1. General Background The "Lisbon strategy" (European Council 2000) aims to create by 2010 a more dynamic and competitive knowledge- based economy in Europe. Implementing lifelong learning (LLL) has a key role in achieving the goal, with enterprise-provided training being one important element. The Continuing Vocational Training Surveys (CVTS) referring to 1993 and 1999 are unique sources for information about enterprise-provided training in Europe. To our best knowledge, so far these rich data have not been used for in-depth analysis of the effects of the national context on enterprise-provided training. 2. Objective It is commonly agreed that country context matters for lifelong learning. It is not clear so far how country context is linked to LLL, what features of the national context interact with LLL, and what mechanisms can be regarded as driving factors for LLL. Our paper seeks to identify central elements and to explore their relationships, while it is focused exclusively on employer- provided training. We use data of CVTS in order to provide a systematic overview of patterns of employer-provided training across 25 European countries. Going beyond descriptive results and based on theoretical considerations, we will analyse the relationship between macro-economic indicators (including e.g. size structure of enterprises), some characteristics of the education system (as explanatory variables) and enterprise-provided training. The results will indicate starting points and future directions for comparative research on CVT systems. 3. Theoretical approach In a paper produced in the context of an (still ongoing) OECD activity Behringer and Coles (2003) identified a range of components of qualifications systems and suggested how they might interact with LLL. This work was located in the theoretical framework of human capital theory, taking decisions of employers and individuals about provision of and participation in LLL as one of the starting points. We apply parts of this concept, while at the same time broadening this approach, e.g. by explicitly incorporating companies' need to provide training (influenced by the pace of innovations, by the condition of the labour market or by the demographic development) and features of the educational system (e.g. the role of initial vocational training). 4. Proceeding and methods We use CVTS 1999 as data source for continuing vocational training (CVT). Regarding influencing factors, the paper restricts itself to data available at EUROSTAT, thus trying to ensure comparability across countries.First, we will display and compare CVTS results across 25 European countries, using spider charts for a graphical presentation of the quantitative key indicators. For a condensed and spanning description of the main national features of continuing vocational training we are using the following key indicators of CVTS: Incidence of enterprises providing training Access to training Intensity of training Costs of training Following Mosley and Meyer (1999) a surface measure of overall performance is used as a composite indicator, besides a typology grouping countries on the basis of CVTS results. Second, we analyse the relationships of indicators at national level and performance of vocational training (both single key indicators and composite indicator). The specific patterns of CVT will be related to features of the education systems and the social and economic context in order to explain differences of CVT by existing differences in the national context. We use different techniques in the search for systematic patterns, preparing the ground for further analysis incorporating in fuller detail characteristics of national VET systems and of the socio- cultural context they are embedded in.