The Comparative Study of Electoral Systems (CSES) is a collaborative program of national election studies around the world devoting itself to comparative electoral research and investigating the behavioral impact of political institutions since 1994. CSES provides survey data on elections for a large set of countries. However, what was missing was a dataset combining all these election surveys from different CSES waves in one comparative dataset which enables straight-forward access and research possibilities for the wider scientific public. A team of WZB researchers, Bernhard Weßels, Heiko Giebler, Josephine Lichteblau, Antonia May, Reinhold Melcher (now Fern-Universität Hagen) and Aiko Wagner, has now compiled such a dataset and it was made accessible via the CSES website on June 1st.
The WZB has been involved in the CSES program from the beginning being the host of its founding conference in 1994 and, since then, repeatedly organizing meetings of the CSES Planning Committee and Plenary Sessions. Members of CSES are national election study teams that agreed to include a common module of survey questions in their representative post-election studies, each for the duration of about five years. The respective module is developed by the CSES Planning Committee, that Hans-Dieter Klingemann and then Bernhard Weßels from the WZB were part of from 1994 to 2014. Providing one of the most important data sources for comparative electoral research, CSES can be regarded as a successful and constantly growing project of the comparative social sciences. Since its foundation, the number of participating countries increased continuously from 25 to over 50 (see figure below).
The WZB team of the German Longitudinal Election Study (GLES) under principle investigator Bernhard Weßels is responsible for the German contributions to CSES. The team only recently published a Harmonized Trend File of CSES Modules 1 to 3 covering 128 national elections of 51 countries from 1994 to 2011. The CSES 1-3 Harmonized Trend File is not simply a merged version of the three single datasets of the three modules containing each variable of each module in their original form. For the CSES 1-3 Harmonized Trend File all micro- and macro-level variables that were part of at least two of the three waves have been cumulated and partly integrated across the modules and participating countries. This means that for those variables there is only one harmonized variable containing all cases of all (or up to) three waves in the harmonized dataset. Coding of these variables has been harmonized as well to ensure that values represent the same substantive content over all waves and countries for which the variable is present. Therefore, the CSES 1-3 Harmonized Trend File provides a sound data source for analyses covering more than one wave of CSES and over time. Moreover, the universal coding of variables between CSES modules 1 to 3 ensures validity and reliability of empirical research using CSES data. With this data publication, the WZB and especially the department ‘Democracy and Democratization’ lead by Wolfgang Merkel follows its self-conception and tradition of common good production in the realm of empirical research with the goal of enabling and fostering high quality research not just in-house but all over the world.
The CSES 1-3 Harmonized Trend File and an extensive documentation are available for download from the CSES website under the “Data Center” link.