Hide metadata

dc.date.accessioned2013-03-12T13:09:36Z
dc.date.issued2010en_US
dc.date.submitted2010-06-15en_US
dc.identifier.citationRoos, Saskia Maren. "I could answer all your questions in Polish... but in Norwegian?". Masteroppgave, University of Oslo, 2010en_US
dc.identifier.urihttp://hdl.handle.net/10852/31347
dc.description.abstractPurpose: The purpose of this study is to examine the power of teacher-based and test-based assessment methods of second-language skills among minority-language students. The research problem of the present study is as follows: In which way do teacher-based and test-based language assessment methods mirror the variation in second-language skills of minority-language students? The research problem has been broken into four research questions. The research questions guided this study. The first questions asked what distinguishes second-language learners from first-language learners with regards to their Norwegian language proficiencies. The second question asked how the variation in Norwegian language skills among second-language learners can be described. The third question asked how to describe the relations between teacher’ questionnaire and standardized psychometric test results. It has further been investigated how the two methods correlate in their results. The forth question asked how the validity of teacher-based and test-based assessment methods of second-language skills regarding minority-language students can be described. Methods and Procedures: Within the framework of a quantitative approach, a non-experimental research has been applied. A sample of minority-language students (N = 111) and majority-language students (N = 287) have been assessed with the teacher-based questionnaire “20 Questions about Language Performance” (Ottem, 2009) and the standardized language test “British Picture Vocabulary Scale II” (Dunn, Dunn, Whetton, & Burley, 1997). The samples have been divided into those individuals who either showed high-performance majority-language skills or low-performance majority-language skills based on the obtained results of the teacher-based questionnaire. The majority language of this study is Norwegian. The differences between the samples according to their language proficiencies and differences between the high- and low-performance groups have been examined with analytic statistical methods. The study revealed statistical significant relations between the two different assessment methods. The examination of the power for the two assessment tools for the sample of minority-language students has been based on these significant relations. Both instruments have been shown to be valid. The additionally applied statistical two-step cluster analysis revealed different groupings. One group combined students for which teacher-based and test-based data showed low Norwegian language. The other group combined the students for which data of both assessment methods showed high Norwegian language skills. Both methods agreed in their results, and therefore were correlated. The examined relations confirmed the findings of analytic statistics. Results: The findings of the analytic statistics are discussed in the context of a theoretical framework. The theoretical framework includes a description of the assessed language skills, issues about the validity of assessment tools and a review of previous research studies regarding test-based versus teacher-based assessment. The study has shown that there are statistically significant differences regarding language skills between the samples of second- and first-language learners. It further has been revealed that minority-language students show an equal variation in their language abilities than their majority-language peers. High- and low-performance students have been identified in both samples. The results suggested that both teacher-based and test-based methods have power in the majority-language assessment of second-language learners. However, teacher-based tools argued for a more ecologically valid assessment procedure. Implications: The study´s samples are not randomly selected. Therefore, the results cannot be generalized to the entire population of minority-language students in Norway. This affects the study´s inferential statistics and the statistical power analysis. Hence, internal validity is limited.eng
dc.language.isoengen_US
dc.title"I could answer all your questions in Polish... but in Norwegian?" : an empirical, quantitiave study about test-based versus teacher-based assessment of language skills among minority-language studentsen_US
dc.typeMaster thesisen_US
dc.date.updated2010-10-06en_US
dc.creator.authorRoos, Saskia Marenen_US
dc.date.embargoenddate10000-01-01
dc.rights.termsDette dokumentet er ikke elektronisk tilgjengelig etter ønske fra forfatter. Tilgangskode/Access code Aen_US
dc.rights.termsforeveren_US
dc.subject.nsiVDP::282en_US
dc.identifier.bibliographiccitationinfo:ofi/fmt:kev:mtx:ctx&ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&rft.au=Roos, Saskia Maren&rft.title="I could answer all your questions in Polish... but in Norwegian?"&rft.inst=University of Oslo&rft.date=2010&rft.degree=Masteroppgaveen_US
dc.identifier.urnURN:NBN:no-25403en_US
dc.type.documentMasteroppgaveen_US
dc.identifier.duo103565en_US
dc.contributor.supervisorJørgen Frost og Ernst Ottemen_US
dc.identifier.bibsys102180881en_US
dc.rights.accessrightsclosedaccessen_US
dc.identifier.fulltextFulltext https://www.duo.uio.no/bitstream/handle/10852/31347/2/MasterThesis_SaskiaRoos.pdf


Files in this item

Appears in the following Collection

Hide metadata