Managing The Environment

Abstract

The literature on network management in the public sector reports positive effects of network activity on agency performance. Current studies show however no differences between specific types of contacts in an agency's environment. The present article adopts an explorative design to study the different types of environmental actors in the networks of nine colleges for nursing studies in the Netherlands. A typology of environmental ties is introduced, and applied to the nine cases. It appears that contacts to different types of actors reflect different levels of ambition in the network management of the colleges. The level of ambition in network management appears to be associated with two indicators for college performance: it is positively associated with diploma rate, and negatively associated with drop-out rates among freshmen. These results are discussed with reference to the current literature in public network management.

Key words:

• Networks
• performance
• public management

Notes

1 An alternative explanation could be that the measurement instrument (the survey) is not refined enough to observe differences in network activity between types of environmental actors.

2 The main task of the national program committees is the determination of competences for their graduates and advising on the standards of accreditation of the programs. However, the committees are autonomous in their agenda and free in cooperating on every matter they perceive useful. The range of issues discussed in the national program committee differs between the networks. For example, the network of colleges for nursing studies frequently discuss their performance, such as dropout rates, enrollment, diploma rates, or program insufficiencies. The frequency with which the national program committee meets varies across the five networks between six and eight times a year.

3 Some colleges may apply binding recommendations regarding the continuation of studies in the first year. This could be an alternative explanation for high dropout rates.

4 This result could explain why Meier and O'Toole (2003a: 692) do not find that network activity discriminates between different categories of actors: network activity reflects increasing ambition levels, rather than separate, orthogonal dimensions of organizations in the environment.

5 We calculated Freeman's actor centrality degree defined as C_D (n_i ) = (d(n_i )/(g–)1), where g is the number of actors and d(n_i ) is the number of alters to which actor i is connected. Two actors are connected when they share at least one affiliation.

6 We are not able to test the latter proposition, because we have no data available at the individual level of students, such as for example a proxy for student achievements in pre-education. Future (multilevel) analyses of large student-level datasets may also control for student characteristics when analyzing performance.