![Sample our Built Environment journals, sign in here to start your access, Latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5926/TOC-Subject-Sample-2021-Built-Environment.jpg"})
Abstract
Broekel T. The co-evolution of proximities – a network level study, Regional Studies. Little is known about how network structures and proximity relations between linked actors evolve over time. This paper argues that a number of networks’ internal proximity structures are interrelated, which may give rise to specific types of co-evolution dynamics. An empirical investigation tests these arguments using information on the evolution of 280 networks of subsidized research and development (R&D) collaboration in Germany. The empirical findings clearly confirm the existence of systematic and dynamic interrelatedness between proximities. In this way, the paper underlines the need to consider such relations when investigating the evolution of knowledge networks.
Broekel T. 邻近性的共同演化—网络层级的研究，区域研究。我们对相互连结的行动者之间的网络结构与邻近性关係如何随着时间演化，理解并不多。本文主张，若干网络的内部邻近性结构是相互有关的，并可能带来特定的共变动态类型。本研究运用德国两百八十个受资助的研发（R&D）合作网络的演化信息，以经验探讨来测试上述主张。研究经验发现，明确地确认邻近性之间的系统性与动态相互关联性之存在。本文以此强调在探究知识网络的演化时，考量此般关係的必要性。
Broekel T. La coévolution des proximités – une étude au niveau des réseaux, Regional Studies. Comment les structures des réseaux et des rapports de proximité entre les acteurs reliés évoluent dans le temps est mal documenté. Cet article affirme que les structures de proximité internes d'un nombre de réseaux sont étroitement liées, ce qui pourrait engendrer des types spécifiques de la dynamique de coévolution. Ces affirmations sont mises à l’épreuve à partir d'une étude empirique qui emploie des données auprès de l’évolution de 280 réseaux de collaboration en matière de recherche et de développement (R et D) subventionné en Allemagne. Les résultats empiriques confirment clairement la présence d'une interdépendance systématique et dynamique entre proximités. De cette façon, l'article souligne la nécessité de considérer de tels rapports lorsque l'on examine l’évolution des réseaux de connaissances.
Broekel T. Die Koevolution von Nähen – eine Studie auf Netzwerkebene, Regional Studies. Hinsichtlich der Frage, wie sich Netzwerkstrukturen und Nähebeziehungen zwischen verknüpften Akteuren im Laufe der Zeit weiterentwickeln, ist nur wenig bekannt. In diesem Beitrag wird argumentiert, dass zahlreiche interne Nähestrukturen von Netzwerken miteinander zusammenhängen, was zu bestimmten Formen der koevolutionären Dynamik führen kann. Diese Argumente werden in einer empirischen Untersuchung anhand von Informationen über die Entwicklung von 280 Netzwerken für geförderte Zusammenarbeit im Bereich der Forschung und Entwicklung (FuE) in Deutschland überprüft. Die empirischen Ergebnisse bestätigen klar die Existenz eines systematischen und dynamischen Zusammenhangs zwischen den Nähen. Auf diese Weise wird in diesem Beitrag die Notwendigkeit einer Berücksichtigung dieser Beziehungen bei der Untersuchung der Entwicklung von Wissensnetzwerken hervorgehoben.
Broekel T. La coevolución de las proximidades; un estudio sobre redes, Regional Studies. Poco se sabe sobre cómo evolucionan con el tiempo las estructuras de las redes y las relaciones de proximidad entre los actores vinculados. En este artículo se argumenta que una serie de estructuras internas de proximidad de las redes están interrelacionadas, lo que podría resultar en determinadas formas de dinámica coevolutiva. Un estudio empírico comprueba estos argumentos utilizando información sobre la evolución de 280 redes de la colaboración subvencionada de investigación y desarrollo (I+D) en Alemania. Los resultados empíricos confirman claramente la existencia de una interrelación sistemática y dinámica entre las proximidades. De esta forma, en este artículo se pone de relieve que es necesario considerar estas relaciones a la hora de investigar la evolución de las redes de conocimiento.
Acknowledgements
The author thanks Pierre-Alexandre Balland for his valuable comments and suggestions made on earlier drafts of this paper. Of course, all remaining errors are the author's alone.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
1. Of course, network embeddedness may impact actors' economic performance and thereby be responsible for the failure of an actor implying the actor leaving the population.
2. The measure ‘is expressed as the ratio of the observed number of closed triads over the number of random expected closed triads' (Ter Wal, Citation2014, p. 603).
3. Large joint projects with subprojects in which multiple organizations participate are disaggregated at the subproject level.
4. The four-digit level appears to be the (subjectively) best choice given the trade-off between network size and ‘technological' disaggregation.
5. Isolates: nodes without any links.
6. As organizations may also be active in multiple technologies, networks may share nodes they have in common. However, these shares rarely exceed 5%. Hence, networks can be treated as independent observations.
7. The number of nodes commonly represents the size of a network. Network density is calculated by the number of observed links divided by the number of potential links.
8. The notion of proximity is unfortunate in this respect as it can be used as a synonym for ‘short distance’ as well as a reference the proximity concept, e.g. ‘cognitive proximity’. To avoid confusion, the notion of ‘distance’ will be used in the empirical analysis.
9. This denotation guarantees that large values indicate larger organizational distances.
10. In this definition, ‘universities' includes universities, universities of applied science and university hospitals.
11. While the mean is usually used for normalization, the presence of some extreme values makes the median yield more robust results.
12. The threshold of seven years is chosen as a balance of ‘long-termness' and number of observations. Moreover, long-term autocorrelation processed cannot be modelled because of the limited number of time periods.
13. ‘dep.: COG' indicates the dependent variable being COG.