Standardising the ‘Good Student’: Adaptability and Algorithms in Danish School Testing

Laura Høvsgaard

    Research output: Chapter in Book/Report/Conference proceedingConference abstract in proceedingsResearchpeer-review

    Abstract

    In 2006, the Danish Parliament passed a bill making national tests a mandatory assessment tool in public schools (Folkeskolen). The annual digital test is one of several initiatives that arose from a review conducted by the OECD in 2004, which concluded that Danish schools need to establish an ‘objective evaluation of students’ in order to improve ‘proficiency standards’. Accordingly, the national test can be seen as a political response to improve the standard of Danish Schools and has become a central metric in school evaluation.
    The national test is a digital testing system which has incorporated an adaptive algorithm into its functioning. This adaptability feature is characterized as producing a more “accurate” test because it adapts each question’s difficulty level to the student’s performance throughout the testing scenario. In effect, proponents argue that one of the test’s strengths is that each student sits an individually “tailored” test. As a calculative device the algorithm is part of what generates the standard for when a student is considered to be ‘good’ and as such it has become a key actor in making the school a more measurable entity.
    While it is most often the case that algorithms are a ‘hidden’ feature within computational systems, in the case of the Danish national test, the algorithm’s adaptability makes it, in fact, a much discussed feature of the testing system. While this paper examines the mechanisms through which the algorithm works,
    it pays particular attention to its adaptability, analysing the broader assumptions and values that are incorporated within it. At the same time the paper investigates the collateral effects of adaptive thinking, that is, it examines what else is being made adaptive (students, teachers, principles, the school, the municipality, the nation) with and alongside the algorithm.
    Original languageEnglish
    Title of host publicationAbstracts : Third Nordic Science and Technology Conference
    Number of pages1
    Place of PublicationGothenburg
    PublisherGöteborg Universitet
    Publication date2017
    Pages97
    Publication statusPublished - 2017
    EventThird Nordic Science and Technology Studies (STS) Conference - University of Gothenburg, Campus Haga, Gothenburg, Sweden
    Duration: 31 May 20172 Jun 2017
    Conference number: 3
    https://socav.gu.se/english/research/third-nordic-science-and-technology-studies-conference

    Conference

    ConferenceThird Nordic Science and Technology Studies (STS) Conference
    Number3
    LocationUniversity of Gothenburg, Campus Haga
    Country/TerritorySweden
    CityGothenburg
    Period31/05/201702/06/2017
    Internet address

    Cite this