Not surprisingly, the IALS reports attracted a great deal of interest from national policy makers and the popular press. It is clear that the study has answered many questions of pressing interest and concern. Yet, as with any well-conceived study, IALS raised many new questions. Key among such questions are those that speculate about the relationship of literacy skill to other skills thought to be important to workforce productivity and labour market success. Many studies, including the Secretary'sCommission About Necessary Skills (SCANS) in the United States, have posited theexistence of a range of skill domains thought to be of economic importance. Yet little,if any, empirical evidence exists to test these notional skill frameworks.

The success of the IALS approach led several national governments to wonder ifthe methods could be adapted to measure a broader array of skills on an internationallevel.

A first meeting to consider the possibility of mounting such a study was hostedby the Swedish Educational Authority Skolverket. Documentation prepared by StatisticsCanada for consideration1 reviewed the prevailing notions of basic skills and offered ahybrid typology with eight distinct domains which might be included in an IALS-type study (Jones, 1996). Statistics Canada suggested that the proposed study would,in each of the eight skill domains, administer a computer-based test to a nested sampleof workers within firms so that explicit statistical linkages would be available to isolatethe impact of observed skill on economic productivity and indicators of firm successsuch as employment growth and profitability.

Interest in this idea was sufficient amongst national governments to organise asecond meeting, hosted by the University of Amsterdam, to discuss the merits of availableconceptual frameworks and to review the validity, reliability and operational feasibilityof related assessment technology. The meeting concluded that coherent conceptualframeworks and satisfactory measurement technology did indeed exist for several, butnot all, of the proposed skill domains. It became clear, however, that the costs andoperational implications of fielding a computer-based test to a nested sample of workerswithin firms were well beyond the financial and technical capability of many of theprospective participants.

This realisation led to a meeting of the International Study Team to consider options.Hosted by the US National Center for Education Statistics in Washington DC, themeeting concluded on pragmatic grounds that:

1. The proposed assessment should use paper and pencil rather thancomputer-based tests.
2. The development of frameworks and associated assessment methodsshould be limited to six of the initial eight skill domains:prose literacy,document literacy,numeracy,teamwork,problem-solving,practical cognition, and,working with information technology.While of interest, other skill domains lacked either satisfactory theoryor measures deemed viable within the context of a paper and pencilhousehold-based assessment.
3. The test should be administered to representative samples of adultsdrawn from households rather than from workers within firms.