Taking Test Results "Into Consideration" When Grading

• In Event: More Than a Mark: Understanding The Social and Cultural Dimensions and Consequences of Grading

Abstract

Objectives: In Sweden, grades are high stakes for students. Measures have been taken to increase the agreement in teachers’ grading, most recently by legally requiring that teachers take results from national tests into consideration when grading. This relatively loose strategy of taking test results “into consideration” has not yet yielded any observable change in teachers’ grades, however, and there is still a large discrepancy between teacher-assigned grades and test results for most subjects (Swedish National Agency of Education, 2019). Although this discrepancy has spawned a heated debate on how to better align teachers’ grading with test results, the reasons behind the discrepancies are not known, making it difficult to implement more focused measures. This study therefore aimed to understand teachers’ rationales for assigning grades that deviate from results on national tests.

Perspective: In this study, teachers’ grading is seen as a sociocultural practice, which is mainly intuitive and implicit (e.g. Bloxham et al., 2016), but still possible to reflect upon and partially articulate. This perspective is contrasted with a psychometric viewpoint, as represented by the national tests.

Methods: This is a small-scale interview study (n=21), investigating teachers’ perceptions of the national tests and how they take test results “into consideration”. Teachers teaching the last year of Swedish compulsory school (year 9) in either English as a foreign language (EFL) or physics were recruited through social media and interviewed online. EFL and physics were chosen because the discrepancy between teacher-assigned grades and test results were most prominent in these subjects. The semi-structured interviews were transcribed, and the transcripts analyzed with qualitative thematic analysis.

Results: Findings suggest that the main reason for teachers to assign grades that deviate from test results is that they do not find the test results valid, sometimes even misleading. Furthermore, it is mainly design features of the tests, which contribute to teachers not considering test results as valid. For example, students who fail one of the subtests in EFL may still pass the test as a whole, due to the compensatory scoring model. This is not acceptable to the teachers, since, according to the national curriculum, students need to meet the minimum requirements for all traits (i.e. speaking, writing, etc.). The study identifies several instances when teachers perceive that psychometric principles of test design conflict with their grading practices, resulting in teachers not taking test results into consideration.

Significance: If teachers are to take results from national tests into account when grading, they need to perceive the test results as valid. This study provides an insight into teachers’ rationales for choosing not to consider test results; findings that have implications for test design and/or for attempts to inform/educate teachers about how to consider test results when grading. The findings may also be used to develop a survey instrument, in order to investigate how representative the findings are among a larger sample of teachers.

Author

• Anders Jönsson, Kristianstad University