Page 212 - Contributed Paper Session (CPS) - Volume 7

CPS2065 Lina Schelin et al.
reasoning (Garfield & Chance, 2000). But, as described in Chance (2002) also
traditional examinations can be used to assess such type of understanding.
For instance, to address the challenges identified by Garfield and Gal (1999) a
test designed to measure students’ conceptual understanding of important
statistical ideas has been developed (delMas, Garfield, Ooms, & Chance, 2007).
In our introductory courses, we use several assessment techniques,
including a written exam. The written exam consists of several tasks, and
during the marking of the exam each task is rewarded with a point. The total
score on the written exam determines a student’s grade on the whole course.
Garfield (1994) states that “the primary purpose of any student assessment
should be to improve student learning”. Although this is something we
definitely agree on, it is not the primary purpose of the written exams in our
courses. The total score on the written exam, is instead used as an indicator of
each student’s success in reaching the expected learning outcomes of the
course and used for grading.
One other important aspect of all assessment is that it should be consistent
and fair (Chance, 1997). This can be difficult, especially when several teachers
are involved in the assessment; hence, the challenge of consensus.
At our department, the marking of the written exams on introductory
courses, is performed individually by several teachers in assessment teams;
usually the teachers correct 1-2 tasks each. During the years, there have been
several cases where we during such marking processes have noted that the
corrections deviate between different teachers, and between different
assessment teams. The tendency is that this is a problem that grows with the
size of the assessment teams. To prevent such situations and to further
develop the written exams, we initiated this project where the main aim was
to reach a higher level of consensus among the teachers at our department in
the assessment process. Prior to the start of the project, preliminary
discussions among the faculty indicated that the assessment teams usually
agreed on the overall grade of an exam, while at the same time they did not
agree on all details. Hence, a shift of focus from awarding each task on the
exam with a point (on a rather fine scale) towards something that is similar to
a rubric would be beneficial. We had previously tried to employ a rubric, with
less successful results. Basically, it was too difficult, also for the more
experienced teachers, to apply. The main reason why we want the shift towards
rubrics is that it enables a holistic grading. In our experience, awarding points
on a fine scale, does not provide a good measure of the students’ ability of
statistical reasoning and thinking. Instead, it awards students with fragmented
knowledge.
We will now present the course and the written examination in more detail
before presenting the project, the results and some conclusions.

199 | I S I W S C 2 0 1 9