The Multiple Dimensions of Managerial Networking

Abstract

Studies in public management show that agencies draw different types of support from different actors and organizations in their environment. If this is true, we would expect that managers differentiate their networking activity toward different types of external actors and organizations. However, empirical studies of the networking activities of managers do not reveal such a differentiation: these studies consistently report the existence of only one factor of managerial networking activity. The present article aims to solve this puzzle by disaggregating managerial networking into multiple scales of managerial networking activity, each related to a specific type of support from the agency’s environment. A cumulative scaling analysis of the network ties of Texas school district superintendents for the years 2002 and 2005 shows the existence of three such stable and homogeneous networking scales, respectively, providing (a) political support, (b) bureaucratic coping, and (c) coproduction. We compare these results with those of the method used in previous studies: factor analysis. We illustrate the potential of cumulative scaling for the analysis of managerial networking by comparing the effect of the managerial networking factor with those of the three networking scales on the pass rates of Latino students on the Texas Assessment of Academic Skills. The article concludes with a discussion of the implications for our understanding of managerial networking.

Keywords

managerial networking performance Mokken Scale analysis factor analysis

Introduction

Networking and networks affect the performance of organizations (Akkerman, Torenvlied, & Schalk, 2012; Brass, Galaskiewicz, Greve, & Tsai, 2004; Meier & O’Toole, 2003; O’Toole & Meier, 1999, 2004a, 2011). Managerial networking activity is conceptualized as the contact frequency of relationships that (high-ranking) managers maintain with external actors and organizations. In the public sector, the characteristics of the externally oriented networking behavior of managers positively affect the performance of their agencies (for instance, Agranoff & McGuire, 2003; Akkerman & Torenvlied, 2011; Bardach, 1998; Hite, Williams, & Baugh, 2005; O’Toole, 1997; O’Toole & Meier, 1999, Meier & O’Toole, 2001; Rethemeyer, 2007; Schalk, Torenvlied, & Allen, 2010). In the private sector, the networking activities of firm representatives (managers and CEOs) positively affect the survival rates, economic output, and innovativeness of their firms (Powell & Smith-Doerr, 1994; Zaheer, McEvily, & Perrone, 1998).

Effects of networks on agency performance are studied from two perspectives. The first perspective focuses on interorganizational networks, as systems of coordination and collaboration. Studies from this perspective refer to “network management” as the strategies that network members—delineated by some boundary rules—employ to govern their interactions and mutual interdependencies (Agranoff, 2006, 2007; Agranoff & McGuire, 2003). The consequences and effects of these strategies are studied for the performance of the network as a whole (Provan & Kenis, 2008; Provan & Milward, 1995).

The second perspective focuses on the networking activity of individual managers with various external actors and organizations—such as suppliers, stakeholders, clients, alliance partners, regulatory agencies, or political institutions. An early empirical study in this field is the analysis of the diverse relationships of agencies for labor market training with external organizations in Sweden and Germany (Hanf, Hjern, & Porter, 1978). A study of urban school management refers to managerial networking as the “external” social capital of these schools (Leana & Pil, 2006). Various studies of Texas school districts show that managerial networking outward into the interdependent environment taps opportunities to exploit resources in the district’s environment and buffer environmental shocks, such as political, economic, and technical demands (Meier & O’Toole, 2003; Meier & O’Toole, 2008; O’Toole, 1997; O’Toole & Meier, 1999, 2004a). The present article builds on these studies.

The idea that the environment of agencies can be managed is broadly recognized in (public) management studies.¹ Recent studies of organizational environments show that organizations are able to draw on different types and sources of support. Bozeman (1987), for example, distinguishes between support from external political and bureaucratic actors, the public media, agency clients, and other stakeholders. The public management literature further stresses the importance of support from political actors and clients of the agency for agency performance, for example, to obtain funding for its service delivery (Lynn, 2007; Moynihan & Pandey, 2005; Rainey & Steinbauer, 1999) or to reduce perceptions of red tape (Torenvlied & Akkerman, 2012). Support from politicians, clients, and stakeholders are key dimensions in the Kennedy School strategic model for government managers (Heymann, 1987; Moore, 1995, 2000). Indeed, Moynihan and Pandey (2005) report positive effects of political support on self-reported agency performance for 83 public managers.

However, despite the recognition in the literature that different external actors and organizations provide different types of support to the agency, the empirical evidence from studies of managerial networking reveal only that different levels of networking activity exist. For example, the studies of Texas school district superintendents consistently show the existence of only one common “networking activity” factor (e.g., Meier & O’Toole, 2003, p. 692; O’Toole & Meier, 2004b, p. 687). A study of directors of colleges for nursing studies shows that one cumulative scale of “networking ambition” exists—with the most ambitious director maintaining strong relationships with many different external organizations (Akkerman & Torenvlied, 2011). Studies on the managerial networking activity of firm managers report similar results. Geletkanycz, Boyd, and Finkelstein (2001) find one common factor for many different measures of the outside directorship of 460 CEOs of manufacturing and services firms in the 1987 Fortune-1000 list. Stam and Elfring (2008) report that only one composite measure exists for the contact frequency of relationships that managers of open-source software manufacturers maintain with 14 different types of external organizations.²

The aforementioned factor analyses imply that managers who are inactive (i.e., have a low score on the factor) tend to network infrequently with few organizations; managers who are highly active (i.e., have a high score on the factor) tend to network frequently with multiple organizations. Hence, the results of the factor analyses obscure potentially present patterns of managerial activity in which managers more frequently network with some pairs of external actors and organizations and more infrequently with other pairs. The latter pattern is the one we expect if managers would differentiate between types of organizations (Bozeman, 1987; Heymann, 1987; Moore, 1995; Moynihan & Pandey, 2005; Rainey & Steinbauer, 1999).

The approach taken thus far in the empirical research literature on managerial networking, in other words, frames the likely importance of managerial networking for performance in terms of the multiple functions that, it is presumed, public managers must engage as they interact with others outside their core agency—tapping various kinds of opportunities and resources in their environment, buffering their organization from various negative shocks. This focus on only one general dimension of managerial networking activity reveals an important link between managerial behavior and organizational performance; at the same time, however, it obscures a great deal of information that is suggested by studies of environmental support. The limitations of much large-n research have typically precluded knowing much about what managers actually do as they operate in their organization’s interdependent environment.³

The first aim of the present article is to explore, theoretically, whether managerial networking activity has multiple dimensions. The second aim of the present article is to illustrate the advances of cumulative scaling techniques by testing our hypothesis on the Texas school district studies, and comparing the results with those of a factor analysis. We do so by studying the association between the scores of school district superintendents on the managerial networking scales and the pass rate of Latino students on the Texas Assessment of Academic Skills (TAAS). We use a nonparametric cumulative scaling technique—Mokken scale analysis—to check which homogeneous cumulative scales of managerial networking can be identified. Whereas factor analysis is based on “classical test theory,” Mokken scale analysis is based on the more recent “item response theory.” Mokken scale analysis pools specific items into scales (in this case, external organizations into “managerial networking activity scales”).

Item-response theory enables us to separately estimate (a) the “difficulty” of contacting for the different external organizations on a managerial activity scale and (b) the time and effort spent by a superintendent on a particular managerial networking activity scale. Each scale retrieved can be interpreted as a specific dimension of managerial networking. A factor analysis could produce biased results when a cumulative scale applies (Embretson & Reise, 2000; Van Schuur, 2003, 2011). Mokken scale analysis is not yet widely used in policy analysis or public management research, although a few notable exceptions can be identified. For example, Jacoby (1994, 2000) uses this technique to find different dimensions of social support for public spending. Schneider, Jacoby, and Coggburn (1997) use Mokken scale analysis to find dimensions of bureaucratic decisions in state Medicaid programs.

We check and test for the existence of cumulative scales in the managerial networking data from the Texas school district studies of Meier and O’Toole for two waves: 2002 and 2005.

Below, we introduce our simple model of managerial networking activity and discuss our hypothesis. We introduce the context of the Texas school districts. Subsequently, we describe the study design and measurement of the contact frequencies between school district superintendents and external organizations. We give a brief introduction to Mokken scale analysis as an alternative to factor analysis. In the Results section, we present three homogeneous cumulative scales of networking activity that are present in the data. We provide an interpretation for these scales and compare the scales with the single factor produced by a factor analysis of the same data. We further illustrate the potential for cumulative scaling techniques by analyzing the associations between the cumulative scale scores of superintendents and school district performance (the pass rate of Latino students) using ordinary least squares (OLS) regression. We compare these associations with the associations found when using superintendents’ scores on the single factor. The article concludes with a discussion of implications for public management theory, the relevance of cumulative scaling techniques for public administration research, and a sketch of other promising avenues for future research.

Investments in Managerial Networking

To understand how different dimensions of environmental support, suggested by the public management literature, tie in with the networking activities of public managers, we further specify the (cost-benefit) mechanism that drives public managers’ behavior toward actors and organizations in the environment of their agency. Clearly, costs are associated with maintaining frequent relationships with external organizations. Agranoff (2006, p. 62) identifies different cost categories in the context of collaborative management, which are also relevant for managerial networking. Agranoff proposes that time and opportunity costs are lost as a result of maintaining relationships with external actors and organizations. Some such organizations are relatively close to the agency, for example, because these organizations are a “natural” partner, or are located relatively close in a physical sense to the agency. Physical proximity appears to be one of the best predictors for communication contacts (Allen, 1984; Krackhardt, 1994, p. 214; Rice & Aydin, 1991). Other costs that follow from network collaboration are a loss of autonomy and a stronger dependency on other organizations’ goals and behavior (Agranoff, 2006, p. 62). Organizations with different missions need to learn about each other to communicate effectively, and this need to learn adds additional costs. Finally, managerial networking is associated with costs because beneficiary relationships with other organizations must be reciprocated, which requires an investment in the provision of information and resources to other external organizations and actors (O’Toole & Meier, 2004b).

Benefits of managerial networking are specified in a general framework by Bozeman (1987), who argues that the “publicness” of the agency environment is shaped by political and bureaucratic actors as well as the influence of the public, agency clients, and stakeholders (see also Heymann, 1987; Lynn, 2007; Moore, 1995, 2000; Rainey & Steinbauer, 1999). Because the support generated by managerial networking activity varies between different external actors and organizations, public managers are advised to be selective in their investment in relationships with external organizations (Heymann, 1987; Moore, 2000). Indeed, a recent qualitative study of managerial networking activity of university managers reports evidence that the relationships which these managers maintain with external organizations are related to their goal-oriented investments (Akkerman & Torenvlied, 2011).

Because not all external organizations provide similar types of resources, information, and support, relationships are useful for different reasons, and some relationships are likely to be more valuable than others. Consequently, goal-oriented public managers likely invest in relationships with those external actors and organizations which serve the goals and interests of their core agency. Hence, we assume that some connection exists between the agency goals and the specific type of support public managers seek with their networking activities. This assumption is logical and plausible. For example, the manager of an agency under tight budgetary constraints will prioritize relationships with organizations that yield financial support. The manager of an agency confronted with negative feedback from clients will prioritize relationships with client interest groups.

The costs of managerial networking and the benefits of relational investments imply that public managers face the choice how much to invest in relationships with specific external actors and organizations. External organizations most likely differ in their (physical and/or mission) proximity to the focal agency, which creates differences in the relative costs and benefits of relationships between different external organizations. We assume that managers balance relative costs and benefits. Hence, some managers will maintain frequent relationships with organizations that provide some type of support, whereas other managers will maintain more frequent relationships with other organizations that provide another type of support.

Public management studies, for example, point at political support as a key dimension of environmental support (e.g., Bozeman, 1987; Meier, 2000; Moore, 1995; Moynihan & Pandey, 2005). Managers may tap political support by maintaining (frequent) relationships with—for example—elected officials, clients, or the media. Which external actors and organizations provide which type of support to the agency depends on the type of agency under study as well as institutional, cultural, and contextual factors.

It follows that specific external actors and organizations could form a “managerial networking activity scale,” which can be interpreted as a specific dimension of support from the focal organization’s environment. This is our core hypothesis.

Hypothesis 1: Managerial networking activity has multiple dimensions, each related to a specific type of support to the agency that can be provided by external actors and organizations.

Most likely, the “managerial networking activity scales” reflect the different types of benefits that organizations can tap from their environment, specified in the general framework of Bozeman (1987), and shaped by political actors, bureaucratic actors, the general public, as well as clients and other stakeholders.

Research Context, Data, and Study Design

The context of managerial networking in the present study is U.S. education policy, in particular, via school districts at the local level. These units are special-district governments with formal independence from other governmental institutions—they have their own taxing powers and elected officials. Districts vary in size (the number and enrollment of schools); and with respect to staff, experienced teachers, and school principals. Local education in the United States has developed into a significantly complex and interdependent setting. Schools are now venues for the delivery of a host of associated services or regulatory programs, from public health (vaccination programs, prevention of sexually transmitted diseases), to substance abuse, to the prevention and control of child abuse, to the achievement of nutritional objectives, to the reduction of adolescent violence, to civil rights, and to the improvement of life chances for disabled children. The “core” educational function has been surrounded by and insinuated into a panoply of other public objectives, and in turn a host of other organizations have become involved in the day-to-day functioning of school district activities.

Because there is ample variation between school districts in the extent to which particular problems and programs require the support of specific types of external actors and organizations, this setting forms an ideal testing ground for the existence of multiple dimensions of managerial networking activity. All these school districts are charged with educating students, but otherwise the districts are quite heterogeneous in terms of size, racial and ethnic characteristics of district clientele, wealth, and other characteristics.

School district superintendents are the chief administrators of school districts, and they are primarily responsible for managing this highly interdependent policy context. They typically report to an elected board but are charged with handling the day-to-day operations of their district, its performance, and both internal and external management. School superintendents have substantial discretion; they set goals for achieving school district policies, recruit and assign other administrators and teachers, and generally set the agenda on local education matters. Superintendents must also manage relationships with external actors who place political and technical demands on the district. These external actors and organizations play a role in both education and the numerous other functions (public health, safety, etc.) in which schools engage. Important ties include (a) their school board, the elected body responsible for local policy making, (b) relevant state-level and federal educational departments and politicians, who define regulations and provide opportunities for funding, (c) local business leaders, who often support locally enacted taxing decisions that are important for school district revenues, and (d) other school district superintendents, who provide peer advice and expertise, transfer innovations, and exchange experiences.

The present study uses data from the 2002 and 2005 waves of the Texas school district studies. These data include considerable information regarding the more than 1,000 school districts in the state of Texas that represent approximately one of every 14 school districts in the United States. The districts range widely on a variety of dimensions, including student composition (race, ethnicity, etc.), resources, setting (urban, rural, suburban), and performance. Response rates for the superintendents in each year ranged from 55% to 67%. Districts responding to the survey were no different from nonrespondents on key variables such as enrollment; enrollment growth; students’ race, ethnicity, and poverty; or test scores (Meier, O’Toole, Boyne, & Walker, 2007, p. 363).

The superintendents of the Texas school districts were asked a range of questions about their goals, administrative style, allocation of time, and—of particular importance for present purposes—the contact frequency of their relationships with the following⁴ external actors and organizations: “parent groups,” “local business leaders,” “teacher associations,” “federal officers,” the “Texas Education Authority,” and “local politicians.” In effect, the superintendents were asked with whom of these actors and organizations they maintain a relationship and how frequently—on a 6-point scale ranging from never to daily. This quantitative approach to measuring the networking activity of managers is a specific approach; other, more qualitative approaches exist (for a concise overview see Scott, 2000).

The dependent variable used in the illustrative analyses of school district performance (in which we compare the effects of different networking activity scales with the effects of one networking activity factor) is the pass rate of Latino students on the TAAS. Comprising 47% of the students in Texas, Latinos are the largest racial/ethnic group in the state’s educational system. The TAAS is a standardized, criterion-based test that all students in Grade 11 must take, and pass, to receive a regular diploma from the state of Texas. This variable has been selected because it represents an important performance metric in these jurisdictions—and is also a challenging objective that requires careful managerial assessment of networking costs and benefits to achieve significant results. The performance measure used is the percentage of Latino students in a district who passed all the (reading, writing, and math) sections of the TAAS (O’Toole & Meier, 2004b).

Cumulative Scaling Analysis

To identify dimensions of superintendents’ managerial networking, we use a nonparametric “item-response” model for scaling analysis, the “Mokken model.” As is the case with reliability analysis—which is a commonly adopted scaling technique in public administration studies—the Mokken model assumes that an item can be included only once in a scale. A participant’s scale score is the sum of her scores on each item in the scale. Other commonly adopted scaling procedures, factor analysis and principal component analysis, assume that an item could contribute to several latent factors. A participant’s factor score is a weighted combination of her scores on each item in the scale (with the weight defined by the “factor loadings” of the items), or the prediction of a participant’s location on the factor using (variants of) regression analysis (Kim & Müller, 1978).

The more commonly adopted scaling techniques of reliability analysis, factor analysis, and principal component analysis are based on “classical test theory,” which makes a strong assumption that the different items in a scale have the same mean and standard deviation. However, problems with correlations (and hence factor analysis) occur when items are not “parallel,” that is, when they have different means and standard deviations (Carroll, 1945; Ferguson, 1941; Van Schuur, 2011, p. 89).⁵ For the frequency of interactions with external organizations or actors, this assumption is also quite unrealistic. The obvious reason is that close external actors or organizations (with respect to their location or mission) are more easily accessible for managers than distant organizations. Hence, we often observe marked differences in average interaction frequency between different types of external actors and organizations. Thus, scaling procedures based on classical test theory are not quite appropriate for the study of patterns in the interaction frequencies of managers with external organizations and actors.

In addition, scaling techniques based on “classical test theory” estimate factor scores and participants’ interdependently and not separately. Hence, each reliable scale or “common factor” found in the data is unique for the test context, for example, the specific survey. The scales cannot be compared across different populations of participants or across different measurement moments in the same population of participants. By contrast, “item response” theory separately models (a) the participants’ scores on a “latent trait” and (b) the item characteristics that measure the trait (Embretson & Reise, 2000; Van Schuur, 2003, p. 143). The latent trait can be—for example—an attitude or ability. Well-known examples are students’ proficiency in mathematics or reading ability. Other examples are voters’ preferences for distinct policy alternatives. In the present study, the latent trait analyzed is the investment in networking relationships by school district superintendents.

The core idea of item response theory is that each item is a separate test of the value of a respondent on the latent trait. Thus, participants differ with respect to their ability, and items differ with respect to their “difficulty” for participants. Both participants and items are scaled on the same dimension.⁶ The concept of difficulty is strictly defined as a methodological concept, and refers to the likelihood that a participant is unable to “pass” the item test. This is why Mokken scale analysis is referred to as a “cumulative” scaling technique: participants who are able to pass an item that has some difficulty attached to it are assumed to be able to pass all items that are less difficult. The following classroom example provides a further clarification. Suppose that we wish to measure participants’ height. We can confront our participants with a large set of gates that vary with respect to the headroom they offer to participants. Each gate is an item. If a participant is able to pass through a gate without bumping her head, she will score a “one” on this item (gate). If, however, she happens to bump her head while passing through a gate, she will score a “zero” on that item (gate). From the response pattern of zeros and ones for participants and gates, we are able to deduce two rank orderings. The first is a rank ordering of participants from tallest to shortest. The second is a rank ordering of gates from highest (least difficult to pass) to lowest (most difficult to pass). Both participants and items are mapped in this order on the same one-dimensional scale. This scale is a “cumulative” scale, in the sense that a participant who passes an item with a particular difficulty is assumed to be able to pass all other items which are less difficult. In our example, if a subject is able to pass through a gate with a particular height, she has been able to pass through all gates that are higher (less difficult).

Although instructive, the example of measuring height by confronting participants with gates of different heights simplifies the real complexities of measuring social constructs. For social constructs, participants and items will be ordered with some level of error; that is, there will be—at least theoretically— participants who pass a particular item, but fail to pass an item that is less difficult. Thus, errors are possible when ordering the frequencies of interaction of superintendents with different types of external actors and organizations in the district’s environment. To form a scale, errors should be randomly distributed, and are modeled using a probability function for participants to “pass” an item of a given difficulty.⁷ For any scale, the observed errors are transformed into a “scale homogeneity” coefficient H, which is the inverse of the ratio between observed and expected errors. For individual items i within a scale, an item homogeneity index H_i can be computed. As a rule of thumb, Mokken (1971) suggested the following interpretation of the strength of a scale—which appears to work quite well after decades of scaling analysis. For H > 0.30, there is a response pattern that reflects the existence of a cumulative scale. For 0.30 < H < 0.40, the scale is considered to be weak; for 0.40 < H < 0.50, the scale is considered to be of intermediate strength, and for H > 0.50, there exists a strong scale. For H = 1, the scale is fully deterministic, and no errors occur.

In addition to random error, there is an additional element of complexity in the scaling procedure for our superintendents’ contact frequencies with different types of external actors. These items are not dichotomous (with response categories: pass/fail), but instead polytomous (with multiple response categories varying from “never” to “daily” contacts). For such items, each response category is translated into a separate “item-step” (Molenaar, Mokken, Van Schuur, & Sijtsma, 2000) and scaled accordingly. Consequently, categories are mixed across items when rank ordered at the one-dimensional scale. For example, suppose that the items “contacts with state legislators” and “contacts with local business leaders” form a relatively homogeneous scale. Then, a possible rank ordering might indicate that that it is more difficult for superintendents to maintain monthly contact with state legislators than it is to maintain daily contact with local business leaders. Such a result would imply (a) that the different categories of contacts with state legislators and local business leaders are indicators for the same one-dimensional latent trait of superintendents, for example, their investment in political support and (b) that contacts with state legislators require more investment or effort for superintendents than do contacts with local business leaders.

The difference between a common factor and a cumulative scale of two items is illustrated by looking at their bivariate distribution. If a common factor between two items exists, we expect the responses of superintendents to be linearly distributed. Figure 1(a) displays a hypothetical distribution of observations for a common factor. A common factor exists if the observations are highly correlated. This is illustrated by the area bounded by the dotted line in Figure 1(a). If the two items form a cumulative scale, we expect a different pattern to emerge. Suppose that, in Figure 1(a), item j is more difficult than item i (for example, because the type i actors are less accessible and require more investments of superintendents). Then, superintendents who score high on item i do not necessarily score high on item j. Depending on their personal or district characteristics, superintendents will vary in the extent to which they are able and/or willing to invest in combinations of contacts to these other actors. Thus, some superintendents (close to the item i = item j axis) invest as much time in maintaining contacts with both types of external actors. However, it may also be that superintendents maintain more frequent contacts with type i than with type j actors. Their responses are distributed below the straight line. Therefore, if a cumulative scale exists, we expect a “triangle” of responses. This is illustrated by the area bounded by the solid line in Figure 1(b).

Figure 1.

(a) Expected bivariate distribution of participants for two items with a common factor (dotted line) and two items that form a cumulative scale (solid line); (b) Empirical bivariate distribution of the items “contact frequency with local business leaders” and “contact frequency with state legislators” for the 2002 survey (n = 614; H = 0.38; p = .29).

What do the data show? Figure 1(b) presents the empirical bivariate distribution of the items “contact with local business leaders” and “contact with state legislators” from the 2005 survey (n = 614). These are the empirical plots of all combinations of superintendents’ scores on items i and j. Figure 1(b) clearly shows indications that responses are distributed in a triangle, and not as a straight line.⁸ Therefore, a cumulative scale would be more appropriate. Indeed, responses on both items are not strongly correlated (p = .29) but they do form a cumulative scale (H = 0.38).

Results

We apply a data-driven, exploratory design to obtain cumulative networking activity dimensions scales from the superintendents’ contact frequency data.⁹ The cumulative scaling procedure is a technique which could identify multiple dimensions of support to the school district, as reflected by different patterns in the managerial networking behavior of the district superintendent toward different external actors and organizations. The resulting cumulative scales must satisfy three main criteria. The first, and most important, criterion is that each cumulative scale must be at least a weak scale (H > 0.30) and each item i in the scale should have an item homogeneity index H_i > 0.30. This boundary is standard for Mokken scale analysis (Jacoby, 1994, 2000; Molenaar, Mokken, Van Schuur, & Sijtsma, 2000). Given the boundary of H = H_i > 0.30, we start with two items that form the strongest scale and subsequently add new items with the highest value of H_i, until the lower boundary of H is reached.¹⁰ We included the following six items: contact frequency with “state legislators,” with “local business leaders,” with “parent groups,” with “teacher associations,” with the “Texas Education Agency,” and with “federal education officers.” Items included in one particular scale cannot be included in another scale.

We analyzed the superintendent data for each wave separately. Theoretically, this approach may result in different cumulative scales for separate waves when, for example, a scale for one wave may contain a larger set of items than for a different wave.¹¹ If Hypothesis 1 holds, we should be able to construct (a) homogeneous, (b) consistent, and (c) clearly interpretable cumulative networking activity scales for superintendents’ contacts with external actors. From our analysis of these data, three networking activity scales emerge which satisfy these criteria.¹²

Table 1 (for the 2002 wave) and Table 2 (for the 2005 wave) provide information about the homogeneity index H of all different combinations of the six items in a scale (the six different types of external actors and organizations). The combinations of two items that are selected in a scale are printed in bold. We start our analysis with the 2002 wave. For the combination of the items “contact frequency with state legislators” and “contact frequency with local business leaders,” the homogeneity index H = 0.38, which is highest for all combinations of items in this wave. If we add another item to this combination, the H-value drops below the critical threshold of H = 0.30. Second best is the combination of the items “contact frequency with parent groups” and “contact frequency with teacher associations,” with H = 0.35. Two items are left: “contact frequency with the Texas Education Agency” and “contact frequency with federal education officers.” Adding any of these items to one of the other scales worsens the homogeneity of these scales below the H = 0.30 threshold. However, the value of H for the combination of both items is highest for any combination of these two items with another item. For this reason, and because H > 0.30, we pooled both items in the same scale.

Table 1.

The 2002 Wave: Homogeneity Index H_ij for All Combinations of Contact Frequency Items.

	State legislators	Local business leaders	Texas Education Agency	Federal education officers	Parent groups
1a. State legislators
1b. Local business leaders	0.38^a
2a. Texas Education Agency	0.27	0.16
2b. Federal education officers	0.31	0.18	0.31^a
3a. Parent groups	0.31	0.30	0.20	0.23
3b. Teacher associations	0.31	0.22	0.15	0.30	0.35^a

Values for the three managerial networking activity scales.

Table 2.

The 2005 Wave: Homogeneity Index H_ij for All Combinations of Contact Frequency Items.

	State legislators	Local business leaders	Texas Education Agency	Federal education officers	Parent groups
1a. State legislators
1b. Local business leaders	0.30^a
2a. Texas Education Agency	0.34	0.20
2b. Federal education officers	0.29	0.22	0.37^a
3a. Parent groups	0.21	0.32	0.22	0.34
3b. Teacher associations	0.24	0.24	0.18	0.34	0.41^a

Values for the three managerial networking activity scales.

The exploration of the 2005 wave shows a slightly different picture: not in terms of the underlying structure of the scales retrieved, but with respect to scale homogeneity. The combination of the items “contact frequency with teacher associations” and “contact frequency with parent groups” produces the highest scale homogeneity (H = 0.41). The combination of the items “contact frequency with state legislators” and “contact frequency with Texas Education Agency” produces the second-highest value of scale homogeneity (H = 0.37). If we add any of the two remaining item to this combination, the H-value drops below the critical threshold of H = 0.30. Thus, only the combination of the items “contact frequency with state legislators” and “contact frequency with local business leaders” remains. This combination of items, however, is not a homogeneous scale because H = 0.30. Remarkably, this combination of items produced the highest value of homogeneity index H in the 2002 wave.

Thus, we obtained five combinations of our six items that form weak or intermediate scales of managerial networking activity for the 2002 or 2005 waves. Most scales retrieved have a modest or intermediate strength, but are acceptable scales considering the criterion of H > 0.30. One scale found is intermediate: the “parent groups” and “teacher associations” managerial activity scale for the 2005 wave. We interpret this managerial activity scale as a “coproduction” dimension of support. Education is a coproduced good, and superintendents need both cooperation from teachers to implement their goals and active involvement by parents in the education of their children. We interpret the “state legislators” and local business leaders” managerial activity scale as a dimension of “political support.” Superintendents need local political support to issue tax levies without public resistance as well as support for bond referenda for capital construction. Similarly, they need legislative support to generate greater state funds for education. Both state legislators and local business leaders are essential in providing this support to the school district. The managerial networking activity scale formed by the items “Texas Education Agency” and “federal education” can be interpreted reflecting a dimension of “bureaucratic coping.” Both state and federal bureaucracies are sources of regulations and jointly operate an accountability system for the school districts.

In conclusion, we find that the analysis does produce three scales of managerial networking activity that correspond to meaningful functions and scalable activities of managers. Thus, although the three scales are not quite strong, Hypothesis 1 holds: multiple scales of managerial networking activity are documented, each directed toward a specific type of support that can be interpreted substantively. Managerial networking, in short, is not some undifferentiated behavior but, when explored by this scaling technique, corresponds to distinguishable managerial responsibilities and patterns of relationships. Having provided a more substantive underpinning for the scales found in the superintendents’ networking activities, we can further substantiate the results with a few additional analyses. We start by analyzing the differences between the different types of organizations that form one scale in terms of the superintendents’ contact frequency (that is, the “difficulty” of the items’ categories).

Figure 2 shows the ordering of all the categories in each of the three managerial networking activity scales. The figure makes clear how the categories of the two items in each scale differ in terms of the methodological concept of “difficulty.” The difficulty of the category represents the likelihood that a certain contact frequency (the category) with a specific type of external organization (the item) occurs. We present the results for the 2005 wave (the rank orderings for the 2002 wave are similar). Each scale of managerial networking activity holds 12 categories because each scale is composed of two 6-point items. On these scales, categories from each of the two items are ranked: from “least difficult” to “most difficult”—that is, from those categories representing the least relational investment to those categories representing the most relational investment made by superintendents (in terms of contact frequency). This ordering is represented in bold: from category 1 (least difficult) to category 12 (most difficult). Above the numbers in bold, the ordering is plotted of all categories from each of the two items in the scale. For example, in Figure 2, the fifth and sixth category of the state legislators are most difficult (rank 11 and 12, respectively). Hence, it is least likely that Texas school district superintendents maintain contacts with state legislators more than once a week, or on a daily basis.

Figure 2.

Item categories ranking in difficulty for the four network activity scales in 2005.

Reading from the right to the left for the political support scale, Figure 2 shows that categories (5) and (6) for state legislators are ranked higher than category (6) for local business leaders. This ranking implies that Texas school district superintendents who have daily contact with local business leaders are unlikely to have more than weekly contacts with state legislators. Figure 2 also shows that category (3) for state legislators is ranked higher than categories (3) and (4) for local business leaders. Hence, fewer Texas school district superintendents have monthly contact with state legislators than weekly contact with local business leaders. In short, it turns out that the Texas superintendents maintain relationships with state legislators on a less frequent basis than relationships with local business leaders—even though both types of external actors and organizations yield the same type of support to their school district. Superintendents clearly devote most of their networking activity in contacts with local business leaders. If, however, superintendents need to invest even more in political support—for example, because they believe these relationships are likely to have most impact on their district—they maintain more frequent relationships with state legislators, who are more “distant” and thus more costly to them than are local business leaders.

Comparable conclusions can be drawn with respect to the relative differences in the contact frequency of relationships with the Texas Education Agency (more frequent) and federal education officers (less frequent). Texas school district superintendents who have daily contact with the Texas Education Agency are not likely to have more than weekly contacts with federal education officers. Superintendents clearly devote most of their bureaucratic networking activity to the Texas Education Agency, and only maintain more frequent relationships with federal education officers if they need to invest more in bureaucratic coping. The interpretation is that relationships with federal education officers are more costly to superintendents than is the Texas Education Agency, which provides a combination of support and monitoring to school districts. Federal officers perform a similar function as the Texas Education Authority, but at a more distant (higher) level in the multilevel system of educational governance.

Figure 2 shows that for the coproduction scale, the contact frequency of superintendents’ relationships with parent groups and teacher associations are more similar than for the other scales. Because superintendents maintain yearly contacts (2 is ranked fifth place) with teacher associations when they have at least monthly contacts with parent groups (3 is ranked fourth place), superintendents devote more of their networking activity on this scale with parent groups. More frequent contacts with teacher associations develop only when frequent contacts with parent groups are already exploited. This seems logical because school districts are likely to benefit more from intense relationships with parent groups than with teacher organizations, and parent groups are probably more easily accessible than are teacher organizations. Texas does not have strong teachers’ unions, and thus superintendents have significant discretion in this area.

All in all, we conclude that the different types of external actors and organizations within each scale indeed differ in the contact frequency of relationships that superintendents maintain with these organizations. Thus, the assumption of equal means and standard deviations of items, required for reliability analysis and factor analysis, does not hold. The observed differences in contact frequency between the types of external organizations, captured by the Mokken scaling analysis, are logical and well-justified through substantive argumentation.

Three Managerial Networking Scales or One Factor?

In the previous section, we obtained some indications for the convergent validity of the three managerial networking scales: to what extent do the different types of external actors and organizations in each scale capture the same patterns in managerial networking behavior in terms of homogeneity and substantive plausibility? Below, we examine the correlations between the three different scales of managerial networking activity as a further check on the discriminant validity of the three scales: to what extent do the three scales capture different patterns in managerial networking behavior? We checked the correlations across networking dimensions in Table 3 for the years 2002 and 2005. Clearly, the different dimensions of managerial networking activity are associated, but not highly. The correlations are generally low, between p = .26 and p = .36. The highest correlation exists between bureaucratic coping and coproduction for 2005 (p = .36). Thus, we have further confidence that the three scales of managerial networking activity measure different dimensions of support from the organization’s environment.

Table 3.

Pearson Correlations Between the Three Managerial Networking Activity Scales.

	2002 (n = 597)		2005 (n = 599)
	2	3	2	3
1. Political support	0.26	0.33	0.33	0.34
2. Bureaucratic coping	1.00	0.28	1.00	0.36
3. Coproduction		1.00		1.00

Because previous studies have consistently found only one factor accounting for the networking activity of superintendents, we compare the results of the previous cumulative scaling analyses with the results of a factor analysis. Such a comparison provides more detailed information about the empirical difference between the common variance found in the superintendents’ contact frequencies with the six different types of organizations, and their managerial networking activity toward the specific combinations of types of external actors and organizations suggested by the cumulative scaling analysis. We performed a factor analysis using the set of six items included in the analysis of the cumulative scales: state legislators, local business leaders, Texas Education Agency, federal education officers, parent groups, and teacher associations, following precisely the same steps as O’Toole and Meier (2004b) in their analysis.

Table 4 shows the results of the factor analysis, which exactly replicates the results reported by Meier and O’Toole in all their previous studies. For each year, only one factor is present in the data with an Eigenvalue larger than one.¹³ Table 4 shows that all six types of external actors and organizations (the items in the analysis) load positively on the first factor. The factor loadings of the six items vary between 0.37 and 0.53, but no clear pattern can be found—other than that all six items contribute to the common variance in the managerial networking activity. The uniqueness of the six items, which is the proportion of variance of the item that is not accounted for by all of the factors taken together, is fairly high. This might be interpreted as an indicator that the six items are not strongly related to the managerial networking factor.

Table 4.

Results of the Factor Analysis.

	2002 (n = 597)		2005 (n = 599)
	One-factor solution Eigenvalue = 1.22		One-factor solution Eigenvalue = 1.39
Contact frequency items	Factor loadings	Uniqueness	Factor loadings	Uniqueness
State legislators	0.51	0.74	0.47	0.76
Local business leaders	0.41	0.83	0.43	0.81
Texas Education Agency	0.37	0.83	0.42	0.79
Federal education officers	0.45	0.78	0.53	0.72
Parent groups	0.46	0.78	0.52	0.71
Teachers associations	0.49	0.76	0.50	0.73

To explore the differences between the three managerial networking scales and the single factor, Table 5 presents the correlations between the superintendents’ score on the three scales and their factor score. The three scales aggregate the networking activity of the superintendents in a different way, and using different combinations of external organizations, than does the single factor. Still, Table 5 shows that the managerial networking factor is strongly correlated with all three managerial networking scales, but not completely (correlations range between p = .67 and p = .78). Given the modest correlations between the three scales, as presented by Table 3, we conclude that—although the single factor captures common variance between the three different scales—these three scales capture different behavioral patterns of the superintendents’ networking activity themselves.

Table 5.

Pearson Correlations Between the Managerial Networking Activity Factor and the Three Managerial Networking Activity Scales.

	Factor
	2002 (n = 597)	2005 (n = 599)
Political networking	0.71	0.70
Bureaucratic coping	0.67	0.75
Coproduction	0.77	0.78

Does It matter for Performance? An Illustration

We have shown that, using an appropriate technique, the behavioral patterns of managerial networking by Texas school district superintendents match the predictions of the public management literature on environmental support. However, as always, the “proof of the pudding is in the eating.” Does it matter for explaining school district performance whether we use the three different scales as explanatory indicators for networking activity or the single factor? Actually, it matters a great deal. Recall, for example, Moynihan and Pandey’s (2005) study which shows that managers’ perceived political support positively affects the performance of their agency, whereas their perceived support from clients negatively affects performance. Differences between the three managerial networking scales in the sizes or direction of associations with performance could exist. Because a thoroughgoing explanation of school district performance is beyond the scope of the present article, we present a performance analysis below only to illustrate the new potential offered by the three managerial networking scales.

Table 6 presents the results of an OLS-regression of school district performance including both the managerial networking dimension scores and the factor scores as explanatory variables. We use a comparable design to that employed by O’Toole and Meier (2004b) and also a comparable performance dependent variable. School district performance is here defined in terms of the percentage of Latino students who pass the TAAS tests. Year is included as a dummy variable and clustered standard errors are computed to control for the nesting of years in school districts. The analysis controls for student composition, expenditures, and student to teacher ratio. The first model in Table 6 estimates the effect of the single managerial networking factor. We find that the factor score has a significant positive effect on the pass rates for Latino students.¹⁴ Hence, the general networking activity of Texas superintendents positively affects the performance of their school district for this subpopulation of students, while controlling for various potentially confounding variables. This result replicates those reported by the previous studies of Meier and O’Toole, as well as the explained variance, which is around 65% (O’Toole & Meier, 2004b).

Table 6.

OLS Regression of Pass Rates for Latino Students (Unstandardized Coefficients; Clustered Standard Errors Between Parentheses).

	One-factor model	Scales main effect model	Scales interaction effect model
Networking activity variables
Factor score	0.86* (0.49)
Political Networking Scale		0.18 (0.27)	0.18 (0.27)
Bureaucratic Coping Scale		0.47* (0.25)	0.54** (0.26)
Coproduction Scale		−0.03 (0.28)	0.03 (0.28)
Political × Bureaucratic			−0.33* (0.17)
Control variables
Year = 2005	−27.23*** (0.74)	−27.14*** (0.76)	−27.16*** (0.75)
Percentage Afro-American students	−0.06 (0.04)	−0.06 (0.04)	−0.06 (0.04)
Percentage Latino students	0.01 (0.03)	0.01 (0.03)	0.01 (0.03)
Percentage low-income students	−0.20*** (0.04)	−0.20*** (0.04)	−0.20*** (0.04)
Total expenditures^a	0.25 (2.15)	0.88 (2.22)	0.59 (2.26)
Student to teacher ratio	0.09 (0.27)	0.10 (0.27)	0.09 (0.27)
Instructional expenditures^b	0.70 (0.72)	0.71 (0.72)	0.74 (0.70)
Constant	87.99*** (6.11)	85.75*** (6.17)	85.44*** (6.08)
N	1,127	1,127	1,127
Adjusted R²	.64	.64	.64

Note: Cases are clustered in districts.

Expenditures in US$10⁹.

Instructional expenditures in US$10³ per pupil.

p < .10. **p < .05. ***p < .005.

The second and third model in Table 6 show quite different effects for the three different scales. The main effects model, which includes the superintendents’ scores on each of the three managerial networking activity scales, shows that superintendents’ bureaucratic coping has a positive effect on the pass rates for Latino students. Neither political support nor coproduction seem to affect school district performance. The results, however, change when we introduce an interaction effect for political support and bureaucratic coping in the final model. Although this interaction effect is negative and significant at p < .10, the main effect of bureaucratic coping becomes more significant. Thus, a positive effect of bureaucratic coping on performance is moderated by political networking activity. This analysis suggests that, although it helps superintendents to seek support from the bureaucratic environment of their school district, it is bad for school district performance to invest much in both types of support. It is beyond the scope of the present article to examine this relationship in more detail—although we find exactly the same negative interaction effect in data that were collected in a different educational context than the Texas school districts (Akkerman & Torenvlied, 2010).¹⁵ The negative interaction effect of political support and bureaucratic coping most likely explains why we observe only a small effect of the factor score on the pass rates for Latino students.

Conclusion and Discussion

The most important message of the present article is that managerial networking is not a one-dimensional phenomenon. The multidimensional nature of managerial networking activity logically follows from the assumption that managers have only limited time and resources available and weigh the relative costs and benefits of maintaining a relationship with an external actor or organization (cf. Agranoff, 2006, 2007). When managers are selective in their networking activities, different dimensions of managerial networking should emerge. We indeed find evidence for the existence of multiple dimensions of networking activity in a large-n, two-wave data set on the frequency of managerial networking contacts maintained by Texas school district superintendents.

Three dimensions of networking activity emerge from the Texas school district data, thus showing distinctive activities aiming at (a) political support, (b) bureaucratic coping, and (c) coproduction. These dimensions are consistent and stable across years, and well-interpretable for the context in which school district superintendents operate. External organizations differ in the extent to which they receive more investments from superintendents. The dimensions elicited in the present analysis have parallels to several of those developed recently from managerial data gathered in a different country (Akkerman & Torenvlied, 2010). It would appear, therefore, that the networking dimensions developed here may have importance beyond the field of public education in Texas, and beyond this particular data set.

The findings in the present article provide further empirical support for the idea in public management that agencies are able to draw on different types, and different sources, of support in their environment (Bozeman, 1987; Heymann, 1987; Lynn, 2007; Moore, 1995, 2000; Moynihan & Pandey, 2005; Rainey & Steinbauer, 1999). The present article also illustrates how a scaling approach—one that is virtually absent in public management research—can be used to explicate behaviorally distinct patterns that map onto managerially distinct functions. The technique is cumulative scaling, based on item-response theory. Almost all other public management studies use reliability analysis or factor designs, despite indications that a cumulative scaling technique is sometimes an appropriate alternative, as shown in the present article.

The illustrative analysis of school district performance reveals that the single networking factor and the three networking scales have distinctive effects on the pass rates of Latino students. Managerial networking activity in general (the single factor) has a positive but only slight effect on school district performance. Although one dimension of managerial networking activity (bureaucratic coping) is strongly and positively associated with school district performance, this effect is moderated by political networking activity. Apparently, too much managerial activity on both dimensions is bad for performance. This finding provides further empirical support for the approach taken, and may stimulate the development of more substantive explanations for the nonlinear effects of managerial networking activity, as recently proposed in the literature (Hicklin, O’Toole, & Meier, 2008; Provan & Sydow, 2008, p. 704). The finding also bears resemblance to the contradictory effects of political support and client support on the performance of public agencies (Moynihan & Pandey, 2005), and has been replicated on managerial data gathered in a different country (Akkerman & Torenvlied, 2010).

Our results thus shed new light on extant theories of managerial networking. Current approaches make elaborate distinctions between the functions of relationships with external actors and organizations (exchange of information and resources, building trust, coordination of activities), but pay relatively little attention to the types of networking contacts needed to build such relationships. Thus, the present study provides a clear direction for further specification of explanatory models of managerial networking—by focusing on the deliberate choices managers make to initiate and maintain contacts with specific types of external organizations. Such a model can provide a much stronger theoretical underpinning to the correlates between managerial networking and performance than has been done in earlier work, which has mainly focused on general networking activity and/or generic network structures. Paired with empirical analyses of the dimensions of managerial networking, the explanatory model can be contextualized for specific agencies, policy sectors, or institutional settings. Thus, the multiple dimensions model of managerial networking can relate the constrained, goal-oriented networking behavior of public managers to specific indicators for the performance of their agency.

To provide deeper and more nuanced explanations for variation in agency performance, at least two steps need to be taken in future research. The first is to identify different types of managers, based on their networking activity on the separate dimensions. Some managers, for example, may be active in political networking but less so in their relationships with bureaucratic actors. Other managers might instead focus their networking activity toward organizations that are important for coproduction. The determinants of differences in managers’ networking strategies are important in efforts to link agency goals to networking activity. A second line of analysis would be to test more systematically how strongly specific networking approaches by managers may affect various indicators of agency performance. It would seem, therefore, that some important lines of additional empirical work are opened up by the type of analysis conducted here.

The public managers under study, the school district superintendents, work in a specific context and results cannot be fully generalized to other public organizations. Therefore, the present analysis needs to be replicated in different institutional contexts—for example, in different countries and also, data permitting, in other fields of policy and management aside from public education. The current investigation provides strong evidence that distinct scales of managerial networking can represent different important functions as managers ply their trade outward in the interdependent environments of their own organizations. Identifying such patterns in other empirical settings would be a significant contribution to knowledge.

Footnotes

Acknowledgements

The authors are grateful to Wijbrandt H. van Schuur and three anonymous reviewers for their useful and detailed comments and suggestions on previous drafts.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: VIDI program of the Dutch Organization for Scientific Research and the “high potentials” program of Utrecht University.

Notes

Author Biographies

René Torenvlied is a Professor of Public Sector Management at Leiden University, the Netherlands, and Scientific Director of the Institute of Public Administration at Leiden University. His research interests include (formal modeling of) political-administrative processes, delegation and accountability in democratic systems, networks and networking in public management, manifestations of red tape, as well as public performance in the fields of education, health, industrial relations, and security.

Agnes Akkerman is an Associate Professor in the Department of Political Science and Public Administration of Nijmegen School of Management, Radboud University Nijmegen, the Netherlands. She is principal investigator of a major research project into the consequences of industrial conflict, in particular the contagion of industrial conflict and the impact of industrial conflict on work relations. Her further research interests include public sector networks and socio-economic policy making.

Kenneth J. Meier is the Charles H. Gregory Chair in Liberal Arts and Distinguished Professor of Political Science at Texas A&M University. He is also a Professor of Public Management in the Cardiff School of Business, Cardiff University (Wales). In addition to his major research agenda on empirical studies of public management, he is interested in race and public policy, methodological innovations in public administration and the relationship between democracy and bureaucracy.

Laurence J. O’Toole, Jr. is the Margaret Hughes and Robert T. Golembiewski Professor of Public Administration as well as Distinguished Research Professor in the Department of Public Administration and Policy, School of Public and International Affairs, at the University of Georgia (USA). His research interests include public management and organizational performance, networks, and environmental and education policy and management.

References

Agranoff

(2006). Inside collaborative networks: Ten lessons for public managers. Public Administration Review, 66, 56-65.

Agranoff

(2007). Managing within networks: Adding value to public organizations. Washington, DC: Georgetown University Press.

Agranoff

McGuire

(2003). Big questions in public network management research. Journal of Public Administration Research and Theory, 13, 295-326.

Akkerman

Torenvlied

(2010, July 23-24). Pupils, principals, and politics: The embeddedness of Dutch primary schools in their local political-administrative environment. Paper presented at the conference on “Democracy, Policy, and Governance: A European Perspective,” Texas A&M University, College Station.

Akkerman

Torenvlied

(2011). Managing the environment: Effects of network ambition on agency performance. Public Management Review, 13, 159-174.

Akkerman

Torenvlied

Schalk

(2012). Two-level effects of inter-organizational network collaboration on graduate satisfaction: A comparison of five inter-college networks in Dutch higher education. American Review of Public Administration. Advance online publication. doi:10.1177/0275074011425418.

Aldrich

H. E.

(2008). Organizations and environments (Stanford Business Classics reprint from 1979). Stanford, CA: Stanford University Press.

Allen

T. J.

(1984). Managing the flow of technology: Technology transfer and the dissemination of technological information within the R&D organization. Cambridge, MA: MIT Press.

Bardach

(1998). Getting agencies to work together: The practice and theory of managerial craftsmanship. Washington, DC: Brookings Institution Press.

10.

Bozeman

(1987). All organizations are public. San Francisco, CA: Jossey-Bass.

11.

Brass

Galaskiewicz

Greve

H. R.

Tsai

(2004). Taking stock of networks and organizations: A multilevel perspective. Academy of Management Journal, 47, 795-817.

12.

Carroll

J. B.

(1945). The effect of difficulty and chance success on correlations between items or between tests. Psychometrika, 10, 1-19.

13.

Embretson

S. E.

Reise

S. P.

(2000). Item response theory for psychologists. Mahwah, NJ: Lawrence Erlbaum.

14.

Emery

Trist

(1965). The causal texture of organizational environments. Human Relations, 18, 21-32.

15.

Ferguson

G. A.

(1941). The factorial interpretation of test difficulty. Psychometrika, 6, 323-329.

16.

Geletkanycz

M. A.

Boyd

B. K.

Finkelstein

(2001). The strategic value of CEO external directorate networks: Implications for CEO compensation. Strategic Management Journal, 22, 889-898.

17.

Hanf

Hjern

Porter

D. O.

(1978). Local networks of manpower training in the Federal Republic of Germany and Sweden. In Hanf

Scharpf

(Eds.), Interorganizational policymaking: Limits to coordination and central control (pp. 303-341). Beverly Hills, CA: Sage.

18.

Hardouin

J. B.

(2004). MSP / LoevH stata modules. Statistical software components. S439402. Chestnut Hill, MA: Boston College Department of Economics.

19.

Heymann

P. B.

(1987). The politics of public management. New Haven, CT: Yale University Press.

20.

Hicklin

O’Toole

L. J.

Jr. Meier

K. J.

(2008). Serpents in the sand: Managerial networking and nonlinear influences on organizational performance. Journal of Public Administration Research and Theory, 18, 253-273.

21.

Hite

J. M.

Williams

E. J.

Baugh

S. C.

(2005). Multiple networks of public school administrators: An analysis of network content and structure. International Journal of Leadership in Education, 8, 91-122.

22.

Jacoby

W. G.

(1994). Public attitudes toward government spending. American Journal of Political Science, 38, 336-361.

23.

Jacoby

W. G.

(2000). Issue framing and public opinion on government spending. American Journal of Political Science, 44, 750-767.

24.

Kim

Müller

C. W.

(1978). Factor analysis: Statistical methods and practical issues. Beverly Hills, CA: Sage.

25.

Krackhardt

(1994). Constraints on the interactive organization as an ideal type. In Heckscher

Donnellan

(Eds.), The post-bureaucratic organization (pp. 211-222). Beverly Hills, CA: Sage.

26.

Lawrence

P. R.

Lorsch

J. W.

(1967). Organization and environment: Managing differentiation and integration. Boston, MA: Harvard University Press.

27.

Leana

C. L.

Pil

F. K.

(2006). Social capital and organizational performance: Evidence from urban public schools. Organization Science, 17, 353-366.

28.

Lynn

L. E.

(2007). Public management. In Peters

B. G.

Pierre

(Eds.), Handbook of public administration (pp. 13-23). Thousand Oaks, CA: Sage.

29.

Meier

K. J.

(2000). Politics and the bureaucracy. Policymaking in the fourth branch of government (4th ed.). Fort Worth, TX: Harcourt.

30.

Meier

K. J.

O’Toole

L. J.

Jr. (2001). Managerial strategies and behavior in networks: A model with evidence from U.S. public education. Journal of Public Administration Research and Theory, 11, 271-293.

31.

Meier

K. J.

O’Toole

L. J.

Jr. (2003). Public management and educational performance: The impact of managerial networking. Public Administration Review, 63, 689-699.

32.

Meier

K. J.

O’Toole

L. J.

Jr. Boyne

Walker

R. M.

(2007). Strategic management and the performance of public organizations: Testing venerable ideas against recent theories. Journal of Public Administration Research and Theory, 17, 357-377.

33.

Meier

K. J.

O’Toole

L. J.

Jr. (2008). Management theory and Occam’s razor: How public organizations buffer the environment. Administration and Society, 39, 931-958.

34.

Mokken

R. J.

(1971). A theory and procedure of scale analysis. With applications in political research. Hague, Netherlands: Mouton.

35.

Molenaar

I. W.

Mokken

R. J.

Van Schuur

W. H.

Sijtsma

(2000). MSP version 5. Groningen, Netherlands: Progamma.

36.

Moore

M. H.

(1995). Creating public value: Strategic management in government. Cambridge, MA: Harvard University Press.

37.

Moore

M. H.

(2000). Managing for value: Organizational strategy in for-profit, nonprofit, and governmental organizations. Non Profit and Voluntary Sector Quarterly, 29, 183-204.

38.

Moynihan

D. P.

Pandey

S. K.

(2005). Testing how management matters in an era of government by performance management. Journal of Public Administration Research and Theory, 15, 421-439.

39.

O’Toole

L. J.

Jr. (1997). Implementing public innovations in network settings. Administration and Society, 29, 115-138.

40.

O’Toole

L. J.

Jr. Meier

K. J.

(1999). Modeling the impact of public management: Implications of structural context. Journal of Public Administration Research and Theory, 9, 505-526.

41.

O’Toole

L. J.

Jr. Meier

K. J.

(2004a). Public management in intergovernmental networks: Matching structural networks and managerial networking. Journal of Public Administration Research and Theory, 14, 271-293.

42.

O’Toole

L. J.

Jr. Meier

K. J.

(2004b). Desperately seeking Selznick: Cooptation and the dark side of public management in networks. Public Administration Review, 64, 681-693.

43.

O’Toole

L. J.

Jr. Meier

K. J.

(2011). Public management: Organizations, governance, and performance. Cambridge, UK: Cambridge University Press.

44.

Peng

M. W.

Luo

(2000). Managerial ties and firm performance in a transition economy: The nature of a micro-macro link. Academy of Management Journal, 43, 486-501.

45.

Pfeffer

Salancik

G. R.

(1978). The external control of organizations: A resource dependence perspective. New York, NY: Harper and Row.

46.

Powell

W. W.

Smith-Doerr

(1994). Networks and economic life. In Smelser

N. J.

Swedberg

(Eds.), The handbook of economic sociology (pp. 368-402). Princeton, NJ: Princeton University Press.

47.

Provan

K. G.

Kenis

(2008). Modes of network governance: Structure, management, and effectiveness. Journal of Public Administration Research and Theory, 18, 229-252.

48.

Provan

K. G.

Milward

H. B.

(1995). A preliminary theory of interorganizational network effectiveness: A comparative study of four community mental health systems. Administrative Science Quarterly, 40, 1-33.

49.

Provan

K. G.

Sydow

(2008). Evaluating inter-organizational relationships. In Cropper

Ebers

Huxham

Smith

(Eds.), The Oxford handbook of inter-organizational relations (pp. 691-716). Oxford, UK: Oxford University Press.

50.

Rainey

H. G.

Steinbauer

(1999). Galloping elephants: Developing elements of a theory of effective government organizations. Journal of Public Administration Research and Theory, 9, 1-32.

51.

Rice

R. E.

Aydin

(1991). Attitudes toward new organizational technology: Network proximity as a mechanism for social information processing. Administrative Science Quarterly, 36, 219-244.

52.

Schalk

Torenvlied

Allen

(2010). Network embeddedness and organizational performance. The strength of strong ties in Dutch higher education. Journal of Public Administration Research and Theory, 20, 629-653.

53.

Schneider

S. K.

Jacoby

W. G.

Coggburn

J. D.

(1997). The structure of bureaucratic decisions in the American States. Public Administration Review, 57, 240-249.

54.

Scott

(2000). Social network analysis: A handbook. London, UK: Sage.

55.

Stam

Elfring

(2008). Entrepreneurial orientation and new venture performance: The moderating role of intra- and extraindustry social capital. Academy of Management Journal, 51, 97-111.

56.

Torenvlied

Akkerman

(2012). Effects of managers’ work motivation and networking activity on their reported levels of external red tape. Journal of Public Administration Research and Theory. Advance online publication. doi:10.1093/jopart/mur076.

57.

Van Schuur

W. H.

(2003). Mokken scale analysis: Between the Guttman Scale and Parametric Item Response Theory. Political Analysis, 11, 139-163.

58.

Van Schuur

W. H.

(2011). Ordinal item response theory (Quantitative Applications in the Social Sciences series). Los Angeles, CA: Sage.

59.

Zaheer

McEvily

Perrone

(1998). Does trust matter? Exploring the effects of interorganizational and interpersonal trust on performance. Organization Science, 9, 141-159.