One of the most important problems in statistical inference is testing of statistical hypothesis. Usually, the underlying data and the hypotheses are assumed to be precise. But, in many situations it is much more realistic in general to consider fuzzy concepts. This paper is devoted to the problem of testing hypotheses when the both hypotheses and data are fuzzy. We first extension the notion of fuzzy p-value which is appropriates for this case and then present an approach for this testing by comparing the obtained fuzzy p-value and fuzzy significance level, based on a comparison of two fuzzy sets.
