14.2.4.4
Setting appropriate titles |
|||||||
|
|||||||
Traditionally, university language essay exams have featured quite a wide choice of questions, in the belief that this was fairer to all students as it catered for their inevitably varied interests and knowledge. However, this can reduce severely the reliability of essays as a testing technique (see Module 13, section 13.1.4.2): the wider the choice of questions, the less reliable the assessment instrument becomes. For this reason, rather than just giving an essay title such as 'Write an essay on the role of the German President', the following will provide more control over what is written, making the task a more reliable indicator of the student's ability to write in the FL and improving the reliability of marking across the group being tested:
Activity 12
Click on 'Commentary' for feedback. This activity emphasizes the importance of ensuring that in setting language essays we are not testing things we do not mean to assess. The following title for an essay at intermediate level suggests how this might be done:
It might be objected that such an approach tells students what to write; but, if the main aim of the essay is to test the ability to write in the FL, we should not be bringing in extraneous factors such as general knowledge. In formulating such questions, it is obviously important to use notes rather than whole sentences so as not to provide too much linguistic help with the task. However, a title such as this is far preferable to a bare 'Compare and contrast the English and German school systems', which places unequal demands on exam candidates, in particular. Moreover, an 'unpacked' title such as the one above, or the earlier one about the German Presidency, allows the tutor to exercise greater control over what is written. This might offend some on intellectual grounds, but it does make the task a more reliable indicator of students' ability to write in the FL, and this is, after all, what we are supposed to be ascertaining when setting FL writing tasks.
|
|||||||
|
|||||||
|
|||||||