Abstract
This thesis highlights first and foremost the language teaching controversies in Finland and Iceland,
applying a critical eye on the outcome of the mandatory language teaching of Swedish in Finland and
Danish in Iceland. But its aim is also secondly to describe, present, and test fundamental concepts and
criteria which are thought to underpin good language testing. Discussing the importance of the
interface and different perspectives between Language Testing (LT) and Second Language
Acquisition (SLA) research, the thesis interconnects these two main research questions in order to
explore learners' test-taking scoring specifically and in general assess the validity and reliability of the
language test exclusively carried out. It is well needed research, since excellent testing of how much
the students know and do not yet know helps the learning process immensely. Poor testing – on the
other hand - may however result in negative backwash for the test taker.
The main purpose of this dissertation is to test the Scandinavian language comprehension in Finland
and Iceland on Finnish and Icelandic youths (17-19 years of age) having nearly completed their
language education in respectively Swedish or Danish. After attending this mandatory language
education for about six years, the Finnish learners understood 32% of the formal language in the news
report and 36% of colloquial dialogue. The Icelanders understood 41% of the news report and 25% of
the dialogue. This mean that average outcome of the test in both countries is in general deficient after
undertaking the mandatory language education of respectively Swedish or Danish. However there are
huge differences within Finland concerning test scoring to take into consideration. In Vaasa, the mean
average was 52% and in Mikkeli it was down to 24%. This should be compared with Iceland's national
scoring average of 25%. Linguistic background of test takers or language learners does therefore not
seem to have a large impact. Rather, it seems that regionalism, accessibility to native speakers of the
target language, and also the sense of purposefulness in the education seem to top linguistic
background. The results indicate also that it is not any direct difference between understanding
dialogues containing slang, normal colloquial tempo with an irregular or informal syntactical structure
and understanding a formal speech containing more advanced terminology but has a slower speed of
speech thus being more structured.
When it comes to testing and teaching languages, technological equipment is important for the
outcome. This became evident on the dialogue portion for Iceland. The equipment used recording the
Swedish test for the Finnish-speaking test takers turned out with better quality than the Danish test for
the Icelandic-speaking test takers (for reasons presented in the study). It is, however, not likely at all
that it would make more than 10-20% difference in average understanding, probably leaving Iceland
with an average mean on five instead of three out of 12 possible points. With this in mind, Iceland and
Finland have approximately the same (about 30-35%) average understanding of their mandatory
Scandinavian language which they are to learn.
This thesis suggests that it is essential to examine not only final linguistic products of tests but also
that language tests be more communicatively relevant for the second language learning-process. The
future of SLA requires proper student evaluation in order to illuminate the issues hindering second
language advancement.