Introduction A large amount of research argues that grades are influenced by subjectivity and that factors other than achievement, such as different student characteristics, teachers’ grading practices and systematic differences within and between schools, exert an influence on grades (Brookhart, 1994; Cliffordson, 2004; Klapp Lekholm & Cliffordson, in press; Pilcher, 1994). Although the relation between cognitive ability and achievement is strong (Gustafsson & Balke, 1993) different non-cognitive factors, such as students’ attitudes, interest, motivation, out-of-school experiences, family values and gender, all seem to be of importance for understanding achievement and grade differences (Wang & Lin, 2005). In a previous study, (Klapp Lekholm & Cliffordson, in press) two dimensions in grades were identified at both an individual and a school level; a subject-specific dimension (achievement) related to grades and national tests scores in three core subjects, and a common grade dimension (non-cognitive factors) related only to grades and which cut across the different subjects. The result also showed that there were gender differences in the common grade dimension. Background There is a substantial amount of research on the importance of teachers’ expectations of student behaviour in the assignment of grades. In order to obtain high grades, students should, for example, be interested, motivated and cooperative. In particular, student interest has been shown to be an important cognitive and emotional resource for school outcomes (Hidi et al., 2004) and students may have both domain-specific interests (sports, school-subject) as well as more general interests in learning. In most industrial countries, research has found that girls perform better in school and obtain better grades (Nycander, 2006). Murphy (2000) has also suggested that out-of-school experiences give students different opportunities to perform in school, and that boys and girls encounter different expectations from parents, teachers and society at large. The main aim of the present study was to investigate how different student characteristics and gender influence the subject-specific and common grade dimensions. Additionally, interest was focused on the investigation of the mediating effects of gender via student characteristics on grades.

Method Data derives from The Gothenburg Educational Longitudinal Database (GOLD) which contains register data compiled by Statistics Sweden for all individuals born between 1972-1992. The subjects were 99 070 students born in 1987, who left compulsory school in 2003. This is the whole population and full information is available for subject grades and national tests scores in Swedish, English and mathematics, as well as for gender. The student questionnaire data used in this study comes from the ETF-project (Evaluation Through Follow-up) where a 10% nationally representative sample has been collected. The methods of analyses used were Confirmatory factor analysis (CFA) and structural equation modelling (SEM). As measures of model fit, the χ² goodness-of-fit test and Root Mean Square Error of Approximation (RMSEA) were used and missing data analyses have been conducted (Muthén, Kaplan & Hollis, 1987). Mplus version 3 (Muthén & Muthén) was used for the estimation and testing of the models. STREAMS (Gustafsson & Stahle, 2005), a modelling front-end environment, was used to execute the analyses.

Results and conclusions The result indicates that grades are influenced mostly by students’ achievements in the different subjects, but that grades are also influenced by different non-cognitive characteristics, such as interest and parental support. The result also indicates that student interest is the characteristic that exerts the greatest influence on the common grade dimension, and that it influences grades positively, whilst general interest in learning accounts for the entirety of gender differences in the common grade dimension. Thus, one important mechanism behind the advantage in grades for girls seems to be that they express a greater amount of interest in schoolwork than boys.