Since the beginning of the 21th century, security-oriented policy agendas have become the modus operandi in Western societies (Beck, 2008), which try to save their belief in the future paradoxically through imagining that future as an ongoing potential danger zone (Reinhard, forthcoming). The schooling sector is no exception to this development, as the global rise of so-called early warning systems clearly indicates. Such systems equally operate through ongoing ‘dangerization’, which is the imagination of learning mainly through categories of menace and adverse probabilities. In 2005, all 50 US governors agreed to a common calculation of high school graduation rates. Five years later, the federal government required the US States to include longitudinal data to that calculation, which ‘follow’ individual students from early childhood into the workforce (Balfanz et al., 2009). Over the past decade, pressure increased to use that data to install early warning systems, mainly aiming to reduce the number of high school dropouts. Promotors of such systems hereby frequently pointed to research which claimed that ‘at-risk students’ could (easily) be identified through indicators such as poor attendance, behavior, course performance, or negative experiences at school (Ryan, 2011). This paper analyses documents as well as scientific literature on the use of early warning systems in US schooling, focusing in particular on ‘socio-technical imaginaries’ (Castoriadis, 1987; Jasanoff & Kim, 2015; Williamson, 2017), that is imaginations about the future of schooling evoked by the utilization of predictive early warning analytics. As the analysis shows, while the material under study – and, thus, the present discourse on early warning systems – is motivated by the socio-technical imaginary of preventing high school dropouts on the basis of data-driven personalized targeting, potential paradoxes have only scarcely been discussed. One example is the emergence of self-fulfilling prophecies, such as the potential reinforcement of discouraging school experiences through the personalization of risks and, followingly, internalization of possible failure on the part of students, which may ultimately increase the very risk factor of discouraging school experiences that it strives to overcome. This paper therefore points to an enigmatic imbalance in the debate on early warning systems, calling for more critical observations of the amalgamation of security-oriented policy agendas and the socio-technical imaginary of preventing high school dropouts.
