Irrespective of the student’s background and the local context (s)he may have grown up in, it is assumed that at the end of compulsory schooling that same student will be holding some competences relative to mathematics, especially since these are considered necessary for continuation of the education and future careers, irrespective if the student chooses STEM or not (OECD, 2016). In that sense while we speak of inclusion and exclusion of students of different background (i.e., immigrant students, low SES) into the mainstream education, we often disregard the extent to which the curriculum of the mainstream system is attuned and up to date with some of the major ideas in policy or the labour market.

The latter is especially discussed when scores of the international large scale assessments are announced and rank-ordering of the countries becomes one of the major concerns (Johansson, 2016; Torney-Purta & Amadeo. 2013). However if we try to go beyond the league tables and observe the differences in the context of the national curriculums of the countries involved, the rank ordering becomes irrelevant.

Trends in International Mathematics and Science Study (TIMSS) assess achievement in mathematics and science in grades 4 and 8. TIMSS strong foundation in the curriculum model of the assessment allows more freedom in comparing the patterns in the country scores and the national education systems. In TIMSS, the curriculum is broadly defined as the major organizing concept in view of how educational opportunities are provided to students and is observed through three set of lenses: the intended curriculum, the implemented curriculum, and the achieved curriculum (Mullis & Martin, 2013). These exemplify the mathematics the society intends for students to learn, not disregarding the fact that what is actually taught in classrooms, teachers’ characteristics and their prior education, as well as the way the content is presented to students may greatly diverge across the countries.

In the sense of the content, that is being provided to the students, during the fourth grade assessment, these include three content domains: numbers, geometric shapes and measures; and data display (Mullis & Martin, 2013). Cognitive domains of knowing; applying; and reasoning are the sets of behaviours expected of students as they deal with the mathematics content.

The present study analyses differential item functioning (DIF) between seven European countries (Croatia, Finland, Norway, Poland, Serbia, Slovenia and Sweden) using data from TIMSS 2011 and 2015 for the fourth grade assessment in mathematics. Selection of countries was grounded on several criteria. Firstly, we wanted to capture systems of different education traditions and trajectories. Secondly, we wanted to contrast countries from different parts of Europe. Finally, we wanted to include in the analyses those countries that appear relatively successful in regards to their general achievement in the TIMSS survey.

Data from TIMSS 2011 and TIMSS 2015 were used for the study purposes. The analyses included students from Croatia, Finland, Norway, Poland, Serbia, Slovenia and Sweden who participated in the TIMSS grade 4 surveys. Two samples were made for each cycle, gathering in total around 30 000 students from the above named countries in each cycle. 1 PL item response modelling (Mantel - Haenszel (MH) procedure) was used to detect differential item functioning (DIF) for the 178 items within the mathematics test on TIMSS 2011 and 179 items in the mathematics test for the 2015. Not reached and omitted responses were treated as missing data. The Test-Curriculum Matching Analysis data were used for the test curriculum-matching of the seven countries. They were in range from 47 (Croatia) to 59 (Norway) average percent correct on all the items (Mullis, Martin, & Hooper, 2016)). In addition, curriculum content analysis for the countries in focus was also carried out regarding the content covered by the curriculum and the desired outcomes for the specified grades.

Curriculum content analyses points to several difference among the examined countries. While in some countries the curriculum is tied to each grade (e.g., Serbia, Croatia), in the others it is connected to a three year period (e.g., Slovenia, Finland) or pre-defined time points during schooling (e.g., Norway). The curricula also differ in its emphasized outcomes. While the curricula in the Scandinavian countries stress more on the reasoning and problem solving skills, curriculums in Serbia and Croatia emphasize on the knowledge domain and application of what is learned. Initial analyses of the 2011 cycle data show strongest differences between the students, across the seven countries we have examined, may be found if students are solving a task belonging to the number content domain, connected to the topic of fractions and decimals. The task is of low reading demand and belongs to the knowing cognitive domain. A task holding such characteristics is relatively more difficult for students in Finland, and relatively easier for students in Croatia. Analyses for the 2015 show the same pattern with strongest differences for the number content domain, fractions and decimals as the topic, low reading demand and knowing as the cognitive domain. If we observe the relative difficulty of such type of tasks in the 2015 data, again it seems they are relatively easier for students in Croatia. Task examples will be discussed with the help of released items for TIMSS 2011, and in the light of the curriculum characteristics in each of the countries.

Johansson, S. (2016). International large-scale assessments: what uses, what consequences?. Journal of Educational Research, 58(2), 139-148. Mullis, I.V.S. & Martin, M.O. (Eds.) (2013). TIMSS 2015 Assessment Frameworks. Boston College: TIMSS & PIRLS International Study. Mullis, I. V. S., Martin, M. O., Foy, P., & Hooper, M. (2016). TIMSS 2015 International Results in Mathematics. Boston College: TIMSS & PIRLS International Study. OECD (2016). Equations and Inequalities: Making Mathematics Accessible to All, PISA, Paris: OECD Publishing. Torney-Purta, J., & Amadeo, J.-A. (2013). International Large-Scale Assessments: challenges in reporting and potentials for secondary analysis. Research in Comparative and International Education, 8(3), 248-258.