 |
|
| |
|
Developing people for
a career in Government Statistics Pete
Collinson, (Office for National
Statistics) Contents
Introduction
Background
What the GSS looks for at recruitment
Gaps in skills offered by graduates
Possible solutions
Introduction
1. Central Government is the largest recruiter of statisticians
in the United Kingdom. The group of statisticians
responsible
for the production of the majority of
government statistics belong to the Government Statistical
Service (GSS). Within the
decentralised GSS there are some 5,000 people
working in 30 or so departments. Although this paper
focuses on
statistics graduates it should be stressed
that the recruitment schemes outlined below also recruit
graduates with different
backgrounds (for example general
administrators, economists and psychologists).
2. The intended audience for this paper includes lecturers in
statistics and undergraduates.
It aims to address:
Background
3. Statistics graduates may enter the GSS through three different
recruitment drives:
4. This paper focuses on the CSSB competition, being the
largest of the recruitment schemes, though some of the
conclusions
apply equally to the other two.
back to top
What the GSS looks for at recruitment
5. In spring each year the GSS organises its annual recruitment
of ASs and ST0s. ASs are members of the Fast Stream
whereas ST0s are not. The skills looked
for at recruitment are therefore different. As members of
the Fast Stream ASs
must show promise as statisticians and
managers and have the potential for rapid promotion. ST0s
are recruited for their
technical skills but must also have other
skills to be effective.
6. In July 1997 the GSS issued a competence framework (attached
to this paper for reference) which is used to guide the
recruitment and development of statistical
specialists across government. It describes the skills
needed to be a
government statistician but recognises that
some people will be stronger in some areas and weaker in others.
The work of
the GSS is wide and varied and therefore the
GSS needs people with different skills. To illustrate this
it is useful to
describe some common GSS roles:
7. Some statisticians do, of course, take on more than one
role. Most of the competences described in the framework
are
needed to perform any role but the
methodologist, for example, needs stronger technical skills than
the statistical
manager. Thus the framework can be used
by departments to specify the types of recruits they are looking
for and CSSB
attempts to match the people with these skills
to departments. In summary, the competence framework is a
detailed
statement of what the GSS expects from
graduates.
back to top
Gaps in the skills offered by graduates
8. There are many excellent applicants for posts in the GSS.
However, a large proportion fail to meet the required standard
in the selection process and some of those who
get recruited do not realise their potential on-the-job. If
we focus on the
1997 CSSB competition then there were 370
applicants of which the top 138 were selected for interview.
Of these 3 1,
were recommended for the AS Fast Stream and a
further 28 for the STO grade. This pass rate does not
compare badly
with other recruitment schemes but the number
of AS Fast Streamers recruited rarely satisfies the demand.
One possible
reason for nearly four-fifths of graduates
interviewed failing to meet the required standard could be that
there are gaps in
the preparation of people for a career in
government statistics - other possible reasons include the wrong
people applying
or unrealistic expectations.
9. The following focuses on gaps which could realistically be
filled during a university education. We would not expect,
for
example, graduates to have general management
skills (strategic thinking, decision making, business planning,
marketing,
quality control, delegation and empowerment,
and motivating others), though we would expect a graduate to have
at least
given it some thought.
Technical skills
10. Graduates are expected to have at least a second class degree
and are thus competent in statistical theory. Reports from
CSSB assessors (government statisticians and
university lecturers in statistics) however suggest that a
significant number
of graduates have difficulty with applying
statistical techniques to practical situations, do not fully
comprehend the
techniques they are describing and cannot
describe statistical techniques in the language of the customer.
Some also have
difficulty in presenting data to illuminate
key messages.
11. There are some technical skills which are not always
covered in university courses but are important to the work of
government statisticians. Some are
unique to government, such as national accounting, and
universities preparing
their students for a variety of careers would
not be expected to include such topics. Others are not
specific to government
such as index number theory.
12. Students can often choose course topics. Those which
are important to the GSS but not always covered by students are
time series analysis and surveys methods.
It would also be useful for students to gain experience of
working with large
datasets with incomplete fields, repeated and
inaccurate entries. Finally, only learning one statistical
software package
can restrict graduates when faced with a
variety of packages in the work place.
General skills
13. This is a broad category and the intention here is to be
specific. Presentation skills often gets quoted by managers
and
trainers as a development need for most
graduates. In recent years there has been a marked increase
in performance and
this may reflect courses including more
project work with students expected to present the results to
their colleagues.
14. People and consultancy skills are essential if
statisticians are to become valued by their customers and add
value to the
policy process. Graduates recruited to
the GSS report that they are unprepared for the demands of the
working
environment. Common themes are: managing
conflicting requests from customers and establishing customer and
project
requirements; establishing working
relationships; dealing with difficult interpersonal situations
and feedback; influencing
and negotiating with colleagues and data
suppliers.
15. Communication skills cover both written and oral
communication. The need for general development in this
area may
reflect the lack of practice due to the nature
of statistics degrees. For example those who apply for the
General Fast
Stream (recruitment of general administrators)
are often more analytical and better equipped to communicate
ideas in a
clear, concise, well structured and logical
way. This may reflect the greater emphasis on essays and
tutorials in some of
the degrees studied by these applicants.
But statisticians need these skills if policy customers are to
understand and value
their advice.
16. Self-development. The GSS has developed
mechanisms to support the development of its graduates but it is
the
responsibility of the individual to manage
their own career and development. This requires new
recruits to seek
responsibility, have drive and determination
and to work on their own initiative as well as becoming efficient
learners.
This message can come as a surprise and is the
focus of early training.
17. Broader perspective. Some graduates have a
narrow focus on their specialism and do not have a broader
perspective
and appreciation of current social and
economic issues.
back to top
Possible solutions
18. It is common for universities to concentrate on statistical
theory and its application. This is, of course, important
for the
development of the statistician's toolkit.
But the interpersonal and communication aspects are crucial if
statisticians are to
influence policy and get their statistics used
by society at large. Some universities are addressing this
need by introducing
project work, presenting data and report
writing modules, and one year placements. Here are some
other suggestions
which might help:
Summary
19. This paper has addressed the fact that the GSS is not seeking
to recruit a single type of graduate. The competence
framework is, however, a good guide for those
wishing to prepare students for a career in government statistics.
Universities do provide a number of excellent
graduates but there are a significant number who do not meet the
minimum
requirement. Most who do apply are
clearly competent in statistical theory, but lack the ability to
demonstrate that
they understand the techniques they are
applying and being able to communicate effectively with customers.
A possible
solution could be to combine theory and the
interpersonal aspects of being a statistician in degree courses.
January 1998
back to top
