The utility of low-stakes assessment with the example of the Berlin Progress Test

In a culture where grades are the aim of a course or a study year there seems to be no rationale for assessments where student performance has no consequences. However, such low-stakes assessments serve several purposes: for students as formative assessment, for faculty as evaluation tool, for policy as large-scale assessment and for society as research tool. Although they seem low-stakes for participants, low-stakes assessments can have severe consequences for teachers, faculty, institutions or policy (Breakspear, 2012; Cole, 2007; Cole & Osterlind, 2008). If test-taking effort is not taken into account, the validity of results can be threatened (Akyol, Krishna, & Wang, 2018; Brown & Walberg, 1993; Butler & Adams, 2007; Eklöf, 2010; Penk, 2017; Thelk et al., 2009; Wise & DeMars, 2005; Wolf & Smith, 1995). As a consequence, low-stakes assessments may not serve their purposes properly. However, low-stakes assessment is only useful if it serves its purpose. Progress tests can serve all of the purposes described above and they represent both moderate-stakes and low-stakes assessment. Therefore, research findings concerning moderate-stakes versus low-stakes progress tests can be compared and thus the special aspects of low-stakes assessments can be worked out. In this cumulative habilitation I first introduce progress testing: why progress tests were developed, what they look like, where they are used, how the purposes of low-stakes assessments are fulfilled and what the stakes of progress tests mean. Furthermore, I discuss moderate- and low-stakes progress tests in terms of the findings for each component of the model of Utility of Assessment Methods (van der Vleuten, 1996) to outline the special features of low-stakes assessment, which are the subject of the studies in my research. These studies pursue the following aims • developing and validating a short scale for identifying students with low test-taking effort, • investigating the construct validity of a low-stakes progress test after eliminating non-serious test-takers, • finding strategies that are related to the acceptability of low-stakes progress tests