This systematic review maps and reports on 21st century Swedish and international research into the field of formative assessment in compulsory school. Our analysis shows that theoretical studies are increasingly positioning formative assessment as a context-bound social practice, rather than a set of general methods. Meanwhile a few large meta-studies, featuring remarkable effect sizes of increased learning on behalf of students, have contributed to policy decisions advocating large-scale implementation of formative assessment practices in many countries around the world. Researchers seem to agree that classroom implementation of formative assessment practices needs to be supported by processes in which professional learning is given space, but in reality such processes are often lacking. It has been seen that the lack of peer learning among teachers and school leaders often means that pseudo-formative practices evolve, where the understanding of formative assessment tends to be instrumental and linked to ritual work with a general method package. We have also seen that ICT-related tools for formative assessment are gaining ground. When formative assessment is regarded as time consuming from the teacher point of view, ICT tools are seen as one way to streamline instructional processes. It is evident in some of these studies that computers/mobile devices tend to be understood as independent actors, who in themselves are regarded as formative, i. e. as being the ones analysing students’ "learning" (answers to questions) and providing feedback. As a result, we predict that important issues for future research will concern, for instance, which type of feedback that can be provided by computers, and with what quality it can be provided. The consequence of digital tests and digital feedback for teachers’ professionalism is of course another important issue, since formative assessment serves to develop not only the students but also the teachers’ understanding of how learning processes can be customized to meet students’ needs. A third interesting question, linked to the former, concerns who develops the software used for this purpose. Generally, it can be concluded that empirical studies conducted at compulsory school level are few, and that we often rely on studies conducted in higher education when commenting on beneficial effects of formative assessment on student learning. Meta studies dealing with general effects of formative assessment are problematized, in part because the umbrella term formative assessment involves so many and disparate phenomena that it is problematic to speak of one overall effect. Swedish research into formative assessment is rather scarce, especially considering how large impact formative assessment as a set of methods/approaches seems to have in Swedish schools. Nevertheless, the results from the various Swedish studies point to a certain coherence. Studies of documentation, mainly through individual development plans, dominate the research field, along with a few classroom studies of teachers’ formative assessment work. There are also studies that focus on national - and to some extent local/municipal - governance of the school, and the consequences of various types of control for teachers’ assessment work. The need for different types of studies that include the perspectives of multiple levels of actors, is evident: a collaboration between research based on qualitative and quantitative data are needed. Intervention studies that measure the impact of different aspects of formative assessment need to be complemented by studies of how teachers, principals and local/municipal policy levels analyse and use assessment information to shape the instructional processes that will ultimately lead to increased learning/higher goal attainment among students. In addition, surveys and studies of Swedish teacher training would be desirable, given that international research shows that too few educators within teacher education programs have the in-depth knowledge required for adequately making teacher students assessment literate. Our review also shows that a very small percentage of studies take the students' perspective, wherefore we know very little about how they perceive and are affected by various methods which are classified as formative assessment.
Stockholm: Vetenskapsrådet , 2015. , 91 p.