This
module is an introduction to statistics, particularly as applied in oncology
and palliative care. The overall aim is to explain why statistics is
important in these fields and to explain some of the basic concepts of
statistics. At the end of the module you should be familiar with many of
the statistical terms and techniques you will encounter when reading the
literature. You should also be able to analyse some of your own
data. The course makes extensive use of the statistical package Minitab,
and through the course you will acquire the skills necessary to start using
this program.
Mode of Study
The first week of study invites you to read a research paper and to try to decide where the statistical issues arise. In the second week you will get to know Minitab, either installing it on your PC or using it via RAS (unfortunately there is no longer a version of Minitab for the Mac). Thereafter each week has a similar format.
In each week there will be one, or occasionally two study documents for you to read. After reading each document there will be a set of exercises to do. These will involve the use of Minitab. These are not formally assessed, they are simply there to help you to consolidate your knowledge. Learning statistics is not straightforward and you will often find that you will have to go over the study document several times in order to complete the exercises successfully. You can check your answers in the Solutions Sheet which accompanies each set of exercises. However, you are encouraged to make a thorough attempt before looking at the solutions!
There are three sets of true/false questions, after week 5, after week 7 and after week 12. These are formative assessments and do not contribute to your mark for the module. However, I strongly recommend that you do these tests.
There are two weeks in which no formal work is set.
In the last week of the course there is a test on all the material in the course: this will be similar in style to the formative assessments set after weeks 5, 7 and 12. You will also be given an extract or extracts from the report of a real study and asked more extended questions on the statistical aspects of the study. The summative assessment for the module is based equally on these two components.
Other sources
The course is intended to be self-contained, in that the study documents cover all the material you need. However, it is often the case that being able to read another author’s view of a subject can help to elucidate points of difficulty, or can simply consolidate your understanding.
To this end, in the study documents, and at other points in the course, references are given to relevant chapters, sections and pages of the book by Martin Bland, 'An Introduction to Medical Statistics' (3rd edition), Oxford University Press, Oxford, 2000 (ISBN 0 19 263269 8). This book has the extra advantage that it contains many multiple choice questions, with fully explained solutions. Doing these can add to your understanding of the subject.
Another useful book is that
by MJ Campbell and D Machin, 'Medical Statistics:
a Commonsense Approach' (3rd edition), Wiley, Chichester, 2000 (ISBN 0 47
198721 2). There is a fourth edition of this book, published in 2007,
albeit with a slightly different title (Medical Statistics: a Textbook for
the Health Sciences) and a third author (authors now Machin, D, Campbell,
MJ and Walters, SJ): ISBN 978-0-470-02519-2. This also looks good but is
rather longer than the third edition and one of the advantages of the earlier
editions was their brevity. I leave it to you to decide - the third
edition is perfectly satisfactory.
A rather different style of book, comprising worked examples, which is a useful
complement to Bland is:
Bland, Martin and Peacock, Janet, ‘Statistical questions in evidence-based
medicine’, Oxford University Press, Oxford, 2000 (ISBN 0 19 262992 1).
Week 1: Why
do you need statistics in medical research?
Week 2:
Introduction to Minitab
Week 3: Basic
descriptive statistics
Week 4: The
idea of a population and its parameters
Week 5:
Properties of the Normal distribution
Week 6:
Samples, estimation and standard errors
Week 7:
Confidence intervals: what they are & how to find them
Week 8: No
formal work this week
Week 9: The
idea of a hypothesis test
Week 10: Tests to compare
two groups: t-tests
Week 11: Binary data and
the c2 test
Week 12: Principles of
sample size calculations
Week 13: No formal work
this week
Week 14: Survival analysis
Week 15: No formalwork this
week
Week 16: Summative
assessment