Article
published in RSS News, on December 2007..
Stephen Senn argues that viewing university
students as 'customers' can be dangerous
“I went to lectures by Fred Hoyle and Herman Bondi and a wonderful Russian called Besicovitch, an expert on almost-periodic functions…If you wanted to learn Hardy, which is what he was supposed to be lecturing on, you had to read the book in your spare time because most of his [Besicovitch’s] examples were taken from his current research. I don’t think this was a bad thing.”
This is John Nelder speaking of his Cambridge
days (Senn, Statistical
Science, 2003) and the book to which he is referring is A Course of Pure
Mathematics by GH Hardy. It is only one example plucked not at all at
random and John Nelder was very far from being an
average student, yet I wonder whether any student of mathematics or statistics,
or for that matter any educational expert, university administrator or
politician these days, could be found to defend as acceptable an approach to
education that appears to place this degree of responsibility on the student.
However, I personally agree with John Nelder
that hearing someone lecture on their own research, whilst being expected to
read, learn and digest in one’s spare time some possibly quite different
material in order to pass a largely orthogonal set of exams, is not necessarily
a bad thing. And certainly what I do feel is that if Besicovitch
was taking a less than ideal approach to teaching Hardy, the opposite, that of
regarding all interaction with students as being a preparation for their exams,
which is what I sometimes fear we are drifting towards, in the long run does
students no favours at all.
The folk saying is that when the policemen look young to you, you know
you are growing old. Perhaps the fact that I am disappointed by certain
attitudes these days is a sign that I am finally in my dotage. All that now
remains is for me to complain about the incorrect use of ’quartile’ when the
correct word would be ‘quarter’ and that people no longer know how to use
statistical tables.
I want to make it quite clear, however, that I am not complaining about
the mental ability of students these days, nor even
about their education at school. These may or may not be what they were but
this is not relevant directly to the argument, and I, unlike John Nelder, was a very average student and am in no position to
cast stones. It is not lack of ability I am complaining about but lack of
intellectual independence and responsibility. And in any case, I believe it is
not the students alone who are to blame; it is partly, at least, the
educational institutions.
Furthermore, to the extent that my unease is due to student attitudes, these attitudes are not
those of the majority. Many students I encounter continue to have an awareness
of their responsibility for their own education and will go on to make fine
statisticians. It is rather that there
is a significant minority who lack this sense of responsibility. Probably this
minority has always existed. The difference now is that their attitudes are
regarded by some segments of society as reasonable and we ourselves as
academics are not immune. I take a recent case where a university was being
sued by a student for failing to prevent his plagiarism
as a straw in the wind.
Let me give some examples of the erosion of self-reliance. Nowadays
students have extensive printed notes to cover the course. This is in itself, I
believe, a good thing or at least it is potentially a good thing. In mathematics
and statistics note-taking can involve much copying, and this is a
time-consuming but unexciting and un-stimulating activity that can interfere
with the business of actually understanding the material. Having notes provided
frees the student to concentrate on the lecture. However, increasingly I find
that my lectures are being assessed according to the degree to which they
conform to the notes. I, who regard my lectures as a pleasant and illuminating
variation on a theme, am having to learn from (some of)
my students that they are received as distracting noise.
The joke goes that when three students were asked to memorise the
telephone directory the mathematician said “Why?”, the lawyer said “How long
have I got?” and the medical student said “Will the yellow pages also be in the
exam?”, but these days I find that “Will this be in the exam?” is a common
question regarding almost anything I do, whatever the class. This is backed up
by an ever-increasing demand for specimen exam questions and model answers. I
had always refused to provide students with the latter, arguing that they are
for communication between examiners, not with students, but now they are
available to all.
When I was a student you were expected to read books. It was through
reading such books that you came to realise that a topic was not defined by its
notation since the same subject could be presented in initially mysteriously
different ways by different authors.
I can even remember the day I discovered by comparing two books that there
was no universal agreement as to how the sample variance was defined. One
author divided the corrected sum of squares by n and another by (n-1).
(I prefer the latter convention but there are contrary arguments in favour of
the former.) These differences led to apparently different formulae for the t-statistic, which were, in fact,
identical. At the time this struck me as being both unhelpful and something of
a scandal but in retrospect it was just part of my education. After all, the
form of the t-statistic we use today
is that promoted by Fisher, which in any case is not the same as that used by
‘Student'. Realising what are essential differences and what are not is part of
your education.
Since I opened by quoting Nelder on Hardy,
let me quote Hardy himself, from his brilliant eulogy to mathematics as a
creative discipline, A Mathematician’s Apology.
“I hate ‘teaching’ and have had to do very
little such teaching, such teaching as I have done having been almost entirely
supervision of research; I love lecturing and have lectured a great deal to
extremely able classes.” (p149)
I am sure that to many academics and many readers of RSS NEWS this
remark will be perfectly intelligible but I wonder how many of the bureaucrats
who now run our universities could make any sense of such a statement.
This is not to denigrate teaching. Teaching school-children is a
difficult task, crucially important to the well-being of our society and
disgracefully undervalued by it. But, "when I was a child I spoke as a
child I understood as a child I thought as a child; but when I became a man I
put away childish things." It is an irony of our society today that whilst
simultaneously increasing the degree of autonomy we allow young adults in the
social sphere, we place less and less responsibility on them for their own
education.
It seems to be increasingly the case, for example, that students bring
their parents along for interview at the university. If you do not find this
extraordinary, ask yourself whether you think their parents should accompany
them to the voting booths also and supervise the marking of their ballots?
University has become an extension of school. In fact we are moving towards a
society in which students must be allowed to address their lecturers by first
name, because they are ‘adults’, but their lecturers will have to be vetted
because they are in contact with ‘children’.
Part of the problem, I think, is that we are now encouraged to think of
students as ‘customers’. However, the current students on any degree course are
not the only customers for that degree. Universities are accrediting
institutions. They are thus maintainers of standards. The customers of any
university also include all past and future graduates. Consider how different
the response would be if you asked students if it should be made easier to get
an upper second or first but then explained that any reforms would only be
instituted to benefit future generations once the respondents had graduated. In
fact, Nobel prize-winning economist Clive Granger has suggested that we should
not be judging courses on feedback from students but from employers (Empirical Modelling in Economics, CUP,
1999)
Although I have some sympathy with this, I think it goes too far. If Nelder led me to Hardy, Hardy leads me to AE Housman. In A
Mathematician’s Apology, Hardy refers to Housman’s famous lecture at
Cambridge in 1933 The Name and Nature of Poetry and the outrage that he
personally felt, as a passionate advocate of creativity, at Housman’s assertion
that ability as a literary critic was more, ‘charily bestowed,’ than the gift
of poetry. Well, criticism is an important part of statistics and I have some
sympathy with Housman’s view that it is not necessarily easy, whether applied
to literature or to science.
However, there is one thing on which Housman and Hardy would have
agreed entirely: the value of a university education is not justified in terms
of its potential practical application. Compare Hardy’s, ‘I have never done
anything useful’ with Housman in his Introductory Lecture (1892) at
University College London: ‘The partisans of Science define the aim of learning
to be utility. I do not mean that any eminent man of science commits himself to
this opinion: some of them have publicly and scornfully repudiated it, and all
of them, I imagine, reject it in their hearts.’
As it turns out neither Hardy nor Housman was right in his claims for
lack of utility. Hardy did not foresee the application of his work in
cryptography and Housman, who could hardly have been expected to predict the
development of global positioning systems, referred to astronomy as a “squeezed
orange” as regards its utility to the art of navigation. However, they were
surely right in maintaining that the predictable potential application of a
university subject is not what constitutes its value.
Of course, I believe, as I am sure do many readers of RSS NEWS,
that statistics is a subject with considerable application and of great
practical importance to society. However, I also believe that it is an
intellectually stimulating subject of great beauty and some philosophical depth
that attempts to provide (with some difficulty and not entirely successfully) a
bridge between mathematics and empirical science. This is also what I want my
students to feel. There is a danger for
our subject if we stress only its utility. In claiming utility we are
incomparably better placed than the medievalists and yet I feel there are many
reasons why we should show solidarity with them against a previous Education
Secretary who took them to task for their lack of relevance.
So I would like to finish my senile grumble by taking some positions
that many will no doubt find outrageous but which I believe have some value.
However, as befits a statistician, my mind is not closed on the subject and I
invite readers of RSS NEWS to tell me that I am wrong.
· Content is of primary importance. Ways and means of delivering that content are not unimportant but they are secondary.
[Stephen Senn
is a professor of statistics at the