Assessing the validity of the Manifesto Common Space Scores

RILE estimates based on party manifesto data suggest that political parties leapfrog on the left-right scale over time. This implausible finding has raised questions about the efficacy not only of RILE for estimating left-right positions but of coded party manifestos for political science research in general. The recently developed Manifesto Common Space Scores (MCSS), which reduce leapfrogging by accounting for the election-specific character of party manifestos, provide alternative estimates for parties left/right-positions, but little is known about their validity. This study shows that MCSS estimates exhibit greater convergent validity relative to RILE estimates when compared to other measures of parties left/right-positions. It also finds that MCSS has greater construct validity relative to RILE estimates in two prominent cases (Greece and Italy). Overall, the findings underscore the election-specific character of party manifestos and demonstrate that MCSS is a useful alternative measure of parties' left-right positions.