In recent years a relevant issue in the analysis of regional innovation policies supporting collaboration
between firms and other organizations (e.g.: other firms, universities, research centres, etc.) is to
evaluate the impact of such policies on the personal network of the targeted firms in a given region. To
deal with this issue, it becomes of primary importance to have a reliable estimates of the size of
innovation networks.
In this work, we propose a novel use of capture-recapture approach to estimate the size and the
composition of the personal networks of firms involved in regional innovation networks. Capturerecapture
is a statistical method widely used to estimate the size of hidden population using two or more
samples drawn from the population. The number of individuals observed in a sample and the number of
those observed in both samples is used to estimate the number of those not selected in any sample. The
method can be also used if two or more incomplete and overlapping lists of ascertained cases from
different sources are available.
We apply the proposed method on data from a survey conducted on a sample of 536 companies located
in Friuli Venezia Giulia (a region in the north-east of Italy) who participated in the POR-FESR 2007-2013
call, named “Innovation, research, technology transfer and entrepreneurship”. The collaboration ties
they claimed to entertain with other organizations (also not regionally bounded) before and after their
participation to the call will be considered as the two lists on which the proposed capture-recapture
approach is applied. The idea is to account for unobserved ties (and thus to estimate the network size of
companies) using the number of ties observed only before or only after the participation to the call and
the number of ties common to both.
