with dementia. A tool designed to measure the quality of informal care for old frail adults with or without dementia was translated into Luganda. A committee of experts reviewed and finalized the translation, which was pilot-tested and then used to measure the quality of dementia self-care. We consecutively enrolled 105 caregivers of elderly people with dementia; the median age was 35 years (Interquartile Range 26-47 years), and 67% were females, taking care of a grandparent (44%) or a parent (34%). We used confirmatory factor analysis to assess for structural validity and computed correlation coefficients and Cronbach's alpha to assess for discriminant validity and internal reliability, respectively. The three-factor model applied to the 20 items, adequately fit the data (Comparative Fit Index = 0.88, Tucker-Lewis Index = 0.87, Root Mean Square Error of Approximation = 0.08; 90% Confidence Interval (0.06-0.09), Standardized Root Mean Square Residual = 0.089). There was good discriminant validity, and correlation coefficients between dimensions/scales and the Dementia Knowledge Assessment Scale scores were low. There was good internal reliability with all items Cronbach's alpha ranging from 0.69 to 0.89. Our findings demonstrated that this culturally adapted, shorter measurement tool is valid and reliable. The tool can be used by researchers, health workers, and agencies to assess the quality of self-care for elderly people with dementia in Uganda.