Network Structure and Spinoff Effects in a Collaborative Public Program

Abstract

Despite a growing body of literature on program implementation networks, most studies have focused on understanding the network structures involving the implementation of initial or inceptive public programs. Little attention has been paid to what network structures actors create to pursue spinoff effects. Spinoff effects are defined as offshoot projects that take place ex-post the completion of the initial public projects or programs. Framing postproject collaboration between communities (villages) and organizations for spinoff projects as social networks, this research examines the network structures that drive postproject collaboration. The network data on postproject collaboration came from a field survey of 62 communities from Nepal that were engaged in building ties with organizations to mobilize resources for spinoff projects. The results from bipartite exponential random graph models applied to the network data show that postproject collaboration ties are influenced by network centralization around communities, or a greater variance among communities in the number of ties with organizations, reflecting differences in the communities’ needs for spinoff projects. In addition, communities use network bridging, or indirect ties to other communities through the partner organizations, to access new information valuable to spinoff projects. Given the widespread use of public programs, these findings provide important insights to communities and managers as they advocate postproject collaboration to realize spinoff effects and thereby to sustain the impact of public programs.

Keywords:

• bipartite ERGMs
• social networks
• spinoff projects
• two-mode network

Notes

Notes

1 Higher-order network tie configurations in one-mode networks are based on partial conditional dependence, also known as social circuit dependence (4-cycle) assumption, in combination with a Markov dependence assumption. The social circuit dependence assumes that two tie-variables, X_ij and X_hm, that do not share an actor are conditionally dependent if ties exist between i and h, and between j and m, creating a 4-cycle. For example, if i works with h, and j works with m, then the presence of collaboration between i and j is likely to affect whether h and m also collaborate. Under a Markov dependence assumption, two possible ties are conditionally dependent when they are incident on a common actor. Higher-order network configurations in two-mode networks are an extension of this assumption to two-mode networks. For details, see Koskinen & Daraganova (2013, pp. 57–58) and Wang (2013, pp. 123–124).

2 The required matching cash was 2.5% for construction and 3.0% for operation and maintenance of the total costs of the project.

3 The five districts are Kavre, Sindhupalchok, Dhading, Makwanpur, and Sindhuli.

4 The way in which ties are assumed to be interdependent within a given network is crucial. The p* model is considered the more general model for networks, allowing for a wider array of network statistics in the equation. It is based on a notion of dependence referred to as “partial conditional dependence” (Prell, 2012, p. 48).

5 The Markov chain Monte Carlo (MCMC) procedure involves computer simulation to create a distribution of graphs, whereby graphs are simulated based on a starting set of parameter values, and subsequent refinements to these values are made by comparing the distribution of graphs with the observed graph (Prell, 2012, p. 211).

6 The graph features used for GOF are all the graph features included in the fitted model and other standard graph statistics, such as reciprocity and lower-order outstar and instar degree distributions, not in the fitted model (Robins & Lusher, 2013, pp. 181–183).

7 Force-directed graph drawing places nodes with similar path lengths to one another closest in the graph. In this sense, spring embedding is fairly similar to multidimensional scaling (MDS). The advantage of spring embedding is that it allows the nodes to be distributed more evenly across the space, so that the graph is much easier to read (see Borgatti, Everett, & Johnson, 2013).

8 UCINET’s NetDraw is a visualization package. When the network data are brought into NetDraw, it produces a visual image of a network with actors and ties between them. While one can manipulate shape or size of the nodes based on their attributes, network graphs are basically a method of exploration (Prell, 2012, pp. 83–84).

9 The two-mode core-periphery analysis in UCINET produces blocked adjacency matrices with a 4 × 4 core and periphery groups and the ties among the actors in the groups. The analysis also calculates a density matrix with density values for the core and the periphery. For details, see UCINET program (Borgatti et al., 2002).

10 The MPNet labels for the configurations in Figure 1 are as follows: Community network centralization = XASA; organization network centralization = XASB; alternating 2-paths (network closure) = XACA; organization attribute (receiver effect) = XEdgeB; community attribute (sender effect) = XEdgeA; Edge parameter (any community linking with any organization) = XEdge (see Wang et al., 2014).

11 The goodness-of-fit (GOF) test is based on 10 million digraphs, from which samples are drawn at 1 per 10,000 digraphs. The estimated model achieves GOF when the simulated networks do not significantly differ from the observed networks, which is indicated by the GOF t-ratios less than 0.1 in absolute value for the configurations included in the model estimation (Wang et al., 2014). GOF t-statistics for the observed graph statistic are derived from the difference of observation and sample mean divided by standard deviation from the simulated distribution (see Wang et al., 2014).

Additional information

Notes on contributors

Manoj K. Shrestha

Manoj K. Shrestha is Associate Professor of public administration and policy in the Department of Politics and Philosophy at the University of Idaho. His research focuses on network governance and network effectiveness in the management of local public goods, local economic development, and water resources.