Logical Framework Analysis

Introduction
Logical Framework Analysis (LFA) provides a tool, among others, to assist in setting out the logic of the project intervention model in a structured and coherent manner. In so doing, it facilities specification of: (a) project objectives, along with the criteria by which these could be gauged; and (b) the assumptions, upon which causal linkages between various levels of objectives are premised. As such, the logical framework ("logframe") is especially suited as an aid to both project design (formulation) and monitoring and evaluation (M & E) work.

Key Elements of the Logframe
In its simplest form, the logframe consists of a 4 x 4 matrix in which the rows represent the different hierarchical levels of objectives (inputs, outputs, purposes and goals in ascending order). (See Figure 1 below.) The vertical logic is a set of means and ends interrelated by the if then condition. Thus:

(i) If inputs are made available, then outputs would be produced:
(ii) if outputs are produced, then the purpose of the project will be achieved, and so on.

An important element of the vertical logic is the set of assumptions, risks and conditions upon which the movement from lower to higher level objectives is based. For instance, even if inputs are made available, factors outside management's control could have an influence on whether outputs would be attained. Similarly, project outputs may not lead to purposes being achieved if certain assumptions (e.g. about people's behaviour, such as farmers' response, or market conditions) are not valid. The test of good project design thus rests in making the best possible judgement in the assumptions-risks-conditions set.

The horizontal logic of the logframe permits a statement as to the criteria and means by which project achievement and success could be gauged. This is done through the inclusion of OVI (objectively verifiable indicator) and MOV (means of verification) columns. The former should always be quantifiable in clear units of measurements (e.g. no. of hectares, tons of produce, dollars, etc.) which the latter should refer to either primary or secondary sources that could provide data of acceptable quality.

For initial establishment of the project logic, the 4 x 4 matrix generally suffices. But more detailed planning and M & E design would benefit from a 5 x 5 matrix, in which an additional row (project activities) and column (project targets, which may be physical or financial) are added. (See Figure 1.)

Summary of Definitions

Inputs - Resources necessary to perform project activities (e.g. funds, personnel, equipment, other materials).

Activities - Tasks and operations carried out by project personnel to transform project inputs into outputs (e.g. conducting farmer training sessions, preparation of training materials, designing small-scale irrigation systems).

Outputs - Specific results produced by the management and use of project inputs (e.g. number of: farmers trained, sets of training manuals produced, ha rendered irrigable).

Purpose - What is expected of a project in development terms as a result of the outputs produced. While purpose is the motivation behind a project's outputs, it falls outside direct project management control, and often equates with end of project status (e.g. improved crop yields/productivity, total production increase, area rehabilitated).

Goal - The ultimate objective for which the project is undertaken. Its realization depends critically on the interaction of various external conditions with the project's purpose(s), upon which project authorities have little or no control and equates with final impacts, which may sometimes be manifest outside the project investment period (e.g. improved incomes, nutritional status, decreased child mortality, environmental degradation stopped).

OVI - Set of criteria which demonstrates in concrete terms the results achieved. Each indicator must be objectively verifiable in that different independent observers would come to the same conclusion as to the status of achievement (e.g. number of sessions held, tons of seed distributed, percent of farmers adopting, ha of land reclaimed).

MOV - The means, methods, and sources which permit OVI to be measured in precise terms (e.g. examining monthly records of farmer cooperatives, reviewing quarterly reports of district extension supervisors, periodic or ad hoc field checks/farm surveys).

Practical Hints
In drawing up a logframe, the following points are noteworthy:

(a) Start by specifying only the first and last columns (objectives and assumptions) leaving the other columns and rows blank, using a 4 x 4 matrix.
(b) Add the OVI & MOV columns later, only after the project logic is already clearly established.
(c) The activity row and target column in a 5 x 5 matrix will be useful only if detailed planning of the project is undertaken. This will also be important for M & E purposes since monitoring cannot be effective unless project activities are clearly spelt out and targets are available against which realization can be compared.
(d) Do not confuse a logframe's outputs with agricultural production. The former can be any project achievement such as km. of roads built, number of people trained or hectares of land cleared/planted through direct use of project inputs. Agricultural production is often a purpose which may result, for example, from the effects of having trained a certain number (i.e. output) of farmers in an improved farm technology.
(e) The assumptions in the last column should always link one level of objective to a specific objective from the previous level. They should not be inserted at random.
(f) For project design work a number of iterations of the logframe is often necessary before arriving at the final one. Examining and re-examining the assumptions, risks and conditions, and reviewing whether the project scope is adequate for attainment of successive levels of objectives would be required.
(g) Team effort, preferably of an inter-disciplinary nature, and a certain amount of "brain storming" is highly recommended.

Figure 1:


See also an example of a logical framework matrix prepared for an animal health normative framework: