Previous research by Klapp and colleagues (e.g. Klapp 2015; Klapp, Cliffordson & Gustafsson 2016; Klapp, 2018) identified a negative grading effect for low-achieving 7-9th grade students (13-15 years old) whereby individuals who previously received grades (in 6^th Grade) show poorer grade development than similarly capable peers who were not previously graded. This effect was found to be greater for boys than for girls. Other research has, however, drawn contrary findings – that not grading students results in worse outcomes (Sjögren, 2010). However, both of these findings were obtained from Swedish data collected in or before 1980. Since then Sweden has undertaken several substantial reforms of the educational system.

These reforms included extensive decentralisation of the education system with transfer of responsibilities from central government to local municipalities, as well as the introduction of several new curricula (Wikström, 2006). From the start of the academic year 1994/95 the new curriculum shifted the focus from knowledge acquisition to critical thinking, problem solving, and cooperation (Pettersson, 2004; Wikström, 2006). Academic year 1994/95 also saw the replacement of the previous norm-relative grading system with the new criterion-referenced grading system. Further reforms in 2011 also introduced a new grading scale which included an explicit fail grade.

Under the previous norm-referenced system students were assessed on a national scale, with centrally administered standardised tests, which would inform the grades that teachers gave their students. However, these tests were not available for all subjects and were delivered infrequently. Teachers also misunderstood and misapplied norm-referenced grading principles to class groups rather than at the national level. Since the reform in 1994/95, students are no longer graded relative to the national cohort, but by whether they have achieved learning goals according to fixed criteria. This means that the number of top grades awarded are no longer limited, but that everyone could theoretically achieve them if the learning targets are reached. In the previous norm-relative grading system this was not possible; additional effort did not necessarily result in improved grades, especially if a student’s peers had also made similar additional effort. Changes to the curriculum targets and assessment methods will have resulted in changes in the teaching environment which might impact students.

In Sweden, the main functions of grades are to enable selection to higher education, for official performance feedback, and for motivational purposes. The introduction of an explicit fail grade where there previously was none, arguably increases the stakes, particularly given that a fail grade has negative implications for admission to higher education. This stricter grading system places more demands on the student and potentially impacts student motivation and performance.

Previous research findings are explained in terms of competition for limited grades either contributing to a negative learning environment or as a motivating factor for different individuals (Klapp, 2015; Sjögren, 2010). It has also been speculated that by removing the passive comparison-by-grades inherent to the norm-relative system, criterion-referenced grading might encourage active outward grade competition, potentially exacerbating the negative competitive learning environment (Wikström, 2006). At the same time, the prospect of a fail grade with substantial repercussions for the future might provide internal incentive to avoid failure.

These multiple changes to the education system may have substantially changed the motivational and learning environment for students, with consequences for their performance and educational outcomes. In light of these changes the research concerning the negative grading effect needs to be updated to determine the impact of these changes. Therefore, the present study plans to investigate whether the negative grading effect can be found in more recent data to determine if the effect is also present in the current criterion-referenced grading system.

Data from the Evaluation Through Follow-up (UGU) database before and after the 2011 reform will be used to compare the grade progression of individuals who have previously received grades to those who have not. The UGU database is a Swedish large cohort database for evaluation of the school system. The database contains regular representative samples (10 % of the population) of students from birth cohorts between 1948 and 2004, in which students are followed at regular intervals throughout their school path. The database contains survey and test information and includes school, family, and social background data as well as register data of school grades and national test results. The cohorts relevant to the present study are cohorts born in 1998 (N=9180) and 2004 (N=5190), which were in grade 6 in the academic years 2010/11 and 2016/17 respectively. These cohorts are well-placed on either side of the academic year 2012/13 in which the new grading scale was first implemented, and allows for the fluctuating effects of novelty and uncertainty of the new system to stabilise. Initially, descriptive statistics will be estimated. Then, several multivariate multiple regression models will be estimated in several steps. First, a basic model with variables graded/not graded, achievement level, gender and SES will be estimated. Then cross product terms will be used in order to investigate interaction effects between the variables thus a saturated model will be fitted. Finally, a model with only the significant interaction effects will be estimated. Test of linearity and the assumption of homoscedasticity of the residuals will be done. Missing information in the variables will be handled by using the missing data modeling in the Mplus program (Muthén, Kaplan & Hollis, 1987). To take account of possible clustering of students in schools (school level), the complex option offered by the Mplus program will be used. Cohen´s d will be computed as a measure of effect size (Ellis, 2010). The analyses will be conducted in the SPSS program, version 27 (2020) and in the Mplus program, version 8 (Muthén & Muthén, 1998-2019).

Analyses of the more recent data are expected to show that a negative grading effect is also present in data from the criterion-referenced grading system. The effect may be even more pronounced for low-achieving students than in the previous norm-relative grading system, given the existence of a fail grade in the present criterion-referenced grading system, where there was not one before. Similar trends in the data in relation to sex and cognitive ability are expected as those previously reported in the norm-relative grading system. As there is extensive debate about whether the change from a norm-relative to criterion-referenced grading system has had a positive or negative impact on grades, particularly via the mechanism of affecting student motivation, it is unclear whether the effects are likely to be larger or smaller overall than previously identified. This investigation and findings will be important because many education systems around the world use a criterion-referenced rather than a norm-relative grading system. To our knowledge there is an absence of literature on this topic for criterion-referenced grading systems. Significant findings would justify subsequent analyses such as longitudinal implications or the inclusion of personal characteristics such as academic self-concept. The findings of this research could also shed light on the potential impact of adding an explicit fail grade to the grading scale. Such results could indicate that further research is required into the impact of grade structures on motivation and performance.

Klapp, A. (2015). Does grading affect educational attainment? A longitudinal study. Assessment in Education: Principles, Policy & Practice, 22(3), pp302-323, DOI: 10.1080/0969594X.2014.988121 Klapp A. (2018). Does academic and social self-concept and motivation explain the effect of grading on students´ achievement? European Journal of Psychology of Education, 33(2), 355-376. Klapp, A., Cliffordson, C., & Gustafsson, J.-E. (2016). The effect of being graded on later achievement: evidence from 13-year olds in Swedish compulsory school, Educational Psychology, 36(10), pp1771-1789, DOI: 10.1080/01443410.2014.933176 Muthén, B., Kaplan, D., & Hollis, M. (1987). On structural equation modelling with data that are not missing completely at random. Psychometrica, 52, 431-462. Muthén B., & Muthén, L. (1998-2019). Mplus user´s guide (8th ed.) Los Angeles, CA: Author. Pettersson, (2004). The National Tests and National Assessment in Sweden. Stockholm Institute of Education, Stockholm: PRIM-Group. Sjögren, (2010). Graded children – evidence of longrun consequences of school grades from a nationwide reform. Working paper 2010:7, IFAU (Swedish Institute for Labour Market Policy Evaluation). Uppsala: IFAU. Wikström, (2006). Education and assessment in Sweden, Assessment in Education, 13(01), pp113-128, DOI: 10.1080/09695940600563470