The aim of the study was to modify the Amotivation Toward Exercise Scale-2 by dropping the outcome amotivation beliefs subscale while retaining the capacity, effort, value, and task characteristics amotivation beliefs. Data from 515 Greek nonexerciser adults aged 19–64 years and 399 Greek nonexerciser older adults aged 65–82 years, men and women, supported the latent four-factor structure of the Revised Amotivation Toward Exercise Scale-2 scores within each sample separately. The hierarchical model was also supported. Measurement invariance of Revised Amotivation Toward Exercise Scale-2 scores was supported across groups of gender, age, body mass index, and educational levels within adults and older adults. Configural measurement invariance supported a replicable factor structure across adults and older adults. Metric and strong invariance was supported across middle-aged and older adults but not between younger and older adults. Full measurement invariance was supported between younger and middle-aged adults.