Farmer participatory research (FPR) is difficult to monitor and review because it uses a 'process' approach; it is done in variable, unpredictable situations; it produces some outputs that are hard to measure objectively; and it involves different types of stakeholder, each with their own aims and ideas of success or failure. For these reasons, FPR projects tend to have permeable boundaries, with sometimes important spillover effects, and less than direct relationships between inputs and outputs.

This paper examines how far a conventional project management tool, the logframe, can be adapted to the monitoring and review of FPR. Normally used in 'blueprint' projects, the logframe presents some difficulties in handling those with 'process' characteristics. But it has a number of strengths: it requires clear specification of purposes, anticipated outputs, activities, and the relationship among them, as well as performance indicators and means of assessing them. Also, it is becoming almost universally adopted by funding agencies, so organisations using FPR may in future have to structure their proposals and monitoring activities in logframe format.

FPR aims to achieve one or more of three outputs using participatory methods:

• to develop improved agricultural technologies in response to farmers' needs;
• to develop the human resources of the farmers and collaborating organisations;
• to develop the institutional capacity of farmers' groups and collaborating organisations.

For the first of these, the relationship between the project activities and its outputs is fairly direct, making this fairly easy to monitor. It is less direct with the second and third, but can still be captured by suitably adapted project management tools.
Because of their mandates and philosophies, NGOs and public sector organisations differ in how they view 'participation'. Government agencies are concerned largely with (i). NGOs, on the other hand, are interested in participation mainly as a way to empower the poor. NGOs are therefore concerned at least as much with (ii) and (iii) as with (i).

This paper offers a generic logframe that readers can adapt to suit their own FPR projects. For each output, it provides illustrative performance indicators that might be relevant to each of the two most important groups of stakeholders (researchers and farmers). The paper also suggests some means of verification for each of the indicators. Readers should select, adapt and add to these to suit their own situations.
An overall conclusion is that, whilst logframes can be constructed to cater for many of the requirements of monitoring and reviewing FPR, they have to be updated frequently to incorporate 'process' changes, and become cumbersome with the more empowering dimensions of FPR. Here they can usefully be complemented by more inductive techniques such as process documentation and monitoring.