This dissertation examines pain, emotional well-being, affect variability, pain variability and mindfulness in community-dwelling older adults diagnosed with osteoarthritis (OA) of the knee. Osteoarthritis is one of the most common chronic illnesses. Its sufferers experience a great deal of pain and a potentially substantial decline in emotional well-being. This data comes from an ongoing research project, Everyday Quality of Life in Osteoarthritis (EQUAL; R01 AG046155), which examines quality of life among African American (AA) and non-Hispanic White (NHW) older adults with a diagnosis of OA. Subjects aged 50 and older complete a comprehensive baseline assessment, as well as an experience sampling method (ESM) procedure. Baseline measures include Philadelphia Geriatric Center Pain Scale, Freiburg Mindfulness Inventory, Center for Epidemiological Studies Depression Scale, as well as the short form Spielberger State Anxiety Scale. Variability was examined via the ESM procedure which includes responses to mood and pain questions 4 times daily over 7 consecutive days. Structural equation modeling (SEM) was used to examine whether traditional summary measures of pain and overall well-being are predictors of within-day pain and affect variability. Second, the role of mindfulness as a moderator was examined. Path analyses indicated that baseline pain, negative affect, and number of pain locations predict positive affect variability; negative affect predicts negative affect variability; and negative affect predicts pain variability. Mindfulness moderated the effect of emotional well-being on pain variability, but did not moderate the remaining three hypothesized relationships. These results highlight the complex nature of pain and affect in older adults suffering from OA, and how variability and mindfulness may affect that relationship.