Doing your Dissertation

A practical guide to conceiving,planning and executing your dissertation:
Role of the Supervisor

The initiative for requesting supervisions lies entirely with you,the student.The
supervisor's role is to give advice and help about the nature and standard of the work,
and direct you to useful literature and appropriate methodology.But remember,the
ultimate responsibility remains yours.Do NOT expect your supervisor to read draft and
re-draft of every little piece of your work,and above all,do NOT embarrass your
supervisor in the latter stages by asking whether you will be successful.When
submitted,the thesis is referred to internal and external examiners who will make an
independent judgment of your work in its entirety.

Working with your Supervisor
It is a good idea to find out how your supervisor likes to work.Try at all costs to avoid
rambling supervisions with vague discussions;both of you need to use time
economically.Having thought of an idea and roughly defined it,your next step will
probably be to prepare a draft outline for your potential supervisor to read before the
next meeting,remember that every time you give your supervisor something to read,
s/he will need a few days to read it.After the first consultation,supervisions should
almost always be based on something written by you,so that the supervisor can give
you feedback on your ideas and proposed methods.The initiative should be taken by
you rather than your supervisor.
Do bear in mind that the staff have many diverse claims to their time,and that this
state of affairs continues throughout the summer.Contrary to popular belief,university
staff are fully occupied with clinics,research,lecture preparation,administrative work
and writing commitments all the year round,and therefore if you wish to see your
supervisor,you should make an appointment.Agree methods of getting in contact with
your own supervisor:email is usually the best way.Your supervisor will almost
certainly be supervising a variety of other projects;therefore,you should not assume
that s/he can immediately recall the last discussion you had together about yours.
Appended to this document is a supervision record sheet,which you can copy and fill in
as a record of each supervision session.We suggest that students be entitled to 6 to 8
supervisions each(totaling about 4 to 6 hours),including 3 sessions for statistical
advice.This is fewer than you think,so be sparing with your supervisor's time and
make each one as profitable as you can.At the very least,you should consult your
supervisor before data collection and before submission of your dissertation,
giving sufficient time for feedback comment.
Supervisions always need to be prearranged.In particular,if your thesis stretches into
the next academic year,your supervisor will be extremely busy.You should,in any#p#分頁標題#e#
case,never expect on-the-spot supervisions.Taken from the Good Practic.

Writing for your Supervisor
Supervisors cannot judge how your work is going if you just hand over a sub-section
that you have written up.You need to have a draft of your contents so that it is quite
clear what follows on from what.You may also need to add a note to show if and how
the section in question is incomplete.
Your supervisor will probably not wish,nor have the time,to read all your thesis as you
prepare it.It would be a good idea,therefore,to discuss this and establish which
sections s/he wishes to see.Examiners,however,read all of it!Also,supervisors are
not to be expected to proof read or to correct spelling/grammar.Students are advised
to buy-in or otherwise arrange such services if needed.
Even if your supervisor reads your entire thesis and you make the suggested changes,
this is no guarantee that it is of a pass standard;after all,the work is yours and its
quality is dependent on your output.

Planning Ahead
When you are planning your schedule,please do remember that your own time and
application is only one of the determining factors.You need to take into account your
supervisor's time as well as your own when you are estimating overall time needed.
Data collection always takes longer than you expect,as does e.g.,the writing of a
literature review,and even the completion of your reference list,so preferably be on
the safe side and double your first estimate.
There will be periods when your supervisor is not available,either because s/he is
heavily committed with other duties such as examining,or on leave for work or holiday
reasons.Do make sure that you and your supervisor are aware of each other's periods
of absence.
Part-time students need to be sure that they have time off for supervisor meetings and
for data analysis.Part-time students may also find email attachments an efficient way
of sending information to their supervisors.
Again,considering your own holiday plans,you are advised not to have more than two
week's holiday.If you are collecting data,and you intend to take a break before you
finish,this must be discussed with your supervisor.
Up to three of your meetings with your supervisor and/or research methods lecturer
may need to be for statistical advice and planning:at the design stage,at the results
stage,and perhaps to discuss any computer output and its interpretation.
With regard to use of statistics packages,students may wish to seek help from the
Faculty of Education Computer Unit.Technical support is there to help with this,but
please note the support from computational assistants will be subject to availability.
Students will need to make an appointment when they first need to use the equipment,
or if they think they need more than brief or intermittent assistance.#p#分頁標題#e#
Some students have,in the past,expected very considerable amounts of time for
statistical help.This is not possible,and the arrangements above should be adequate
for M.Sc.theses.Data Collection
You will,no doubt,be eager to start collecting your data soon after the summer exams
are completed.This may involve setting up any equipment you need and booking up
subjects.Given that you have consulted your supervisor,and have the requisite ethical
approval,data collection should then proceed in accordance with your project outline.
Some projects need controls.If yours involves testing normal subjects for this purpose
or any other,do not assume these can easily be found from the pool of people around
you.Staff are too busy,so please do not ask them.Your fellow students will also be
busy completing their own work,so if you do ask them,make sure it is clear how much
time you require and ensure that they are willing to co-operate.A notice on the board
asking for volunteers(or in'This Week,Next Week')is probably a good method.
Beware,too,of assuming that people are normal hearing,even if they think they are.
Test them to make sure.
You may need an assistant for testing;again it is your job to find one,and don't devise
a project which would make heavy demands on another person's time,unless you are
absolutely sure you can find such a person who is willing.