|
On Preparing for and Sitting Unseen Exams
Geoffrey Sampson
The time of year is approaching when undergraduates begin to revise for their unseen exams.
For many, it may be quite a while since they sat formal exams; for some undergraduates who have
come to us by non- standard routes possibly these may be the first traditional exams they have sat.
Therefore it seems worth circulating some suggestions about how to approach this particular kind
of activity. If revising for and sitting exams is for you by now a frequent routine, you may not
have any use for these hints, but experience suggests that some of the suggestions are not obvious
to all students, so we offer them for you to make what you wish of them.
Revision
Well before the exam period you will want to begin revising the subjects to be examined, in order
to organize your many sources of information on each subject - lecture notes, coursework essays,
publications or notes made from reading publications - into a structure that makes sense in your
mind and to remind yourself of the details that fit into that structure.
There are undoubtedly different ways to revise effectively: one way is the precis spiral. Say
you face exams in four subjects, A, B, C, and D. Take all your information sources on A, and precis
them down into an organized written report that will stretch over quite a few pages. It needn't
consist of continuous prose, and perhaps shouldn't: notes, tables, `concept maps' and the like are
all useful formats, depending on the nature of the material. Anything that adds visual
memorability - perhaps even colour coded notes - is likely to help.
Once you have produced a report that covers the whole of subject A (which will probably take some
days), you will have had enough of A for a while, so go on and do the same thing for subject B, and
then subject C, and D - you might want to choose a sequence which puts different kinds of subject
next to one another, to make it less boring.
Then, once you have finished this first round of precis-ing, it is time to go back to A: take your
report and precis it down to a shorter document, maybe just a page or so - try to organize the
information as systematically and logically as possible, so that your new shorter document makes
sense in its own right (of course no-one but you will ever read it, so you can use whatever
abbreviations, symbols, etc. you like). Go on to make shorter precis's for B and C and D. You may
even have time for a third round, where the information in the four short documents is again precis'd
down to something little more than a postcard for each of the four subjects.
If your ultimate postcard-length summaries are themselves systematic, you will end up with an
intellectual structure that is so concise and clear that you will virtually know it by heart, but
which is linked by a chain of summaries, which you have yourself prepared, to the full richness of
the material in your original set of information sources. This puts you in a strong position to
identify and deploy relevant material when you see the actual questions you have to answer.
Above all, revision should be about writing, not (or not just) about reading. The active process of
writing has a magic power to link facts and ideas into your thinking and memory, which the passive
process of reading does not do. One can take a book on a subject one wants to learn, feel one is
reading through it attentively, and then at the end wonder `what was all that about?' and feel as
blank as before; but if you read with notepad at hand, making summaries of the significant points -
not necessarily long screeds, but perhaps several lines per chapter - you won't end up with that
blank feeling.
Practice Exams
Sitting a formal unseen exam is a rather unusual kind of human activity, and like any other
specialized activity you need to practice beforehand rather than just assume 'it'll be all right on
the night'.
You will know from looking at last year's papers what the format of your exams is going to be;
in CSAI most if not all of them are 90-minute papers in which you answer two questions. Some time
well before the period of the actual exams, take an old paper, lock your door against interruptions,
set yourself up with writing materials and a watch on the table, and practice answering the paper
within the time, not overrunning and not breaking off for cups of coffee. It is so much easier to
do something like this `for real' if you've done essentially the same thing `for practice' before.
Of course it's a good idea to test yourself in different ways on all the subjects you are revising,
but these will often be brief informal `turn the notes over and write down the essential points'
tests in the middle of precis-making. At least once, though, you should try doing a formal timed
practice exam in some subject, to give yourself the feel of what the exercise is like.
Mens Sana In Corpore Sano
Mental work like answering exams is done best by people who are feeling healthy and fresh. You
probably will be burning the midnight oil sometimes during the revision period, but don't take
that right up to the exam time itself - by a couple of days before you sit a paper, if you haven't
revised enough already you probably won't improve your performance by revising more. There is a lot
to be said, after putting in solid weeks of revision, for taking a couple of days immediately before
the exam period as days off/holiday: the freshness when you sit down to writing your scripts may do
more to improve your performance than a few more hours looking at books. At the very least, go for
a walk in the morning of a day you have an exam to sit. As the exam period progresses, eat properly,
get some exercise, and go to bed at sensible hours.
When you go to sit the papers, you may have all sorts of troubles; a close relative may be seriously
ill, you might have girlfriend/boyfriend or money worries - but for the few hours you are sitting
in the exam room, set these things aside. You can do nothing to solve them during those few hours,
so treat them as nonexistent: allow yourself to enter a daydream world where everything is hunky-dory
and your only task is to find ways to show your abilities to the examiners to best advantage. If
conversations with someone you know often leave you seething, avoid getting into a conversation
with that person just before an exam.
Organize yourself so that you come to the exam with a spare pen, and ink if you use a fountain pen;
if you find yourself without a working writing instrument in the exam room, the invigilators may be
able to help, but it all creates stress that you don't need (and wastes your time).
Exam-Writing Technique
Here are some suggestions about the actual process of writing the script:
- Read the questions. Yes, it sounds silly; but candidates sometimes seem to glance at the paper,
spot one question that they are happy to have a go at, and jump straight into answering that one
without taking the time to read through and consider the alternative possibilities which the paper
offers. CSAI papers commonly require two questions to be answered out of only three, but within those
three questions there are often alternatives so that the true set of choices is larger. Spend a little
while weighing up which combination of questions would really be the best ones for you to choose.
- Manage your time. If you have 90 minutes, that isn't very long; divide it up in a sensible way,
say five minutes for reading the paper and deciding which questions to answer, 35 minutes to answer
each question divided into ten minutes for organizing an answer plan and 25 minutes for turning the
plan into a written-out answer, and 15 minutes for reading your script through and catching errors
or adding improvements. Jot down, or remember, at what times you should have reached these various
milestones - you might depart from them a bit, but don't let yourself get into difficulties by
becoming so caught up in doing one thing that you lose sight of how much time is left for the other
things to be finished. Above all ... :
- Don't spend so much time answering your 'favourite question' that you can write only
scrappy notes for the other question you choose. Unless the rubric on a particular paper says
differently, each
question contributes equal weight, and you should put broadly similar time into each of your answers.
It might well be true that you can write a really marvellous answer to question 1, but if you spend
so much time doing that that you can produce only some token jottings in response to question 2 then
the latter will get a very low mark which will be averaged with the question-1 mark to give,
undoubtedly, a poorer overall result than if you had reined yourself in slightly on question 1 to
make time for a respectable answer to question 2.
- Plan each answer. Jot down a skeleton answer-plan, on a page which you will cross
out as rough work, before writing the actual answer to be read by the examiners which puts flesh
on the bones of the skeleton. That is the way to produce an answer which reads as a systematic,
logical exposition rather than a jumbled stream of facts or ideas. Developing your answer-plan does
take some time away from constructing the real answer, but it is time well spent. Tweak the plan to
get it right: cross bits out and shift the topics round on paper, until you feel it is organized
logically - much of the point of the plan is to create an opportunity to organize your answer into
the best shape.
- Answer the question on the exam paper - not the one you were expecting to find on
the paper. We don't set 'trick questions', but experience shows that candidates who, during their
revision, have worked out questions that are likely to come up on a topic, and prepared what they
should say if those questions do come up, then fail to notice when they see the actual paper that
the questions set on those topics ask for somewhat different information or ideas than the ones
they anticipated. It is a standard dilemma for examiners to know what mark to give to a really
good answer to a question that wasn't the question set; we usually feel sympathetic, but have to
give considerably less than the same candidate could undoubtedly have achieved if he had read the
question.
- A special case of the last point: questions will often be divided into multiple
sub-questions, and in some cases candidates have to answer any one of the sub-questions, in other
cases they have to answer all of them. Make sure you notice which is required!
- Lard abstract answers with concrete examples. This is a point which obviously
depends on the topic and may be inapplicable to some topics; but in any subject where it is
possible to illustrate theory with concrete examples, an answer which does so has far more impact.
And in particular, if there is a `stock example' which the textbooks or the lectures standardly
quote, give a different example if you can. Quoting a stock example just shows that you have
remembered it. Quoting a different example (provided it is a true example of the issue it is used
to illustrate) shows that you have understood that issue well enough to identify an example for
yourself; it is much more impressive.
- When you discuss ideas/techniques associated with specific individuals, mention
their name and if possible give an indication of the book or article title - another way in which
it is better to make answers concrete rather than abstract.
- Use all the time available. Candidates who have finished are permitted to
leave the exam room early (though I think there is a rule against leaving in the closing minutes,
to avoid disturbance to candidates who are scribbling against the clock). Many candidates take
advantage of this and do leave early; this always surprises me, and seems very unwise of them. If
you manage your time properly you should complete your answers well before the close of the exam,
but there is no reason to leave then. Obviously you will want to read through your answers at least
once to catch silly slips that anyone can make; but in my own exam-sitting days I was glad to take
the chance to read through a second and a third and even a fourth time (if time allowed), weighing
up whether some slight change here or there might improve what I had written. It might be a change
to wording, or to some technical notation, or correction of a spelling mistake or improvement of
the punctuation: making each of these aspects of your answers as good as you can get them all helps
maximize your marks, and after you leave the exam you have no more chance to improve them.
- Students often wonder what 'level' they should write at - how much should they
assume, what things should they spell out? If their script is to be judged by their teacher, he
knows everything (well, almost), so it isn't clear that any particular facts need to be spelled
out. But if one were to say, in that case one should write as if for Auntie, she maybe doesn't
even know the difference between software and hardware - obviously it will not be possible to
spell everything out down to that level.
A good compromise here is to write as if for a CSAI teacher who doesn't teach this particular
topic; who certainly will know very well what 'software' and 'hardware' mean, but will not be
familiar with the technical details of the individual course. And this is not an artificial
recommendation. The ultimate arbiter of the standards to which we mark exam scripts is an
External Examiner, who visits us from another university to check that we are doing everything
as we should and that our assessment standards are in line with national norms. The External
Examiner will read a sample of scripts to check that our marking is OK, and there is only one
External for the BA and one for the BSc degrees; so for any particular exam the chances are that
the External will not be a subject expert in that topic.
- It probably isn't a good idea to be witty in an exam script. It won't get you
extra marks, and what seemed like a good joke in the heat of the moment may fall very flat and
irritate an examiner working through a pile of scripts late at night. (You may not enjoy sitting
exams; we are not wild about marking them, but it has to be done.)
- Write legibly. One sometimes hears people boasting 'no-one can read my
handwriting'; if that fits you or even comes close to fitting you, train yourself quick to
write clearer. We have to mark large numbers of scripts to a tight timetable; if an examiner
finds himself going cross-eyed trying to make out the words of one script, it will be very
difficult for him to assess the answers at their true value - apart from the irritation, if
writing is so unclear that the words have to be puzzled out one by one then it is hard to put
the separate words together in one's mind and grasp the overall meaning.
- Poor handwriting written with a fountain pen is usually quite a bit more legible
than poor handwriting written in ballpoint.
Answering Mathematical Questions
Here are some suggestions (courtesy of Carolyn Brown) on the special issue of exams in
formal/mathematical subjects:
- Try the question, even if you think you don't understand it. Read it through slowly and don't
panic!
- There are often marks for stating what the appropriate proof method is - for example, I would
give some credit to a student who wrote 'the proof proceeds by structural induction' and a little
more credit to a student who wrote 'the proof proceeds by induction on the structure of expressions'.
Even when confused, students tend to know this much about a question.
- At each step in a proof, say what you are doing. It is difficult for an examiner to follow a
proof which consists of disjoint, unconnected equations. It is easy enough to say things like
Base Case: ... There are two possibilities:
First ...
Second ...
This completes the base case.
Inductive step: ... (stating clearly what the inductive hypothesis is and where it is used)
Concluding line: "this completes the proof"
This also allows the examiner to allocate some marks even if your equation happened to have
a mistake in it. An unexplained, wrong equation will get no marks.
- Read through the whole proof after you have written it. Be sure that statements make sense
as English sentences. Be particularly careful to use phrases such as 'therefore', 'implies', 'if
and only if' correctly.
- When you make steps which are obvious (a common, simple such step is deducing A = C when
A = B and B = C) try to think why the step seems obvious to you. Almost every step in a proof
should be accompanied by a explanatory reason, even if it is only a statement such as 'by arithmetic'
or 'by properties (or by transitivity) of equality'.
And finally ... DON'T WORRY
Do bear in mind that we examiners are not 'out to get you'. We would like to see our students
do well - we really would. We try to think of questions that will make it easy for candidates
to display their abilities and knowledge; designing exam questions is a difficult task, but
(whatever you may occasionally suspect) we truly are not being vindictive and setting out to
trip you up! In my experience, whenever unclarities or doubts arise in the examining process about
whether something a candidate has written is appropriate or not, the candidate is virtually
always given the benefit of the doubt. If you have worked reasonably during the year and revised
sensibly, you should get respectable results. Sometimes people get better results than they
expected and sometimes, alas, they get worse - but it really isn't worth getting into a tizz
and worrying about taking exams. In ten years time it probably will make little difference what
exam marks you got this year; and worrying won't get you better marks! Treat exams as an interesting
part of the experience of young-adulthood, to be taken seriously, but not too seriously.
Good luck!
|
