Data Population: The unit of analysis in our project are EU legal rules. The relevant population has been discussed in the context of the operationalization of the dependent variable. In brief, it encompasses the following: the primary law of the European Union; directives and regulations emanating from the EU legislators; secondary legislation by the European Commission insofar as it constitutes a decision on a request for a derogation by a member state; international agreements governing the relations of the European Union with third countries; domestic legislation of third countries aiming at the adoption of EU rules irrespective of whether it is based on international agreements or voluntary action. Our coverage is comprehensive regarding intra-EU differentiation. Concerning the relations with third countries, we will devote particular attention to the EEA and Switzerland. Switzerland will serve as an in-depth study of the factors and mechanisms accounting for the external validity of EU legal rules. Hence, in geographical terms, our dataset will cover 31 European countries. Data sources and data gathering: For the dependent variable we rely on the Official Journal of the European Communities/European Union which contains all legal acts falling within our population except for Swiss legislation (this is available in digital format). As indicated, further research is required into Swiss-EU relations before we can determine with certainty the relevant data sources. For the independent variables we rely on various datasets and surveys cited earlier including the Comparative Manifestos dataset, the Euromanifestos dataset, the Chapel Hill expert survey series, the Ray-Marks-Steenbergen expert survey, the Benoit-Laver expert survey, the DEU dataset, the Eurobarometer, the European Values Study, the World Values Survey, the Political Constraint Index and the Shapley-Shubik Index. For macro-economic data we will consult the statistical information from various organizations including the United Nations, World Bank, the International Monetary Fund, the Organisation for Economic Co-operation and Development, and the Statistical Office of the European Union. Finally, certain independent variables are coded alongside the dependent variable based on information from the Official Journal. Methods of data analysis: We seek to determine the variation in the extent of differentiated integration at three levels of analysis: system, country, and policy. At the systemic level, we seek to explain the change over time in the extent of Europe-wide differentiation. At the country level, we try to explain the variation across countries. And at the policy level, we seek to explain the variation across policy areas. Our dataset can be used for different kinds of macro-quantitative analyses, depending on the measurement level of the dependent variable, the data structure, and the hypotheses to be tested. In general, we will employ bivariate and multivariate regression analyses at multiple points in time (cross-section design) or over time (panel design). For example, we will employ count models to estimate the impact of country and policy factors on the number of derogations, opt-outs and opt-ins of a given country or policy area in a given year. Count models, such as Poisson or negative binomial regressions, are generalized linear models and are used when the dependent variable represents a count (yi = 0, 1, 2,..., n) of a relatively rare event (Cameron and Trivedi, 1998). Erhebungsverfahren: Akten- und Dokumentenanalyse standardisiert Erhebungseinheiten: EU legal rules, Swiss legal rules Auswahlverfahren: complete inventory count Untersuchungsdesign: quantitative empirical study