Checklist : Protocols for Experiments with Farmers

Experiments with Farmers: Checklist for Preparing Protocols

Trial Protocols A protocol for a trial is a detailed and complete description of what happens in the experiment. It is first produced during planning of the trial and is revised as necessary as the activity progresses. There are three reasons for producing a detailed written protocol for any trial: 1.

Before starting the trial the protocol can be shared with anyone who might be able to comment on it and improve the effectiveness of the research. Consider eliciting comments on the protocol from:  people who know the farmers and farming systems in the area where you are working  people who understand the subject area (soil fertility, animal feed, etc)  people with experience of the methods you plan to use. Ask any one who might having something to offer. The worst that can happen is that you get no response! Remember that people working outside your immediate location may well have valuable experience that may benefit your research.

2.

Success of the trial can not depend on the continued presence of any single person. If a detailed protocol is written up then it will be possible for the trial to continue to a successful conclusion even if the principal investigator leaves.

3.

After the trial is completed the protocol is a record of what was done, to be referred to in any reporting and to be archived with the data.

The Checklist This is a checklist - a list of points to remember when preparing a protocol. To use it, consider each point in the list and make sure you have addressed it in the protocol. Different institutions require protocols prepared in different formats, but any useful protocol will have to cover most of the points in this checklist. The checklist does not give you the answers, or tell you how to carry out the trial. It is just a reminder of the points to decide on. The checklist is designed to help develop protocols for single trials or experiments. It does not help in designing projects, which will consist of several linked activities (planning and training activities, surveys, experiments and so on). It is designed for planning experiments that involve farmers. Other types of research activities require protocols with different information.

Prepared by: Members of the AHI Workshop 'Participatory Experimentation', Nairobi, 28 June- 3 July 1999 Compiled by: Richard Coe, ICRAF, Nairobi, Kenya Version: 31 August 1999

1

Abstract A summary of the trial is useful. Note the format of the whole protocol my have to conform to institutional requirements.

1

Reference or Number Use a unique reference or number of this study. Needed to keep track of this experiment and not confuse it with other related studies.

2

Title Choose a short, memorable title that people quickly learn relates to your study.

3

Location Part of the identification of the study is where it takes place. Details (eg exactly which villages) come later.

4

5

Investigators 4.1

Team Leader (Principal Investigator) and his/her Institution Remember the PI is the person responsible for the design and implementation of the work. However the 'team' operates, there has to be one leader who takes overall responsibility.

4.2

Team members and their institutions It may be useful to list their professional area.

Background and Justification You have to justify spending money on this work! In each of the following sections you must make it clear:  What the farming problem is  How your work will help solve the problem  What the next step (when this experiment is concluded) is expected to be  Target group and why chosen  Who the work will make an impact on The information given must be specific to the location you are working in and the problem you are working on. 5.1

Summary of literature Review of what is known about the problem and possible solutions. This will often come from other locations within the country and elsewhere .

5.2

Past research by you and others in this area Results from PRAs, surveys, other experiments etc.

5.3

Links to other parts of the same project Describe how this trial links to other activities of the same or related projects, such as other trials, farmer training activities or planing meetings. Describe existing institutional linkages and relationships with farmers and farmer groups.

Version: 31 August 1999

2

6

5.4

Hypotheses Even initial and exploratory studies have hypotheses! Hypotheses are statements which you believe to be true and when this is confirmed by the study allow the work to progress. They may concern biophysical, social or economic processes, or the links between them.

5.5

Potential Impact If the work goes as planned and hypotheses are confirmed (or not), what will the effect be? Who will the beneficiaries be (both on and off the farm)? How will they benefit and by how much? Will any people suffer a negative impact? How sustainable will the impact be?

Objectives The whole of the rest of the design ( and hence the protocol) depends on the objectives. List them clearly, completely and in sufficient detail to leave no doubts about any aspect of the study later. Include a description of who the resulting information is aimed at. The objectives must be consistent with each other and capable of being met with a single trial. Farmers should have been involved in deciding the objectives, and that process described in Section 5. If that has not yet been done it is probably too early to write a detailed protocol.

7

Methods Give enough detail here for the protocol to be useful for :  Anyone to see what you plan to do, so that suggestions for improvement can be made.  As the permanent record of what should be done, to be referred to during implementation. It should be good enough for this even if the PI leaves. Farmers will be involved in deciding many of the details of the protocol. In each section described how that was done (if it is completed) or how it will be done. 7.1

Trial type There are several ways of summarising the type of trial. ICRAF has use Type 1, 2 and 3. The Contractual-Consultative-Collaborative-Collegial description is useful. Try to find a simple description that summarises the approach used.

7.2

Duration Be realistic! The start date must be far enough in advance to make proper preparations. The trial must be long enough to get outputs (eg for farmers to realize benefits) but short enough to keep everyone interested. The appropriate length will depend on the objectives.

7.3

Location Describe how and why locations are selected. Methods may be random, stratified , selected to follow a known gradient. Description of locations may well be hierarchical eg Districts chosen because of the mandate area, locations chosen randomly, sublocations chosen to cover a range of distances from the main road. Describe sampling schemes and sample sizes.

7.4

Farmers Describe how farmers (or households, fields ,etc) are selected to be in the study. Examples are volunteers at a village meeting, introductions made by extension staff, contact farmers from a previous study, farms chosen during a mapping exercise,…Section 7.10 points out it is important to find out who the farmers are (eg gender, household type, wealth category) and how these relate to the general population. Say how many farmers will be involved.

7.5

Treatments Describe both the treatments to be compared and the method of arriving at these. Make it clear what is determined by farmers and what by researchers. If the farmer makes the decision, find out what they are comparing (it may not be physically adjacent and it could be from another season).

Version: 31 August 1999

3

If "farmer's practice" is included as a treatment make it clear exactly how this is defined and if it varies between farmers.

7.6

Layout Describe where the treatments will be applied and how these are chosen. Describe plot location within a farm, sizes, method of allocating treatments to plots. Make it clear what is determined by farmers and what by researchers.

7.7

Inputs Describe what inputs (eg seeds) are needed and how (and on what terms) these will be supplied

7.8

Management Who is responsible for deciding on management activities (e.g. planting, weeding, spraying, harvesting)? Who is responsible for carrying them out? Describe each management decision and who is making it. Do not generalize. Distinguish decisions about the management (eg how many times to weed) from carrying out the work (eg doing the weeding)

7.9

Non-experimental Variables Describe key variables and if they are fixed. If so, at what level and by who. How are farmers involved in deciding the level to fix NEVs?

7.10 Data collection Data may be collected on 'response variables' such as  agronomic performance  economic performance  farmers assessment  ecological impact  … It is also necessary to record 'design variables'. These are such things as:  location of participants (GPS)  household and farmer characteristics  layout details (what treatments selected by farmers and why, what plots or niches used and why)  levels of non-experimental variables For each describe:  What variables or indicators will be used  Who decides on them and how farmers were involved  The weights they should be given in data analysis (perhaps also elicited from farmers)  The measurement tool (eg survival by researcher count, farmer's assessment by questionnaire)  The measurement unit (plant, plot, farm, village,…) and sampling scheme Describe the monitoring process (eg visits to the farmers that do not involve any planned data collection) 7.11 Data Management Describe who will be collecting data. If farmers are doing data collection explain how this will be organised and the training necessary. Describe how and where data will be looked after. How will it be computerised? Who will have access to it? How and where will it be archived?

7.12 Data analysis, reporting and feedback Describe methods to be used for analysing, interpreting and reporting the data. Describe farmer involvement (including how results will be reported to farmer-collaborators).

Version: 31 August 1999

4

8

Implementation Plan 8.1

Outputs List 'hard' outputs

8.2

List of tasks Include such activities as monitoring and evaluation, reporting and reviewing - things which take time but get forgotten when planning

8.3

Timing

8.4

List of partners/team members

8.5

8.6

9

Roles of all partners Make it clear exactly who is responsible for what Budget

References

10 Version Keep the protocol up to date, both while it is being revised during planning and when details change during implementation.

Version: 31 August 1999

5