The aim of the paper is to assess heterogeneity of the innovation
process. Using exploratory factor analysis on micro data from the third
Community Innovation Survey in 13 countries, we identify four factors
that that can be interpreted as research, user, external and production
ingredients of innovation. All too often it is assumed that the
differences between the rates at which these factors are found in firms'
innovation strategies can be accounted for by differences across sectors
and/or countries. To put this proposition under scrutiny, we partition
variability of the innovation process into components identified by the
different levels. The analysis shows that sectors and countries matter
to a certain extent, but far most of the variance is given by
heterogeneity among firms within either sectors or countries. On the
other hand, a grouping of firms produced by cluster analysis ac-counts
for a much higher share of the variance, which implies that the most
relevant contextual fac-tors cut across the established boundaries
between sectors and countries. We discuss the implica-tions of these
findings for the literature on national and sectoral systems of
innovation, and for the way in which evolutionary economics has analyzed
the role of selection.
Keywords: Innovation, heterogeneity, sectoral systems of innovation, factor analysis, variance com-ponents analysis.
UNU-MERIT Working Papers ISSN 1871-9872