​Results from a set of questionnaires designed to compare methods of obtaining membership functions were analyzed to identify general trends in fuzziness. It was found that fuzziness in comprehension is ahvays reflected in the answer; concepts such as difficulty, imprecision, and shape of membership functions are clearly interrelated; and the transition between crisp extremes through fuzziness has characteristic manifestations. Emphasis is placed on the practical implications of these ideas, applications include developing standard and type H fuzzy sets, improving the completeness and consistency of available data, and supporting the development of knowledge-based systems using fuzzy sets.