The purpose of this study is to examine the variability in sensory test of tactile results using Semmes-Weinstein monofilament (SWM). At present, several methods for measuring the tactile sensitivity are clinically used in diabetic peripheral neuropathy screening. One of these methods is a touch test that uses a device with nylon SWMs, i.e., SWMs embedded in a plastic handle. A small pushing force is applied at the handle to bow the filaments. Because of its ease and simplicity, the SWM test is conducted at the patient's bedside in a hospital. However, previous studies have reported some problems with this test. Studies have shown variations in the measured data, and it is uncertain whether these variations are caused by mechanical properties of the nylon fiber or by the motion of the operator's hands. We carried out two experiments to examine the effect of (1) the variability caused by the human operator conducting the SWM test on the test results and (2) the number of compressions of the SWM on the test results. In experiment 1, we measured the velocity of the operator's hand motion and the buckling force of the SWMs. The results showed variability in the hand motion of the operator conducting the SWM tests. In experiment 2, we measured the buckling force of the SWMs under a controlled velocity. We compared the buckling force of the SWMs through a number of trials. These results showed that the buckling force gradually decreases as the number of test cycles increase. In conclusion, we find that the accuracy of the SWM tests is a factor of the number of test cycles. Additionally, manual training for standardizing skills of medical staff members needs to be developed. Furthermore, the characteristics of the SWMs deteriorated over time. In future work, we aim to find a solution to minimize the variability in the SWM test results and develop a new testing system that uses tactile sensibility for diabetic peripheral neuropathy screening.