Purpose/Hypothesis : Models of anatomical structure have been shown to enhance learning. In rehabilitation, accurate palpation skills are vital for many clinical and research scenarios. In kinematics studies, specific landmarks on body segments are used to define local coordinate systems which are used to describe motion. Accurate landmark identification is critical for the accuracy of quantifying motion. The purpose of this study was to determine if using a skeleton model enhanced learning in first-year DPT students when setting up coordinate systems for the shoulder for the first time. We hypothesized students who were educated with a skeleton model would demonstrate better learning in the cognitive, affective, and psychomotor domains than students who were not.Number of Subjects : 27 first-year DPT studentsMaterials/Methods : First-year DPT students participated in a kinematics research learning experience. Education included (1) an introduction to kinematics and electromagnetic motion capture, (2) a review of anatomical landmark palpation, and (3) training in and execution of establishing local coordinate systems for the thorax, scapula, and humerus segments. Student pairs were randomly assigned to participate in the aforesaid education alone (n=13) or education with training on a skeleton model (n=14). Student learning was measured in three domains. Recall of anatomical coordinate system landmarks on each segment was evaluated by multiple choice items (cognitive). Confidence with kinematics set-up procedures was assessed by Likert-scale items (affective). Anatomical landmark location as digitized on the three segments was exported from data files (psychomotor). Cognitive and affective domain scores were compared between groups, Mann-Whitney U test, ?=0.05. Cartesian coordinates of anatomical coordinate system landmarks (3 per segment) were used to create triangles. The lengths of triangle sides were used to determine differences in variability between groups, LeveneÕs test, and inter-rater reliability of each group, ICC(2,1), for each segment.Results : No significant difference was seen between groups in the cognitive (z= -1.5,p=0.13) or affective (z=1.3, p=0.19) domains. Variability (SD) in triangle side length ranged from 1.8 to 28.5 mm, (both from the thorax). There was no difference in variability between groups. Inter-rater reliability coefficients ranged from 0.89 (scapula) to 0.99 (humerus) with no differences between groups.Conclusions : Additional training using a skeleton model did not affect cognitive, affective, or psychomotor domains of learning of first-year DPT students in this study. The magnitude of variability within groups provides insight to how coordinate system set-up, and ultimately motion, may vary when different, novice individuals perform the set-up.Clinical Relevance : The additional time and cost associated with the use of a skeleton does not result in additional learning or mastery in this approach to kinematics set-up education. However, the variability observed could affect resultant descriptions of motion.