Contents
Background
The MEANS Project is
a two-year project funded by the Department for Education and
Employment. It is based at the RSS Centre for Statistical
Education in the University of Nottingham and has core partner
institutions in the University of Nottingham Trent, the
University of Sheffield and Sheffield Hallam University. The
Project Director is Dr Anne Hawkins and the Project Manager is
Peter Holmes.
It is one of many Discipline Networks in Higher
Education which have been set up by the DfEE to promote
better co-operation between employers and universities in the
content and approach to undergraduate courses. Work on the first
of these networks began in April 1995; the Statistics network
began its work in April 1996.
These networks are the latest of many initiatives
that have been taken in linking education with industry by the
Department for Employment over the past 10 years. The thinking
behind this latest initiative is that imposed change does not
work and that university lecturers are more likely to be
influenced by peers in their subject specialism. Although the
background to setting up the networks lies in employment
considerations, there is a broader view of what constitutes
proper education; there is not a narrow view of education being
merely training for work. Each discipline network is a network of
people from education and employment who are trying to improve
undergraduate education in their subject. Each network is trying
to build up information about resources that will help to achieve
this.
Aims of the
Project
The aims of the MEANS Project are to:
a. identify the statistical skills and knowledge
which are needed by people whose work includes statistical duties,
and by those who work with them;
b. identify examples of good statistical training/education
and assessment practice;
c. promote a closer correspondence between
training and assessment in higher education and employment needs,
based on the findings of a and b;
d. provide a forum for discussion, dissemination
and research collaboration, making full use of electronic as well
as traditional means of communication;
e. lay foundations for enabling the network to
extend into more higher education contexts where statistics is
taught.
Web
pages and contacts.
Our Web pages are being developed as part of the
pages from the RSS Centre for Statistical Education at the
University of Nottingham. They are at:
http://www.maths.nott.ac.uk../means.html
We are establishing an electronic network using the mailbase
facilities. To register on our mailbase send a one line message
of the form join means Annie Boddy to mailbase@mailbase.ac.uk (Replace
Annie Boddy with your own name.) The purpose of the
network is to facilitate the exchange of views on the subject of
the MEANS Project and to encourage
greater co-operation between employers, employees and academic
institutions to improve teaching, training and asessment in
statistics. This covers service courses and in-house courses in
statistics as well as specialist undergraduate statistics courses,
If you wish to have your name put on our mailing list, or to
get in touch with us for any other reason, then you can email:
Anne Hawkins at ash@maths.nott.ac.uk
or Peter Holmes at peterhomes@ntu.ac.uk.
Our postal address is:
MEANS Project, RSS Centre for
Statistical Education
University of Nottingham, Nottingham NG7 2RD
Tel: 0115 951 4911. Fax: 0115 951 4951
MEANS
Workshops - A
Meeting of Minds? Peter Holmes (MEANS
Project)
Employers, employees and university lecturers at
our three workshops - in Nottingham, Woking and London - have had
animated discussions on matching education and assessment with
employment needs. Full reports can be found on our Web pages.
Here is a brief outline of some of the main points.
Employers - assessing their needs
The employers were generally more interested in
attitudes and personal skills than in specific technical
competence. A good grounding in basic statistics was rated higher
than deep theory. Companies differed greatly, from those who
provide training to those who do not; from those who have
specific expectations about the statistical activities of their
employees to those where senior and middle management still need
to be convinced of the usefulness of statistics.
A common requirement of employees is that they
should demonstrate good communication skills - both oral and
written. Statisticians often need to work in a team with non-statisticians
and may also have to communicate in non-technical language to
clients. Interactions of this sort require a maturity of outlook
and the ability to ask intelligent questions and to express
complex ideas simply.
Employers welcome courses with more practical
orientations and placement years (in universities that have them)
were felt to be extremely beneficial. They are generally
disappointed at the lack of practicality in new recruits,
something that shows up particularly at interview. In general
employers are critical of sanitised and over-theoretical courses.
Employees - reflecting on their experience
The employees were encouraged to compare the needs
of their current jobs with their experience as university
undergraduates. Those who had had a placement year said that it
had been very useful. They also appreciated courses which were
genuinely practical. There was some disappointment with pseudo-practical
courses.
Some felt that the university first course had not
convinced them of the usefulness of statistics or given them a
sufficiently global overview. For some, this had come on MSc
courses.
Many recalled being spoon fed on their own courses
and being assessed for their ability to recall theory rather than
exercise their statistical skills and understanding. However
several appreciated the depth of their courses for helping them
gain important perspectives on their present work.
Very few had learned how to write reports as part
of their statistics course In fact, some had had to produce
reports but had not been taught how. Some had acquired the
necessary skills in other subjects such as Economics or
Management Science. All would have welcomed something more
positive by way of help to develop their communication and
consultancy skills.
Most recent graduates found that they were having
to apply their statistics in new areas, and there is a clear need
to learn on the job. Some had needed to learn how to learn, a
skill that they had not acquired at university.
University Lecturers - responding to the needs
Most of the university lecturers were aware of the
points being made by the employers and employees, and were trying
to respond. They were, however, very aware of constraints on
their freedom to act in this way - some of which are given below.
Outside specific practical data-handling courses
and projects, any practical work has to be geared to the specific
topic of the course (e.g. multivariate methods, time series).
Teaching practical and consultancy skills is very
labour intensive - unless students can do some self-assessment or
research students/staff can be involved in the process. One
department insists on groups of three students working round a
computer to encourage discussion and hence to improve students'
teamwork and communication skills. First year service courses may
be too early as they precede any perceived needs by the students.
However, such courses are often constrained in time and content
by the other departments' requirements for later years.
Some novel developments were described - see the
articles by Neville Davies and Richard Castle in this newsletter.
Some universities were using packages such as Maple and Derive to
enable mathematically weaker students to cope. The lecturers and
the employees had conflicting views on students' attitudes. The
lecturers detected a reluctance by students to choose practical
modules, or courses which incorporated the development of
personal interactive skills. They saw a trend towards students
choosing 'easier' modules to get a degree. The employees were
less convinced, asking could the lecturers do more to make their
courses interesting and show that they were important.
Constraints
1. Most specialist statistics courses are
associated with mathematics degrees. Mathematics courses are
largely linear and this expectation carries over into statistics
courses, sometimes where it is not appropriate.
2. Lack of mathematical ability in students means
that some do not appreciate mathematical explanations. This is
not completely solved by using symbolic manipulation packages.
3. Pressure of time. Practical work and computer
data analysis can be time consuming. Students may choose to take
perceived easier options. This is particularly the case in
modular degrees.
4. Size of classes. It is more difficult to run
practical and group sessions with large classes. One option is to
lecture to large classes and have smaller groups in follow-on
seminars.
5. Examinations and assessment. It is much more
time-consuming to mark a set of reports on practical projects
than to mark the same number of examination papers. One
possibility is to give students data to analyse before they come
into an examination. Another is to have students using computers
and real data in their examinations.
What can MEANS do?
Encourage co-operation.
Put employers and universities in contact with
each other.
Solicit data, open-ended problems and case studies
from companies and government sources for use in undergraduate
teaching.
Find out how software packages (statistical such
as Minitab, SAS, SPSS as well as symbolic such as Maple,
Mathematica and Derive) are being integrated into undergraduate
courses.
Identify, and circulate information on, innovative
assessment procedures.
Maintain contact lists so that educationists and
employers can communicate with each other.
Keep up a dialogue with those who left university
some time ago.
Improve employment prospects of statistics
graduates by encouraging universities to develop numerate skills.
Maintain a directory of contacts in industry/business/commerce/government
who have an interest in, or experience of, ways of matching
education, assessment and employment needs.
Have Web pages of case studies of effective
approaches.
Produce a list of people from industry willing to
give lectures/courses.
Find academics willing to spend time in industry.
Workshop
in Glasgow - 20 February 1997
John McColl at the University of Glasgow is
organising our next MEANS Workshop. It will
be held on Thursday 20 February in the Department of Statistics
from 10.00 am to 4.00 pm.
The aim of this workshop is:
To bring together academic teachers of undergraduate statistics
with employers of graduates who have to use statistics and some
of these graduates to: explore the expectations and experiences
of the different groups; identify any areas of mismatch; explore
ways in which any mismatch is being/could be met; begin to
develop case studies to give insight into the nature of the
mismatch problems that could themselves be used for possible
future teaching/training.
If you wish to come to this workshop please let
John McColl know. If you know of other people who we can invite
please also let us know. We are particularly keen to get
contributions from a wide range of employers and young graduates
recently started in employment. We would also like contributions
from those who have been involved in any type of employment/education
co-operative venture.
For further information contact John at John@stats.gla.ac.uk ,
telephone 0141 330 4749. The mail address is John McColl,
Department of Statistics, University of Glasgow, Mathematics
Building, University Gardens, Glasgow, G12 8QW
Jobs
Up for Grads! Anne
Hawkins (RSS Centre for
Statistical Education)
One way of finding out about what employers want is to look at
what they say they want. To this end, the MEANS project team has
been collecting job advertisements that have appeared on
electronic bulletin boards, in newspapers and in journals.
Clearly, it is not sufficient to rely merely on the wording of
initial advertisements, because these vary considerably in the
amount of detail given. Some, but not all, employers provide
further particulars for potential applicants. These might take
the form of a standard information pack that can be downloaded
from a web-site, or obtained through a personnel officer or some
other administrator. The amount and type of information available
to applicants is, however, particularly difficult to assess if,
as is a common practice, the advertiser invites informal
enquiries by phone or e-mail. It is, of course, not entirely
clear who benefits more from this - the employer or the
prospective employee - or for that matter when an 'informal
enquiry' ceases to be such and becomes part of the interview
process.
Because of the inconsistent approaches observed, a number of
the advertisers were asked to elaborate on their initial
advertisements. It is perhaps of note that a number of those
followed up did not feel that they had anything more to add. This
was sometimes the case even when the original advertisement was
very vague and lacking in specific detail.
It was fairly obvious that some employers were
reluctant to specify their requirements in advance. Sometimes
this was because they had not really thought them through. On
other occasions it was more a case that the employers had some
idea 'what they did not want' but otherwise were prepared to 'see
what came out of the woodwork'. In the best dictates of
occupational psychology, this rather haphazard way of recruiting
employees can not be considered satisfactory or desirable.
At the other extreme, there are prospective employers who
provide applicants with large information packs. This is
presumably in the expectation that, before attending for
interview, candidates will acquire an in-depth knowledge of the
structure and organisation into which they hoped to be accepted.
Other employers have systematic and extensive evaluation
procedures that include posing candidates practical problems
similar to those that might be encountered later in the work-place.
Sometimes, such problems are posed at the interview. However,
some employers require candidates to prepare suggested solutions
prior to attending for interview, or even to submit their
solutions in advance of the short-list being drawn up.
Our database now holds more than 200 job advertisements,
classified by educational and experiential pre-requisites, by the
type of duties that would be expected of the successful applicant,
and by any personal qualities or skills deemed necessary by the
advertisement. Although the data collection will continue, in
order to span any seasonal variations, interim analyses and
reports will be posted on the MEANS web-site, and a
final report will appear in the next edition of this newsletter.
To date, the bias has been towards collecting data about
advertisements for posts that explicitly mention statistical
duties as part of the job description. The next phase of this
investigation will be to consider the wording of advertisements
for jobs where the incumbent will clearly be required to
undertake work of a statistical nature even though the
advertisement does not make this explicit. Market research posts
could be one example of this.
Information has also been gathered about the sort of job
requirements that pertain for more highly qualified, or more
experienced, applicants. The rationale for including this is that
such information may provide insights into the upper limits
of expectation that employers have of recent graduates.
Not all recent graduates move directly into 'employment', but
instead enrol on courses leading to higher qualifications,
although sometimes such post-graduate studies are undertaken as
part of research assistant posts or the like. Attempts have
therefore been made to discover whether academic 'employers' or
supervisors are satisfied with the levels of knowledge, skills
and understanding with which their potential recruits (recent
graduates) present. This is an area where it is relatively
difficult to obtain information. Firstly, the advertisements for
such posts generally carry little explicit information, each
being assumed (not necessarily correctly) to be one of a
considerable number of similar studentships. Secondly,
recruitment is often carried out on the basis of one academic
personally recommending one of his or her current cohort of
students to a colleague who has secured funding for a particular
post. The 'selection' of a student is then made from a known
source of undergraduate training that is presumably acceptable to
the 'employer'. In any event, there is a natural reluctance on
the part of academics to question the level of undergraduate
preparation, when they themselves are not only responsible for
post-graduate supervision but also for under-graduate training.
Issues related to 'pots calling kettles _.' or 'motes and beams'
may be influential factors in this context. Perhaps, however,
this type of recruitment of a 'would-be specialist' for a 'specialised
position' simply tends to breed a certain air of complacency on
the part of the educator or supplier of the graduate, and the 'employer'
or consumer. Some interesting insights have been gained from this
part of the study, albeit that these are often of a more
anecdotal kind.
One of the things that is emerging from the investigation,
which complements information gleaned from MEANS seminars and
other avenues of enquiry undertaken by the project team, is a
perception of the degree to which employers are willing, or
expect, to provide in-house training for their recent graduates.
Another issue that is receiving attention is the variability in
the ways in which employers state what they are looking for. This
may indicate that in some cases job advertisements are a 'voice'
that is being under-exploited when it comes to the employers
telling the educators (and their prospective employees) what they
require.
Email
mentoring Neville Davies (Nottingham Trent University)
The original idea of using email mentors came from Janice Derr,
Statistical Consulting Director, Pennsylvania State University,
USA. It has also been implemented during 1996 by Dr Ian Gordon of
The University of Melbourne, Australia. Nottingham Trent
University has made email mentoring part of the second level
course Communicating Statistics in semester 1, academic
year 1996/1997. Course notes are written in the html language,
and management of the students' activities is done entirely from
my web pages. Details, including other activities that are part
of the Communicating Statistics course, can be found at the url
http://www.maths.ntu.ac.uk/nd/www/communicating/lectures.html
This shows the opening screen, giving access to the notes, url
links and course management details. Standard html files, with
hot-word dynamic links to related pages, can be accessed from
this web page.
The idea behind email mentoring is that students can make
contact with professional statisticians who have previously
agreed to allocate some time to the task. The only cost for such
professionals should be the time taken to carry out the
communication. The students are asked to find out the kinds of
statistical activities that their mentors get involved in,
including formal or informal consulting, special presentations
and generally communicating the subject to others. The students
are warned not to ask trivial or irrelevant questions and are
encouraged to keep regular contact. The activity is assessed from
a report that the students are asked to write up, paying
particular attention to the way they relate how the email mentors
communicate the subject to others.
I have a pool of around 20 email mentors from within the UK,
USA and Australia and pairs of students are allocated to each
mentor. I have tried to match student interests with the main
work of each mentor. For example, several students on the course
are studying for a degree in Business and Quality Management, and
their email mentors are in industry and/or work in the area of
quality management. The students were asked to make initial
contact early in the semester, to give plenty of time to allow
for electronic difficulties and so forth.
The html language allows direct emailing to hot-linked names,
and the students have access to a web page that lists all the
mentors and their allocated students. The following section gives
verbatim the web page that contains the guidelines that I have
used for the questions the students should ask.
Guideline questions to ask email mentors
This assignment requires you to make contact with a
statistical email mentor. These people are all practising
statisticians, and your task is to find out about their work, and
write a brief report about it (3-4 pages). You might ask them
about:
the nature of his/her work;
communication with clients;
how the statistician goes about problem-solving;
problems the statistician faces at work;
typical tasks, how the week is filled in;
the type and amount of non-statistical work done;
pluses, minuses of the job;
views the statistician may have about the nature of
statistical consulting;
technical issues;
anything else thought to be relevant.
This list is intended as a loose guide only and you can raise
any issue you like. The mentors have kindly agreed to do this,
because they see its educational value. But they are all busy
people, and you should not pressure them. They are not expected
to send you more than two brief messages per week. How you
structure your enquiries is up to you, but you should obviously
not simply send them the above list and wait for responses. You
should introduce yourself, explain your experience in statistics
so far, and take it from there. Some of the mentors may be
willing to meet you in person; you may request that, but please
be understanding if they cannot. The mentors will not be able to
discuss confidential matters. The write-up should be a synthesis
of what you have found out, based on the total email
correspondence. You should not simply cut and paste their answers
into a document. You may quote them, but only briefly and only if
it is particularly appropriate. You should regard the assignment
as an exercise in report writing as much as anything else. The
mentors will receive a copy of your assignment.
Measuring success
It is very early days to be able to make any objective
assessment of the success or otherwise of this project. However,
several pairs of the students have been visibly enthused by the
interest their mentors have taken in their studies. The students
on the BA Quality Management have been particularly keen on the
use of their email mentor. They are hoping that the connection
with Australia may even provide them with job opportunities! I
intend to make available on the world wide web all the reports
that the students write up. Naturally I shall send a copy of the
report to the mentors and ask their permission before making the
information publicly available on the internet.
My own feeling is that this part of the course will provide
valuable experience for the students in the practice of
communicating with professional statisticians. Hopefully this
will help the students themselves with the less than easy job of
communicating statistics
Choosing
the right sandwich filling Margaret and Peter
Rangecroft (Sheffield Hallam University)
Sheffield Hallam University has a long tradition of offering
sandwich courses, that is courses which incorporate a substantial
period in industry. What follows is a description of how this
course component is implemented in the degree programme in
Management Sciences, one of several such programmes in the School
of Computing and Management Sciences.
This programme consists of three named degrees:
Applied Statistics;
Business Systems Modelling;
Computing and Management Sciences
Each of these is a four year course, the third year of which
is spent in paid employment, known as 'the placement year'.
It is felt that this placement year is a vital part of the
students' education, providing them with valuable work experience.
In addition to workplace skills such as time-keeping, working as
part of a team, keeping to deadlines and communicating
effectively, they begin to get a feel for practice as well as
theory. We also find that when they return for their final year
in college they are not only better able to see applications of
what they are studying but are much more mature. All these
factors contribute to their employability at the end of the
course, which is probably one of the reasons why the School has
such an enviable employment record.
The students are placed in a wide variety of
establishments, both in the UK and abroad. These include:
| The pharmaceutical industry |
Financial services |
| The utilities (gas, electricity, water) |
Manufacturing |
| Software houses |
The civil service |
| The armed forces and police |
Research and consultancy |
| Retailing |
Within these companies the students' roles are many and varied.
They might include such jobs as collecting and analysing data,
mathematical and statistical modelling, programming (SAS, C++,
etc.), risk analysis, project management, logistics, design and
analysis of clinical data, to name but a few. The variety and
depth of experience is determined by the employer, as is the
degree of supervision they provide.
So what are the benefits for the employer? Many employers come
back to us year after year to recruit our students. They often
see the placement year as an extended interview and, if a student
is found to be suitable, may offer her a permanent job on
graduation. In addition, they hope that the student will bring
fresh ideas. For small companies in particular, short term
contracts such as these offer a great deal of flexibility. The
fact that the rate of pay for a placement student is generally
lower than the graduate equivalent is an added inducement. For
the student, however, earning a salary instead of living off a
grant is riches indeed. Many take this opportunity to try to
become solvent!
Placing the students in suitable employment is the job of the
placement unit, a team of academics and administrators with
special responsibility in this area. In addition they, with the
help of other academic members of staff, supervise and support
students during their placement year. This involves at least two
visits to the student's workplace to meet with both the student
and her line manager, once early in the placement year and again
towards the end.
With increasing competition for employment, we have found it
advantageous to spend time and effort in preparing the students
for placement. During their first year in college they will have
learnt the rudiments of CV writing and this is followed up early
in the second year by further work on CV's and interview
technique. The process of applying for, and getting, a placement
is ongoing throughout the second year. The effort invested in
this process has to date paid off, with all students finding a
placement. However, we are not complacent. Although most
companies come back each year for placement students, the
increase in numbers of students and the desire to widen the
student experience mean that we are constantly seeking new
placements. So, if you think that you could offer one of our
enthusiastic and able students a placement with your company give
us a ring on 0114 2533149 and we will do our best to help you.
Being
interviewed for the Government Statistical Service Andrew
Ledger (National Heritage)
The Government always needs statisticians - you
can do that if you can't decide on anything else". This was
the worldly advice which I was given while in my first year at
University by a wise final year student.
Having finished with education in 1991 - studying
a BSc in Mathematics, Operational Research, Statistics and
Economics followed, by an MSc in Econometrics and Mathematical
Ecomomics, I ended up thinking about a career towards the end of
1992. Working for the Civil Service did hold some kind of appeal
for me as I started to look for jobs and adding to that the fact
that in my academic area (mostly statistics and economics) the
Government seemed to be the largest recruiter - the decision was
made. I applied for the entry scheme in October 1992, having
filled out an application form which was obviously designed to
test dedication at an early stage. The recruiting process meant
that, after this initial hurdle, those successful were invited to
the dreaded Civil Service Selection Board (CSSB) in February/March
1993.
The CSSB consists of a day and a half of
interviews, tests and group exercises. Having recently been a
guinea pig for some of last year's sections, I can confirm that
they are clearly carefully thought out. However, many of the
exercises are highly artificial - e.g. put five strangers
together and get them to discuss a topic which they have probably
never even thought about before, while being scrutinised by a
panel of three interviewers. Despite being artificial, this kind
of exercise can help to tell a lot about people - some people
naturally assume control (which may or may not be a good thing),
while others meekly wait for their turn to make a point,
sometimes a key point which others have completely missed.
The individual tests were very wide ranging - from
rearranging bits of a sentence to make sense, or spotting the
last in a sequence of dominoes, right through to having to make
recommendations on a proposed statistical survey. Most of these
are tightly time constrained such that candidates can never fully
complete and check everything; a good thing for forcing people to
make value judgements, but in my view suffering from the same
problem as some exams - he who writes quickest has an advantage.
The parts of CSSB which everyone was dreading were
the interviews - one classed as statistical, the other as general.
For me, I decided that honesty was the best policy and began my
statistical interview by confessing that the statistics which I
had studied (mainly in 1989-ish) was mostly just a distant memory.
This turned out to be a good move, since I was asked just very
general questions whereas those who went along trying to prove
they knew lots of technical statistics were given a hard time.
Then, I did wonder whether this might just result in a large
black cross against my name, but presumably having the right
qualifications counts for something, and, as long as the
interview confirms that the candidate does have the background
understanding which the degrees imply, then that is fine.
My impression was that the final, general,
interview is the most important part of the whole process. I find
it hard to believe that a series of artificial exercises and
disjointed tests can tell as much about the suitability of a
candidate as an hour long chat with them. This interview covers
general discussions about topical issues (including one with a
statistical theme) and tests the key abilities to think quickly,
see both sides of an issue and follow lines of argument to a
logical conclusion. Thus, I ended up discussing whether
advertising for smoking should be banned, whether parents should
be able to choose the sex of their babies, and many other
genuinely interesting subjects.
My overall conclusion from the CSSB was that
statistical knowledge was not perceived as being all that
important and common sense appeared to be much more valued. I
imagine that the reason for this is basically that the entry
requirements already ensure that candidates are, if not
statistical experts, at least highly numerate. I would also add
that the posts which I have subsequently held within the
Government Statistical Service have suggested that that is the
right judgement. At times I have lacked the statistical knowledge
which people would expect of somebody aiming to become a
statistician - but these times are rare. It is much more common
to have to work out what is going on with some figures - where do
they come from and roughly what do they mean is often more
important than elaborate detail.
I should finish by giving a reassurance that the
GSS is not just something which people should join in the absence
of alternatives - it offers a uniquely diverse range of jobs and
the chance to have a say in the areas in which you want to work.
I am grateful for my third year friend's advice because I think
it probably did stick with me at the back of my mind, but a GSS
career is certainly more than a default option.
The
best kept secret Richard Castle
(University of Brighton)
The Teaching
Company Scheme (TCS) seems to be one of the best kept secrets.
Outside university engineering departments, few academics have
heard of it or appreciate the benefits and opportunities it
offers; yet it has been in existence for 20 years.
The TCS is a Government-sponsored scheme which actively
encourages partnerships between universities and businesses which
will strengthen the competitiveness and wealth creation in the UK.
The scheme works by supporting two year projects designed to
realise a company's aims. Traditionally, many of the projects
have been based on engineering problems, but the scheme is not
limited to such projects. There are many oportunities for
statisticians to be involved in projects such as quality
assurance or pattern recognition.
A company must identify a development which is strategically
important to their future, which requires skills that are
currently unavailable within the company. They would then contact
a university department which can offer the necessary skills.
Together they would draw up and submit a proposal to the
Government. Successful applicants can then appoint a high calibre
graduate (teaching company associate) to work full time in the
company who will be supervised by industrialists within the
company and academics. An initial two-year scheme is funded
through a budget worth £65000 for each associate employed. Small
companies attract 70% funding from the Government, leaving the
company to find pay the remainder (less than £10000 per year).
The funding includes monies to pay for the skills of an academic
on a consultancy basis (25 days per year).
The scheme is generally considered an advantage to all the
parties involved. Naturally, the company acquire new skills at a
greatly subsidised rate. The associate benefits from working on a
high profile project, and has a special training allowance
included in the budget. Furthermore many associates progress
within the company after the project has ended. The academic
benefits by learning to apply his skills on real and original
projects. The Government considers the TCS a success because it
generally leads to increased profits and often to increased
employment.
At present, I am the academic supervisor for a Teaching
Company project at Drallim Industries in Bexhill. The company is
developing an instrument for monitoring pneumatic machinery which
can predict machine wear before there is catastrophic failure.
Naturally, this instrument generates large amounts of data, and
pattern recognition techniques are used to monitor the machinery.
This requires both mathematical and statistical skills which,
prior to the TCS, the company did not have in-house. The Teaching
Company scheme is proving an excellent means for acquiring the
necessary skills.
To
make you think. Peter Holmes (MEANS
Project)
The MEANS Project
is only one of many Discipline Networks on Higher Education. Some
started a year earlier than ours and the DfEE published a paper
on Issues from Round 1. Here are some quotations -what do you
think?
Students, and sometimes staff, are
increasingly being driven by the need to obtain paper
qualifications. ... The development of strategic, rather than
deep, approaches to learning, militates against the objectives
which we share ....
Much innovation is seen as an additional
burden on top of an excessive teaching load. Sometimes the
rational solution is less teaching and more guidance or
assessment, but often this is beyond the control of the
individual academic, and seen as inappropriate by one's peers.
At times the mismatch between assessment
processes and objectives is striking: graduates enter the world
of work on the basis of a qualification awarded largely on the
ability to write essays on unseen questions: a task almost no one
needs to do in the workplace. However, alternatives - self
assessment, group assessment, project reports etc are rarely used
because they are seen as unreliable or unduly time consuming. To
sacrifice validity in the interests of reliability can hardly be
commended, but is the basis of much practice.
Academic journals are a major influence on
academics' values and ideas, but it is clear that many are very
reluctant to devote space to pedagogical issues.
Do you agree or disagree?
If you agree what can be done?
Write to us at the office in the RSS Centre for
Statistical Education in the University of Nottingham
or
join our electronic mailing list and send your
message to means@mailbase.ac.uk
or
email us at: peterhomes@ntu.ac.uk.
To see some of the discussion on the mailbase discussion list,
check out the archive.
Some
comments from employers and employees
At the present stage of the project we have seen
only a few employers and they have been mostly from industry -
not from business, commerce or government agencies. here are some
of the comments we have received:
Planning, design of experiments and correctly
interpreting data are important skills.
Employees need to know how to use statistics in
improving quality.
They need to be able to communicate with non-statisticians.
The need an appreciation of how research,
scientific and in statistical methods, is done.
They need a good perspective on the interplay
between data and model.
(On introductory courses for biologists). They
are often taught before the practical importance of statistics
can be appreciated ... by the time a scientist embarks on a
research career much of this reluctantly learned and partly
understood material is forgotten.
Students do not help each other, they do not
exchange opinions about what they are doing. Each works in a nice
isolated cell. They only care about meeting their own deadlines
and scoring the best marks (i.e. competition). They are not used
to working as a team, to accepting responsibility, to working to
a deadline. They often suffer a general lack of communication
skills. They are not prepared to interact with non-statisticians.
Statisticians need to work together in teams.
As with all generalities there will always be
exceptions and some of the above comments may seem overstated.
Nevertheless these comments have many implications not only for
our current undergraduate courses but also for the way we assess
the students on them.
Here are some comments from employees.
These employees come from the same companies as
the employers quoted above.
My undergraduate course was very dry. I was not
helped to establish a problem as a statistical problem; I was
taught to solve statistical problems.
The undergraduate course should give students
problems to solve and fit in a real environment.
I needed to learn how to communicate "This
is what statistics can do for you".
Exams are too much on regurgitation. We must
assess "Can they do it?".
There was no consideration of ill-defined
problems or development of listening skills.
I needed help to learn how to think round a
problem.
' A good sound theoretical background is
important.
I would have found it helpful to have been
exposed to practical problems.
I had a surprise when I started as a statistical
consultant. I knew how to do an ANOVA, but I didn't know I didn't
understand the ideas behind it until I had to explain to clients.
The biggest problem I have had is in learning how
to explain statistics to both statisticians and non-statisticians,
so more interaction in a course would be helpful.
Back to MEANS Homepage
Please email rsscse@ntu.ac.uk
with any comments or corrections.
