Research laboratories are increasingly coming under pressure to evaluate the economic impact of their research. After the multiplication of technology discrimination devices bound to foster the innovative efforts in firms, public policy makers are seeking to evaluate the impact of these means on the economy.
Existing evaluation indicators are poorly equipped to grasp the very characteristics of innovative activities. A set of impact indicators has been developed on the basis of an empirical analysis of twenty-two collaborations. The links with some determinants such as the profile of the firm, the characteristics of the project and the type of innovation have been investigated. Since intrinsic structure of the data is not at all linear, our data analysis extensively uses Kohonen maps which are encountered to realize a rapid and robust classification and to visualize the classes, their differences and homogeneity. Policy maker and R&D (Research and Development) program administrators should be especially sensitive to those impacts that depend on manageable determinants.