The objective of this paper is to present the initial results of a study aimed at showing the feasibility of using kinematic measures to distinguish skill levels in manipulating surgical tools. Through a simulated surgical task (dissection of a mandarin orange), we acquired motor performance data from three sets of subjects representing different stages of surgical training. We computed the average lateral, axial and vertical tooltip velocities for each of the two main subtasks ('Peel Skin' and 'Detach Segment'). For each subject, we defined a 6-element vector to describe the kinematic measures extracted from the two tasks and used Principal Components Analysis (PCA) to extract the two dominant contributors to overall variability to simplify the presentation of the data to the trainer. We found that the first two principal components accounted for approximately 90% of the variance across all subjects and tasks. Moreover, the PCA plot showed good intrasubject repeatability, consistency within subjects with similar levels of training, and good separation between the subject groups. The results of this pilot study will allow us to design a future intraoperative study.