A correspondence study field experiment (CSFE) was conducted in Denmark, Germany, the Netherlands and the United Kingdom from November to December 2020. The pre-analysis plan has been published and is available via Related Resources. This data set contains the anonymised original data from the audit experiment that was conducted in the United Kingdom from 2nd of November 2020 to 18th of December 2020. Manipulated variables include social class, ethnic background and partisanship, which are all manipulated in the treatment emails. Social class is signalled by the sender's occupation, ethnic background is signalled by their name and partisanship is signalled by the sender saying they support the MP's party. Measured variables include district size, district structure, party, and legislators’ personal characteristics, i.e. their ethnic and class backgrounds. The latter are only used for exploratory analyses. This experiment was pre-registered and full details of the experimental design can be found in the pre-registration documentation. Please see Related Resources.The growing psychological detachment of West European electorates from political parties provides one background of the proposed research. This trend is indicated by a number of specific developments such as decreasing levels of party identification, dwindling numbers of voters who continuously vote for one and the same party, and also decreasing trust in political parties (Dalton and Wattenberg 2000; Dalton 2016; Mair 2013). These developments progressively accumulate to a situation in which political representation is no longer structured through long-term stable attachments between distinct national political parties and distinct national coalitions of voters. The rise of non-partisan electorates raises pressing questions about ways to sustain responsive democratic governance in future decades. As partisanship was a key mechanism linking citizens and the state throughout the 20th century in Western Europe we must question how responsive democratic government be sustained in the 21st century in the absence of it. Theories of individualized (dyadic) political representation provide a second background to the proposed research. Conceptual work on this issue pictures individual MPs as a potential mechanism for connecting citizens and the state. It portrays individualized forms of representation as a key alternative to European style collectivist representation. While the latter ideally is assumed to pair national coalitions of voters with national disciplined parties, the former has been characterized as a dyadic relationship between distinct constituents and individual representatives that might significantly affect voter satisfaction with the democratic process (Esaiasson et al. 2017; Karvonen 2010; Colomer 2011). Empirical work on this issue stresses related evidence on non-European systems such as the US-American case, in which individual legislators indeed structure the representative process to significant degrees, in which direct interactions between individual legislators and constituents are expected and frequent, and in which most politics is said to be local (Powell 2004; Uslaner and Zittel 2006). Advancing from these two backgrounds, the proposed research asks about the extent to which dyadic representation provides a viable supplement to partisan mechanisms of representation in European contexts.
The data was collected via an audit experiment which involved sending policy queries to legislators via emails where fictitious constituents varied regarding their ethnicity, gender, class status, and partisanship. Full details are available in the pre-analysis plan https://osf.io/b7rz9
