 |
|
| |
|
University Teaching and Assessment Peter Holmes (MeaNs Project)
Most of the information used in this talk comes from the MeaNs workshops or survey of
undergraduate courses
Contents
What Universities are aiming to do in their
statistics courses
a. Main Courses
b. Service Courses
c. Special Courses
d. Employer Input
Specific and Transferable Skills
a. Within the university
b. Within Employment
Teaching and Assessment
a. Some techniques of assessment
b. The General Picure
Some Constraints on changing current courses
What Universities are aiming to do in their statistics
courses
1. Main Statistics Courses
There is a difference between specialist statistics courses and
service statistics courses. The purpose behind the main
statistics courses reflects the diversity of paths that graduates
can follow. One university summarised as follows. Statistics and
mathematics graduates have a wide range of possible careers. The
main ones are research, medical, actuarial, government, pharmacy
and teaching. The specific requirements in these differ so the
department needs to look at the general skills needed. So we look
for a strong theoretical foundation. Statistics graduates need to
do much more in statistics than, say, the psychologist. But it is
also important to emphasise the applied side, so they do project
work, laboratory work and work in teams to solve problems. In
some specialist areas there are specialist courses. Psychology
includes one third of a year in statistics. They use a computer
package, emphasise collection, design and interpretation.
The need for theoretical background was confirmed by a young employee in the pharmaceutical industry and by others in our workshops. a good theoretical background is essential in the undergraduate course. Others said that if students don't learn theory at the university it will not get learned. You may only appreciate the significance of theory after you start work.
But also, time and time again there were references to
practical work being included in the university courses. All but
three of the respondents to our university survey claim to have
courses offering practical work or solving real problems. And
many of the graduates expressed appreciation of this when it was
there and disappointment when it was not.
At my university I was taught from examples and used computers.
This did motivate me, and I did have to mix with other students.
More students chose the practically oriented courses than chose
the core mathematical theory He would have liked his
undergraduate course to have had more working through examples
from the beginning. He had not been taught much on the
presentation of statistical data. One lecturer said that these
things were part of the courses at his university, but there was
a problem with the students' poor quality of English.
There are difficulties in including open projects and practicals in all courses because university courses tended to have specific labels (multivariate methods, time series etc.) and the real life problems included in these would have to use the techniques being developed in the course
University courses are also dependent on the students they can recruit. One university representative said that students there generally do not like number crunching so the emphasis now is on the ability to take and use data, use commercial software and then communicate to non-statisticians. Another commented that with growing computer power, there is a need to emphasise interpretation rather than calculation
Just as universities are dependent on the type of student they attract, so students are dependent on the type of lecturer they get. One young graduate commented that she felt that the lecturers were in a rut, with little enthusiasm.
back to top
2. Service Courses
Service courses, where still done through the statistics
departments (and often they are not), cause concern to those
involved. Most of the evidence we have received on them has been
negative.
The major problems here are motivation, attendance and modularisation. Statistical methods need to be a core part of the course - if they are optional they will be avoided. An employer queried the value of the 10 week service course -The timing of service courses is important; in many cases they are too early. But first year service courses are often constrained by other departments' ideas of what should be in second year courses.
Social sciences, many statistics courses are more mathematical than the students expect.
back to top
3. Special Courses
Some universities are responding to employers' needs by setting
up very specific courses (even more at the postgraduate level.
Fifteen of the responding departments do offer courses aimed at
specific areas of employment - but they do not necessarily have
great contact with employers in that field whilst developing them.
As examples: Strathclyde have a Statistics in Industry course while St. Andrews have a new degree available in Quantitative Ecology. Generally, the main effect has been to provide new ideas for projects and case studies. De Montfort University Medical and Health Statistics, Industrial Statistics. The context here is to get what companies want and use industry to validate. We are told that it is not always easy to recruit students for these courses.
back to top
4. Employer input
a. to courses
Fourteen of these (36 responding out of 85 contacted) departments
say that they have developed courses in general consultation with
employers. Although 15 departments reported ongoing contact with
local employers, little specific change in content was reported.
There is some evidence of greater input and co-operation in a few universities.
b. to students
The main input to students comes through placement and is widely
welcomed and found beneficial. The placement of students in
industry is relatively widespread with around half of those who
replied saying that this was a practice that they used.
One workshop participant commented that if done well the placement year is more than just a time shift. The tutor has an important role in making it part of the learning experience. There is also a spin-off into third year courses where problems from the placement year can be used. It is in the placement year that many students learn to work under pressure to time restrictions and have to prioritise their work. There are more comments below about the use of placement in developing transferable skills and developing maturity.
back to top
Specific and transferable skills
Wherever we have gone to employers of statisticians we have heard
a plea for more rounded students having general skills in
communication (including reading), teamwork, numeracy, logical
thinking, practical problem solving etc. as well as personal
qualities such as enthusiasm. There has been less emphasis on the
need for particular statistical skills - though it may be in some
cases, that these are taken as read. Universities are
trying to take this on board, and develop these skills as well as
the specific statistical skills needed. But there are
difficulties.
One young graduate commented that it was the practicalities of the job that were missing. These include: discussion with others; having to use SAS; communication skills; why statistics is useful to solve problems.
back to top
1. Within the university
General
Almost all departments provide courses intended to develop
communication, problem solving, interaction and personal skills.
Usually this is by using project and coursework, often group-based.
Only a few say that they specifically teach these skills; it may
be that most require the students to do the type of work where it
is hoped that they develop these skills without being
specifically taught them. Similarly, most departments claim to
develop collaboration and interaction and personal skills and to
enhance the ability to read and interpret published results from
other contexts. However, only about one third of the replies
suggested courses were available which were aimed at working with
non-statisticians.
Input from employers in the form of data or real problems for teaching purposes was reportedly used by nearly all of the respondents, often for projects or case studies.
Courses that develop skills in statistical consultancy generally appear to involve solving real life problems. Approximately a quarter of the respondents claimed that their courses did in fact develop this skill. Areas of good practice include written communication; group projects with peer assessment; case studies; operational research output interpreted; 5 minute interviews on interpreting computer output.
The lack of courses that developed practical and problem solving skills was often mentioned by employers. In these courses there is little on problem solving. I would like the students to develop interviewing and consultancy skills - particularly in questioning - and learn general methods of thinking through a problem.
Integrated or modular
If we are to teach transferable skills should we integrate them
into the whole of the statistics course or can we put them
substantially into a module in that course (e.g. a practical/project)
or do we make them more context free and have them as part of
general university teaching , not necessarily taught by
statistics lecturers. Integration may not be easy, but
modularisation also has its difficulties.
There are real practical difficulties in building cross-subject co-operation into undergraduate practicals. There are also associated assessment problems - but the experience itself is very important.
There were many comments about the difficulties when the project was an optional course. Students work on individual and group reports - if courses requiring this are optional then students do not choose them. Another university also said that they have an option on consulting but students do not choose to do it. Separately we have heard this comment from many universities.
One university said it was trying to incorporate transferable skills, including group presentations, into courses. There is resistance to this by the students even though this was done as a possible improvement on a previous 'Personal Skills' course.
Another commented that a good course on learning helps you identify your own weaknesses. Students can easily pick up bad habits. Some thought it best to build interpersonal skills into beginning courses and continue throughout. Students from other disciplines could be used to check the ability to communicate.
One example of integration was from a university which had all students on the Applied Statistics degree take a unit in the first year designed to enhance their communication skills and their knowledge of business, set as far as possible in the context of the sort of environment they are likely to find during their placement. The statistics units also include oral presentations.
Cross discipline
There were some suggestions that it would be useful for
statistics students to be involved in problems that arose in
other disciplines. It would make them more aware of the effect of
context on getting, analysing and drawing inferences from the
data. Those who had tried this were more aware of the
difficulties (theoretical as well as organisational).
Group projects and interaction with departments such as psychology were frowned upon by the Mathematics and Computer Science divisions of one department meaning that, as a result, such things were becoming rarer
back to top
2. Within employment
One Scottish university places students for 3-4 months
between the third and fourth years. It is possible that the
Office of National Statistics and other employers could use
summer students like this. As part of the statistics course,
four times a year students had to do a practical, report on it
and receive feedback. Report writing skills were developed in the
context of the data with not too much detail.
The demands of an employer can come as a shock to postgraduates. It would be useful if lecturers became more aware of the skills and practices used by employers of statisticians, so that they can better prepare their students for entry into the workplace
The University of Surrey, amongst others, runs a sandwich
course with a year out between years 2 and 3 of the course. The
find that students come back much more mature for their year out.
They are better at report writing and at seeing into a problem.
The record of getting students into employment from these courses
is very good. Another discussant said he had a very positive view
of sandwich courses. The extra maturity helps students improve
their final grade.
back to top
Teaching and Assessment -
Competition and Co-operation
The more we have discussed such things as the importance of
transferable skills, with the need to communicate, work in a team
etc. the more we have come up against a conflict with assessment
and the need, at the end of the course, to grade a student as an
individual. There is a real conflict which has teaching aimed at
getting students to be interested in the subject, learn it for
its own interest and the student looking at the form of
assessment and saying 'but is it on the examination?' A similar
conflict lies in the use of practical and project work in teams,
developing communication and team skills with the assessment of
that work and the individual feeling that they are in
competition when it comes to marks, though the activity is
designed to develop co-operation.
We had many comments on aspect of this during the project.
back to top
1. Some techniques of assessment
Here is a selection of comments we heard on this topic.
We need to teach our students to think and make sure our assessment is not just a test of memory. Teaching for a common sense approach to data may well be more time consuming. Theory is best tied to data, so that students know when it is and is not appropriate, and balanced with practical applications.
Motivation is strongly linked with assessment. Visits have to be written up as an assessed report.
We have computer laboratory work and a final year project using real data which many students do not like. Many are happier with a fairly dry theorem/example approach.
Students can learn by having to explain. Could we do this in an undergraduate course by having groups of, say, five explain to each other. We might have to disentangle this from assessment.
We are experimenting with different forms of examination questions e.g. comment on two different analyses of a set of data
In the first year course in Pure and Applied Statistics there is group discussion; students present material to each other and mark each other. Some of the staff feel that there is a need to have the students working in groups and then the students share out the group's mark according to the degree to which each person has contributed. There is resistance to this from some other colleagues who argue that this is not statistics and not mathematics
One graduate said that after a practical his presentation of findings was video recorded as part of the assessment. He found his course helped him to become a problem solver rather than a statistician.
back to top
2. The General Picture
Our survey of undergraduate courses showed that the methods of
assessing students’ statistical competencies were generally
the conventional formats expected. Most mentioned the use of
examinations, coursework and project reports. The projects could
be by individuals or group-based. Also included as the bases for
assessment were posters, presentations and case-studies together
with laboratory examinations, 'take-away' assignments and
portfolios of practical work.
The methods used for assessing transferable skills were along
the same lines. They usually involved projects, posters and
presentations but also included integrating the assessment within
statistics assignments. Only one department mentioned using an
examination paper where the students had to comment on a
published paper in a professional journal and on reports of data-analysis.
Students at Sussex carry out group exercises which then include
peer assessment. We have also heard of cases where peer
assessment requires the students to share a total mark between
individuals in the group. Students at Bournemouth simulate
business exercises and these are also assessed.
back to top
Some constraints on changing current
courses
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 a mathematical level of
understanding. This
is not completely answered
by using symbolic manipulation packages.
3. Pressure of time. Practical work and
computer data analysis can be time consuming. Students may choose
to take the
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 in large classes. One option is to
lecture to large
classes and have smaller
groups in seminars.
5. Examinations and assessment. It is much more
time consuming to mark a set of reports on practical project 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 data in their examination.
