Objective: Older adults often spontaneously engage in compensatory strategies (CS) to support everyday task completion, but factors that influence success of chosen CS remain unclear. This study examines whether real-world prospective memory (PM) task completion is better predicted by CS count or a CS quality rating. Method: Seventy mid-life and older adult participants were presented four novel, real-world PM tasks via remote assessment and encouraged to use their typical CS. The examiner captured detailed information about planned CS at task presentation (T1) and utilized CS at follow-up testing (T2). From this information, count (CS Count; quantity of CS) and quality (CS Quality; rating of CS thoroughness and utility) scores were coded separately for the planned and utilized CS. PM task performance accuracy was also coded (PM Accuracy). Results: Hierarchical regressions revealed planned CS Count and Quality did not predict PM Accuracy. In contrast, the utilized CS Quality predicted a significant amount of PM Accuracy variance over and above CS Count, global cognition, and age (R2 = .47, ΔR2 = .24, ΔF = 29.36, p < .001, f2 = .45). Furthermore, utilized CS Quality accounted for a similar amount of variance in PM Accuracy when utilized CS Count was removed from the model. Conclusions: This study's CS coding system can capture and quantify the quality of strategies, which uniquely predicts real-world PM performance. This coding system may provide researchers with a nuanced CS measure and lead to improved CS interventions designed to support everyday PM performance, such as targeted CS trainings.
Keywords: Compensation; aging; ecological validity; everyday functioning; prospective memory.