Women’s substantive representation (SRW) benefits from women’s descriptive representation (DRW). This assumption has captured the attention of many scholars who tried to establish whether women ‘make a difference’ in parliament (Celis 2006; Childs 2004). However, empirical studies remain inconclusive on the existence of a direct causal link between DRW and SRW (Dodson 2006; Beckwith & Cowell-Meyers 2007). They claim that the relationship is probabilistic rather than deterministic (Dodson 2006: 8), since political opportunities and institutions mediate the translation of DRW into SRW (Childs & Krook 2008). Most of this previous scholarship is also methodologically biased, in that it tries to establish causality with case-study research. Comparative perspectives on the issue are a rarity, and often use datasets with different indicators (Mateo Diaz 2005). This paper aims to study the relationship between DRW and SRW from a comparative perspective. It tries to answer two questions: Does DRW lead to SRW? And under which conditions does DRW translate into SRW? The first question discusses whether DRW is a sufficient/necessary condition for SRW, and raises alternative explanations for the emergence of SRW. The second question assumes the probabilistic character of the relationship between DRW and SRW and looks at the configurations of factors that shape this relationship. To answer these questions, we use new data on legislators’ attitudes and behaviour from the 2009 PARTIREP survey in 77 national and regional parliaments in Europe. This survey offers uniform measures for SRW, and allows to compare the impact of different institutions and political opportunities: newness of institutions, power of parliaments, left-right division in parliaments, electoral systems and the presence of women’s movements and equal opportunity structures. The SRW in parliaments is considered at an aggregate level. A qualitative comparative analysis is particularly suited to deal with an intermediate N design of 77 parliaments (Rihoux 2006).
