This paper presents a modality-independent evaluation methodology to study users' acceptance and satisfaction of biometric systems. It uses a survey questionnaire for data collection, and some data-mining tools for their analysis. We have applied it on two biometric systems developed in our research laboratory. The results from this survey show the necessity of taking users' point of view when designing and evaluating biometric systems. A panel of 100 volunteers was more satisfied from the keystroke system than the face one. Users surprisingly considered that its perceived performance was also better, even if the used face system has a better performance with an EER of 8.76% than the keystroke one with an EER of 17.51%. The robustness of a system against attacks, computation time required during the verification phase and its easiness to use have been identified as important factors influencing their opinions regarding the tested systems.