Should We Do Field Experiments with Parliamentary Elites? A Critical Review of the Debate and a Considered Answer

Experiments with parliamentary elites are a most recent addition to the tool kit of the legislative studies. In this paper we review the current debate and discuss promises and pitfalls. We conclude with a proposal on experimental designs that best are suited to avoid pitfalls and maximize promises. Field experiments with parliamentary elites commonly use e-mail interactions between fictitious voters and legislators to better understand to whom legislators are responsive, and thus what kind of biases they follow for what reasons. This approach provides several advantages compared to traditional modes of research on political representation such as survey or standard observational research (Grose 2014). It allows factual observations in light of counterfactuals, and thus increases the validity of our causal inferences. Furthermore, this approach permits focusing on actual behavior instead of survey-based self-reported behaviour that among elites, in particular, may suffer from strategic misrepresentation. Independent of these promises, field experiments with parliamentary elites raise serious ethical questions and also pose risks with regard to the field that we wish to work with in our profession. Specifically, with field experiments involving fictitious identities, we risk disturbing democratic representatives, who have important task to attend to, distorting elite-voter interactions if elites become skeptical whether or not voters approaching them are “real” or fictitious stimuli, and shattering trust relations between researchers and elites, which will harm future cooperation. In this paper, we discuss these ethical dilemmas, outline considerations from the literature and sketch a design with we aim to maximize authenticity and minimize deception. Hereby we aim at developing a field experiment, which emulates and strengthens interaction between voters and legislators rather than disturbs or distorts this relationship.