BACKGROUND AND PURPOSE: The aim of this study was to evaluate whether postgraduate physical therapy students studying manipulation could learn to accurately produce specific forces during palpation of an intervertebral joint. SUBJECTS: The 12 subjects (7 female, 5 male), aged 26 to 36 years (X = 29.5, SD = 2.9), had each completed a 4-year degree course in physical therapy and had worked between 3 and 10 years in clinical practice. All subjects were enrolled in a 12-month postgraduate manipulative therapy diploma course. METHODS: Subjects in the experimental group (n = 6) trained to apply specific forces of 1, 5, 10, 15, 20, and 25 kiloponds using bathroom scales. They practiced for 10 minutes per day for 30 days. Their ability to produce these forces on command was measured using a force platform as they applied posteroanterior passive accessory intervertebral joint movements to the lumbar spine of the healthy subjects. This testing was done prior to training (pretest), immediately after training (posttest), and 1 month following cessation of training (retention test). The control group subjects (n = 6) had no training with scales but were also students of the postgraduate manipulative physical therapy course. RESULTS: In comparison with the control group, the experimentally trained group showed reduced error in force production both immediately after training and 1 month later. This improvement was significant for the retention test. For the retention test, the experimental group subjects were also tested on the trained task (ie, their ability to apply specific forces to the scales). They developed higher levels of accuracy than did the control group. CONCLUSION AND DISCUSSION: Experimental training, therefore, was an effective addition to normal training, suggesting that therapists can learn to quantify applied forces, with significant implications for communication and evaluation of joint behavior.