Upper secondary exit exams bridge the gap between school and university (Noah & Eckstein, 1992). This double role of marking successful passing of one stage and acting as a gatekeeper for another is especially salient in countries where the number of applicants accepted to higher education exceeds those eligible, a problem sometimes solved by ex post selection during early tertiary studies (Declercq & Verboven, 2018). Yet, when the exam plays a prominent role in student admission, the high stakes of the exam are acute.

The positive role of centralized exit exams in improving student performance is well documented (Bishop, Mañe, & Bishop, 2001; Jürges, Schneider, & Büchel, 2003), apparently through a process where students transfer the extrinsic requirements of the school to their personal intrinsic motivation (e.g., Ryan & Deci, 2000). Some authors, however, stress the adverse effects exams can have on instruction and on especially weaker students’ achievement and motivation under the pressure of testing (Amrein & Berliner, 2002; Reardon et al., 2009).

In Germany, students’ success (grades) in the exit exam has been found to have a stronger impact on their future earnings in Länder with a central rather than local exam (Schwerdt & Woessman, 2017),  corroborating earlier results (Piopiunik et al., 2013). But discussion on the possible impact of centralized exit exams has not only centered on their effect on student performance but also on their potential for increasing standardization or comparability of school grades (Maag Merkl & Holmeier, 2015). When exploring the shift from course-based to state-wide exit examinations in Germany between 2007 and 2011, Maué (2016) found a significant difference in students’ teacher given grades related to gender, ethnicity, and family background. She also found that the relation between course grades and achievement test increased through time, possibly through a corrective effect of the state-wide exams.

In their investigation of statewide exit exams in 16 OECD countries, Klein and van Ackeren (2011) point out that the ‘‘statewide exit examination’’ describes diverging procedures, and the differences in the design and standardization of statewide examinations are even more distinct when examinations are contrasted in an international comparison while summarizing the results of earlier literature as “assumed to affect and regulate work at both the school- and classroom level by setting minimum standards (qualification), promoting quick implementation of new syllabi (innovation), helping to improve instruction (professionalization), and increasing commitment (motivation).”

The Finnish matriculation examination has a long history (Kaarninen & Kaarninen, 2002) but has seen major reforms during the past twenty years (Kupiainen et al., 2016; Kupiainen et al., 2018). The ensuing increase in exam choice has led to the weakening of the exam’s compliance to John Bishop’s (1998) construct of Curriculum-Based External Exit Exam Systems (CBEEES), which he sees to be the most germane for advancing student achievement. This endangers the comparability of the grades of the different subject-specific tests, countered in the Finnish university admission by differentiating the credits awarded for the grades awarded in the different exams.

In the present study, we approach the Finnish matriculation examination through three questions: (1) How well do the different subject specific exams capture students’ general readiness for tertiary education? (2) How well do the different exams capture students’ motivation and performance in the respective subjects across their upper secondary studies? (3) Does the matriculation examination serve its purpose of correcting between-school differences in school-based grading?

The questions are especially acute now as from year 2020 on up to sixty per cent of students will be accepted to university based just on their matriculation examination results, credited according to a formula awarding varied credit for the different subject-specific exams.

The data is drawn from a longitudinal study of the approximately 13 500 students in the Helsinki metropolitan area who entered lower secondary education in autumn 2011. Following the dual structure of the Finnish upper secondary education, 8 109 of them were accepted to one of the metropolitan area’s 67 academic track schools. Of them, the present study regards the 6 172 students (43.4 % boys) who passed their matriculation examination in spring 2017 after three years of academic track upper secondary studies, representing 87.1 per cent of the students who entered the metropolitan academic track schools in 2014. Dropout from the academic track is very small and students are allowed to extend their studies to four years, so about ten per cent of the rest of the students are expected to have graduated in spring 2018. Of the 39 exams of the matriculation examination, we will focus on the 27 exams taken by at least 50 students each in the current sample of 6 172 matriculates. In accordance with the upper secondary syllabus, the number of courses covered by the different exams in the matriculation examination vary from 3 in Health Education to 13 in Advanced Mathematics. All exams offer choice for the student regarding the specific content and level of difficulty of the tasks (e.g., 10 questions out of which the student has to answer 8) with two more demanding integrative tasks providing more credit. The exams are graded with a seven-point scale with a discontinuity between failed (0 = improbatur) and 2–7 (approbatur to laudatur). School grades follow a continuous seven-point scale from 4 (failed) to 10 (excellent). In the analyses, we will use both the grades for individual courses and grade means for the mandatory and specialization courses, which act as the basis for the subject-specific exams their matriculation examination. The results will be presented at the descriptive level, using ANOVA for group-level comparison. As the share of students sitting for the different exams in the 62 schools differs, multi-level analysis has not been a valid option. Students’ course grades were provided by the schools and the matriculation examination grades by the Finnish Matriculation Board. The research project providing the data for the present study has the approval of the Ethical Board of the National Institute for Health and Welfare.

Students’ success in the different exams was strongly related to their success in the other exams they chose for their examination. However, there were clear differences between the general level of attainment of students receiving the same grade in the different exams, implying that the steps taken to strengthen their comparability has not succeeded in full (cf. Kupiainen et al., 2018). Students’ grades in the subject-specific exams also correlated strongly with their grades in the successive courses of the respective subject across their upper secondary studies, independent of the number or courses allocated to the subject in the syllabus and acting as the basis of the respective exams. The grade-mean /exam correlation was highest (r = .809) for A-level English, included in almost every students’ examination. The high correlations (r > .650 for 15 of the 27 most popular exams) can be seen to imply the key role of a general cognitive component and mastery motivation behind both. Also, in all subjects, students’ grades on average differed from the first course on linearly according to the grade they were to get in the matriculation examination, meaning that students ending up with the highest grades had received, on average, the highest grades in all the courses of that subject from first semester-period on. Consequently, both the matriculation examination and school graders can be seen as valid indicators for students’ readiness for tertiary studies. However, the considerable between-school differences in the relation of students’ school grades and their success in the respective exam in the matriculation examination found in the study speak strongly for maintaining the tradition of the centralized matriculation or exit examination despite its current problems. In the presentation, we will extend our discussion to the criticisms presented against the examination and its use in student selection to higher education.

Amrein, A. L., & Berliner, D. C. (2002). High-stakes testing & student learning. Education policy analysis archives, 10, 18. Bishop, J. (1998). The effect of curriculum-based external exit exam systems on student achievement. Journal of Economic Education, 29(2), 172–182. Bishop, J. H., Mañe, F., & Bishop, M. (2001). How external exit exams spur achievement. Educational Leadership, 59(1), 58–63. Declercq, K., & Verboven, F. (2018). Enrollment and degree completion in higher education without admission standards. Economics of Education Review, 66, 223-244. Jürges, H., Schneider, K., & Büchel, F. (2003). The effects of central examinations on student achievement: Quasi-Experimental evidence from TIMSS Germany (Ifo Working Paper No. 939). Retrieved from CESifo Group Munich http:// www.cesifo-group.de/portal/page/portal/DocBase_Content/WP/WP-CESifo_ Working_Papers/wp-cesifo-2003/wp-cesifo-2003-05/cesifo_wp939.pdf Kaarninen, M., & Kaarninen, P. (2002). Sivistyksen portti. Ylioppilastutkinnon historia. Keuruu, Finland: Otava. Klein, E. D., & van Ackeren, I. (2011). Challenges and problems for research in the field of statewide exams. A stock taking of differing procedures and standardization levels. Studies in Educational Evaluation, 37(4), 180–188. Kupiainen, S., Marjanen, J., & Hautamäki, J. (2016). The problem posed by exam choice on the comparability of results in the Finnish matriculation examination/Das Problem der Wahlmöglichkeiten von Prüfungsfächern hinsichtlich der Vergleichbarkeit von Ergebnissen der zentralen Abschlussprüfung in Finnland. Journal for Educational Research Online, 8(2), 87. Kupiainen, S., Marjanen, J. & Ouakrim-Soivio, N. (2018). Ylioppilas valintojen pyörteessä. Suomen ainedidaktisen tutkimusseuran julkaisuja. Ainedidaktisia tutkimuksia 14. https://helda.helsinki.fi/handle/10138/231687 Maag Merki, K., & Holmeier, M. (2015). Comparability of semester and exit exam grades: Long-term effect of the implementation of state-wide exit exams. School effectiveness and school improvement, 26(1), 57-74. Maué, E. (2016). Achievement—and what else? The standardisation of semester grades due to the implementation of state-wide exit examinations. Studies in Educational Evaluation, 51, 42-54. Noah, H., & Eckstein, M. (1992). Comparing secondary school leaving examinations. Examinations: comparative and international studies, 3-24. Piopiunik, M., Schwerdt, G., & Woessmann, L. (2013). Central school exit exams and labor-market outcomes. European Journal of Political Economy, 31, 93-108. Reardon, S. F., & Galindo, C. (2009). The Hispanic-White achievement gap in math and reading in the elementary grades. American Educational Research Journal, 46(3), 853-891. Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary educational psychology, 25(1), 54–67. Schwerdt, G., & Woessmann, L. (2017). The information value of central school exams. Economics of Education Review, 56, 65-79.