External sources for innovation in public organisations

Abstract

This article's objective is twofold: First, to provide a framework for understanding the role of external information sources in service innovation, with particular attention to innovation in public organisations. The second part of the article's objective is to test the proposed framework. Method of analysis is based on data at European level from the 2010 Innobarometer Survey on Public Innovation. Data is used for estimating a bivariate probit with sample selection, where the selection variable indicates whether or not organisations have recently implemented service innovations and the estimated variable indicates whether organisations have introduced services that are new to the whole public sector. Results indicate a significant relation between the use of both internal and external sources of information and the implementation of innovations by public organisations. External sources are shown to be more relevant for the implementation of services that are new to the public sector.

Keywords:

• information sources
• innovation
• knowledge flows
• external sources

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. There are some market-like incentives for public innovation; they are based on the idea that public organisations which show a better performance will be rewarded with a bigger budget. Competition among public organisations can exist to enhance a budget through improving effectiveness, efficiency and productivity (Hartley, 2005; Roste, 2005). However, this view of public management as replicating private firms’ incentives oriented toward market goals is outside the framework of analysis of this study as it has been shown that public sector organisations do not share with private firms the same incentives toward their final objectives (Recascino, 2002).

2. The PCA variable reduction problem assumes a linear relation between variables. Also, the objective of PCA is to reproduce as much variance as possible from the original variables, though variance is a concept applied only to continuous numeric variables. See Linting et al. (2007).

3. For a detailed and technical overview of CATPCA – also known as nonlinear PCA – see Meulman, Van der Kooij, and Heiser (2004).