Conference Agenda
| Session | ||
31 SES 02 B JS: Language & Assessment - Joint session (NW 31 and NW 09)
Joint Paper Session | ||
| Presentations | ||
31. LEd – Network on Language and Education
Paper Test-taker Perceptions of EFL School-leaving Test Difficulty: the Role of (In)formal Learning Contexts Charles University, Faculty of Education, Czech Republic (Czechia) Presenting Author:Language tests are not only psychometric instruments designed to measure specific language skills, but also social and experiential events embedded in particular educational and institutional contexts (Cheng & DeLuca, 2011; Jin, 2023). Although contemporary test validity theory increasingly recognizes the role of test-taker experience as part of a comprehensive validity argument (Jin, 2023; O’Sullivan, 2011; Weir, 2005), empirical research still tends to privilege statistical indicators of test quality, including item difficulty parameters. This emphasis is especially visible in standardized national examinations across Europe, where extensive psychometric monitoring of item functioning is routinely conducted by test developers and national authorities (EACEA & Eurydice, 2009), yet students’ own perspectives on test difficulty remain underexplored. Studying subjectively perceived test difficulty is important because test performance is shaped not only by psychometric properties of the test items but also by how test-takers experience and interpret test tasks in situ. The test items may be experienced as difficult not because they require higher proficiency, but because they overload attention or clash with learners’ expectations. Perceived difficulty influences strategy use, as tasks experienced as overly difficult tend to elicit guessing or disengagement, whereas tasks perceived as manageable foster deeper processing and more systematic strategy use (Atai & Hashemi, 2008). Thus, learners with comparable proficiency may perform differently on the same items due to differences in perceived task difficulty. Previous studies showed that familiarity, educational context, and experiential frequency strongly influence difficulty perceptions (Garcia-Ponce et al., 2022; Gardiner & Howlett, 2016; Kim & Elder, 2015; Oruç Ertürk & Mumford, 2017; Prestwood & Weiss, 1977). Analyzing perceived difficulty therefore helps explain subgroup differences and performance patterns that remain opaque when relying on psychometric outcomes alone. Most empirical studies on perceived difficulty have been conducted in university or professional testing contexts. In contrast, upper-secondary school-leaving examinations, despite their high-stakes role as exit qualifications and gatekeepers to further education across Europe, remain underrepresented in this line of research. One of the few exceptions is Stopar and Ilc’s (2016) study of the Slovenian Maturita exam, which revealed divergence between students’ perceived and psychometric difficulty measures in some tested items. While highlighting linguistic features that contribute to increased subjective difficulty ratings, this study did not explore how learners’ educational and social backgrounds shape their perceptions of test task demands. Addressing this gap, the present study adopts a mixed-methods design to examine students’ perceptions of item difficulty in EFL school leaving test (Czech English Maturita test). It seeks (1) to identify which test items or subdomains are perceived as easy or difficult by the test-takers and why, and (2) to examine how these perceptions relate to students’ educational, experiential, and social backgrounds. By integrating quantitative questionnaire data with qualitative accounts of test-taker experience, the study contributes to developing the concept of perceived difficulty. Besides theoretical implications, the study engages with European and international debates on test validity, fairness, and comparability in standardized school-leaving language assessments. Methodology, Methods, Research Instruments or Sources Used The study employed an explanatory sequential mixed-methods design (Creswell, 2012) combining a small-scale EFL mock test (15 items selected from published past Maturita tests to cover different language skills and task demands), item difficulty rating scales and student background questionnaire, followed up by qualitative interviews with selected testees to solicit information about their lived experience of test item difficulty to help explain the patterns identified in the quantitative phase. Participants were 50 final-year upper-secondary students (Year 13) from four Czech secondary schools, representing academic and vocational tracks. Given a relatively low difficulty of the EFL Maturita test as compared to curriculum specifications, schools were purposively selected from the lower two quartiles of median achievement in the previous national EFL examination to ensure inclusion of students likely to experience the demands associated with solving the mock test. School principals were asked to nominate classes preparing for the English Maturita, preferably those expected to find the test challenging. The sample showed substantial variation in educational achievement and family background, providing a suitable basis for analysing perceived difficulty. Data collection took place in April 2025, approximately one month before the national exam, and followed a two-step procedure. In the first step, students completed a 15-item EFL mock test under time constraints comparable to the national examination, followed by item-level difficulty ratings (1 = very easy to 5 = very difficult) and a 20-item background questionnaire covering exposure to English, English self-concept, exam preparation, and family background. In the second step, 16 students were purposively selected for semi-structured interviews based on their answers provided in the first phase with an aim to achieve a heterogeneous sample with maximum variation in test performance, perceived difficulty, and learning backgrounds. Interviewees reflected on their emotional reactions, identified easy and difficult items, described solution processes and obstacles, and discussed their in- and out-of-school English learning experiences. Quantitative data were analysed using descriptive statistics, correlations and OLS regression models in R. Qualitative data were transcribed verbatim and analysed thematically using a combination of deductive and inductive coding (Gibbs, 2007). Throughout the analysis, analytical memos (Saldaña, 2013) and comparative matrices were systematically developed on the basis of the emerging codes and themes, thereby supporting interpretation and data exploration. Generative AI tools were used as a support during the analysis process (Nguyen-Trung, 2025). Ethical approval was obtained from the authors’ university, and informed consent was secured from all participants. Conclusions, Expected Outcomes or Findings Sampled students perceived the mock test overall as moderately easy (M = 2.19). Listening was rated as the least difficult subdomain (M = 1.94), while Reading was perceived as the most difficult (M = 2.33). Perceived difficulty aligned with objective performance, as items and subdomains rated as more difficult had lower success rates. Regression analyses showed that the higher frequency of out-of-school exposure to English was the most consistent predictor of lower perceived difficulty across test subdomains, remaining significant even when other background factors were controlled for. More positive attitudes toward English, stronger English self-concept, and higher prior achievement as measured by school grades were also associated with lower perceived difficulty, although these relationships attenuated in multivariate models. In contrast, socioeconomic background, gender, exam preparation, and private tutoring participation showed no robust independent effects. Consequent qualitative interviews analysis revealed that students with extensive out-of-school exposure to English (particularly through films, series, music, gaming, or reading) described greater listening ease, faster lexical access, and an intuitive sense of grammatical acceptability, which reduced perceived difficulty across subdomains. Not only the amount but also the nature of exposure mattered: active engagement, such as rewatching without subtitles, looking up unknown words, or intentionally using English in everyday contexts shaped students’ perceptions of lower difficulty, whereas passive exposure often failed to lead to desired outcomes. School-related factors acted indirectly, with consistent instruction, exam familiarization, and stable teaching reducing uncertainty and anxiety. Unlike school-based EFL tests designed to evaluate and support student learning progress, national school-leaving exams have far-reaching consequences in cases of failure. Given that not only objective item features but also perceived task difficulty shape student test-taking behavior (Atai & Hashemi, 2008), this study calls for integrating the learners’ perspective into school-based language learning. References Atai, M. R., & Hashemi, S. M. R. (2008). Task-based Strategy Assessment: The Effect of Task Difficulty on Listening Strategy Use of Advanced EFL Learners. The Journal of Asia TEFL, 5(3), 199–223. Cheng, L., & DeLuca, C. (2011). Voices from test-takers: Further evidence for language assessment validation and use. Educational Assessment, 16(2), 104–122. https://doi.org/10.1080/10627197.2011.584042 Creswell, J. W. (2012). Educational research: Planning, conducting, and evaluating quantitative and qualitative research (4th ed). Pearson. EACEA & Eurydice. (2009). National testing of pupils in Europe: Objectives, organisation and use of results. Publications Office. https://data.europa.eu/doi/10.2797/18294 Garcia-Ponce, E. E., Mora-Pablo, I., & Segovia-Hernadez, J. G. (2022). Role of EFL Learners’ Perceptions of Task Difficulty in Complexity, Accuracy and Fluency: An Exploratory Case Study. Porta Linguarum Revista Interuniversitaria de Didáctica de Las Lenguas Extranjeras, (37), 123–142. https://doi.org/10.30827/portalin.vi37.15855 Gardiner, J., & Howlett, S. (2016). Student perceptions of four university gateway tests. Gibbs, G. (2007). Analyzing qualitative data. SAGE Publications, Ltd. https://doi.org/10.4135/9781849208574 Jin, Y. (2023). Test-taker insights for language assessment policies and practices. Language Testing, 40(1), 193–203. https://doi.org/10.1177/02655322221117136 Kim, H., & Elder, C. (2015). Interrogating the construct of aviation English: Feedback from test takers in Korea. Language Testing, 32(2), 129–149. https://doi.org/10.1177/0265532214544394 Nguyen-Trung, K. (2025). ChatGPT in thematic analysis: Can AI become a research assistant in qualitative research? Quality & Quantity, 59(6), 4945–4978. https://doi.org/10.1007/s11135-025-02165-z Oruç Ertürk, N., & Mumford, S. E. (2017). Understanding test-takers’ perceptions of difficulty in EAP vocabulary tests: The role of experiential factors. Language Testing, 34(3), 413–433. https://doi.org/10.1177/0265532216673399 O’Sullivan, B. (2011). Language testing. In L. Wei, Z. Hua, & J. Simpson, The Routledge Handbook of Applied Linguistics (2nd edn, pp. 259–273). Routledge. https://doi.org/10.4324/9781003082644 Prestwood, S., & Weiss, D. (1977). Accuracy of perceived test-item difficulties (p. 29) [Technical report]. MINNESOTA UNIV MINNEAPOLIS DEPT OF PSYCHOLOGY. https://apps.dtic.mil/sti/citations/ADA041084 Saldaña, J. (2013). The coding manual for qualitative researchers (2. ed). SAGE Publ. Stopar, A., & Ilc, G. (2016). Item and task difficulty in a B2 reading examination: Perceptions of test-takers and CEFR alignment experts compared with psychometric measurements. Círculo de Lingüística Aplicada a La Comunicación, 67, 318–342. https://doi.org/10.5209/CLAC.53487 Weir, C. J. (2005). Language testing and validation: An evidence-based approach. Palgrave Macmillan. 31. LEd – Network on Language and Education
Paper From Learning to Read to Reading to Learn: Building a Coherent Reading Assessment Framework in Primary Education 1: ISEI-IVEI - Basque Institute for Educational System Evaluation and Research, Spain; 2: UPV-EHU - University of the Basque Country, Spain Presenting Author:Reading development is often described as a transition from learning to read in the early grades to reading to learn in later primary education (Chall, 1983) involving the progressive integration of multiple components such as fluency and language comprehension (Scarborough, 2001; Duke & Cartwright, 2021). However, this developmental perspective is not always reflected in educational assessment practices, as educational systems rarely rely on coherent assessment frameworks that allow monitoring this developmental trajectory with conceptually aligned and educationally meaningful instruments. This challenge is particularly salient in bilingual education systems where reading is taught and learned in two languages and assessment must be conducted in both of them, especially when one of these languages is a minoritised language, such as Basque in the Basque Autonomous Community in Spain. Recent international evidence from PISA has shown that reading fluency remains a significant predictor of reading comprehension across languages and education systems, and that this relationship is particularly strong in multilingual contexts (OECD, 2023). The aim of this study is to examine whether a coherent assessment framework can capture key dimensions of reading development from the early stages of decoding and fluency to later reading comprehension in primary education, and to explore what such a framework reveals about students’ reading development and differential educational outcomes. More specifically, the study addresses two complementary questions: (a) how the components of reading fluency (accuracy, speed and prosody) are organised and related to early reading comprehension in Grade 2, and (b) how reading comprehension is structured and can be interpreted in Grade 5, when reading increasingly functions as a tool for learning. In both cases, the study also examines how reading performance is associated with contextual and student-level variables such as socioeconomic background, home language, language of schooling, gender and age. From a broader perspective, the paper argues that a developmentally grounded approach to reading assessment provides a transferable model for other European education systems facing similar challenges in monitoring reading development across languages and addressing differential educational outcomes in linguistically diverse contexts. Methodology, Methods, Research Instruments or Sources Used The proposed framework is based on two complementary large-scale studies assessments conducted in a bilingual education system. The Grade 2 assessment has been in operational use for five academic years (2021/22-2025/26) and is offered to schools as part of a voluntary formative assessment programme. The analyses reported here are based on data from 19,214 students from 343 primary schools. The assessment is administered individually by trained teachers and is based on an oral reading task using a narrative text. Teachers record decoding errors and prosodic features using a standardized observation protocol. From this procedure, indicators of decoding accuracy, reading speed and prosody are obtained, together with a short measure of early reading comprehension. The Grade 5 instrument was developed in the previous academic year (2024-2025) and implemented for the first time in the current school year, with data from a sample of 1,052 students. It is administered under standardized classroom conditions and consists of three texts (narrative, transactional and expositive) and sets of questions designed to assess reading comprehension at three levels (literal, inferential and global), complemented by indicators of reading fluency (accuracy, speed and prosody). In both studies, school participation was voluntary. The analyses were designed to address both developmental and educational questions. Descriptive and comparative analyses were used to examine performance patterns and group differences associated with socioeconomic, linguistic and demographic variables. In addition, latent-variable models, including structural equation models and item response theory models were used to estimate the internal structure of the instruments, latent reading comprehension indexes, and evaluate the relationships between fluency and comprehension. The hierarchical structure of the data (students nested within schools) was taken into account in the analyses. Conclusions, Expected Outcomes or Findings The results of this study show that a developmentally grounded assessment framework can capture both the structural organisation of reading skills and their changing educational relevance across primary education. The Grade 2 and Grade 5 assessments together provide a coherent picture of the transition from learning to read to reading to learn. In Grade 2, reading performance is best explained through foundational components. Decoding accuracy, reading speed and prosody form a coherent fluency construct that strongly predicts early reading comprehension, confirming the role of fluency as a bridge between decoding and understanding. At this stage, systematic differences related to gender and relative age, as well as variability between schools, indicate that reading development is still strongly influenced by maturity and instructional context. By Grade 5, reading comprehension functions primarily as a general educational competence. Although the assessment distinguishes between literal, inferential and global comprehension, these processes are highly correlated and largely reflect a single underlying factor. Differences in performance are mainly associated with socioeconomic background, home language, language of schooling and language proficiency, showing that reading comprehension is increasingly shaped by broader linguistic and social factors. While decoding accuracy and especially reading speed remain related to comprehension, prosody no longer contributes independently. Overall, the findings support a differentiated assessment approach, with a focus on fluency in the early grades and a more global interpretation of comprehension in later primary education, providing a transferable model for linguistically diverse education system. References Arco-Tirado, J. L., Fernández-Martín, F. D., Hervás-Torres, M., Jiménez-Fernández, G., Calet, N., Defior, S., Neitzel, A. J., & Slavin, R. E. (2024). A best-evidence synthesis and meta-analysis on effective reading programs in Spanish. Review of Educational Research. https://doi.org/10.3102/00346543241297668 Cabell, S. Q., & Hwang, H. J. (2020). Building content knowledge to boost comprehension in the primary grades. Reading Research Quarterly, 55(S1), S99–S107. https://doi.org/10.1002/rrq.338 Cenoz, J. (2009). Towards multilingual education: Basque educational research from an international perspective. Multilingual Matters. Chall, J. S. (1983). Stages of reading development. McGraw-Hill. Duke, N. K., & Cartwright, K. B. (2021). The science of reading progresses: Communicating advances beyond the simple view of reading. Reading Research Quarterly, 56(S1), S25–S44. https://doi.org/10.1002/rrq.411 Kuhn, M. R., Schwanenflugel, P. J., & Meisinger, E. B. (2010). Aligning theory and assessment of reading fluency: Automaticity, prosody, and definitions of fluency. Reading Research Quarterly, 45(2), 230–251. https://doi.org/10.1598/RRQ.45.2.4 Messick, S. (1989). Validity. In R. L. Linn (Ed.), Educational measurement (3rd ed., pp. 13–103). Macmillan. OECD. (2023). What can we learn from PISA’s reading fluency test? (PISA in Focus, No. 121). OECD Publishing. Pearson, P. D., & Hamm, D. N. (2005). The assessment of reading comprehension: A review of practices—Past, present, and future. In S. G. Paris & S. A. Stahl (Eds.), Children’s reading comprehension and assessment (pp. 13–69). Lawrence Erlbaum Associates. Scarborough, H. S. (2001). Connecting early language and literacy to later reading (disabilities): Evidence, theory, and practice. In S. B. Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research (Vol. 1, pp. 97–110). Guilford Press. Snow, C. E., & Van Hemel, S. B. (Eds.). (2008). Measuring reading comprehension: Science and applications. National Academies Press. 31. LEd – Network on Language and Education
Paper Does LUFO Work? Findings from a Study on Formative Assessment Tool for Improving Disciplinary Literacy Among L2 Learners University of Helsinki, Finland Presenting Author:This paper presents the impact of LUFO model on learning outcomes. LUFO (abbreviation from the Finnish title: LUkutaidon FOrmatiivinen arviointi) is a new model for formative assessment of disciplinary literacy. LUFO combines the principles of formative assessment feedback (Black & Wiliam, 1998; Hattie & Timperley, 2007) theoretical frame of reference of multiliteracy (Cope & Kalantzis, 2000, 2009, 2015; Kress, 2003) and disciplinary literacy (Fang & Coatoam, 2013; Moje, 2015; Shanahan & Shanahan, 2012). LUFO has several interrelated goals. First, it aims to provide students with structured and targeted feedback on subject-specific literacy skills. Second, LUFO offers teachers a concrete and accessible method to enhance linguistically responsive pedagogy (Lucas & Villegas, 2013). The development of LUFO is described in detail in previous publications by the author of the present study. LUFO study project is conducted by the author and consists of three phases. The first phase explored Finnish teachers’ perceptions of factors influencing assessment and learning. The findings indicated that L2 teachers do not apply formative assessment guidelines as intended, thus missing the possible positive effects on learning. The results have been published in a Finnish peer-reviewed journal (Saari & Hildén, 2023). In the second phase, LUFO formative assessment model was developed using a cyclic design-based research approach, combining findings from the first sub-study with relevant research literature. The manuscript describing this developmental process in detail has been provisionally accepted for publication in peer-reviewed international journal (publication expected in 2026). This paper describes the third and final phase of the LUFO study project, which examines the potential impact of LUFO on learning outcomes. The specific research questions are: 1. Does LUFO has effect on quantitative learning outcomes? 2. Are there differences in the effects of LUFO between L1 and L2 learners, or between genders? 3. How do learners experience studying with LUFO? In Finland, L2 learner’ literacy development is weaker compared to other students (Leino et al., 2023), and current extra support provided has not been able to narrow the learning outcome gap (Ståhlberg et al., 2023). Adequate literacy is crucial for academic progression, employment, and overall societal participation. Ensuring the success of multilingual students’ educational pathways has significant societal implications. However, the assessment of L2 learners has been sparsely studied, and existing research mainly addresses summative assessment at the end of basic education. Formative assessment has been shown to be an effective tool for enhancing learning outcomes (Black & Wiliam, 1998; Hattie & Timperley, 2007). Yet, there is limited research on its specific effects on L2 learners’ literacy development. LUFO study aims to fill this gap in knowledge. The three LUFO manuscripts will constitute the author’s doctoral dissertation. In this paper, the term L2 learner refers to a multilingual student learning Finnish as a second language syllabus at school, in the initial stages of his or her language studies. While aimed for multilingual students, LUFO principles and guidelines are likely to be applicable also for other subgroups of learners. Methodology, Methods, Research Instruments or Sources Used Data collection will take place in spring 2026. The target sample size is N=380. The participants are 5th grade students studying in 16 study groups in four schools located in Helsinki, Finland. These 16 groups are randomized to eight proband groups (N=190) and eight control groups (N=190). Around 60% of these students are L1 learners and 40% L2 learners. The study timeline consists of two courses of environmental studies. First, both groups will study Course A with standard instruction, after which all participants will take Exam A. Second, the teachers of the proband groups are trained in the LUFO model. Next, both groups will study Course B and take Exam B. After Exam B, the control group teachers will also receive training in the LUFO model. The outcome variables include quantitative and qualitative measure. The main outcome, Exam B grades of probands and control groups, are compared with Mann-Whitney U test. Participants will be grouped according to L1/L2 status and gender (girls or boys), allowing comparisons both between language groups and within gender-specific subgroups Exam A results will be used for assessing individual-level changes and for controlling possible group differences in baseline level. In addition, a sub-sample of L2 learners (N=5) from proband group will be interviewed. These participants will be chosen by random sampling. The interviews will be transcribed and analysed for revealing students’ personal experiences on the pros and cons of LUFO. The results will be published in a manuscript, which will be submitted to an international peer-reviewed journal later in 2026. Conclusions, Expected Outcomes or Findings results of this paper will assess LUFO model’s effect on learning outcomes and provide insight of the learners’ perspectives. Currently (January 2026), there are no results yet because data gathering is ongoing. These results will be available in April 2026, after which the analysis phase will take place. Thus, it is highly likely that at the time of ECER 2026, the analysis phase has been completed, and the author will be able to present the final results. If there will be some unexpected delays, preliminary results will be used in this presentation. A notable strength of the LUFO, compared with similar studies, is the sample size estimated to be sufficient to detect the possible effects. The sample size was determined based on an a priori power analysis conducted at the design stage. LUFO contributes on assessment and literacy research. It is especially aimed to strengthen L2 learners’ disciplinary literacy skills. In addition to theoretical results and viewpoints, the findings from LUFO provide applicable every-day practices for teaching. LUFO results also add to the research of assessment methods aimed for L1 learners. Although LUFO project is not yet completed, some important findings can already be presented. Previous LUFO studies showed that formative assessment guidelines are not applied as intended, which results to missing the potentially advantageous outcomes. However, teachers might be motivated, if provided with an effective, low threshold tool. LUFO study aims to provide reliable, research-based insights for improving educational practices. It holds significant societal potential both in Finland and internationally. References Black, P., & Wiliam, D. (1998). Assessment and classroom learning. Assessment in Education, 5(1), 7–74. https://doi.org/http://dx.doi.org/10.1080/0969595980050102 Cope, B., & Kalantzis, M. (2000). Multiliteracies : literacy learning and the design of social futures / edited by Bill Cope and Mary Kalantzis. Routledge. Cope, B., & Kalantzis, M. (2009). “Multiliteracies”: New Literacies, New Learning. Pedagogies: An International Journal, 4(3), 164-195. https://doi.org/10.1080/15544800903076044 Cope, B., & Kalantzis, M. (2015). A pedagogy of multiliteracies : learning by design (1st 2015. ed.). Palgrave Macmillan UK. https://doi.org/10.1057/9781137539724 Fang, Z., & Coatoam, S. (2013). Disciplinary Literacy : What You Want to Know About It. Journal of Adolescent & Adult Literacy, 56(8), 627-632. https://doi.org/https://doi.org/10.1002/JAAL.190 Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112. https://doi.org/10.3102/003465430298487 Kress, G. (2003). Literacy in the New Media Age. Taylor and Francis. https://doi.org/10.4324/9780203299234 Leino, K., Sirén, M., Nissinen, K., & Puhakka, E. (2023). Puoli tuntia lukemista: Kansainvälinen lasten lukutaitotutkimus (PIRLS 2021). Koulutuksen tutkimuslaitos: Tutkimuksia, 1-91. https://doi.org/10.17011/ktl-t/37 Lucas, T., & Villegas, A. (2013). Preparing Linguistically Responsive Teachers: Laying the Foundation in Preservice Teacher Education [Article]. Theory into practice, 52(2), 98-109. https://doi.org/10.1080/00405841.2013.770327 Moje, E. B. (2015). Doing and Teaching Disciplinary Literacy with Adolescent Learners: A Social and Cultural Enterprise. Harvard Educational Review, 85(2), 254-278,301. Saari, E., & Hildén, R. (2023). S2-opettajien käsityksiä formatiivisesta arvioinnista oppimisen tukena. In T. Mäkipää, R. Hildén, & A. Huhta (Eds.), Kielenoppimista tukeva arviointi. AFinLA-teema. Nro 15 (pp. 142–161). Suomen soveltavan kielitieteen yhdistys AFinLA. https://doi.org/https://doi.org/10.30660/afinla.121249 Shanahan, T., & Shanahan, C. (2012). What Is Disciplinary Literacy and Why Does It Matter? Top Lang Disorders Vol. 32, No. 1, 7–18. (Wolters Kluwer Health | Lippincott Williams & Wilkins) Ståhlberg, L., Lotta, U., & Hotulainen, R. (2023). Lukutaidon yhteys suomi toisena kielenä ja kirjallisuus (S2) -oppimäärän valitsemiseen toisella ja seitsemännellä luokalla. NMI-bulletin, 2023(1), 54–72, 54–72. 31. LEd – Network on Language and Education
Paper Language Curriculum Tracks and Educational Inequality: Matriculation Exam Grade Outcomes for Children of Immigrants in Finland 1: INVEST Research Flagship Centre, University of Turku, Finland; 2: University of Eastern Finland Presenting Author:Topic: Language Curriculum Tracks and Educational Inequality: Matriculation Exam Grade Outcomes for Children of Immigrants in Finland Background and Motivation The growing presence of children of immigrants in schools who speak the language of instruction as a second language is a global phenomenon that has expanded considerably in Europe in more recent decades compared to traditional immigrant-receiving countries such as the United States, Canada, and the United Kingdom (Crul & Mollenkopf, 2012; OECD, 2018; UNHCR, 2019). Education systems across these contexts have responded with diverse strategies to support these learners, yet the evidence on the effectiveness of such approaches remains limited. Finland offers a particularly structured model in which a second-language (L2) curriculum in Finnish or Swedish runs parallel to the mainstream first-language (L1) curriculum, or what is referred to as the mother tongue subject. Students are generally expected to transition from the L2 to the L1 track as their acdemic language proficiency develops (European Commission, 2019). While this arrangement was designed to facilitate inclusion, it has drawn criticism on several fronts. Research indicates that L2 curricula tend to deliver simplified content, which in turn limits students’ exposure to academically demanding material (Siarova & Essomba, 2014; Thomas & Collier, 2002; Valdés, 2004). Segregated instruction has also been linked to restricted interaction with native peers and hindered social integration (Fandrem et al., 2024; Öztürk et al., 2024). Furthermore, L2 learners are often subject to social, racial, and political biases that are closely tied to language (Flores & Rosa, 2019; Flores et al., 2018). Taken together, these structural challenges are reflected in persistent achievement gaps: children of immigrants across OECD countries, Finland included, continue to underperform academically relative to their native-born peers (OECD, 2023). Despite the body of literature that has documented these achievement gaps and related concerns, systematic comparative evidence on educational outcomes between second-language and mainstream curricula remains scarce. This gap in the research calls for closer examination of how language curriculum tracks shape the academic outcomes of children of immigrants, particularly in well-structured systems like Finland’s. Research Focus and Questions This study examines whether Finland's L2 curriculum contributes to educational inequality among children of immigrants by shaping their academic outcomes. The Finnish matriculation examination, a nationally standardized exam administered at the end of general upper secondary education for students on the academic track, offers a particularly suitable context for this inquiry. As a key certification and a gateway to higher education, it captures a critical point at which curriculum track differences may have already shaped students' academic trajectories. Therefore, the study seeks to answer following research questions: RQ1. Whether taking the compulsory Finnish/Swedish exam as part of the L1 or L2 curriculum is associated with grades in other subjects in the matriculation exams. RQ2. Does the curriculum type (L1 vs. L2) moderate the relationship between Finnish/Swedish grades and grades in other subjects? In doing so, RQ2 seeks to disentangle whether differences in academic outcomes are driven by language proficiency itself or by the structural nature of the curriculum track. Methodology, Methods, Research Instruments or Sources Used Data and Methods Data This study draws on high-quality national population register data from Finland. Our analytical sample consists of 21,924 individuals aged 22 and below, covering matriculation exams taken between 1996 and 2024. The sample is restricted to students whose both parents are of migrant background, allowing us to focus on the population most likely to be affected by language curriculum track placements. Methods To address RQ1, we employ ordinary least squares (OLS) regression to estimate the association between curriculum type and grades in other subjects in the matriculation examination. Average grades in other subjects, measured on a continuous scale from 0 to 6, serve as the dependent variable, while the type of Finnish/Swedish exam taken, coded as 0 for L1 and 1 for L2, is the key independent variable. For RQ2, we extend the model by introducing an interaction term between Finnish/Swedish grades and curriculum type. Here, the dependent variable remains average grades in other subjects (0–6 scale), Finnish/Swedish grades (also on a 0–6 scale) serve as the key independent variable, and curriculum type (L1 vs. L2) is entered as a moderator. This specification allows us to assess whether the association between Finnish/Swedish performance and performance in other subjects differs depending on which curriculum track a student followed, thereby helping to disentangle the effect of language proficiency from the structural influence of the curriculum track itself. Across both analyses, we control for parental social class, measured using the EGP classification (Erikson, Goldthorpe & Portocarero, 1979), parental national origin, sex, and length of residence in Finland. Conclusions, Expected Outcomes or Findings Findings and Conclusion Findings The findings reveal that children of immigrants who take the Finnish/Swedish exam under the L2 curriculum achieve significantly lower grades in other subjects compared to those in the L1 track. The result holds up even when accounting for parental social status, parental national origin, sex, and length of residence in Finland, which rules out the possibility that family background and other factors are driving the difference. This aligns with existing research suggesting that L2 curricula deliver simplified content, which may in turn limit the development of academic language and literacy skills that are essential for performance across subjects. Perhaps more notably, even when L1 and L2 students achieve identical Finnish/Swedish grades, L1 track students substantially outperform their L2 counterparts in other subjects. This indicates that the disadvantage is not solely a matter of language proficiency. Rather, the simplified nature of L2 curricula appears to constrain students’ broader academic development, likely by reducing their exposure to complex texts and the kind of analytical reasoning that transfers across domains. Conclusions These findings indicate that being placed in the L2 curriculum creates growing disadvantages for children of immigrants over time. Instead of helping students catch up, L2 placement appears to limit their academic skills in ways that remain even when their language proficiency is on par with L1 students. This calls into question whether simplified curricula actually prepare students well enough for academic success, or whether keeping the content challenging while offering more language support would be a better way to support both language learning and educational fairness. References References Crul, M., & Mollenkopf, J. (Eds.). (2012). The changing face of world cities: Young adult children of immigrants in Europe and the United States. Russell Sage Foundation. Erikson, R., Goldthorpe, J. H., & Portocarero, L. (1979). Intergenerational class mobility in three Western European societies: England, France and Sweden. The British journal of sociology, 30(4), 415-441. European Commission: European Education and Culture Executive Agency, Noorani, S., Baïdak, N., Krémó, A., & Riiheläinen, J. (2019). Integrating students from migrant backgrounds into schools in Europe : national policies and measures, Publications Office of the European Union. https://data.europa.eu/doi/10.2797/9799 Fandrem, H., Jahnsen, H., Nergaard, S. E., & Tveitereid, K. (2024). Inclusion of immigrant students in schools: The role of introductory classes and other segregated efforts. International Journal of Inclusive Education, 28(4), 440-455. https://doi.org/10.1080/13603116.2021.1950222 Flores, N., & Rosa, J. (2019). Bringing race into second language acquisition. The Modern Language Journal, 103, 145-151. https://doi.org/10.1111/modl.12523 Flores, N., Lewis, M. C., & Phuong, J. (2018). Raciolinguistic chronotopes and the education of Latinx students: Resistance and anxiety in a bilingual school. Language & Communication, 62, 15-25. https://doi.org/10.1016/j.langcom.2018.06.002 OECD (2023), PISA 2022 Results (Volume I): The State of Learning and Equity in Education, PISA, OECD Publishing, Paris, https://doi.org/10.1787/53f23881-en. OECD. (2018). The resilience of students with an immigrant background: Factors that shape well-being. OECD Publishing. https://doi.org/10.1787/9789264292093-en Öztürk, N., Hassani, S., & Schwab, S. (2024). Insights into students’ experiences of social inclusion and exclusion during attending language pull-out classes. International Journal of Inclusive Education, 1-14. https://doi.org/10.1080/13603116.2024.2367537 Siarova, H., & Essomba, M. À. (2014). Language support for youth with a migrant background. Policies that effectively promote inclusion. Sirius Network Policy Brief Series, (4). Thomas, W., & Collier, V. (2002). A national study of school effectiveness for language minority students' long-term academic achievement. UNHCR. (2019). Access to education for refugee and migrant children in Europe. https://www.unhcr.org/neu/wp-content/uploads/sites/15/2019/09/Access-to-education-europe-19.pdf Valdés, G. (2004). Learning and not learning English: Latino students in American schools. Teachers College Press. | ||