Q Methodology, a useful tool to foster multi-actor innovation networks performance. Louah, L. 1 and M. Visser 2 1 Université Libre de Bruxelles (ULB), Line@Louah.com 2 Université Libre de Bruxelles (ULB), Marjolein.Visser@ulb.ac.be Abstract: We address in this paper opportunities of Q Methodology for empirical agricultural innovation studies. In the systems perspective on innovation, multi-actor innovation networks are seen as a key strategy to successful innovation. Given the several types of actors involved, the scientific and policy literature points at the need for ‘innovation brokers’ to build capacity for collective innovation and prevent innovation network failures. This paper aims at introducing Q Methodology as a fitting and promising tool to assist these systemic facilitators to probe more deeply into the mechanisms of social learning and collective cognition. Q Methodology is a mixed method that provides quantitative structure to individuals' opinions via factor analysis, based on a clear methodological structure and process. It has gained popularity in a range of ‘messy’ studies to analyze and typify the diversity of worldviews on complex and socially contested issues. Increasingly considered as a well-established method to address rural research questions, its use in agricultural innovation studies is still missing. After providing a deal of practical information about the conduct of Q methodological research, we thus offer to reflect on the usefulness of Q Methodology in fostering multi-actor innovation network performance. Keywords: Q Methodology, mixed method, stakeholder analysis, agricultural innovation system, innovation broker, multi-actor innovation network Introduction Q Methodology has a rich, if little known, history. In 1935, the psychologist and physicist William Stephenson – a doctoral student of Charles Spearman – published a letter in Nature (Stephenson, 1935); the letter announced that he had reconceptualized correlation analysis in such a way that in place of correlating tests in relation to random variables expressing traits, he had developed a method to correlate whole aspects of persons. What Stephenson introduced as an objective study of human subjectivity would grow into the scientific method Q Methodology (hereafter referred to as Q). Considerably developed and codified by the political scientist Steven Brown (1980), Q has been used in a wide range of studies applications seeking to uncover and analyze similarities and differences in the subjective viewpoints of individuals (McKeown, 1990). In the eight decades since it was first proposed by Stephenson, Q has spawned both an increasing community of active practitioners, and recurrent severe critiques (eg Burt & Stephenson, 1939; Cattell, 1951; Kampen & Tamás, 2014), which the Q community in turn considers as repeated substantial misunderstandings of its mathematical and practical aspects (Brown et al., 2015). However, the last 15 years have witnessed a further increase in published Q studies – and a decline in published criticisms –: 92 publications per year in the years 2001-2013 compared to 35 in the years up to 1991-2000 (Brown et al., 2015). According to Donner (2001), Q is particularly well-suited for topics where it is necessary to recognize social complexity and, consequently, has slowly gained popularity in a range of ‘messy’ environmental issues (eg Addams and Proops, 2000; Cuppen et al., 2010; Curry et al., 2013; Hermans et al., 2012; Visser et al., 2007, 2011).