In summary, it seems that technology evaluation by projects is mostly
limited to biological aspects and to variables that are conventionally assessed
in agriculture and forestry research. Those aspects of the technologies
that are more specific to agroforestry, e.g. tree/crop compatibility, component
arrangements and management or service functions, were evaluated
by relatively few projects. This may in part be due to the fact that assessment
of these is not as straightforward as, for example, counting tree survival.
Service functions can in most cases only be evaluated through indirect and
often subjective indicators, and relationships may not always be clear
because of other intervening factors. There are currently no easy
methods available to projects to assess the performance of most
agroforestry-specific variables.