**3. Methods**

### *3.1. Simulation and Sample*

One hundred and forty-five individuals (managers and consultants of which 54 women) with an average age of 43 years old participated in the study. Data was collected in five behavioural simulations that involved different participants. All participants were enrolled in postgraduate education at a business school or in a professional development program. We used a multi-party behavioural simulation (Schruijer and Vansina 2008; Vansina and Taillieu 1997; Vansina et al. 1998) in which seven or eight (depending on the number of participants) stakeholder parties engage in within-group as well as between-group interactions to deal with a complex regional development situation involving economic, social and environmental factors in the St Petersburg area, including the island Kotlin.

The participating parties were: a local authority (with an interest in the socio-economic situation on the island), a shipyard (the largest employer on the island facing a severe problem related to unemployment and decrease in business opportunities), an island-based yacht club (located on a scenic piece of land on the island and interested in developing their yacht club), a bank (interested in long-term investments), a group of young and rich entrepreneurs (with an emotional tie to the island), a Finnish yacht club (interested in new sailing routes for their members), a yacht club near St Petersburg (also wanting to expand their activities) and a technical school (associated with the Shipyard) (for more details see Vansina et al. 1998). Each party received the briefing describing their interests and had the freedom to develop their own strategy and approach.

At the onset of the simulation participants were allocated to a party based on their expressed preferences. Then each party received a booklet containing information concerning economic (e.g., risk of bankruptcy for the most important employer on the island), social (e.g., unemployment and social unrest) and environmental (e.g., water pollution) challenges in the region, as well as specific information concerning their own interests as a party.

The simulation proceeded in real time during a full day and few hours in the next morning and was followed by a joint debriefing. We allocated the first time slot of the simulation to the study of briefs within each party and develop their party strategy for interacting with the other stakeholder parties. No specific roles were assigned to the individual participants. After the within-party preparation, one-hour intergroup visiting and plenary meetings were alternated during the day. Intergroup visiting slots consisted of the possibilities to visit other parties (with a maximum of 3 parties being present at any particular time and place). During plenary meetings, all parties could be present and send their representative to the table, while the constituencies were allowed to sit behind their representatives and send notes.

The participating parties (36 groups having 3 to 4 members each over five simulations) were asked to fill out surveys concerning within-party (task and relationship conflict) as well as between-party dynamics (the perceived conflictuality and collaborativeness of all other stakeholder parties) three times during the simulation (before the interactions started, during interactions and after the simulation ended). In order to capture the entwinement between expectations and experienced social interactions, only the first and the second evaluations were used for further analyses as the intergroup interactions only started after filling out the first questionnaire.
