A critical need remains for a practical and low-cost methodology to bridge the gap between methodologies for assessment of literacy at the national survey level. At the program level, it must be able to be used effectively by those in developing countries with limited funds but major literacy problems. Etic (external, quantifiable, comparison-oriented) measures are clearly important in understanding how people acquire literacy, how educators and policymakers view literacy, and how economic and societal systems interact in an increasingly interdependent world. Collection of international statistics on literacy poses problems that specialists have debated. With the greater need for direct measurement and increased technical capacity for such measurement in developing countries, clarification of issues is occurring in language policy and multilingualism; classification; international comparability of data; and measuring the consequences of literacy. The Education for All approach that emphasizes measuring learning achievement and widening the view to beyond the 3Rs raises definition issues. National and local program needs are artificially separated by goals and tools typically used in these two domains. Confusion exists with regard to the terminology used to describe methods for literacy assessment. Challenges to validity of literacy assessment tools are cross-cultural comparison, targets and "minimum" levels of achievement, and technical/statistical challenges. (Contains 51 references.) (YLB)