The students who graduated in the spring of 1997, after three years’ of upper secondary school, was the first cohort who graduated with the new, criterionreferenced, grades. The new grading system is characterised by having defined goals for the grade levels. The purpose of this study is to investigate the relation between the new upper secondary school grades in English, Mathematics and Swedish, and the SweSAT subtests, administrated in the autumn of 1996 and spring of 1997, taken by the students who graduated from a Swedish upper secondary school in 1997. The result shows that there is a positive correlation between grades and SweSAT scores. The subjects English and Mathematics have a stronger correlation with the SweSAT, compared with Swedish. The English B-course have the strongest correlation with the subtests WORD, READ and ERC, and with the overall SweSAT score, while the Mathematics A-course have the strongest relation to the subtests DS and DTM. The same results show for both the SweSAT administered in the autumn of 1996 and the test administered in the spring of 1997. The students are ranked similarly by the SweSAT test scores as by their grades from the courses English A and B, Mathematics A, B, C, D, E and Swedish A and B. In two of the courses included in the study, English C and Swedish C, students with the grade level IG (fail) are ranked higher by their SweSAT scores than students with the grade level G (pass). The students who scored the highest on the SweSAT have also the highest grade level (pass with special distinction) in the courses English B and/or Mathematics E. Students from the Natural Science oriented programme are in majority among these students. The students who score the lowest on the SweSAT have the grade IG in English A and/or Mathematics A. When performing a factor analysis, three factors were identified. One factor was found in the
subtests WORD, READ, ERC and English grades. Another factor was found in the subtests DS, DTM and the Mathematics grades. The third factor was found only in the grades from the courses Swedish A, Swedish language B, and Swedish literature B. The aim would be to modify the SweSAT by introducing an additional dimension, a suggestion is to focus on the factor that is defined by the Swedish courses. However, making such a modification is a considerable step to take, and it is only by research that new and necessary knowledge can be obtained, that will make it possible to change the SweSAT in a longer perspective.
Umeå: Institutionen för beteendevetenskapliga mätningar, Umeå universitet , 2004. , 51 p.