What do we mean by appraisal?
The terms ‘appraisal’ and ‘evaluation’
are often used interchangeably. However, in this guidebook
we use them to refer to two different forms of assessment.
Appraisal is the ex ante process of deciding how well a scheme
or strategy will perform. Evaluation is the specific application
of appraisal to the ex post assessment of completed projects.
In both cases the question is: “How well does this scheme
or strategy meet the objectives which we have set?”
Appraisal and evaluation need to be conducted in a consistent
way. We consider evaluation in Section
15 .
Why do we need appraisal?
Choices in transport policy are rarely easy. One scheme may
offer a greater reduction in congestion, but at the expense
of the environment; another may be environmentally more effective
but restrict access for key groups of residents. Appraisal
is a means of assisting the decision-maker to make effective
choices between such options. As shown in the box, appraisal
can contribute to a number of the steps in the logical structure
(Section 6).
What is an appraisal framework?
It is essential to assess the problems, solution or strategy
being considered against the full set of policy objectives
(Section 7). Since these
objectives represent different commodities and concepts, it
is simpler to consider them separately. An appraisal framework
enables this; at its simplest it is a table in which each
column is a scheme or strategy option, and each row is an
objective. One option should be doing nothing, to provide
a base for deciding whether doing something is worthwhile.
Other options can be variants of a particular scheme (e.g.
light rail), different policy instruments (e.g. light rail,
bus service increases, fares) or different combined strategies.
How are objectives and problems reflected in the
appraisal framework?
All objectives should be represented in the framework, and
each should be covered in a similar way. One of the easiest
ways of doing this is to use the indicators selected for assessing
performance against objectives (Section
7). These also provide a means of assessing the scale
of problems, at least in aggregate. However, problems are
usually associated with particular places or times, and this
requires a more detailed framework. Treatment of equity issues
also requires more detail and disaggregation. A choice is
therefore needed on whether the framework is to be a relatively
simple, aggregate one for the whole city, or one which provides
more detail for different locations, times of day and groups
of individual.
How can an appraisal framework be used to help make
decisions?
Appraisal is a technical process, but it is also a key input
to participation and decision-making. Decision-makers can
simply choose between options using detailed appraisal framework
tables of this kind. However, the choices and trade-offs become
increasingly complicated as the framework gets larger. Two
techniques are available to help overcome this.
Cost-benefit analysis uses money as the comparator. Changes
in amounts of travel, travel time, accidents and the environment
are assigned money values, based on observations of the choices
which people make. Costs and benefits are each calculated,
relative to doing nothing, for each future year. The net benefit
is then discounted to the present day and summed over all
years in the appraisal period to give, as a single indicator
of performance, a net present value of the benefits. The appraisal
period will normally be longer than the plan period, to allow
for longer term impacts. A scheme with a positive net benefit
is worth building; the option with the highest net benefit
is the best. The main weaknesses in this approach are the
assumptions required to value attributes like noise and accidents,
the difficulty of appraising impacts on future generations,
and the fact that the final value appears to determine the
decision, rather than encouraging discussion.
Multi-criteria appraisal overcomes some of these problems
by allowing the decision-maker to assess the weights to be
assigned to different indicators, objectives and impact groups.
In this way, differing views on the relative importance of,
say, noise and accidents can be reflected. At its simplest,
this can be used to generate a single weighted score for each
option, allowing the user to decide which is the best option,
given the weights they have selected. It is also possible
to test sensitivity to variations in weights, and thus identify
the solution which is the most robust to changes in weights,
or which is least likely to be rejected by someone who takes
a very different view on the relative importance of the objectives.
ECOCITY developed a multi-criteria appraisal method based
on a mixture of qualitative and quantitative indicators and
benchmarks, which could be generated without recourse to models.
Further guidance on appraisal methods is given in the PROSPECTS
Methodological Guidebook.
How can appraisal methods deal with uncertainty?
Uncertainty can arise in the scenarios against which strategies
are being tested (Section 3),
the structure, timing and sequence of the elements of the
strategy (Section 11), the
predicted impacts of the strategy (Section
12 ), and the weights to be assigned to the individual
indicators (see above). One of the simplest ways of treating
appraisal is sensitivity and robustness testing. A strategy
is tested against variations in a scenario, or with the model
assumptions varied, or with differing weights. If the performance
of a strategy is very sensitive to these changes, it is less
robust, and therefore more risky. If other strategies (or
schemes) perform better than it when the assumptions are varied,
it may be better to select them. As with other elements of
appraisal, there are more complex ways of analysing uncertainty.
|