Level 1 indicates very low literacy skills, where the individual may, for example, have difficulty identifying the correct amount of medicine to give to a child from the information found on the package.

Level 2 respondents can deal only with material that is simple, clearly laid out and in which the tasks involved are not too complex. This is a significant category, because it identifies people who may have adapted their lower literacy skills to everyday life, but would have difficulty learning new job skills requiring a higher level of literacy.

Level 3 is considered as the minimum desirable threshold in many countries but some occupations require higher skills.

Levels 4 and 5 show increasingly higher literacy skills requiring the ability to integrate several sources of information or solve more complex problems. It appears to be a necessary requirement for some jobs.

Between 2,000 and 3,000 adults (5,660 in Canada) in each of the eight countries --Canada, Ireland, Germany, the Netherlands, Poland, Sweden, Switzerland, and the United States-took part. All the respondents were tested in their national language and in their own homes. The prose and document literacy scales each comprised of 34 tasks and the quantitative literacy scale included 33 tasks. All the tasks were of varying difficulty. The attached sheets show some sample tasks, the levels and scores.

If economies require increasing numbers of highly skilled workers to expand, then growth will be affected by existing practices of employers, 'individuals and governments: IALS has shown that instead of enlarging the pool of highly skilled workers, the tendency is to increase the skills of the already skilled. The reserve employment pool, made up of the unemployed and those working in declining industrial sectors, is low-skilled. Policies directed towards providing more educational opportunities and increasing skills in that pool must he a necessary part of any industrial growth strategy.

The distribution of literacy is also a good predictor of the magnitude of differences between social groups, making literacy an essential element for promoting social cohesion. Therefore, any view of literacy which is focussed on economic objectives alone is untenable.

At the broadest national level, the IALS findings are consistent with those of its predecessor, LSUDA, conducted in 1989. There is considerable variation among Canadians in their literacy skill and the pattern of these variations shifts according to the different yardsticks used (region, language, age, educational attainment, occupation, and so on). Indeed, the differences between the two studies are not large. On the one hand, there appears to be little to support earlier predictions and present concern of a rapid erosion of either educational quality or of the adult skills base. On the other hand, some improvement was expected. Since LSUDA was conducted in 1989, those leaving the labour force have been replaced by an incoming cohort of young people who are collectively much better educated and more literate. The fact that no appreciable overall improvement was detected by IALS suggests that skills are being affected by other processes-processes that are of policy concern and need to be better understood.