Countries that desire to report literacy-related data face difficult choices, given that assessment resources are always limited. Most countries prefer to publish inaccurate data than to not publish data at all, and hence choose to rely on indirect ("proxy") measures (e.g., "number of years of schooling") despite their many limitations. Such measures are inexpensive and easy to collect.

Given that the data reported to international agencies are problematic, and given that UNESCO's statistical experts are forced to publish estimates and projections in many cases, users of published literacy statistics should be aware of their many limitations.

It is likely that for most countries, published literacy statistics overestimate true literacy levels. In the United States, for example, it was assumed for many years that over 90% of the citizens are literate. Yet, in 1993 a large scale survey examined literacy skills using functional tasks (i.e., large samples of respondents had to answer many questions relating to a wide range of realistic texts). It was discovered that up to 50% of the population had trouble dealing with many basic literacy tasks.

Literacy statistics published by international agencies may not reflect true literacy levels. Yet, they do show general trends and do enable some comparisons of one country to another, or of one sector to another within a country. Thus, published statistics do provide a rough picture that is of some value, if used with caution.

Users who are interested in statistics pertaining to literacy in their own country should become familiar with the way literacy is assessed and reported in their country. We urge you to seek answers to questions such as:

• How does my national statistical bureau or division define litearcy?
• In my country, what data are collected during a census about literacy?
• What limitations do I see with the way literacy is defined or measured?
• How does the statistical bureau compute the statistics that are actually reported to the U.N. ? (e.g., are the statistics based on people's answers to one question? are they an average of responses to several questions?)
• Do published statistics make sense in light of other known data? (e.g., given published data from local direct assessments of literacy, from performance of children on school-related tests of reading, etc.)

This Introduction was designed to help you become an informed user of reported literacy statistics. Yet, given space limitations, only basic concepts and ideas were included, only selected points were covered, and texts were simplified to provoke you to think more about key points. Please seek further information about key issues discussed here to make sure you get the fuller picture. A good place to start is the 1997 book: "Adult Basic Skills: Innovations in Measurement and Policy Analysis" (edited by A. Tuijnman, I. Kirsch, and D. Wagner).

We hope that the discussions in this section and in the suggested resources, together with any information you can obtain about assessment of literacy in your own country, will enable you to better understand the literacy situation in your country and in other countries.

back to top

Back to Introduction or move on to Basic Statistics