Do students think they can predict what they will be asked to do in an exam?
As part of a larger project on the predictability of the Higher Irish Leaving Certificate examinations, students were asked whether they thought they could predict which questions were asked in exams and how this impacted upon their learning. Students believed that certain subjects were more predictable than others. Furthermore, the predictability of an exam affected the learning and revision strategies undertaken by students (and their teachers).
HOW THE RESEARCH WAS CARRIED OUT
Eighty-one students were interviewed from 12 schools (7 secondary schools, 3 community schools and 2 vocational colleges). Thirteen group interviews were carried out, each consisting of 5 to 8 students.
FINDINGS – IMPLICATIONS FOR TEACHERS
Students stated that there were elements of the exams that they thought were predictable.
- Students thought that some subjects were more predictable than others: The students in general considered biology, geography, and economics to be less predictable due to the large volume of content learnt on these courses making it difficult to predict the topics that were in the exam. However, English and French were viewed as more predictable because material could be prepared and rote learnt in preparation for oral and written exams.
- Predictability focused pupil learning: Students revealed that elements of exam predictability helped them; identify the key content for efficient revision, pinpoint key definitions, gain a better understanding of the types of answers that examiners accepted, and develop the test-taking skills needed for the exam.
- Different techniques were used to make the exams more transparent: The interviews revealed that both students and teachers employ various techniques to increase exam predictability and reduce the element of surprise. Techniques reported were:
- Test conditions: Sitting tests and mock exams in examination style conditions helped students to familiarise themselves with formal exam procedures.
- Past papers: Using past papers helped students to familiarise themselves with question formats and structures.
- Mark schemes: Students stated that using mark schemes helped them to structure their responses to questions, recognise the format of sought answers and understand how exam questions are marked by examiners. Familiarisation with mark schemes enabled students to identify the most ‘valuable’ questions (with the most marks). Students used this indicator to help them appropriately allocate their time during revision and while answering questions in an exam (notably in biology and geography).
- Using textbooks: Students used course books to learn the end of chapter summaries and answer the past paper questions. Participants reported that the course book helped them focus on what they needed to learn for the exam.
IMPORTANT TO NOTE
- Students stated that both they and teachers tended to narrow the curriculum taught for content heavy subjects due to the limited time frame. This was often to the detriment of students developing a better understanding of the content.
- In general, students believed that exams were more about testing recall than understanding. Consequently, students tended to adopt techniques that benefitted their ability to memorise content for recalling information in an exam. However, students recognised the need to develop an understanding of the content in order to attain the highest grades.
- A separate part of the project, which is in preparation for publication, looked at whether students’ views of predictability were related to their exam results.
Jannette Elwood (Queen’s University Belfast)
Therese Hopfenbeck (OUDE)
Jo-Anne Baird (OUDE)
The work was supported by the State Examination Commission, Republic of Ireland.
More information on this project can be found at: http://www.education.ox.ac.uk/research/educational-assessment/ Jannette Elwood, Therese Hopfenbeck & Jo-Anne Baird (2015): Predictability in high-stakes examinations: students’ perspectives on a perennial assessment dilemma, Research Papers in Education, DOI: 10.1080/02671522.2015.1086015