In this paper a model of technological evolution based on a twin chara
cteristics representation and on a population approach has been used t
o discuss the possibility of obtaining a single measure of technical c
hange which could be used for all possible types of evaluation. Possib
le changes for a technological population include changes in position,
changes in density, corresponding to a change in the intensity and ty
pe of competition; changes in the shape of the technological populatio
n corresponding to the formation of particular trajectories, or to the
fragmentation of an existing population following from a process of s
pecialization. In view of these evolutionary trends it is considered i
mpossible to obtain a unique measure of technological change which is
useful for all types of evaluation. The heterogeneity/dispersion of th
e technological population and the presence of qualitative change limi
t the accuracy and the meaning of measurements of technical change. Th
is means that only in particular circumstances (e.g. long periods of o
bservation and populations of unchanging shape, or developments along
a particular trajectory) would such measurements be reasonably accurat
e and meaningful.