Field researchers will be involved in several kinds
of trials and must regularly spend time at those they are
managing. They need time to analyze data and write
reports. They require transportation to be able to move
among widely dispersed trial sites. All these factors
modify the number of farmers each researcher can include
in FMTs. For a team of three to five researchers
operating in one recommendation domain a minimum number of
farms participating in on-farm trials is probably around 20.

ANALYSIS AND INTERPRETATION OF RESULTS

 There are four different kinds of evaluation that can
be made from farmer-managed trials:
1) the farmers' personal evaluations, from which they
 decide whether to accept the technology, reject it, or
 continue testing it;
2) a survey evaluation by researchers in the year of the
 trial to ascertain what farmers say about the
 alternative tested;
3) a technical evaluation made during the year of the
 trial from data collected by researchers and/or
 farmers; and
4) an evaluation of acceptability made the year following
 the trial, based on farmers' active acceptance or
 rejection of the alternative.

Farmers' Evaluation

 This is the most important evaluation in an FMT.
Farmers know their land, climate, and families' needs and
capabilities and have a built-in method of evaluating new
technology based on their interpretation of what they see
or measure in the trial (in some cases taking into account
opinions from neighbors). From the time they need to
obtain any new inputs (e.g., fertilizer, seed) or follow a
new procedure, their evaluation mechanism is in operation.
It is not always necessary for farmers to measure the
harvest from a plot in order to make a decision about the
acceptability of an alternative. When farmers do measure
the harvest, they will be more interested in their own
means of measuring than those of the researcher.
Researchers' needs should not interfere with farmers'
evaluation.