Testing of Norwegian as a second language for adult immigrants has reached a high level of standardisation and professionalisation in Norway. There are standardised and officially recognised tests developed according to assessment theory and international assessment practice at several levels: Norskprøven for voksne innvandrere (Test of Norwegian for adult immigrants), developed by Skills Norway (Vox, Norwegian Agency for Lifelong Learning’s new name from January 1st 2017) measures at CEFR-levels A1, A2, B1 and B2 and Test i norsk—høyere nivå (Test in Norwegian—advanced level) developed by Norsk språktest (Folkeuniversitetet/ University of Bergen) measures at levels B2 and C1. The theoretical construct of both tests is communicative competence, and they measure the four language skills: reading, listening, oral communication, and written production. The focus of this paper will be on Norskprøven.
Norskprøven is a partly digital test; listening, reading and writing are measured on computer, whereas listening and reading are multistage tests that are scored automatically. Writing is done on a computer, but scored by human raters. The oral test is a traditional face-to-face paired examination. Norskprøven measures the four skills in separate tests so that candidates may take the tests at different levels in line with their profile of proficiency. This is particularly important for low-educated learners with low levels of literacy (Carlsen, 2016).
Around 20 000 candidates take Norskprøven every year. The candidate population is heterogeneous, consisting of low-educated learners with limited literacy, as well as highly educated learners in need of an advanced level of Norwegian to enter higher education or start to work as doctors, dentists or similar. The teaching of Norwegian is divided into three tracks depending on the learners’ educational background. Refugees, asylum seekers and reunited families are obliged to follow 550 hours of Norwegian tuition and 50 hours of ’knowledge of society’, but those who need it may get as many as 3000 hours free of charge. Norskprøven is administered twice a year at around 300 different test centres around the country. Writing is marked centrally in accordance with standardised assessment criteria and assessment routines. The oral test, however, is scored locally at the different test centres. Skills Norway has developed an online training system for oral examiners and raters.
Norskprøven is a compulsory achievement test for candidates who get free language tuition (refugees, asylum seekers and reunited families). It has become more and more common that employers require a certain level on the test for employment purposes, and it is a requirement for university admission (B2 in all four skills). As a reaction to the increase in immigration, the conservative/right-wing government introduced several restrictions with the purpose of “making it less attractive to come to Norway”. Among them, it was decided that from January 1st 2017, applicants for Norwegian citizsenship need to pass a language test (Norskprøven, oral test at A2-level). Similarly, in the near future, it will be a requirement to pass the oral test with an A1-level in order to get permanent residency and family reunification.
Norskprøven is developed according to international standards of test construction. The test construct is communicative competence as described in Bachman (1990) and Bachman and Palmer (1996). It also builds on the CEFR descriptors (Council of Europe, 2001) and the Curriculum of Norwegian for adult immigrants (Vox, 2012). We work hard to ensure the validity of the test, constructing items carefully for all skills. All items for the reading and listening tests are piloted on several thousand candidates, in incomplete linked designs, to allow for statistical analyses and construction of test measuring at levels A1/A2, A2/B1 and B1/B2. At each level (e.g. A1/A2), parallel tests are constructed. To determine cut-scores between CEFR-levels, standard setting procedures have been applied. (Moe & Verhelst, Forthcoming). Since the test is digital, important error sources in marking and punching of results are eliminated as marking is automated and scores stored immediately. To ensure the reliability and validity of the performance tests of written and spoken production, Skills Norway develop rating grids with rating criteria and level descriptors that all raters should follow carefully as they rate performances. Writing test performances are rated centrally by 80 experienced raters. Skills Norway meet with the raters annually and monitor them closely during and after rating. The oral test performances are assessed locally by teachers of adult immigrants. Skills Norway have developed an online training resource with mini-lectures as well as benchmark candidate videos, accompanied by thorough verbal explanations to the scores given. It has always been of great importance to the test developers that the test has a positive washback effect on teaching and learning, i.e. that it stimulates activities that promote language learning. This is one of the reasons why it is important that the tests measure a broad construct and that all the different skills are measured.
The main challenge with respect to the tests of Norwegian for adult immigrants lies, as we see it, in the way the tests are used; in 2016 more than 20,000 candidates took Norskprøven, yet test certificates open few doors in practice. It would have been an advantage if the job market recognised these tests as proof of the language proficiency to a greater degree, instead of assessing their skills themselves. Assessing a person’s language competences is a professional skill, and not all employers are equally capable of making valid judgments.
Another problem in the area of assessment of adult learners is the lack of standardisation of tests for university entrance. A study from 2006 revealed that the distinct locally developed exams were quite different, and the rating scale of higher education (European Credit Transfer and Accumulation System, ECTS, grading scale, A–F) is used quite differently across education institutions (Andersen, 2006). The correlation between the tests developed locally and the standardised test at advanced level developed at Norsk språktest was found to be poor. A later study investigated the language requirements for university entrance, which was decided by the Norwegian Association of Higher Education Institutions (Universitets-og høgskolerådet, UHR) for the AL-test and for the Trinn 3-examen. The results of this study show that the pass score for the AL-test is stricter than that for the Trinn 3-examen (Carlsen 2008, Forthcoming).
The authors of this paper have earlier expressed their concern about a policy where democratic rights are linked to skills and knowledge, as this may undermine the very idea of a democratic society and equal rights for all citizens (Carlsen & Moe, 2013). It is even more detrimental to link human rights, like the right of refugees to have a safe shelter and to be reunited with their families, to performance on a language test. We are concerned to witness, both in Norway, Europe and abroad, an integration policy, which to an increasing degree is based on restrictions and demands, which we fear may lead to segregation and alienation rather than to integration.
Cecilie Hamnes Carlsen works at Skills Norway, coordinator of Norskprøven for voksne innvandrere.
Eli Moe works at Skills Norway with Norskprøven for voksne innvandrere, and at the University of Bergen with the national tests in English for young learners.
References
Andersen, R. (2006). Korrelasjonsundersøkelsen 2006. Sammenlikning av opptakskravene til høyere utdanning som gjelder for Trinn 3-eksamen i norsk og Test i norsk – høyere nivå. Unpublished report. Norsk språktest.
Bachman, L. F. (1990). Fundamental Considerations in Language Testing. Oxford: Oxford University Press.
Bachman, L. F., & Palmer, A. S. (1996). Language Testing in Practice: Designing and Developing Useful Language Tests. Oxford: Oxford University Press.
Carlsen, C. H. (2016). Giving LESLLA learners a fair chance in testing. Presentation at LESLLA Symposium. Granada, September, 8th– 10th 2016. Retrieved from http://wdb.ugr.es/~leslla2016/wp-content/uploads/Carlsen_LESLLA_2017.pdf
Carlsen, C. (2008). Oppfølging av utenlandske studenter. Retrieved (February 4, 2013) from http://www.folkeuniversitetet.info/avd_filer/ls/spraaktest/Fase_2_Korr.und.pdf
Carlsen, C. H. (Forthcoming). The adequacy of the B2-level as university entrance requirement. Language Assessment Quarterly, Special issue on the CEFR for university entrance purposes.
Carlsen, C., & Moe, E. (2013). Assessing Norwegian. In A. Kunnan (Ed.), Companion to Language Assessment Vol. IV, 2029-2037. Wiley-Blackwell.
Council of Europe. (2001). Common European Framework of reference for languages: Learning, teaching, assessment. Cambridge, England: Cambridge University Press.
Moe, E., & Verhelst, N. (Forthcoming). Setting standards for multistage tests of Norwegian for adult immigrants. In S. Blöemeke and J. E. Gustafson (Eds.), Standard Setting – International State of Research and Practices in the Nordic Countries. In the series Methodology of Educational Measurement and Assessment. Springer.
Vox. (2012). Læreplan i norsk og samfunnskunnskap for voksne innvandrere [The National curriculum in Norwegian and Social Studies for Adult Immigrants]. Retrieved from http://www.vox.no/contentassets/f6594d5dde814b7bb5e9d2f4564ac134/laereplan_norsk_samfunnskunnskap_bm_web.pdf
