Pergamon ElectoralStudies, Vol. 15. No. 2, pp. 149-166, 1996 Copyright © 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0261-3794/96 $15.00+0.00 S0261--3794(96)00009--1 What Voters Teach Us About Europe-Wide Elections: What Europe-Wide Elections Teach Us About Voters* CEES VAN DER EIJK Department of Political Science, University of Amsterdam, 1012 -DL Amsterdam, The Netherlands MARK FRANKLIN Department of Political Science, University of Houston, Houston, TX 77204-3474, USA MICHAEL MARSH Department of Political Science, Trinity College, Dublin, Republic of Ireland With four sets of European parliamentary elections now behind us, it is appro- priate to review the prevailing interpretation of such elections as second-order national elections, a view first put forward by Reif and Schmitt in 1980. While the second-order model has yielded important insights into the way European elections can be understood as manifesting national political processes, more recent research has fruitfully turned the model on its head, and focused on what European elections can tell us about national elections and the nature of the voting act. Indeed, the use of individual-level survey data to study elections to the European Parliament has for the first time truly shown us the importance of institutional and political context in conditioning turnout and party choice. Findings of recent research suggest that the second-order features of European elections should be thought of as contextual variables that can affect other elections as well. Copyright © 1996 Elsevier Science Ltd. The fact that elections to the European Parliament (EP) are different from elections to national parliaments in European Union (EU) member countries has been evident from the time of the very first of these Europe-wide elections, held in Jtme 1979. Turnout in such elections is low, major parties generally do badly (compared to their performance in adjacent national elections) and small parties often do better than in national elections. Writing in the immediate aftermath of the 1979 European elections, Reif and Schmitt (1980) proposed a way of thinking about these elections that would account for their characteristics. This was to regard them not primarily *The authors would like to thank Hermann Schmitt and, especially, Christopher Wlezien for helpful comments.