We use the label of community-based projects to describe project types where CSOs are in a position where they socially drive the project and have transformative effects on methods and / or outcomes. CSOs are involved in all kinds of activities from dissemination, data collection, provision of expertise to agenda setting and defining the research methods. Their motivations focus mainly on the solution of practical problems relevant to their central organisational goals, however may also include increasing their networks and external connections, raising the legitimacy of key issues, and the improvement of knowledge. Often, these motivations are shared equally by the researchers. However, the strong value placed by the CSOs on the project puts the researchers at risk of not being able to sufficiently feed into decision making, possibly leading to problems in producing scientifically valuable research results. A protected area of actions for researchers is thus important in ensuring that the researchers' involvement and priorities are sufficiently emphasised within the project.
Within a particular community-based research project there were several CSOs and local communities involved. The project aimed to determine the dangers and consequences of an industrially-caused environmental disaster for the members of affected local communities. The project funding required the participation of CSOs and local communities, and the process of setting up the proposal was initiated by the communities and CSOs. They had a vital interest in the research results and it took the consortium (consisting of 5 research partners, 2 communities, and 5 CSOs) one and a half years to finalise the proposal after the idea was initiated. In the course of the project the communities participated in all of the project tasks, were responsible for parts of the data collection, and were also involved in data analysis. Within our interviews some of the researchers expressed concern regarding the scientific quality of the data collected by the communities; they also feared not being able to publish such results. Moreover, the communities banned some researchers from cooperating with industrially-sponsored colleagues. Clarification of the researchers' needs and expectations before the start of the project (especially their need for a 'protected' space within which to conduct their research objectively) would have avoided many of the problems that arose.