be gauged and related to a silvo-pastoral extension programme? Can it be
done 'in house' or does it require detailed monitoring exercises that may be
expensive and for which there may be few resources available? What options
exist for using other locally generated data and relating it to the extension
programme? These are common problems in evaluating many agroforestry
interventions, in that it is difficult to know how much of the change to
attribute to the project and how much is due to other factors [Scherr and
Muller, 1990].