Since 2018, Swedish teachers are expected to pay particular attention to pupils’ results on national tests when grading. There are, however, still large discrepancies between national test results and teachers’ grades, which differ between schools and subjects. The purpose of this study is to investigate possible reasons behind these differences, from a teacher perspective. The study is based on semi-structured interviews with teachers who teach grade 9, either in physics (n=9) or English (n=12). The interviews have been transcribed and analyzed with thematic content analysis. The most important reason for differences is that the test results are considered misleading by the teachers. Although most reasons for considering the test results as misleading relate to test design, some of the reasons relate to the pupils. The only reason for differences that does not relate to test results being considered misleading, is that some pupils receive instruction at the end of the semester based on their shortcomings identified in the test. Furthermore, since the teachers tend to compare test results and own grades in a 1:1 ratio, they may discard the entire test if it is considered misleading. The fact that test results are presented as a single score or grade, therefore, seems to contribute to teachers not paying particular attention to test results when grading.