The observation of participation in educational institutions or in learning activities has become an essential instrument of evidence based educational policy in most of the European member states. In connection with the concept of 'lifelong learning' the interest in monitorings on adult learning participation and furthermore, within the proceeding globalisation, the interest in international comparisons and benchmarking processes of educational systems increased within the last few years (Grek 2008; Ioannidou 2007). As a consequence, there is a new demand for information on educational systems. A response to this demand ist the implemenation of new instruments in order to acquire more and high-quality informationen on educational processes and outcomes, especially on an international level. These innovations have impact on the traditional national statistical systems. This effect and its relvance for a coneptual framework of continious monitoring can be exemplarily illustrated by the transition of monitoring structures on continuing education and training (CET) in Germany. From 2011 (prospectively) the German national monitoring system of continuing education, the "Continuing Education Reporting System" (CERS) (Berichtssystem Weiterbildung), is going to be replaced by the European monitoring system "Adult Education Survey" (AES), which is going to be compulsory for all European member states. The CERS was implemented in 1979 and can also be understood as a reply on the failure of the institutionalization of CET within the German educational system, which can be characterised as heterogeneous regarding legal regulations as well as institutions offering CET. Since 1979 the CERS is carried out every three years by questioning a (changing) representative sample of the German population about participation in CET (BMBF 2006: I; Rosenbladt/Bilger 2008: 16). All in all until 2007 the CERS provided the most profound statistical information on adult learning activities on an individual cross sectional basis in Germany. In 2007, Germany took part voluntarily in the first wave of the AES survey. Simultaneously, the national reporting system, CERS, was continued, too, in order to detect differences in definitions and operationalisation of CET (see Rosenbladt/Bilger 2008: 11-12). This comparison revealed that profound conceptual differences exist between the German national reporting system (CERS) and the European monitoring on CET (AES) (see Gnahs 2008: 32). Taking the above mentioned conceptual changes and its expected implications into account, our analysis focuses on the changes in operationalisation of CET within those two monitoring concepts. The aim of our analysis is to identify possibilities and limitations of continuing trend analyses for indicators of CET in Germany in the transition to the new, international monitoring concept of the AES. In detail we are interested in the developement of participation rates in non-formal and informal CET. By a very profound comparison of the two surveys (CERS and AES) the impact of changes in operationalisation of CET on the developement of participation rates in CET is (partly) traceable. Thus, effects of the transition and therefore changes in the concept of operationalisation can be considerd in a long-term trend analysis.

Data basis: 1. Continuing Education Reporting System (CERS) of Germany. Survey waves of 1988, 1991, 1994, 1997, 2000 and 2003. 2. Adult Education Survey (AES) 2007, data for Germany. Trend analysis requires comparability and continuity of the relevant data. Therefore, the data of the respective reference years is subject of a detailed assessment. Creation of the time series thereby takes into account two aspects of fitness, referring to operationalisation in the questionnaires and coding in the data sets: * Comparability of variables of interest within the respective inquiries of the CERS (1988-2003) * Breaks and connectivity in the transition from CERS to AES Thus, trends in CET from 1988 to 2007 can be descriptively analyzed, tracking back the development of participation rates in non-formal and informal CET in combination with socio-demographic characteristics.

Research on the transition from CERS to AES and problems of conceptual connectivity already exists (Gnahs/Kuwan/Seidel 2008; Rosenbladt/Bilger 2008). Our analysis deepens the state of research by covering further variables that were neglected so far. In addition, the comparison is widened by including previous survey waves. In doing so, the dimension of change in the operationalisation of CET and socio-demographic variables in the respective surveys is covered by the long-term perspective of trend analysis. Possibilities and limitations of analysis are identified and assessed at their consequences for the quality of research results. A partial result is a manageable data set combining the respective surveys. In spite of quite considerably changes in the concept of monitoring CET in Germany, a continous trend analysis is possible. However, the devil is in the details, especially when it comes to assessing the impact of changes in operationalisation. These decisions can be supported by plausibility checks based on the data, nevertheless there is always a normative element of assessment by the researcher. These decisions take place in a stress ratio of two ideal objectives - maxium possible connectivity and maximum possible continuity, which are going to be considered and discussed within our analysis.