I think that Liz may be confusing two separate but related
projects. The Liverpool John Moores Project was based around
students returned to HESA as "leavers".  The results of
this project, and a related one based at Keele University,
were presented at a seminar in London in February of this
year.  Full details are in HEFCE Research publication 97/29
"Undergraduate non-completion in higher education in
England" published last December

The second project (the current one) involved a pilot group
of 6 institutions who volunteered to take part in response
to a general invitation issued by HEFCE in April of this
year. (It's not that I like work but my VC has a particular
interest in this area)  The pilot institutions were given
similar data sets to those issued recently together with a
set of instructions.  There was a single consultative
meeting with HEFCE in London at which it was agreed that
the instructions/details of the calculations needed to be
considerably expanded from those issued to the pilot group.
The pilot group also insisted that sector wide figures
should be provided for comparison purposes because we all
guessed that that would be the first question we would be
asked.

Like Liz I have only checked the unusual entries and have
assumed that normal progression students were correct.  I
have found that all those entries matched our HESA return
including the 72 students with a coding error on the HESA
return which has been correctly interpreted in accordance
with what we returned, even if that info can now be seen to
be wrong.  It took me less than a day to check the outliers
and in all cases I found the HEFCE stats to accurately
reflect what we had returned and to have accurately tracked
students who had taken a year out, even those who had been
given a new HESA ID on their return.

I believe that HEFCE are to be congratulated on the hard
work they have put into developing this analysis which
should put the end to all the arguments about "flunk
rates" in the PUSH Guide. It isn't a pure cohort analysis
but it provides a method of getting consistent figures
which can be used to compare institutions.  I believe
that this is something which HEFCE are being pressed by
DFEE to produce. HEFCE are fully aware of the difficulties
of comparing institutions in this way and were very careful
to ensure that even within the pilot group discussions no
institution was allowed to see data for other named
institutions, all tables and diagrams had had all
institutional names omitted.  I believe that HEFCE intend
to continue to develop this method of analysis of the HESA
data with the aim of eventually producing a theoretical
non-completion rate which will allow for the influence of
subject mix when comparing institutions.

Hope this clears up any possible confusion between the
projects

Ric Halliwell
On Wed, 26 Aug 1998 16:11:36 -0400 (EDT) PLDEHASL

<[log in to unmask]> wrote:

> Dear Jenny,
>
> Liverpool John Moores University was one of the pilot institutions. The
> original work took a year and involved questionnaires to students who
> had left as well as analysis of progression data. There were two
> research assistants employed to do the work.
>
> Like you I have the data from HEFCE on my desk. I do not propose to
> check all of it, I simply do not have the resources. I am checking all
> the dubious groups e.g we are recorded as having 59 students who left
> with uncertain qualifications. When I checked them they almost all
> turned out to be students who had failed their final exams.
>
> I also intend to spot check a sample of the total who are recorded as
> having left to ensure that they agree with our own records. If there is
> any systematic error I would expect to be able to pick it up. My
> opinion of the data to date is that it is not very different from our
> own internal figures. There is nothing unexpected so far, and HEFCE
> appear to have gone about identifying students quite carefully. I
> suspect that they may be understimating the number transferring to
> other HE institutions, but no method is foolproof and it is not data
> that can be checked by individual HEis.
>
> I would be interested in how others are tackling this issue!
>
> Regards
>
> Liz Haslam
> Planning and Development, LJMU
>
>
> On Tue, 25 Aug 1998 17:04:08 +0100 (BST) "J.C.Ovenden"
> <[log in to unmask]> wrote:
> >
> > Apologies for cross posting.
> >
> >  Interesting and welcome though these figures on ft uk-domiciled degree
> > students are, are any colleagues concerned at the amount of work involved
> > in detailed checking of the figures by 2nd October?
> >
> >  Can any of the six institutions involved in the pilot project comment on
> > the manpower input required to check data in the pilot phase of the
> > project?
> >
> > Perhaps it is just post holiday blues!!
> >
> > Jeremy Ovenden
> > Student Planning Data Office
> > University of Kent at Canterbury
>
>
>

----------------------
Dr. Ric Halliwell
Senior Assistant Registrar
Academic Office, University of Hull
Hull, HU6 7RX
01482 465948 (direct line)
[log in to unmask]



%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%