Data Envelopment Analysis (DEA) is a nonparametric technique for measuring and evaluating the relative efficiency of a set of decision making units using observational input and output data. A fundamental DEA model is formulated as linear programming and the input and output data are assumed to be quantitative positive values. However, in the practical application, there are some uncertainty in the input and output data such as an observational disturbance and subjective data. In this paper, we consider ambiguous data expressed as a fuzzy categorical variable and propose a DEA model for a noncontrollable fuzzy categorical input variable. The proposed model gives a reasonable efficiency score, which compensates for the lack of information caused from the discrete categorization, and has robustness against the change of the boundaries and the fuzziness of categorization.