Oral exams are high-stake final tests that hold longstanding traditions in school systems around the world, including in the Scandinavian countries. However, they have attracted very little empiric research, and knowledge about how oral exams are realised within the Norwegian context are scarce. This paper examines how disciplinary oral competence exams (DOCEs) are carried out in four Norwegian secondary schools. Guiding the analysis is the following research question: What are the phases of disciplinary oral competence exams in Norwegian secondary schools, and what characterises the different phases?

In Norway, all students complete one DOCE at the end of lower secondary school and 2-3 DOCEs at the end of upper secondary school, depending on the study program they attend (The Norwegian Ministry of Education and Research, 2019a). The Directorate of Education and Training (DET) regulates the procedures for DOCE and there are national guidelines for schools’ application of these exams. The subject a student is examined in is determined through a process of random selection (a draw) and students receive notice about which subject 48 hours before the DOCE takes place (Forskrift til opplæringslova, 2006, § 3-29). According to national guidelines, the DOCE may last up to 30 minutes per student (Forskrift til opplæringslova, 2006, § 3-18) and two examiners partake in the oral exam (The Norwegian Ministry of Education and Research, 2014). The DOCE includes a student presentation and a subject conversation between student and examiners. The presentation should not make up more than 1/3 of the time spent on the exam (The Norwegian Ministry of Education and Research, 2014). During the subject conversation the examiner should facilitate the conversation in a manner that “ provides students theopportunity to demonstrate competence in as large a part of the subject as possible during the oral exam” (my translation) (The Norwegian Ministry of Education and Research, 2019b).

Local municipalities are responsible for producing, implementing and grading the DOCE (the municipality for lower secondary and the county authority for upper secondary). The local authorities typically delegate these responsibilities to schools or teachers. Research examining Norwegian policy documents regulating DOCEs have underlined the lack of  uniform procedure for carrying out the DOCE (Isager, 2020). As uniform procedures during DOCEs are essential for allowing comparison between the results of all students, this makes the Norwegian context particularly interesting. If there is variation in content areas addressed, difficulty of questions or differences in the help examiner provide, this may imply that different students do not have the same opportunity to demonstrate achievement. This may have implications for the reliability, validity and fairness of DOCEs. The reliability of an exam can be explained as the extent to which, if repeated, the second exam result would match the first. This implies an accurate exam grade that is not influenced by the particular occasion or who does the grading. Validity theory, on the other hand, is concerned with how well what is actually assessed, corresponds with what is intended tested (Harlen, 2005). According to theory, teachers must have common understandings of standards and what kind of performances demonstrate those standards to achieve high assessment reliability and preserve assessment validity (Klenowski & Wyatt-Smith, 2014).

By using thematic analysis, this article maps video recordings of 36 DOCEs. The first part of the analysis involves mapping the different phases of the DOCEs, based on the main activities the participants orient towards during the exam. The second part investigates characteristics within the different phases. Preliminary findings reveal that DOCEs consist of five main phases clearly sequenced chronologically. Findings also reveal that the time spent on the different phases varies more than anticipated.

This study is part of the research project Conversation Analytic innovation for Teacher Education (CAiTE) funded by The Research Council of Norway. This paper draws on more than 21 hours of video recordings of disciplinary oral competence exams (DOCEs) collected by members of the CAiTE-project. The 36 DOCEs that make up the data material took place in two lower and two upper secondary schools during 2018 and 2019. Because thematic analysis (TA) is efficient for finding, analyzing and reporting patterns within data, it is suitable method for this study. Another advantage with TA is that the method is flexible and allows for both giving rich descriptions of an entire data set as well as much more detailed accounts of particular aspects (Braun & Clarke, 2006). As the analysis in this paper consists of two stages, this makes TA particularly useful. The first stage involves a mapping of the video data, studying the ongoing activities the participants are involved. Activity is here is understood as something students and examiners do, this involves solving certain communicative tasks (Linell & Thunqvist, 2003). The second stage involves a detailed analysis of the subject conversation and the student presentation. TA may be carried out in both a deductive and an inductive way. As an inductive approach means a strong link between the data and the themes identified (Braun & Clarke, 2006), this approach was preferred for this paper. The software tool NVivo was used for coding the data. In the first stage of coding, essence-capturing words and short phrases that attributed meaning to portions of video data were developed. The codes were means of patterning and classifying the data (Saldaña, 2011). Process codes were developed using gerunds in the code names and basing the code names and descriptions on the main activities the participants oriented towards in the different phases of the DOCE (Charmaz, 2002; Corbin & Strauss, 2008). During the second stage of coding more detailed codes were developed. These codes were essence-capturing for the different parts of the conversation in the different phases. The goal of this was to identify different topics and themes covered by the conversation. The research project is approved by The Norwegian Centre for Research Data (NSD) and all participation in this study was voluntary. The participants have had the opportunity to withdraw at any time within the project period and they have all signed a letter of consent.

Preliminary findings reveal that oral exams consist of five main phases. These are greeting and preparing, student presenting, subject conversing, grading and examiners giving feedback and announcing grade. All of the phases take place inside a normal classroom (refurnished for the oral exam), and the students partake in all phases but one; the grading phase. The data reveals that the external examiner is present in all phases. On the other hand, there are occurrences of the internal examiner leaving the room during the greeting and preparing phase, when the feedback and grade is given, as well as between phases. In 35 out of 36 cases in the data material, the oral exams played out with the five different phases in the same chronology. The coding revealed only one case of differing chronology with regards to phases. The time spent on the different phases on the other hand, varied substantively. In addition, a significant pattern was revealed that connected to the division of time spent on the student presentation and the time spent on the subject conversation. In three out of four schools, more time was spent on the subject conversation than on the student presentation. In the fourth school however, the length of the student presentation varied more than in the other three schools, and could even take up more of the time than the subject conversation. Preliminary findings form the second stage of analysis reveals differences in topics and themes covered during the subject conversation. Together with the variation in amount of time spent on different phases of the DOCE, this seems to affect both the reliability, the validity and the fairness of the DOCE.

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. doi:10.1191/1478088706qp063oa Charmaz, K. (2002). Qualitative Interviewing and Grounded Theory Analysis. In J. F. Gubrium & J. A. Holstein (Eds.), Handbook of interview research: Context & method (pp. 675-694). Thousand Oaks, CA: Sage. Corbin, J. M., & Strauss, A. L. (2008). Basics of qualitative research : techniques and procedures for developing grounded theory (3rd ed. ed.). Thousand Oaks, Calif: Sage. Forskrift til opplæringslova, FOR-2006-06-23-724 C.F.R. (2006). Harlen, W. (2005). Trusting teachers’ judgement: research evidence of the reliability and validity of teachers’ assessment used for summative purposes. Research Papers in Education, 20(3), 245-270. doi:10.1080/02671520500193744 Isager, J. M. (2020). Konstrukt, generalisering og held ved mundtlige eksaminer i gymnasiet. Acta Didactica Norden, 14(3). doi:10.5617/adno.8204 Klenowski, V., & Wyatt-Smith, C. (2014). Assessment for education : standards, judgement and moderation. Los Angeles, Calif: Sage. Linell, P., & Thunqvist, D. P. (2003). Moving in and out of framings: activity contexts in talks with young unemployed people within a training project. Journal of pragmatics, 35(3), 409-434. doi:10.1016/s0378-2166(02)00143-1 Saldaña, J. (2011). Fundamentals of qualitative research. Retrieved from https://ezproxy2.usn.no:2537/lib/ucsn-ebooks/detail.action?docID=665394 The Norwegian Ministry of Education and Research. (2014). Lokal gitt muntlig eksamen Udir-2-2014. Retrieved from https://www.udir.no/regelverkstolkninger/opplaring/eksamen/Udir-2-2014/ The Norwegian Ministry of Education and Research. (2019a). Kunnskapsgrunnlag for evaluering av eksamensordningen. Retrieved from https://www.udir.no/tall-og-forskning/finn- The Norwegian Ministry of Education and Research. (2019b). Regler for muntlig eksamen. Retrieved from https://www.udir.no/eksamen-og-prover/eksamen/muntlig-eksamen/