Version classiqueVersion mobile
OpenEdition Books

The role of German universities in a system of joint knowledge generation and innovation

 | 
Mirja Meyborg

11. The Cases of the Karlsruhe Institute of Technology and the University of Heidelberg

Texte intégral

  • 70 Both universities are further taken as they are under the best-performers regarding their publicat (...)

1This last part of empirical analysis is based upon two German elite universities, namely the Karlsruhe Institute of Technology (KIT) and the University of Heidelberg. Thereby, the KIT is a technical and former elite university, while the University of Heidelberg is a medical and an elite one. These two universities have been selected as most differences have been found between the samples of technical/elite universities and the sample of medical universities. They show differences not only in their overall publication, patenting and collaboration behaviour, but also with regard to the institutional and spatial distribution of their cooperation partners. Thus, it is reasonable to compare two elite universities, whereas one is technically-focused and the other one is medical.70

  • 71 In the following course of this chapter referred to as solely Heidelberg.

2In the following, all relevant former analyses are conducted solely for both chosen universities in order to compare them with each other, and to identify special behavioural patterns for each type of university. First of all, the next figure shows the overall publication and cooperation activity of the KIT and the University of Heidelberg71.

Figure 35: Publication and Cooperation Activity of KIT and Heidelberg, absolute Numbers, 2000 and 2009 (own illustration).

3In the first instance, it can be seen that the overall publication activity of Heidelberg is higher than the one of the KIT. But, it should be mentioned in this regard that in 2008 the KIT has possessed around 240 professors and 18,000 students compared to around 280 professors and 25,000 students in Heidelberg. Hence, in the following analysis, it has to be kept in mind that it is not surprising to have some differences in size regarding their overall publication and cooperation activity.

4Overall, the KIT has increased its number of publications by about 100%, while Heidelberg could even raise their publications by 110%. Regarding the share of co-authored publications, 54% of the overall publications have been co-authored at the KIT in 2000, and around 56% in Heidelberg. Thus, both universities did not yet show big differences regarding their publication and cooperation behaviour. Later on, they could even enlarge their shares of co-authored publication; the KIT has increased its share of co-authored publications by 140%, and Heidelberg by even 170%. Hence, Heidelberg is a bit more engaged in co-authored publications than the KIT which is not surprising as the analysis of all German universities has shown that the technical universities have been less engaged in co-authored publications compared to the non-technical universities, while the medical universities have more likely cooperated with other partners compared to their counterparts.

5Keeping in mind that both universities are extremely involved in close network collaborations as the highly increasing number of co-authored publications indicates, the following figure shows the absolute numbers of co-authors, as well as the values of degree and betweenness centrality, whereas degree centrality points to the distinct number of co-authors and betweenness centrality to the importance of the particular university as knowledge intermediary within the network as already discussed above.

Table 35: Absolute Number of Co-Authors, Degree and Betweenness Centrality of KIT and Heidelberg, 2000 and 2009 (own illustration).

6Above table shows that in 2000, the KIT has possessed 4,460 cooperation partners and Heidelberg even 7,261. During the past years, the KIT could increase its number of co-authors by around 150%, and Heidelberg by approximately 180%. Regarding the value of normalized degree centrality which can be for example used to compare networks of different size, it is obvious that Heidelberg has had a much higher value than the KIT, especially with regard to the last time period. Further, while Heidelberg could even highly raise its value of normalized degree centrality, the KIT has even lost strength in this regard. The same finding holds for the value of betweenness centrality as Heidelberg could again improve itself over the past ten years, becoming more central within the knowledge networks, while the value of betweenness centrality of the KIT has dropped from 0,0757067 in 2000 down to 0,062739 in 2009. Hence, Heidelberg has been more involved in close network collaborations over the past ten years and has been more important as knowledge intermediary within the knowledge networks compared to the KIT. Remembering above finding that the average mean value of betweenness centrality has decreased over the past ten years, it has been demonstrated that, amongst other, the KIT is one of the universities that have had decreasing values in this regard.

7Coming to the institutional distribution of the co-authors of the KIT and Heidelberg, the following figure illustrates to what extent the two universities have been linked to either other universities, research institutes, enterprises or to the subject of study itself.

Figure 36: Institutional Distribution of the Co-Authors of KIT and Heidelberg, absolute Numbers, 2000 and 2009 (own illustration).

8As can be seen from above figure, the highest share of cooperation partners has been allocated to the universities followed by the research institutes, the German universities themselves and finally the enterprises. But how about comparing the shares of each cooperation partner of the KIT and Heidelberg? For doing this, the following table illustrates the particular shares of the co-authors of both universities:

Table 36: Percentage Share of the different Co-Authors of KIT and Heidelberg, 2000 and 2009 (own illustration).

9First of all, the table demonstrates that the percentage share of the other universities and the enterprises has increased over the past ten years. While the group of research institutes has remained almost similar in the case of Heidelberg, it has highly declined in the case of the KIT. Last, the 76 German universities as potential cooperation partners have become more important for the KIT as well as for Heidelberg. However, it is obvious that the KIT has been more involved in close network collaborations with enterprises compared to Heidelberg, and Heidelberg has more likely cooperated with research institutes than the KIT. Hence, the result of the chi-square test in this regard for all technical and medical universities can be confirmed by above finding, too.

10Being aware of the institutional distribution of the co-authors of the KIT and Heidelberg, the next step considers distance patterns. Thus, the following figure firstly shows the spatial distribution of the co-authors of the KIT and Heidelberg within the 1,000 km radius from 2009:

Figure 37: Spatial Distribution of the Co-Authors of KIT and Heidelberg (100 km Corridor) weighted Numbers, 2009 (own illustration).

11As can be seen from the figure, the spatial distribution of the co-authors of the KIT and Heidelberg has differed greatly. While Heidelberg has been highly engaged in local network collaborations, the spatial distribution of the cooperation partners of the KIT is more consistent as it shows a slowly declining trend with the growth of distance. This finding does not hold for Heidelberg, at least for the first 500 km, as the spatial distribution of its co-authors is highly differing. From around 500 km onwards, Heidelberg has also experienced a decreased number of cooperation partners with the growth of distance. Hence, above figure already indicates that Heidelberg as medical oriented university seems to be more likely engaged with local cooperation partners compared to the KIT as technical oriented university. Thus, locality might be especially important for medical universities than for the technical universities.

12However, to perceive a more valid impression in this regard, the following figure firstly shows the spatial distribution of all cooperation partners for 2000 and 2009, either being regionally or supra-regionally located:

Figure 38: Spatial Distribution of the Co-Authors, KIT and Heidelberg, absolute Numbers, 2000 and 2009 (own illustration).

13As it has already been presented through the overall illustration of the spatial distribution of the German university co-authors, the KIT as technical university has always possessed more linkages to supra-regional cooperation partners, while Heidelberg as medical university has more likely cooperated with regional located cooperation partners over the past ten years. In 2000, the KIT has possessed around 63% cooperation partners which has been supra-regional located, while in 2009 this percentage share has decreased by six percentage points. However, the KIT is still more engaged in supra-regional collaborations. In contrast, Heidelberg has possessed around 52% co-authors which has been regional located in 2000 and has even increased this percentage share by two percentage points. Heidelberg has more likely cooperated with regionally located cooperation partners. Hence, above hypothesis considering all German universities can be also confirmed by this finding.

14In order to better estimate which cooperation partners have been regionally located and which ones have been more likely supra-regionally located, the following figure firstly shows the regional distribution of the co-authors of the KIT and Heidelberg for 2000 and 2009:

Figure 39: Regional Distribution of the Co-Authors of KIT and Heidelberg, weighted Numbers, 2000 and 2009 (own illustration).

  • 72 See appendix for the visualization of both networks with the 76 German universities.

15It can be seen that the regional distribution of the co-authors of the KIT and Heidelberg somewhat differs. Within the first time period, most of the cooperation partners of the KIT have come from the German universities themselves followed by research institutes. The other universities and the enterprises have shared third place in 2000. Referring to the distribution of the cooperation partners of Heidelberg, it is the group of German universities that have been at the fore front closely followed by the research institutes, too. With a considerably lower share, the other universities and the enterprises have been followed. Overall, it is the KIT that has been much more engaged in close network collaborations with regionally located enterprises, and it is Heidelberg that have more likely cooperated with regionally located research institutes. However, it is also obvious that within the regional distribution of the co-authors of the KIT and Heidelberg, the share of the German universities themselves has highly increased over the past ten years. Thus, both have been frequently engaged in close network collaborations with themselves (76 German universities)72, while the share of the other universities as well as the share of the research institutes has declined.

16Ten years later, all shares have only slightly changed. In the case of the KIT, it is the German universities and the enterprises which have been increased, while the group of research institutes as well as the group of other universities have become lower. For Heidelberg, it is also the group of German universities and enterprises which have increased over the past years, while the group of research institutes has remained quite similar and the group of other universities has decreased.

17Having presented the regionally located co-authors of both German universities, the following figure now demonstrates the supra-regionally distribution of the co-authors of the KIT and Heidelberg:

Figure 40: Supra-regional Distribution of the Co-Authors of KIT and Heidelberg, weighted Numbers, 2000 and 2009 (own illustration).

18As can be seen from the figure, the KIT and Heidelberg have most often cooperated with partners coming from universities, followed by the group of research institutes and finally the enterprises.

19In 2000, 54% of all cooperation partners have come from universities in the case of the KIT and even 62% in the case of Heidelberg. Around one third have been employed at research institutes in the case of Heidelberg, and 40% in the case of the KIT. Finally, only a minor part has come from enterprises, namely 6% at the KIT and 3% in Heidelberg. From 2000 until 2009, the KIT has increased the number of cooperation partners that have come from universities up to 58% as well as the share of enterprises which has already accounted for 10% in 2009. In contrast, the group of research institutes has lost eight percentage points from 2000 until 2009. Heidelberg has increased its cooperation partners that have come from universities and enterprises, while the share of the research institutes has declined over the past ten years.

20Referring to the overall distribution, the KIT has especially increased the share of supra-regionally located enterprises during the past ten years, even though in terms of absolute numbers, the KIT still cooperates more likely with regionally located enterprises. In the case of Heidelberg, the share of regionally located research institutes has remained quite similar over the past ten years, while the share of supra-regionally located ones has highly declined from 2000 until 2009.

21To sum up, Heidelberg is a bit more engaged in co-authored publications compared to the KIT. However, this finding is not surprising as it confirms the results of the former analysis of all German universities which has shown that the technical universities have been less engaged in co-authored publications compared to the non-technical universities, while the medical universities have more likely cooperated with other partners compared to their counterparts. Further, Heidelberg has been more involved in close network collaborations and has been further more important as intermediary within the knowledge network. It has been also confirmed by the comparison of the KIT and Heidelberg that the technical universities have been more involved in close network collaborations with enterprises, while the medical universities has more likely cooperated research institutes. Finally, it has been discovered that Heidelberg as medical oriented university seems to be more likely engaged with regional cooperation partners compared to the KIT as technical oriented university.

22Finally, it is shown how the KIT and Heidelberg has performed regarding their innovation and collaboration activity. For this, the following table firstly illustrates the absolute number of their co-applicants as well as the values of normalized degree and betweenness centrality:

Table 37: Number of Co-Applicants, Degree and Betweenness Centrality of KIT and Heidelberg, 2003-2004and 2007-2008 (own illustration).

23Above table now illustrates that within the first time period, the KIT has possessed 15 co-applicants and Heidelberg has been at the same level with 14 co-applicants. During the past years, while the KIT has increased its number of co-applicants by around 50%, Heidelberg has even experienced a decline by 30%. Regarding the value of normalized degree centrality which can be for example used to compare networks of different size, it is obvious that within the first time period, both universities have still been at the same level, while the performance of Heidelberg highly decreased over the past years. Hence, in terms of patents, it is the KIT which has been more involved in close network collaborations over the past years and has been more important as knowledge intermediary within the network compared to Heidelberg. However, as the overall activity is still rather small, especially compared to the overall publication activity, it would not be reasonable to show the institutional and spatial distribution of only these two universities. But, the following figure visualizes their overall networks from 1999 until 2008:

Figure 41: Co-Applicant Network of the KIT based on joint Patents, 1999-2008 (own illustration).

Figure 42: Co-Applicant Network of Heidelberg based on joint Patents, 1999-2008 (own illustration).

24As can be seen from the figures, the co-applicant network of the KIT offers a more dynamic development regarding its distinct cooperation partners compared to the one of Heidelberg. Besides, it is obvious that the KIT has rather cooperated with enterprises while Heidelberg has been more engaged with research institutes and universities as already discovered in terms of the publication analysis. Thereby, famous industry partners of the KIT are, amongst other, Bayer, Mitsubishi, Südzucker Mannheim, Fujihunt, Eisenmann, Henkel, Tchibo and Evonik. In contrast, Heidelberg has cooperated with several research institutes as DKFZ Deutsches Krebsforschungszentrum, the Oregon Health and Science University, the Helmholtz Zentrum Munich, the Max-Planck-Gesellschaft and the Seattle Medical Research Institute, amongst other. Of course, both universities have also cooperated with either renowned research institutions or enterprises respectively.

25Overall, it is the KIT which has been more involved in close network collaborations from 1999 until 2008 and has been simultaneously more important as knowledge intermediary within the network compared to Heidelberg. The analysis of the KIT and Heidelberg in terms of patents has finally also confirmed the findings of the broad analysis regarding the whole sample of the German universities.

Notes

70 Both universities are further taken as they are under the best-performers regarding their publication, patenting and cooperation activity.

71 In the following course of this chapter referred to as solely Heidelberg.

72 See appendix for the visualization of both networks with the 76 German universities.

Table des illustrations

Légende Figure 35: Publication and Cooperation Activity of KIT and Heidelberg, absolute Numbers, 2000 and 2009 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-1.jpg
Fichier image/jpeg, 75k
Légende Table 35: Absolute Number of Co-Authors, Degree and Betweenness Centrality of KIT and Heidelberg, 2000 and 2009 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-2.jpg
Fichier image/jpeg, 78k
Légende Figure 36: Institutional Distribution of the Co-Authors of KIT and Heidelberg, absolute Numbers, 2000 and 2009 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-3.jpg
Fichier image/jpeg, 86k
Légende Table 36: Percentage Share of the different Co-Authors of KIT and Heidelberg, 2000 and 2009 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-4.jpg
Fichier image/jpeg, 82k
Légende Figure 37: Spatial Distribution of the Co-Authors of KIT and Heidelberg (100 km Corridor) weighted Numbers, 2009 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-5.jpg
Fichier image/jpeg, 96k
Légende Figure 38: Spatial Distribution of the Co-Authors, KIT and Heidelberg, absolute Numbers, 2000 and 2009 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-6.jpg
Fichier image/jpeg, 87k
Légende Figure 39: Regional Distribution of the Co-Authors of KIT and Heidelberg, weighted Numbers, 2000 and 2009 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-7.jpg
Fichier image/jpeg, 97k
Légende Figure 40: Supra-regional Distribution of the Co-Authors of KIT and Heidelberg, weighted Numbers, 2000 and 2009 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-8.jpg
Fichier image/jpeg, 96k
Légende Table 37: Number of Co-Applicants, Degree and Betweenness Centrality of KIT and Heidelberg, 2003-2004and 2007-2008 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-9.jpg
Fichier image/jpeg, 80k
Légende Figure 41: Co-Applicant Network of the KIT based on joint Patents, 1999-2008 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-10.jpg
Fichier image/jpeg, 200k
Légende Figure 42: Co-Applicant Network of Heidelberg based on joint Patents, 1999-2008 (own illustration).
URL http://books.openedition.org/ksp/docannexe/image/241/img-11.jpg
Fichier image/jpeg, 159k

Lire

Open access

Acheter