Spasticity refers to abnormal increases in muscle tone and reflex activity that occur when central nervous system pathways are damaged. Common in conditions such as stroke, cerebral palsy, and spinal cord injury, this heightened stiffness and reflex excitability interferes with daily movement and rehabilitation. Accurate, repeatable measurement is vital for effective therapy planning. In this work we describe a quantitative approach that records both resistive forces and joint angles using force-sensitive resistors (FSRs) and goniometers during seated trunk tasks. Data from 50 adults (20–60 years) encompassing able-bodied and wheelchair users were analyzed. Statistical testing showed clear force differences among spasticity levels in the lumbar region (F = 28.24, p < 0.001) and thoracic region (F = 8.02, p < 0.01), with significant angular changes only in the thoracic segment. A Support Vector Machine (SVM) model trained on the combined sensor features classified spasticity with 98% accuracy. These findings suggest that sensor-based, machine-learning evaluation can provide a reproducible, objective framework for assessing trunk muscle spasticity, potentially reducing dependence on subjective clinical scales.