Lenz, Thomas, Backes, Susanne, Colling, Joanne, Esch, Pascale, Fischbach, Antoine, Gezer, Elif Tuğçe, Grund, Axel, Keller, Ulrich, Ugen, Sonja, and Faculty of Language and Literature, Humanities, Arts and Education (FLSHASE) > Luxembourg Centre for Educational Testing (LUCET) [research center]
• Luxembourg is a highly diverse country in terms of the socioeconomic, sociocultural, and linguistic composition of its population. This diversity is reflected in the national education system with an increasing share of students speaking a language other than Luxembourgish and/or German at home. In order to deal more adequately with the increasing language diversity of the student population and to counter educational inequalities that presumably result (at least in part) from a curriculum that places high language expectations on a growing number of students, the Luxembourgish government has broadened the educational offer by introducing European public schools (EPS). These schools follow the European curriculum and allow students to select one main language of instruction among the offered language sections (i.e., German, French, and English). • By combining data from different sources (e.g., administrative student data, expert interviews with stakeholders, achievement scores in mathematics from the Luxembourg School Monitoring Programme “Épreuves Standardisées” - ÉpStan), the present report offers preliminary results on EPS in Luxembourg. They consist of (1) the societal demand for EPS; (2) the composition of the student population in EPS; (3) the perception of EPS by school management teams and parents, and tangible education outcomes in the form of (4) educational trajectories; and (5) academic achievement in mathematics among EPS students compared to their peers in schools following the Luxembourgish curriculum. Described below are the key preliminary findings for each of these aspects: (1) Since 2016, a total of six EPS have opened in different locations across Luxembourg and the amount of students attending EPS has increased considerably at both primary and secondary school level. With the number of applicants surpassing the number of places currently available in EPS, it can be concluded that there appears to be high demand for EPS. (2) With students having a low socioeconomic status (SES) and/or students speaking Portuguese at home taking up the offer of EPS less frequently than high SES students and/or students speaking French or English at home, the student population in EPS differs from the student population in schools following the Luxembourgish curriculum (e.g., nationality, language primarily spoken at home, SES). (3) School management teams and parents report a rather positive perception of EPS, with the extended linguistic offer (i.e., possibility to select a language section) being the main reason why parents select EPS for their child. (4) Looking at the educational trajectories of EPS students, preliminary results offer a tentative indication of EPS students showing less school delay than their peers in school following the Luxembourgish curriculum and high continuity in their educational trajectories (i.e., the vast majority of students remains in EPS instead of changing curriculum). (5) With regard to achievement in mathematics at primary school level, the present report indicates that students in EPS perform better than their peers in schools following the Luxembourgish curriculum. At secondary school level, EPS students perform better than their peers in Enseignement secondaire général - voie d'orientation (ESG) and in Enseignement secondaire général - voie de préparation (ESG-VP), while staying below the performance of Enseignement secondaire classique (ESC) students. Although low SES students or Portuguese speaking students in EPS show better achievement scores than their respective peers in schools following the Luxembourgish curriculum it is not yet possible to draw strong conclusions based on these preliminary findings as these student groups currently take up the EPS offer less frequently than their peers considered as advantaged in the context of schooling. Their number is currently too small to allow more robust and in-depth statistical analyses. • The present report’s findings, especially regarding the tangible educational student outcomes, however, must be considered as tentative due to important methodological limitations. Indeed, the small numbers of students in EPS, particularly so for student groups with specific background characteristics (e.g., low SES students, Portuguese speaking students), do not allow separate analyses based on language section, for example. Thus any identified pattern could be sensitive to the inclusion or exclusion of outliers (e.g., students with particularly high or low ÉpStan scores). In addition, the comprehensive EPS school system at secondary school level (i.e., common track) is compared to the ability-based tracked school system of schools following the Luxembourgish curriculum, which limits the interpretability of secondary school data. Regarding the academic achievement tasks in mathematics, it should be noted that they were developed using education standards of the Luxembourgish curriculum. It is thus possible that achievement was underestimated for EPS students (e.g., assessment of mathematical concepts that have not yet been introduced in EPS). To this date, the ÉpStan administered in EPS only assessed academic achievement in mathematics for which a bigger overlap between curricula is assumed than for language subjects (e.g., German, French). Current psychometric shortcomings (e.g., different timepoints of language introduction within the language section in EPS, task development, comparability of tasks) do not yet allow to assess academic achievement in language subjects. • Considering that the ÉpStan do not currently include a measure that operationalises the learning environment, the present report is unable to draw any conclusions regarding which EPS aspect contributes decisively in explaining the observed differences in educational outcomes. Nevertheless, three potential explanations are presented for further exploration: better linguistic fit in EPS (i.e., students learning to read and write in their native or a related language), structural differences between school offers (e.g., primary and secondary education within one institution, the institutionalized quality assurance and flexibility in teacher recruitment in EPS), and the differences in the composition of the student population (i.e., lower uptake rate of the EPS offer by low SES students and Portuguese speaking students). • The finding that low SES students and Portuguese speaking students take up the EPS offer less frequently than their high SES peers and French or English speaking students, and that the EPS student population differs from the student population in schools following the Luxembourgish curriculum, could potentially result out of three main hurdles: namely (1) the application of selection criteria considering that the demand for EPS is surpassing the number of available places (i.e., the linguistic and/or academic profile of applying students is taken into consideration); (2) lacking system knowledge regarding the characteristics of Luxembourg’s education system among all actors involved in education (which makes it difficult to take informed decisions on a student’s education); and (3) potential organizational challenges that hamper the uptake of the EPS offer (e.g., geographical location of the EPS). • In light of the tentative result that students in EPS show better educational outcomes than many of their peers in schools following the Luxembourgish curriculum, two main implications for educational policy can be deduced. First, the student composition of EPS could be diversified in a targeted manner. This could be achieved, for example, by a) encouraging EPS to target student groups considered as disadvantaged in the context of schooling (e.g., low SES students) more effectively, and by b) fostering an encompassing system knowledge (e.g., characteristics, similarities and differences of the two school offers) among all actors involved in education (e.g., teachers, parents, educational advisors, school psychologists) to allow parents to take an informed decision on their child’s education. A second implication would be to introduce certain characteristics of EPS in schools following the Luxembourgish curriculum (e.g., extending the linguistic offer as in the French literacy acquisition pilot project currently implemented in four C2.1 classes). • By progressively integrating EPS into the well-established Luxembourg School Monitoring Programme, the ÉpStan will allow for a more in-depth analysis of potential educational outcome differences between EPS and schools following the Luxembourgish in the future. With the aim of providing reliable data for evidence-based policy making in the field of education, the results from the ÉpStan could in turn be used for the creation of school offers in which all students can make use of their full academic potential irrespective of their individual background characteristics (e.g., SES, language background).