Visualising program logic: Two new graphic conventions

M E T H O D O L O G Y U P D A T E
Visualising program logic: Two new
graphic conventions
Doug Fraser
This article1 introduces two new kinds of graphic representation which have
been developed by the evaluation and policy teams in two Commonwealth
agencies, Environment Australia and Agriculture, Fisheries and Forestry Australia
(AFFA), to assist in understanding the logic of complex programs.
The traditional type of program logic diagram was intended to serve two purposes at
once. One was to provide a basis for detailed analysis of the program, the sequence of
actions and consequences, and the inputs and results required at each stage. The other was
to allow a reader who was unfamiliar with the program and its context to seize in a single
page the most important gist of what the program did and how it was expected to work.
Most people who have used the standard technique have quickly found that the normal
conventions serve both purposes adequately only when the program is a relatively simple
one – ideally, one with a single intervention strategy and a single output. As soon as the
program develops multiple strands, or has an outcomes hierarchy extending beyond a
couple of levels, the traditional boxes-and-arrows diagram falls down on both counts. On
the one hand, any account of the program simple enough to be captured in a single logic
tree on a single page misses much of the content and strategic detail, in particular those
aspects that make the program strategy
unique. On the other hand, the number of
Most people who have used the standard technique have
boxes and the thatch of arrows that
quickly found that the normal conventions serve both
connect them rapidly become so confusing
purposes adequately only when the program is a relatively
(often even to someone who knows the
simple one – ideally, one with a single intervention strategy
program) that the whole purpose of
and a single output.
diagrammatic representation is defeated.
Matters become several orders of
magnitude worse when one comes to deal with environmental and Natural Resource
Management programs of the kind which are becoming increasingly common in Australia
and elsewhere. Such programs use multiple strategies to address complex problems of
management and rehabilitation, involving a multiplicity of causative factors – biophysical,
economic, social and political – and have a variety of objectives, not all of them strictly
compatible. The difficulty is all the greater because in many such cases, however well the
program works, it is unrealistic to expect to identify any clear sustained impact on the
variables affecting resource condition (i.e. the true high-level objectives of the program)
until anywhere between 10 and 40 years after the cessation of program funding.
In such cases, it becomes all the more important to model the complex interactions as
rigorously as the science and the evidence permit, simply to enable program managers and
funders to establish with some confidence within the funded life of the program whether or
not it is on track to achieving its real objectives. To be sufficiently informative, such models
must necessarily be detailed and intricate.
Conversely, the complexity of these programs is such that even those who work in quite
strategic positions within them often have great difficulty in identifying where they fit into
the overall scheme, or exactly where their work leads, without some form of road map.
Imagine how much worse the challenge must be for a new minister coming into the
portfolio for the first time, for a community-based group trying to adapt to the new
program structure, or for a parliamentary committee attempting to come to grips with it in
order to assess its effectiveness.
At this level of complexity, it is no longer feasible to combine both purposes in the same
kind of model. The kind of model needed for detailed planning could only be described in
54
E v a l u a t i o n J o u r n a l o f A u s t r a l a s i a , V o l . 1 ( n e w s e r i e s ) , N o . 2 , D e c e m b e r 2 0 0 1

---

connected sentences, preferably in a matrix format
mind are development of major industries, and
familiar to users of Logframe or of Sue Funnell’s
urban renewal projects.
work. But for the purposes of enabling outsiders,
and even insiders, to visualise the program structure,
The purpose of the models
it was clearly necessary to develop a new kind of
These visual models result from an attempt to
graphic convention which forfeited any pretensions
address three perennial problems with the
to that kind of analytical precision.
traditional inputs-processes-outputs-outcomes
This article deals only with the latter purpose. In
model, which become particularly acute in such
evaluating these depictions, it is important to
contexts.
remember, firstly, that they represent relatively early
The first is that people who look at such a model
stages of a work in progress, and secondly, that they
almost inevitably confuse logical sequence with
do not purport to show the full workings of the
temporal sequence. Because C follows B follows A
program or provide a basis for precise modeling.
in logical order, it is all too easy to suppose that A is
Their purpose is solely to provide
two different ways by which
people can find their bearings.
Conversely, the complexity of these programs is...