Using Experimental Data to Analyse Decision-making Processes in Meetings. A Political Science Perspective

This paper belongs to Thematic Session 1 of the Gothenburg Meeting Science Symposium

Frank Nullmeier, University of Bremen (SOCIUM Research Center on Inequality and Social Policy)
Tanja Pritzlaff-Scheele, University of Bremen (SOCIUM)

Abstract 
In political science, collective decision-making is identified as the core objective of politics. Moreover, the face-to-face meeting is identified as the key element when it comes to processes of political decision-making. Therefore, data collection in studies on decision-making processes often focuses on data from actual political meetings, especially on audio-visual material. However, apart from the difficulties of getting access to these meetings and a permit to videotape them, there are other downsides to the use of video data from real political meetings: Often times, the quality of the data suffers from the fact that too many things are going on in a room at once. Moreover, a study based on audio-visual data from actual meetings usually stays at the level of small-N studies. Therefore, we suggest to supplement real-world meeting data by data from group experiments. This allows us to test and to validate findings on decisive factors, causal relations and typical interaction sequences in meetings.
In our paper, we present an experimental approach to the study of meetings. Combining the analysis of chat-based experiments and face-to-face group experiments, we recommend the integration of data from group experiments as an important step for the further development of meeting research. In an experimental situation, the stages of the decision-making process within a group and the sequences of contributions to this process can be reconstructed in detail. We analytically distinguish four key steps in a decision-making process: Participants in a meeting may discuss the implications of a specific task or the problem to solve. They may proceed by presenting proposals on how to handle the problem and they may signal acceptance or non-acceptance to these proposals. Especially in meetings that do not use voting procedures, the last step of a decision-making process can be identified as the confirmation of a particular proposal as the decision of the group as a whole.
Using these analytical elements, the detailed analysis of experimental data helps to find out under what conditions certain sequences of decision-making emerge. Examples are proposal chains (proposals that produce new proposals), re-narrations (repetitions of a proposal in a slightly different mode by different persons), regressions to the stage of problem discussion, and takeovers of the position of the moderator and initiator of the confirmation process.
The major shortcoming of chat-based experimental data is the exclusion of non-verbal communication and bodily co-presence. Findings from computer-mediated experiments can only be compared to actual meeting data at a very abstract level. Therefore, our experimental data set also includes audio-visual data from face-to-face group decision experiments. The data produced in these experiments, with groups communicating and deciding in physical co-presence, opens up a wide range of opportunities for a detailed validation of findings from actual meetings. Furthermore, research that uses chat-based as well as face-to-face experimental data is able to identify specific advantages of face-to-face meetings when compared to technically mediated forms of communication. We therefore consider both types of experimental data as important resources of meeting science.