Information Technology Staff Load Calculation
A proposal
Heinz Schmidt
Created: 20.4.98
Last revised: 25.2.99
Acknowledgement and References
The following proposal is based on established load calculations that have stood the
test of time:
- Monash Clayton (specifically Econometrics), since at least 1994
- Monash Caulfield (specifically Software Development), 1995-1997
- UNSW, Computer Science, 1992-
- ANU, Computer Science, 1992-, revised 1997
- QUT, Computer Science, 1995-
Context
Both teaching and research load are curently computed in the faculty and partly
in the MUSIS system (student supervsion, timetabling information, School load
coordination etc).
Load predication, monitoring and performance improvement require a well-defined
load model, capturing load partly quantitatively (measured in hours per week)
and partly qualitatively (listing expected duties and roles).
The Monash Staff Engagement Profiles formally require listing teaching and
research student supervision load explicitly as the minimum quantitative
information required.
Ultimately, the allocations and workloads are the responsibility
of the Head of School. In most Schools, this is delegated to a load
coordinator (often per campus). Across Australia and in the faculty,
in most Schools load recommendations are participatory processes, with
groups (by year or area) feeding recommendations into the load allocation.
Fairness and openness of the process requires a well-defined target
(such as the number of subjects, or hours, or supervised students per staff
on average).
From the load formulas quoted from other universities, this number is
typically in the range of 17 to 23 hours per week, varying by year, student load, problems.
Staff generally accept that a perfect system is not possible, that the
quantitative measurements are relatively coarse grained.
Staff expect righly that special credits are given for high performance in
prior semesters such as extra industry projects, extra publications (DETYA classified),
or for staff development such as staff PhD or the induction of junior staff.
General requirements to a load model include
- transparent load in a collegiate system: everyone has access to
the data and knows what is being done in the school by who;
- load expectations: everyone (and particularly junior staff)
get a sense of the load expected in connection with certain duties;
- time management: for everyone (and particularly junior staff)
the load spreadsheet contributes to helping in time management;
- fair load: everyone can develop a sense of fair
load distribution according to well-understood default principles and values;
- traceability and accountability: the management of the school
knows what is going on and can reliably plan, manage and improve the processes;
- career development and documentation: the activities of
individual staff are documented and agreed and the documentation
supports the processes of staff development, appraisal and promotions.
The Load Model
A mixed quantitative and qualitative approach is proposed. For
example, some admin duties will only be listed by the tasks or roles a
staff has in the department and a estimated figure will be put next to
the overall admin duties, essentially distinguishing small, medium or
heavy admin duties (in sum) without attempting to do the impossible,
quantify the individual admin duties (at the task level).
Teaching and research (supervision) load and also certain administrative duties
are treated on an equal footing as hours per week.
(This differs from the old SD load model where during research rampup, researchers agreed
to weigh research less than appropriate to avoid a cultural break between teachers and researchers).
Teaching Load calculation
The basis for the calculation is a subject given 1hr/week full year
(i.e. running over both semesters). We assume that by default
some revisions will be necessary.
Large subjects, such as first year (or Java with 270 students in
1997), are accounted for by extra points under (subject)
administration, see high and medium admin load list.
Special allowance is given to new staff.
| Lecturing (based on 1hr/week full year) |
|
| Lecture new subject 1hr/week full year | 1.5
|
| Lecture with revision 1hr/week full year | 1.
|
| Lecture no revision 1hr/week full year | 0.9
|
| Tutorial 1hr/week full year | 0.75
|
| Subjects run as mostly seminars | 0.5
|
| Normal subject converted to reading | 0.1
|
| New staff allowance | 1
|
| Subject coordination add: |
|
|---|
| Medium subject (100-200) | 1.0
|
| Large subject (>200) | 2.0
|
| Admin assistance in medium/large subject (*) | 1.0
|
Administration
Administrative duties are simply categorised in light to heavy load. A list of such
duties is published. Such duties include year coordination, Honours coordination etc.
| High load | 1
|
| Medium load | 0.5
|
| Light load | 0.125
|
Administration duties are reviewed yearly.
High load admin items
First year core subject coordination;
BComp coordination; Credits committee; Curriculum coordination (incl.
new degree initiatives); Deputy HOD;
Faculty education committee; Faculty Research Committee; HOD (4x high);
Honours coordination; IE project coordination; Malaysia program
coordination; Major studies coordination; On-line course coordination;
OOSIG coordination; Postgraduate course coordination;
postgraduate research coordination (including FRC);
research projects and grants coordination;
Medium load admin items
2nd year core subjects coordination; medium subject coordination (>100);
Newsletter; Open day coordination; TOOLS organisation;
Year 1, 2, 3 coordination;
Light load admin items
SFTHelp coordination;
Research
For research supervision load we follow Prof. Max King's model (Monash Econometrics, Clayton)
equating an hour lecture over the full year with
the supervision, reading, joint writing etc in connection with one
equivalent full-time HDR student.
| F/T HDR (PhD or Masters Research) full year | 1
|
| F/T Honours full year | 0.5
|
| MComp (50%) minor thesis full year | 0.5
|
| MComp (25%) minor thesis full year | 0.25
|
What others do
Monash Clayton, Econometrics
Quoth Max King email, 1995...
From: Max.King@monash.edu.au
> # The teaching formula is fairly simple:
> # 1 point for each contact hour for an entire year
> # I.e. 2hours for one semester (half year) gives one point.
> # Tutorials get weigthed as 0.75 because of their nature.
> # 4th and 5th year classes get a weight of 1.25 because
> # of their more difficult content.
> # Graduate research supervision is 1 for full-time M.Ec. or Ph.D.
> # where the staff member is the only supervisor. Part-time is
> # .5 and shared supervision means the points get shared.
> # Kind regards Max King
Monash Clayton, Mechanical Engineering
Quoth John Sheridan email, Tue, 09 Mar 1999 19:50:04 +1100
The process is fairly transparent and open.
In general there is an attempt to balance all primary teaching evenly
i.e. all lecturing. Thus, most people in Mechanical Engineering lecture
in about 3 subjects, spread over the full year (26 weeks). The exception
is where people are part-time, in which case their load is divided by
their fraction e.g. if
they are half-time their load is divided by 0.5 (or X 2). The units are
"hours" over the year.
Weightings are applied if the lecturer is giving
the subject for the first time or it is a large class. The weightings
are 1.3, where a large class is considered to be one with over 100
students enrolled.
| Lecturing |
|
| Lecture regular subject 1hr/week full year | 1.0
|
| Lecture new subject 1hr/week full year | 1.3
|
| Lecture large subject 1hr/week full year | 1.3
|
| Tutoring |
|
| Tutorial/Lab 1 hour/week | 1.0
|
| Projects |
|
| Final year project supervision (1 student) | 1.0
|
| Final year project supervision (> 1 students) | 1.5
|
| Research Supervision |
|
| full-time higher degree research student | 1.0
|
We seem to do less secondary teaching now i.e. tutoring or
demonstrating in subjects we are not lecturing in. If we do, the credit
is 1 hour per hour in the class/lab. We give 1 hour/week for supervision
of final year thesis projects if one student is involved and 1.5
hours/week if there are 2 or more students working on the project.
Similarly we give one hour/week for full-time p/g student supervision
and half this if part-time.
Overall, the numbers can look quite horrendous if one does all this.
Typically, we had annual (26 week) load of 450+ hours i.e. over 17
hours/week, but this is with loadings and p/g supervision, which might
be considered research rather than teaching.
From information gathered in the US: MIT, Princeton, Michigan, Cornell
and Chicago all seem to have a typical load of about 2-3 subjects per
year. However, they have a different level of involvement with students
due to the greater concentration on "homework" and their use of Teaching
Assistants.
UNSW, CSE
The formula quoted below is still used at UNSW according to a
recent email:
From: Bill Wilson
Subject: Workload formulae
Date: Mon, 08 Sep 1997 09:29:14 +1000
To: geoffw@cse.unsw.edu.au, Heinz.Schmidt@fcit.monash.edu.au
Geoff: A couple of years ago while TimM was at Monash he contacted
me for a copy of the School's Workload formula as input to a possible
formula for use at Monash.
...
Heinz: we still use the same model. 1.0 points = 1 hour of work per
week. I'm forwarding your message, as you will have gathered, to
our School Executive Assistant, Dr Geoff Whale,
...
in case he can provide any further comments for you.
Regards,
Bill Wilson
UNSW formula, obtained 1995, dated 1992
In this formula a subject requires 2 1-hour lectures per week. Unlike
most of the other models the repeat lecture is credited as the basis
with 1 point per lecture hour. Minor revisions are explicitly factored
in under the default ``lecture'' which counts 2.5 hours per week.
Also tutorials are calculated at 1 point per hour, as opposed to the
0.75 per hour at Monash. The overall result is two-fold. Firstly,
lectures get a little more credit for the actual work outside contact.
Secondly, the target per staff may be a little higher than 20 hours
per week elsewhere.
| | Lectures |
|
|---|
| NL | Lecture (new series) | 4.0
|
| L | Lecture | 2.5
|
| RL | Repeat Lecture | 2.0
|
| PL | Parallel Lecture (0 load) | 0.0
|
| | Tutorials |
|
|---|
| T | Tutorial | 2.0
|
| TC | Tutorial, Consultation, Marking | 1.5
|
| | Labs |
|
|---|
| LS | Lab Supervision | 1.0
|
| LT | Lab/Tutorial (lab with marking or add'l
responsibilities | 1.5
|
| | Consultation |
|
|---|
| C | Consultation not subject specific | 1.0
|
| XM | Add'l marking not counted in subject | 1.0
|
| | Subject coordination |
|
|---|
| IS | Smallish subject (<100) | 2.0
|
| IM | Medium subject (100-200) | 4.0
|
| IL | Large subject (>200) | 6.0
|
| AA | Admin assistance in medium/large subject (*) | 1.0
|
| SA | School admin (*) | 1.0
|
| | Supervision |
|
|---|
| SU | Ungraduate thesis | 1.0
|
| SP | Coursework masters thesis | 1.0
|
| SR | Postgraduate research student | 1.5
|
| | Special Duties |
|
|---|
| SL | Study leave | 25
|
| ST | Study time (formal courses only) (*) | 1.0
|
| SD | Special duties as authorized (*) | 1.0
|
| | Allowances |
|
|---|
| NS | New staff allowance | 5.0
|
| NS2 | New staff allowance, 2nd session | 5.0
|
| E | Evening class allowance | 1.0 |
| RA | Research allowance per intl paper last year | 1.0
|
(*) any appropriate number of these may apply.
as of 20 July 1992
ANU
From my time at ANU (92-94) and discussions with ANU colleagues, only teaching
load was allocated by formula. The department is small and everyone is equally
into research, although some are taking a larger admin burden than
others.
At the time, on average, 2 subjects were taken each year, i.e. 1 subject per
semester, and admin burden was negotiated with the HOD in part
and in part discussed and agreed at annual retreats.
Since 1997 teaching load has increased slightly to three 6 credit point units per year average,
with approximately 20 lecture hours per unit per semester.
Major admin items are now figures in as well. So are honours and
research student supervision. Additional weight is given to units new to the
lecturer and (more weight) entirely new subjects.
Large units such as first-year have a extra post of subject coordinator,
which is credited appropriately.
From discussions with the HOD:
"Everyone is DEEMED to be equally into research, but in fact they are not equally so. The
discrepancy is now becoming so evident that I am bing asked to give
some consideration to weighting the previous year's research output
contribution and I propose to use a formula equivalent to that for research
quantum publications."
Major administration duties, such as year coordination are now explicitly counted.
Small admin tasks are not included yet.
The ANU formula used is a weighted sum using the following weights
(excluding the tricky bit of teaching release for percentage of CRC
research). These weights for lecturing load reflect the stable class
sizes in each year across the subjects and may have to be adapted with
changing student load / quotas.
| Subject (6 credit point, 1 semester, 20 lectures) |
|
|---|
| 1st year | 6.0
|
| 2nd year | 5.0
|
| 3rd year | 4.0
|
| 4th year (Honours) | 4.0
|
| Subject novelty |
|
|---|
| repeat | 0
|
| first time for this lecturer | 1.0
|
| new subject | 2.0
|
| Tutorials |
|
|---|
| Tutorial | 0.75
|
| Teaching administration |
|
|---|
| 1st year coordinator | 2.0
|
| 2nd year coordinator | 1.0
|
| 3rd year coordinator | 1.0
|
| Head of School | 10.0
|
| Supervision |
|
|---|
| Honours student | 1.0
|
| Postgraduate research student | 2.0
|
"The result is that most loads on this model are between 19 to 23".
Source: personal communication, 20 Apr 1998, Dr. Chris Johnson, Head of Department, ANU DCS.
QUT
As per fax tabled at CSSE Management Committee meeting, 20.4.98.
The School of CS is organised in teaching areas (Foundations, Systems, Software Engineering etc).
Areas are loosely (self-)organised, recommend lecture allocations, curriculum review, student surveys etc.
Ultimately the load allocation is the responsibility of the Head or delegate.
Load is measured in lecture hours per week. In addition the following
hours are calculated. Note that small numbers of students e.g. in
postgraduate units are to a fair degree compensated for b ythe nature
of the material which is typically more advanced. Funded small
research projects are credited by an appropriate small number of hours
as part of the load caculation.
The baseline per week is 15 to 17 hours.
| Subjects additional weight (hours) |
|
|---|
| New unit (prep) | 1.0
|
| Big unit over 100 100 | 1.0
|
| Big unit over 200 | 2.0
|
| Teaching administration |
|
|---|
| Area coordinator | 2.0
|
| Major coordinator | 2.0
|
| Supervision |
|
|---|
| Higher degree research student | 1.0
|
| Teaching Release |
|
|---|
| previous sem project | 1.0 per 3 students
|
| staff PhD | 2.0
|
| research project | 1.0 - 2.0 per project
|
Source: personal communication, 09 Apr 1998, A/Prof George Mohay, Head of Department, QUT DCS.