A problem with empirical studies of party policy shifts: Alternative measures of party shifts are uncorrelated

Many recent cross-national studies analyse the causes and electoral consequences of party policy shifts, using party position measures derived from election manifestos, expert surveys or voter surveys. However few studies validate their findings by analysing multiple measures of party policy shifts. In this article, data on European parties’ position shifts on both European integration and left-right ideology is analysed, showing that this is problematic because, while alternative measures of party policy positions correlate strongly in cross-sectional analyses, alternative measures of parties’ policy shifts are essentially uncorrelated in longitudinal analyses. Suggestions are offered on how to address this problem.