The purpose of this study was to determine the construct and divergent validity of the Physical Activity Scale for Individuals with Physical Disabilities (PASIPD) in people with spinal cord injury (SCI). The construct validity was examined by relationships between PASIPD and measures of fitness (peak oxygen uptake, peak power output, muscular strength) and activities (wheelchair skills, Utrecht Activity List, mobility range and social behaviour subscales of the SIP68) in 139 persons with SCI 1 year after discharge from inpatient rehabilitation. Divergent validity was determined by comparing PASIPD scores of people with different personal (age, gender, body mass index) and lesion characteristics (paraplegia/tetraplegia, completeness, time since injury). PASIPD scores showed low correlations with fitness parameters (0.25-0.36, p<0.05) and low to moderate correlations with activities (0.36-0.51, p<0.01). Persons with a tetraplegia or longer time since injury had significantly lower PASIPD scores compared to those with a paraplegia (p<0.02; effect size: 0.17) or those with a short time since injury (p<0.03; effect size: 0.30). It can be concluded that the PASIPD showed weak to moderate relationships with fitness and activity parameters. This construct validity is comparable to self-report questionnaires from the general population. The divergent validity of the PASIPD was low. Therefore, the PASIPD should be used with caution in persons with SCI.