This empirical paper examines the conditions under which individuals are likely to engage in learning activities with other participants during processes of innovation in multi-actor, collaborative networks. Learning is here understood as, “the exchange of knowledge and ideas that helps improve a person’s understanding of a problem situation”, while innovation is interpreted as, “creative decision-making and search processes used to design and realize new ideas and solutions that radically transform the way in which we are imagining and doing things in the public sector” (Ansell and Torfing, 2014:4-10).
In theory, collaborations, and particularly learning activities between participants of collaborations, are expected to boost innovation, as more actors and thus more knowledge, information and experiences are incorporated in decision-making processes (Bekkers et al., 2013:13).
Yet, recent studies have shown that in practice not all individuals have the same propensity to engage with each other in learning practices; thereby undermining the innovative capacity of multi-actor collaborations (Bressers, 2014:104). As such, the following research question has become a relevant topic for the public sector innovation literature to be addressed (Ansell and Torfing, 2014:238-239):
Why are individuals more likely to engage in learning practices with some participants than with others during processes of innovation in collaborative networks?
So far, not many scholars have explicitly posited this research question. Instead, scholars have mainly used the case study method to look at aspects of group-learning in collaborative processes of innovation (cf. Waldorff et al., 2014:85). Within these group-level analyses, ‘learning’ has rather been conceptualized as an emergent property of the collective, instead of an accumulation of dyadic learning activities between individuals in collaborations.
As such, the literature has offered explanations for how learning in collaborations affects group-level outcomes and results (in terms of goal achievement, efficiency, etc.), but has failed in explaining the differences in learning manifestations between individuals in collaborations.
Therefore, to contribute to the existing literature we try to find an answer to the aforementioned research question. In this paper, we particularly examine the learning interactions among individuals in a multi-actor, collaborative network established by the Flemish government to radically alter its Spatial Planning policy.
To measure the dependent variable of learning, we make use of the validated measurement scale of the social psychologists Van den Bossche et al. (2010), which captures ‘individual learning behaviour’ as a distinctive analytical term, and we turn it into several Social Network (interview) questions. Moreover, we also ask in the (structured) interviews with all participants of the collaborative network about various independent variables, like: trust, belief homophily, actor importance, societal- and personal meaningfulness, reciprocity, transitivity, etc.
Eventually, we use the statistical network method of Exponential Random Graph Moddeling, which is a methodology that has not be frequently used in public administration research but more in the research fields of neurosciences and disease studies, to make inferences about the learning dynamics among individuals in the Flemish collaborative innovation network.
