Functional ability (FA) is here defined as having no limitations in five questions related to the ICF dimensions "Activity" and "Participation" (ICF=International Classification of Functioning). The aim was to explore how individual and contextual factors were associated with FA in each gender, and to explore to what extent physical activity (PA) and this measure of FA used in large multipurpose health surveys seem to measure different constructs, and how close the association between PA and FA was, and whether it differed within self-rated health (SRH), lifestyle and contextual factors. Methods: All men and women aged 75-77 (N=11684) in five counties in Norway were invited to health surveys in 2000-2003. 49% participated (2626 men, 3146 women) with answers on the five questions on FA. Data were analysed using logistic regression, factor analysis, and Structural Equation Modelling (SEM). Results: Good SRH, no serious pain or stiffness in muscles and joints, interest from others, no psychological distress and no weekly use of analgesics, hypnotics, tranquilisers or antidepressants showed the highest positive associations with FA for both men and women. The SEMestimated correlation of FA and PA items was lowest in healthy elderly (r=0.19 (0.11-0.24)) compared to those with poor health (r=0.40 (0.35-0.45)). FA and PA measured different constructs. Conclusion: According to low correlation between FA and PA in healthy men and women, we suggest that FA and PA did not measure the same construct, and impairments in functional ability did not necessarily mean low level of physical activity if SRH was good.