ONPROFIT ANAGEMENT  L EADERSHIP  vol
59K - views

ONPROFIT ANAGEMENT L EADERSHIP vol

20 no 4 Summer 2010 57513 2010 Wiley Periodicals Inc 405 Published online in Wiley InterScience wwwintersciencewileycom DOI 101002nml20002 The Effectiveness of Nonpro64257t LeadOrganization Networks for Social Service Delivery Bin

Tags : Summer
Download Pdf

ONPROFIT ANAGEMENT L EADERSHIP vol




Download Pdf - The PPT/PDF document "ONPROFIT ANAGEMENT L EADERSHIP vol" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.



Presentation on theme: "ONPROFIT ANAGEMENT L EADERSHIP vol"— Presentation transcript:


Page 1
ONPROFIT ANAGEMENT & L EADERSHIP , vol. 20, no. 4, Summer 2010  2010 Wiley Periodicals, Inc. 405 Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/nml.20002 The Effectiveness of Nonprofit Lead-Organization Networks for Social Service Delivery Bin Chen, Elizabeth A. Graddy Public agencies increasingly contract with nonprofit organiza- tions to lead community-based networks for social service deliv- ery. We explore the role that partnership characteristics play in the effectiveness of these networks. Using data on children and

family services in Los Angeles County, we consider the impact of both the motivations for forming partnerships and the nature of the resulting partnerships on perceived outcomes for clients, interorganizational relationships, and organizational learning. We find that client outcomes and interorganizational relation- ships are enhanced when partnerships are formed to meet cer- tain programmatic and organizational goals. Organizational learning, however, is affected only when partnerships are formed to enhance organizational legitimacy. Partners selected because they share common vision

increase effectiveness, while those selected because there are few alternative partners decrease effec- tiveness. Finally, when partnerships use an interorganizational coordination mechanism, client outcomes are improved. The managerial implications of these impacts for the nonprofit sec- tor are developed. The results lend considerable support to the role of partnership motivation and partner selection in the effec- tiveness of nonprofit lead-organization networks, and specificity about the nature of that role. OTH THE PUBLIC AND NONPROFIT management literatures have focused

considerable attention on formation of interorganiza- tional partnerships for delivery of publicly funded social ser- vices (Kettl, 2006; Bingham and O’Leary, 2008; O’Leary and Note: We appreciate the helpful comments of three anonymous reviewers. We are grateful for the cooperation of the Los Angeles County Family Preservation Program, and especially Rhelda Shabazz and Walter Yu-lung Kiang for their considerable help and expertise. This paper was supported by a research grant from the Professional Staff Congress of the City University of New York.
Page 2
Bingham, 2009). By any

measure, there has been a large increase in such alliances (O’Leary, Gerard, and Bingham, 2006; Gazley and Brudney, 2007; Agranoff, 2007). Different types of collaborative ser- vice arrangements can be distinguished on a continuum of cooper- ation, coordination, collaboration, and service integration (Selden, Sowa, and Sandfort, 2006). One increasingly popular model of interorganizational partnership is the formation of lead-organization networks, in which public funding agencies develop a single con- tracting relationship with a lead organization and then encourage or mandate creation of a

community-based network of service providers (Graddy and Chen, 2006; Poole, 2008; Johnston and Romzek, 2008; Chen, 2008; Provan and Kenis, 2008). Just as it often makes more sense to hire a general contractor to build your home than to try to hire and manage a slew of contractors yourself, it is often in the best interests of a public agency to hire a lead organization to manage the network (Goldsmith and Eggers, 2004). Many nonprofit social service providers are selected by public funding agencies to lead in creating community-based service delivery networks. These net- works can facilitate

information sharing, mobilize resources, and reduce gaps and overlaps in service delivery (Poole, 2008). Yet we are only beginning to explore the effectiveness of these lead- organization network arrangements. In this article, we seek to understand the effectiveness of dyadic partnerships created within the nonprofit lead-organization networks to deliver publicly funded social services. In particular, we explore the role played by the lead organization’s motivation for forming a partnership and by the characteristics of the resulting partnership on collaborative effectiveness.

Understanding the association between partnership characteristics and effectiveness should inform a nonprofit’s decision to form dyadic partnerships within the net- work and enhance the likelihood of its survival. The key question for network effectiveness is “effectiveness for whom?” (Provan and Kenis, 2008). The effectiveness of service deliv- ery networks can be assessed from several perspectives: those of the clients, the organizational partners, the service “industry,” and the community. The specific network outcomes should be assessing the network’s functioning (Provan and Milward,

2001). Agranoff (2007) suggests that ex post judgments of key players be used to evaluate col- laboration outcomes. In a hub-and-spoke arrangement that contains dyadic ties of a single organization, the lead agency is ultimately held accountable for overall network maintenance and functioning (Provan and Kenis, 2008). We focus here on how effective the dyadic partnerships within the network are perceived by its lead organiza- tion in terms of achieving service goals, improving interorganizational relationships, and improving organizational learning. In the next section, we develop the

theoretical connection between partnership characteristics and effectiveness. Then we explore our model with data on the delivery of family preservation services 406 HEN , G RADDY Nonprofit Management & Leadership DOI: 10.1002/nml
Page 3
collected in Los Angeles County. We conclude with a discussion of the implications of our findings for nonprofit lead agencies. Partnership Characteristics and Effectiveness Efforts to understand the determinants of effectiveness in the broader set of partnerships suggest that partner reputation, prior relationships (an indicator of

trust), shared decision making, and similarity between partners are positively associated with alliance outcomes (Saxton, 1997). Arya and Lin (2007) found that organi- zational characteristics, partner attributes, and network structure are related to collaboration outcomes in nonprofit service net- works. Thomson, Perry, and Miller (2008) and Chen (2008) linked some collaborative processes with outcomes. A small num- ber of studies looked at the impact of interorganizational collabo- rations and networks on improved perceived and actual client outcomes of social service programs, such as

mental health (Provan and Milward, 1995), early child care and education (Selden, Sowa, and Sandfort, 2006), and welfare-to-work programs (Ewalt, 2004). This stream of research suggests that network integration, centralization, and stability and the intensity of the collaborative relationship are related to positive client outcomes. It is plausible that the outcomes of a partnership between two organizations will be influenced by the reasons the partnership was formed. The purpose of the relationship should suggest the areas in which it is likely to be effective. But partnerships formed

for the same general purposes may differ substantially in how well the part- ners can and do work together. This suggests that the nature of the partnership is also likely to influence its ultimate effectiveness. In this section we develop the relationship between the lead agency’s motivation for partnerships and their expected outcomes, and then we consider how the nature of those partnerships can mit- igate this relationship. Motivation The literature suggests two broad categories of benefits that lead organizations seek from strategic partnerships (Guo and Acar, 2005; Cho and

Gillespie, 2006): resource exchange , where they seek a resource needed for their activities that they do not possess (Gazley and Brudney, 2007; Arya and Lin, 2007); and organiza- tional legitimacy , where they seek associational advantages from a well-respected or well-connected partner (Provan, Kenis, and Human, 2008). More specifically, in the public service delivery context, we develop two broad motivations that explain formation of strategic partnerships between lead agencies and their partner- ing organizations: to obtain resources to meet programmatic needs, and to achieve

organizational goals. HE FFECTIVENESS OF ONPROFIT EAD -O RGANIZATION ETWORKS 407 Nonprofit Management & Leadership DOI: 10.1002/nml Partnerships formed for the same general purposes may differ substantially in how well the partners can and do work together.
Page 4
Programmatic Needs. Complex programmatic needs in social ser- vice delivery can drive formation of partnerships. Individual orga- nizations are constrained by technological, political, and cognitive limits in the face of complex, many-sided problems. Establishing strategic alliances may afford an organization access

to tacit knowl- edge and complementary skills, new technologies or markets, and the ability to provide a wider range of products and services beyond its organizational boundaries (Dyer and Singh, 1998). Through qualitative interviews with lead-organization execu- tives, we identify five types of resource acquisitions that are likely to be important in helping lead providers of social services meet their programmatic needs: extra provision capacity for case overload, spe- cific service expertise, geographic coverage, local knowledge and client access, and cultural and linguistic

competence. A single social service agency may not have the capacity to handle growing case- loads alone. An individual agency is constrained in the number and types of services it can provide to clients because of agency tradition and specialization. There may be spatial restrictions on a social ser- vice provider’s ability to serve clients scattered over a large geo- graphic area. An organization may lack local knowledge and access to targeted clients. Therefore, some partnerships will be strategically formed for their ability to furnish such complementary resources. We can also explore what

specific kinds of resources lead organiza- tions find most useful in managing collaborative partnerships (Gazley, 2008). Organizational Goals. Lead organizations may seek partnerships that enhance their legitimacy. Legitimacy is defined as actions and be- haviors of a network or an organization that are perceived as desir- able and appropriate by key external and internal stakeholders (Provan, Kenis, and Human, 2008). We identify three such motiva- tions: to meet funding agency expectations or requirements, to en- hance organizational reputation, and to build future relationships.

Local funding agencies are increasingly promoting a partnership plan as a condition for receiving service delivery contracts. If public fund- ing is, or appears to be, contingent on a collaborative plan, then so- cial service agencies, which depend on public contracts, will comply. Partnerships can convey legitimacy in other ways. Institutional theory suggests that strategic alliances can originate from an organi- zation’s motives to improve its reputation, image, prestige, or congru- ence with the prevailing norms in its institutional environment. A nonprofit social service organization

may thus choose to affiliate with a reputable organization to enhance its own reputation (Galaskiewicz, Bielefeld, and Dowell, 2006). This in turn may make it more com- petitive for public funding. Organizations may also seek partners because they want to build future relationships with them, either to enhance their ability to gain future contracts or to achieve other organizational goals. 408 HEN , G RADDY Nonprofit Management & Leadership DOI: 10.1002/nml
Page 5
Outcomes It is reasonable to expect that the motivation for forming partner- ships will affect their

effectiveness. We consider three dimensions of effectiveness: client goal achievement, enhanced interorganiza- tional relationships, and improved organizational learning. These dimensions reflect the types of goals sought by lead agencies from service delivery partnerships. The first—client goal achievement refers to the primary purpose of most public-sector efforts to en- courage interorganizational partnerships: to gain resources that will improve public service delivery. The second—enhanced interorganizational relationships—captures both potential collec- tive and organizational

benefits of partnerships. If partners work well together, this may enhance the social capital in the commu- nity served. Better working relationships among organizations en- hance the opportunity for community problem solving and pave the way for better future relationships (Gazley and Brudney, 2007). Organizations themselves may also benefit if these relationships enhance their legitimacy. The third dimension—organizational learning—most directly benefits the lead organization. If the partnership enhances the orga- nization’s learning, this may increase its capacity to compete effec-

tively for future contracts and improve its ability to achieve its mission and goals. The mechanism for such learning often comes from working with another organization in developing a shared understanding of the problem and reaching a consensus on how to address it (Gray, 2000; Agranoff, 2008). When partnerships within the network are formed to meet pro- grammatic needs, we expect a strong positive association with client goal achievement. Because the partnership is viewed as supplying programmatic components that are necessary for successful service delivery, the expectation is that the

resulting service provision will be enhanced. If lead organizations come to rely on the specific exper- tise and capacity of their partners, this can lead to better collabora- tive processes and thus better outcomes (Doz and Hamel, 1998). Strong collaborative processes should enhance interorganizational relationships; therefore, we also expect a positive association with this indicator of effectiveness. However, organizational learning may not be correlated with partnerships formed to meet programmatic needs. Therefore, we do not expect partnerships motivated by pro- grammatic needs to

necessarily promote organizational learning. When partnerships are formed to enhance organizational legiti- macy, we expect a positive effect on organizational learning and on improved interorganizational relationships. Lead organizations seek to build their collaborative relationships with a respected organiza- tion not only to enhance their own reputation and gain greater legit- imacy but also to develop a foundation for future collaboration. Given these expectations, this motivation should be positively associated HE FFECTIVENESS OF ONPROFIT EAD -O RGANIZATION ETWORKS 409 Nonprofit

Management & Leadership DOI: 10.1002/nml We do not expect partnerships motivated by programmatic needs to necessarily promote organizational learning.
Page 6
with these measures of effectiveness. However, when a lead organiza- tion seeks a partner to position itself to receive funding rather than because it views the partner as necessary to meet client goals, the part- nership is presumably less likely to yield effective client outcomes. Therefore, we do not expect an impact on client goal achievement. To summarize, partnerships formed to meet programmatic needs are expected to

produce improved client outcomes and improved interorganizational relationships. Partnerships formed to achieve organizational goals are expected to produce improved organiza- tional learning and improved interorganizational relationships. Partnership Characteristics This expected relationship between the goals of partnerships and their effectiveness is likely to be mitigated by the characteristics of the partnership. How well partners work together will have an im- pact on their effectiveness. Although the potential benefits for partners are always elaborated, research on interorganizational

col- laborations tends to overlook the cost side. Partners involved in any collaborative endeavor should also be assumed to assess the trade-off of cost and benefits (Graddy and Ferris, 2006; Gazley, 2008). Given the transaction and coordination costs of interorga- nizational collaborations, partnership characteristics that reduce these costs should enhance effectiveness. Key characteristics fall into two categories: specific partner relationships and partnership governance. Organizations that have prior knowledge of each other, or that have similar missions, should be viewed as more

trustworthy, and trust lowers the transaction costs of partnerships. Coordination costs should be lower if there are formal agreements that define the nature of the partnerships and if there are interor- ganizational groups or mechanisms that facilitate joint decision making. Partners. A partnership is an inherently risky endeavor. The inte- gral characteristic is mutual interdependence, and this interdepen- dence implies vulnerability to the behavior of one’s partner. This vulnerability can be mitigated by selecting partners the organiza- tion views as trustworthy (Snavely and Tracy, 2002).

Such trust can be viewed as incorporating both the intentions of a partner and its competence (Das and Teng, 2001). Can and will the poten- tial partner behave as expected? If lead organizations can identify such partners, they can reduce the informational and monitoring costs associated with collaboration and significantly enhance the chance for effective partnerships. We include four proxy indicators of the informational costs associated with identifying desirable partners and monitoring costs with man- aging successful partnerships. First, organizations should seek part- ners that

share their organizational vision or mission. Shared vision 410 HEN , G RADDY Nonprofit Management & Leadership DOI: 10.1002/nml
Page 7
represents a cognitive dimension of social capital that was found to facilitate knowledge transfer between partners (Li, 2005). Partner- ships between organizations with differing visions are difficult to ini- tiate and sustain because fundamentally different missions can create interorganizational conflicts. The expectation of shared mission may be based on knowledge of a specific partner’s goals and values, or on more

general expectations of the sector within which the organiza- tion operates. For example, nonprofit service delivery organizations often distrust the profit motives of business organizations. Second, lead organizations have the best knowledge of partners with whom they have collaborated in the past. Thus lead organiza- tions should be more willing to form future partnerships with such organizations, especially if the past experiences were successful. Gazley and Brudney (2007) find that a manager’s past collaborative experience increases partnership activities. In the

contracting liter- ature, trust derived from previous working relationships is found to improve outcomes (Brown, Potoski, and Van Slyke, 2006). In addi- tion, prior partnering experience allows an organization to build expertise in effective management. This expertise reduces transac- tions costs and thus increases the willingness of an organization to form more partnerships. Finally, there are search costs associated with finding partners. Social network theory introduces an organization’s relationships with other organizations as facilitating or constraining its internal and exter- nal

capacity to join an alliance (Guo and Acar, 2005). Gulati (1995, 1998) argued that an organization’s external capacity to form an alliance is constrained by its social network, and Van Slyke (2007) called it “supply-side imperfection.” In their study of network formation, Graddy and Chen (2006) found that even if lead organizations were willing to comply with a funding requirement to form a provider net- work, they were constrained by the availability of potential partners. In some cases, a lead organization may select a particular partner sim- ply because they are not aware of alternative

service providers. When a lead organization seeks to reduce partnership costs by identifying trustworthy partners, lower monitoring costs, easier com- munication, and more joint decision making should result. All of these should enhance partnership performance and thus improve client goal achievement and interorganizational relationships. Such well-performing partnerships are also likely to provide spillover ben- efits to organizational learning. Thus all dimensions of effectiveness should improve. If, however, the lead organization has high search costs because of capacity limitations

in the community, its own lack of a social net- work, or its high expectations and standards on partners, then all mea- sures of effectiveness should suffer. If lead organizations select partners solely on the basis of their availability, the resulting partnership is unlikely to be associated with improved client outcomes, stronger interorganizational ties, or enhanced organizational learning. HE FFECTIVENESS OF ONPROFIT EAD -O RGANIZATION ETWORKS 411 Nonprofit Management & Leadership DOI: 10.1002/nml If lead organizations select partners solely on the basis of their availability, the

resulting partnership is unlikely to be associated with improved client outcomes, stronger interorganiza- tional ties, or enhanced organizational learning.
Page 8
Governance. Network governance in a lead-organization model is built around a collection of dyadic ties between a lead organization and its partners (Provan and Kenis, 2008). Even with the most compatible partners, there are coordination costs associated with effective partnerships. Working together to meet programmatic needs requires some investment in mechanisms for coordination, management of information, and

accountability. We expect such processes to be facilitated by a written agreement defining the nature of the partnership. A written contract or memorandum of understanding (MOU) reduces ambiguity around roles and expec- tations. Moreover, the process of writing such an agreement can help develop an effective working relationship among the partners. Therefore we expect formal partnerships, evidenced by a written contract or MOU, to be more effective in meeting client goals and to have better interorganizational relationships and organizational learning. The effective ongoing management of

partnerships requires some relationship-specific governance structure. There is evidence that partnerships with central mechanisms of control are more effec- tive in meeting service delivery goals (Provan and Milward, 1995). Therefore we expect partnerships that rely on interorganizational committees to coordinate activities across the partners to be more effective in achieving client goals. Moreover, as these relation- specific assets develop, partners establish trust and thus improve their working relationship (Cohen and Levinthal, 1990). Therefore we also expect a positive impact on

improving interorganizational relationships. We do not expect an effect on organizational out- comes. Our model is summarized in Figure 1. 412 HEN , G RADDY Nonprofit Management & Leadership DOI: 10.1002/nml OUTCOMES Client Goal Achievement Enhanced Inter- organizational Relationships Improved Organizational Learning MOTIVATIONS Programmatic Needs (+) Organizational Goals (+) PARTNER CHARACTERISTICS Shared Visions (+) Same Sector (+) Prior Experience (+) Limited Choice (–) GOVERNANCE Formal Governance (+) Interorganizational Coordination (+) Figure 1. Hypothesized Determinants of

Nonprofit Lead-Organization Network Effectiveness
Page 9
Empirical Specifications The population for this study is the social service agencies in the Family Preservation (FP) program administered by the Los Ange- les County Department of Children and Family Services (DCFS). The aim of FP service interventions is to improve family function- ing when children are at imminent risk of placement in foster care, and to prevent this placement. The FP program in Los Angeles County is the largest of its kind in the United States in terms of number of clients and funding. Los

Angeles County has a population of ten million living in about ninety incorporated cities and numerous additional neighborhoods, many of which are multiethnic communities. Children and family- related social services are both substantial and diverse. DCFS created thirty-eight Community Family Preservation Networks (CFPNs) in defined geographic areas throughout the county. For each area, DCFS contracts family preservation services to a lead agency through a request for proposal (RFP) process. The lead agency receiving the con- tract, which can be either a public or a nonprofit

organization, is asked to partner with one or more additional service providers to deliver a range of services to children and families. Such a lead-organization model is typical of many social and human service collaborative networks sponsored by state and local governments (Poole, 2008; Johnston and Romzek, 2008). The model is designed to address the problems of fragmented, disjointed programming and uncoordinated services offered by specialized providers while also ensuring account- ability for results (Page, 2004). We study here the dyadic relationship formed when lead agencies form these

partnerships. Data Collection We collected the data for this study from late 2004 to early 2005. From our review of program documents and interviews, we devel- oped a survey instrument to collect data. As many as thirty-four lead agencies were slated for study. The survey was mailed to the executive director or the FP program man- ager in each of the lead agencies or networks. The response rate was 76 percent, with twenty-six of the thirty-four lead agencies com- pleting the survey. One lead agency is not involved in any partner- ship and was dropped from the study. All the lead agencies are

nonprofit social service providers. The unit of analysis for this study is the dyadic relationship between a lead agency and each of its network partners. The twenty- five nonprofit lead agencies formed 132 partnerships to deliver up to eleven family preservation services. Variable Measurement The variables are measured from lead agency responses to our sur- vey questions. Our dependent variables—the three measures of HE FFECTIVENESS OF ONPROFIT EAD -O RGANIZATION ETWORKS 413 Nonprofit Management & Leadership DOI: 10.1002/nml
Page 10
partnership effectiveness—were

measured with the three questions that assess to what extent collaboration helped achieve client goal, enhance interorganizational relationships, and improve organiza- tional learning. Respondents were asked to select the number from a seven-point Likert scale (with 1 representing “not at all” and 7 representing “very effective,” “very high quality,” or “to a great extent”) that best indicates their assessment for each partner on each outcome. Table 1 presents the descriptive statistics of their responses. On average, the partnerships were viewed as effective in achieving these outcomes.

Client goal achievement has the highest overall mean (and the lowest variance in responses), indi- cating that on average this was the most successful outcome. Our measure of organizational development—improved organizational learning —was on average the least successful outcome (with the largest variance), but even this outcome was rated highly. All of this suggests that the distribution of the perceived out- come measures may be skewed to the higher end. However, this skewed distribution should not affect the unbiasedness of the ordi- nary least squares (OLS) estimator since the property of

unbiased- ness of OLS is not conditioned on any assumption about the distribution of the dependent variable. Yet the distribution of the 414 HEN , G RADDY Nonprofit Management & Leadership DOI: 10.1002/nml Table 1. Descriptive Statistics ( 132) Mean Std. Dev. Min Max OUTCOMES Client goal achievement 6.45 0.75 4 7 Enhanced interorganizational relationships 6.36 1.03 3 7 Improved organizational learning 5.63 1.56 1 7 MOTIVATION Programmatic needs Extra caseload capacity 2.27 2.07 1 7 Specific service expertise 5.28 2.44 1 7 Geographic coverage 4.08 2.65 1 7 Local knowledge and client

access 3.93 2.51 1 7 Cultural and linguistic needs 3.83 2.75 1 7 Organizational goals Funding agency preferences 3.30 2.47 1 7 Reputation enhancement 5.02 2.13 1 7 Building future relationships 4.87 2.34 1 7 CHARACTERISTICS Partners Shared vision 4.98 2.37 1 7 Same sector 0.80 0.40 0 1 Successful past collaboration 3.92 2.61 1 7 Few available partners 2.97 2.36 1 7 Governance Formal governance 0.77 0.43 0 1 Interorganizational coordination 0.17 0.19 0 1
Page 11
HE FFECTIVENESS OF ONPROFIT EAD -O RGANIZATION ETWORKS 415 Nonprofit Management & Leadership DOI: 10.1002/nml outcomes

measures, and that of the error terms of our models, may affect the statistical inferences regarding our coefficient estimates, because -values of the estimated coefficients are based on the nor- mal distribution of the errors. What researchers usually do is to run the regression, save the residuals from the model, and look at the empirical distribution of the r esiduals (Bryk and Raudenbush, 1992). After we did that, we saw that the residuals were normally distributed. Therefore, we do not think the skewed distribution of the outcome variables is a serious threat to the validity of the

statistical inference. Motivation and some partnership characteristic variables were formed by lead-agency responses to questions about the importance of programmatic needs, organizational goals, and partner charac- teristics in their decision to form specific dyadic partnerships on a seven-point Likert scale (with 1 representing “not at all” and 7 rep- resenting “very important”). Among programmatic reasons to choose a particular partner, specific service expertise was the most important. For those seeking organizational goals, reputation enhancement was the most impor- tant

motivation. Gaining capacity to cope with case overload was the least compelling rationale for forming a partnership. The governance and sector variables were formed on the basis of direct questions about the nature of the partnership. Formal gov- ernance is a dummy variable that assumes a value of 1 if there is a written contract or MOU governing the relationship between the partners. Same sector is a dummy variable that assumes a value of 1 if the lead agency and the partner are in the same sector (public, sec- ular nonprofit, faith-based organization, business). Interorganiza- tional

coordination is the percentage of how often interorganizational committees are used to coordinate activities between the lead agency and its partner. The summary statistics are presented in Table 1, which shows that the majority of dyadic partnerships govern by for- mal governance (77 percent), 80 percent of partners are in the same sector as the lead agency, and on average 17 percent of the time an interorganizational committee is used as a coordinating mechanism. We performed a test of the variation inflation factor (VIF) and did not detect any multicollinearity problems. Estimation

and Analysis Recall that the 132 dyadic relationships in our study are clustered in twenty-five networks. Relationships in the same network are presumably more similar than those in different networks. Thus it is not reasonable to assume independence across pairs within a net- work. Dyadic partnerships operating in the same network are likely to share values on several variables. Some of these variables will not be observed, and their presence in the error term would violate estimation assumptions in the classical multivariate regression model (Bryk and Raudenbush, 1992). Among programmatic

reasons to choose a particular partner, specific service expertise was the most important.
Page 12
Either fixed-effect or random-effect models can be used when observations are clustered into groups (Bryk and Raudenbush, 1992; Singer, 1998). We estimated both models in STATA. The results of the two models are almost identical in terms of the estimated direc- tion of the relationships and the statistical significance of the coeffi- cients. This suggests that our estimates are good across specifications and lends some confidence in our ability to make

inferences. We also ran a Hausman test to compare fixed with random effects in STATA. The results indicate that the fixed effects are preferred over the ran- dom effects (significant -value, Prob Chi smaller than 0.05) for two dependent variables (client goal achievement and improved organizational learning), and random effects are preferred over the fixed effects for the dependent variable of improved interorganiza- tional relationship (insignificant -value, Prob Chi larger than 0.05). Yet the random-effects models impose more restrictive assumptions on the

variance-covariance structure of the error terms, namely that the individual observations in the clusters are indepen- dent. In our study population, some organizations partner with multiple-network lead agencies, and thus the assumption of indepen- dence between the networks is violated. Therefore we present the results from the fixed-effects models in the next section. Results The results in Table 2 reveal that the model is a good fit for the data for all three outcomes, thus supporting the importance of partnership characteristics in perceived effectiveness. The impact of the

variables varies with the outcomes. Programmatic Needs. As expected, partnerships formed to meet service delivery needs are positively associated with improved client outcomes and better interorganizational relationships. Part- nerships formed to increase caseload capacity and obtain desired geographic coverage yield better client outcomes. Forming part- nerships to improve geographic coverage and obtain specific ser- vice expertise also improved interorganizational relations. Obtaining specific service expertise from a partner requires inter- action and learning, and this may promote

communication and co- ordination across organizations. Communication and some joint decision making is normally needed to coordinate geographic cov- erage, and this may explain the positive impact of these partner- ships on interorganizational relationships and organizational learning. Organizational Goals. As expected, forming partnerships to meet the expectations of the funding agency was positively associated with improving interorganizational relationships and organizational learning. The other two organizational motivations, however, did 416 HEN , G RADDY Nonprofit Management &

Leadership DOI: 10.1002/nml
Page 13
not affect these two perceived outcomes. Therefore the impact of this motivation was less than expected. We did not expect partnerships formed to enhance organizational goals to improve client out- comes, but the results suggest they do. Partnerships formed to en- hance the reputation of the organization are positively associated with client outcomes. Perhaps lead agencies formed these partner- ships at least in part on the basis of the expertise of the partners and there were associated spill-over benefits to service provision. Partnerships formed

to lay the foundation for future relationships are negatively associated with achieving client goals. There is no reason to expect this motivation to have a positive effect, and the negative impact may simply reflect the presence of multiple goals on the part of the organization that may undermine its focus on service delivery. Partner Characteristics. Partners selected because of a shared vision have the expected positive impact on interorganizational relationships HE FFECTIVENESS OF ONPROFIT EAD -O RGANIZATION ETWORKS 417 Nonprofit Management & Leadership DOI: 10.1002/nml Table 2.

Fixed-Effects Regression of Partnership Motivation and Characteristics on Outcomes Outcomes Improved Improved Client Goal Interorganizational Organizational Achievement Relationships Learning MOTIVATION Programmatic needs Extra caseload capacity 0.103** 0.058 0.022 Specific service expertise 0.040 0.073** 0.030 Geographic coverage 0.127*** 0.10** 0.097 Local knowledge and client access 0.020 0.060 0.025 Cultural and linguistic needs 0.020 0.044 0.050 Organizational goals Funding agency preferences 0.025 0.11** 0.196*** Reputation enhancement 0.12** 0.067 0.029 Building future

relationships 0.074* 0.015 0.003 CHARACTERISTICS Partners Shared vision 0.007 0.083* 0.22*** Same sector 0.167 0.119 0.120 Successful past collaboration 0.025 0.067** 0.019 Few available partners 0.049 0.09* 0.213*** Governance Formal governance 0.033 0.309 0.55* Interorganizational coordination 0.60* 0.49 0.248 Number of observations 132 132 132 Prob F (38, 93) 0.00 0.00 0.00 -squared 0.74 0.74 0.82 Note: 0.1; ** 0.05; *** 0.01.
Page 14
and organizational learning. Such partners appear to reduce transac- tion costs and thus improve the operation of the partnerships. The impact of

successful past collaborations, however, was not as expected. These partners had a negative impact on interorganizational relationships. Perhaps multiple collaborations with the same partner raise concerns about dependency. This relationship may be a nonlin- ear one, reducing transaction costs up to a point and then raising other concerns that undermine the relationship’s effectiveness. Finally, partners selected because the lead agency believed there were few available qualified partners had the expected negative impact on both interorganizational relationships and organizational

learning. Not sur- prisingly, partners selected largely because they are the only alterna- tive do not produce effective relationships. None of these variables had an impact on client outcomes. Governance Characteristics. Finally, the formal partnership gover- nance affected improved organizational learning. Interorganiza- tional coordination affected only the outcome of goal achievement in our sample. Partnerships that rely more heavily on interorgani- zational committees to coordinate partnership activities are more effective in achieving client goals. The effect was as expected and

presumably reflects the benefits to clients of joint decision-making mechanisms that can coordinate service delivery. Conclusion In an era that often necessitates working together, scholars have devoted little attention to understanding the factors that promote well-functioning lead-organization networks formed by nonprofit organizations. Nonprofit organizations would benefit from more systematic empirical investigation of the factors that influence partnership performance of lead-organization networks. Nonprofit lead organizations in the social service arena form partnerships for

a variety of reasons. In this study, we considered two broad categories of motivation: programmatic needs and organiza- tional goals. Both motivations were found to affect outcomes, but their effects differed. The differences have policy implications for funding agencies seeking to promote lead-organization networks, and for the nonprofit lead organizations. The public sector is usually interested in the ability of partner- ships to improve client outcomes or aid development of the social network of communities to enhance their problem-solving capacity. Our results suggest that both are

influenced by efforts to meet pro- grammatic needs and thus are likely to be affected by contract requirements. The results suggest that RFPs covering larger contracts and broader geographic areas may encourage partnerships among nonprofits that ultimately benefit clients, presumably through more integrated service delivery and richer service choices. Similarly, RFPs 418 HEN , G RADDY Nonprofit Management & Leadership DOI: 10.1002/nml
Page 15
that require multiple services and broader geographic coverage should help communities, if the improved

interorganizational rela- tionships produced by these partnerships create denser social net- works that can be tapped for community problem solving. Partnerships formed largely to meet the requirements or prefer- ences of the funding agency enhance both interorganizational rela- tionships and the lead organization’s own learning. Thus this policy instrument seems to support the social network goals of service delivery partnerships. Unfortunately, it seems insufficient to improve direct client outcomes. The lead agency plays a critical role in effectively developing and sustaining

community-based local networks. For public funding agencies, a well-articulated process for qualification should be in place when selecting lead agencies. It is important that the funding agencies select nonprofits with a proven track record of collaborat- ing with community stakeholders as the lead agencies (Poole, 2008). For nonprofit lead organizations seeking to enhance their own learning via such partnerships, our results suggest that seeking spe- cific partners with a shared vision is very important. For both the nonprofit lead organization and the funding agencies, efforts

to increase knowledge about available partners are likely to yield important dividends, because partners of convenience do not pro- duce successful partnerships. Finally, the results afford only limited support for the impact of governance arrangements on effectiveness. Partnerships making more decisions via integrated mechanisms were viewed as more effective in client outcomes, but this was the only impact. More nuanced struc- ture variables may be needed to further explore this relationship. Our results are interesting, but additional work is needed to understand the determinants of

lead-organization network effec- tiveness. First, it is difficult to discern causality from cross-sectional data. Another concern is the self-reported nature of the data used to construct the dependent and independent variables. Nevertheless, it is important to understand the determinants of perceived effective- ness, because organizational perceptions will determine their will- ingness to enter and invest in such relationships. Organizations may tend to overvalue their collaborative outcomes, so there is also a potential “social desirability bias” in these self-reported data. Finally,

our analysis is focused on a single type of service and a single loca- tion. The effectiveness of service delivery alliances may vary for ser- vices with substantially different characteristics. Los Angeles County encompasses a complex and diverse set of communities. This com- plexity may make partnerships more necessary and seen as more valuable than in more homogeneous settings. IN HEN is assistant professor in the School of Public Affairs at Baruch College of the City University of New York. HE FFECTIVENESS OF ONPROFIT EAD -O RGANIZATION ETWORKS 419 Nonprofit Management & Leadership

DOI: 10.1002/nml The lead agency plays a critical role in effectively developing and sustaining community-based local networks.
Page 16
LIZABETH A. G RADDY is professor of public policy and holds the Jeffrey J. Miller Chair in Government, Business, and the Economy at the School of Policy, Planning, and Development, University of Southern California, Los Angeles. References Agranoff, R. Managing Within Networks: Adding Value to Public Orga- nizations . Washington, D.C.: Georgetown University Press, 2007. Agranoff, R. “Collaboration for Knowledge Learning from Public Management

Networks.” In L. B. Bingham and R. O’Leary (eds.), Big Ideas in Collaborative Public Management . Armonk, N.Y.: Sharpe, 2008. Arya, B., and Lin, Z. “Understanding Collaboration Outcomes from an Extended Resource-Based View Perspective: The Roles of Orga- nizational Characteristics, Partner Attributes and Network Struc- tures. Journal of Management , 2007, 33 (5), 697–723. Bingham, L. B., and O’Leary, R. Big Ideas in Collaborative Public Man- agement . Armonk, N.Y.: Sharpe, 2008. Brown, T. L., Potoski, M., and Van Slyke, D. M. “Managing Public Service Contracts: Aligning Values, Institutions

and Markets. Pub- lic Administration Review , 2006, 66 (3), 323–331. Bryk, A., and Raudenbush, S. Hierarchical Linear Models . Thousand Oaks, Calif.: Sage, 1992. Chen, B. “Assessing Inter-organizational Networks for Public Ser- vice Delivery: A Process-Perceived Effectiveness Framework. Pub- lic Performance and Management Review , 2008, 31 (3), 348–363. Cho, S., and Gillespie, D. F. “A Conceptual Model Exploring the Dynamics of Government-Nonprofit Service Delivery. Nonprofit and Voluntary Sector Quarterly , 2006, 35 (3), 493–509. Cohen, W. M., and Levinthal, D. A. “Absorptive

Capacity: A New Perspective on Learning and Innovation. Administrative Science Quarterly , 1990, 35 (1), 128–152. Das, T. K., and Teng, B.-S. “Trust, Control, and Risk in Strategic Alliances: An Integrated Framework. Organization Studies , 2001, 22 (2), 251–283. Doz, L. Y., and Hamel, G. Alliance Advantage: The Art of Creating Value Through Partnering . Boston: Harvard Business School Press, 1998. Dyer, H. J., and Singh, H. “The Relational View: Cooperative Strat- egy and Sources of Inter-organizational Competitive Advantage. Academy of Management Review , 1998, 23 (4), 660–679. Ewalt, J. G.

“Alternative Governance Structures for Welfare Provider Networks.” In P. W. Ingraham and L. E. Lynn (eds.), The Art of Gov- ernance: Analyzing Management and Administration . Washington, D.C.: Georgetown University Press, 2004. Galaskiewicz, J., Bielefeld, W., and Dowell, M. “Networks and Orga- nizational Growth: A Study of Community-Based Nonprofits. Administrative Science Quarterly , 2006, 51 , 337–380. 420 HEN , G RADDY Nonprofit Management & Leadership DOI: 10.1002/nml
Page 17
Gazley, B. “Intersectoral Collaboration and the Motivation to Col- laborate Toward an Integrated

Theory.” In L. B. Bingham and R. O’Leary (eds.), Big Ideas in Collaborative Public Management Armonk, N.Y.: Sharpe, 2008. Gazley, B., and Brudney, J. L. “The Purpose (and Perils) of Government- Nonprofit Partnership. Nonprofit and Voluntary Sector Quarterly 2007, 36 (3), 389–415. Goldsmith, S., and Eggers, W. D. Governing by Network: The New Shape of the Public Sector . Washington, D.C.: Brookings Institution Press, 2004. Graddy, E., and Chen, B. “Influences on the Size and Scope of Net- works for Social Service Delivery. Journal of Public Administration Research and Theory ,

2006, 16 (4), 533–552. Graddy, E., and Ferris, J. “Public-Private Alliances—Why, When, and to What End?” In N. Verma (ed.), Institutions and Planning Oxford: Elsevier, 2006. Gray, B. “Assessing Inter-organizational Collaboration: Multiple Con- ceptions and Multiple Methods.” In D. Faulkner and M. de Rond (eds.), Cooperative Strategy: Economic, Business, and Organizational Issues . Oxford: Oxford University Press, 2000. Gulati, R. “Does Familiarity Breed Trust? The Implications of Repeated Ties for Contractual Choice in Alliances. Academy of Management Journal , 1995, 38 (1), 85–112. Gulati, R.

“Alliances and Networks. Strategic Management Journal 1998, 19 (4), 293–317. Guo, C., and Acar, M. “Understanding Collaboration Among Non- profit Organizations: Combining Resource Dependency, Institu- tional, and Network Perspectives. Nonprofit and Voluntary Sector Quarterly , 2005, 34 (3), 340–361. Johnston, J. M., and Romzek, B. S. “Social Welfare Contracts as Net- works: The Impact of Network Stability on Management and Per- formance. Administration and Society , 2008, 40 (2), 115–146. Kettl, D. “Managing Boundaries in American Administration: The Collaborative Imperative. Public

Administration Review , 2006, 66 (Special Issue on Collaborative Public Management), 10–19. Li, L. “The Effects of Trust and Shared Vision on Inward Knowledge Transfer in Subsidiaries’ Intra- and Inter-organizational Relation- ships. International Business Review , 2005, 14 , 77–95. O’Leary, R., and Bingham, L. B. The Collaborative Public Manager: New Ideas for the Twenty-First Century . Washington, D.C.: George- town University Press, 2009. O’Leary, R., Gerard, C., and Bingham, L. B. “Introduction to the Sym- posium on Collaborative Public Management. Public Administra- tion Review , 2006, 66

(Special Issue on Collaborative Public Management), 6–9. Page, S. “Measuring Accountability for Results in Interagency Collaboratives. Public Administration Review , 2004, 64 (5), 591–606. HE FFECTIVENESS OF ONPROFIT EAD -O RGANIZATION ETWORKS 421 Nonprofit Management & Leadership DOI: 10.1002/nml
Page 18
Poole, D. L. “Organizational Networks of Collaboration for Com- munity-Based Living. Nonprofit Management and Leadership , 2008, 18 (3), 275–293. Provan, K. G., and Kenis, P. “Models of Network Governance: Struc- ture, Management, and Effectiveness. Journal of Public

Adminis- tration Research and Theory , 2008, 18 (2), 229–252. Provan, K. G., Kenis, P., and Human, S. E. “Legitimacy Building in Organizational Networks.” In L. B. Bingham and R. O’Leary (eds.), Big Ideas in Collaborative Public Management . Armonk, N.Y.: Sharpe, 2008. Provan, K. G., and Milward, H. B. “A Preliminary Theory of Inter- organizational Network Effectiveness: A Comparative Study of Four Community Mental Health Systems. Administrative Science Quarterly , 1995, 40 (1), 1–33. Provan, K. G., and Milward, H. B. “Do Networks Really Work? A Framework for Evaluating Public Sector

Organizational Net- works. Public Administration Review , 2001, 61 (4), 414–423. Saxton, T. “The Effects of Partner and Relationship Characteristics on Alliance Outcomes. Academy of Management Journal , 1997, 40 (2), 443–461. Selden, S. C., Sowa, J. E., and Sandfort, J. “The Impact of Nonprofit Collaboration in Early Child Care and Education on Management and Program Outcomes. Public Administration Review , 2006, 66 (3), 412–425. Singer, J. D. “Using SAS PROC MIXED to Fit Multilevel Models, Hierarchical Models, and Individual Growth Models. Journal of Educational and Behavioral

Statistics , 1998, 24 (4), 323–355. Snavely, K., and Tracy, M. B. “Development of Trust in Rural Non- profit Collaborations. Nonprofit and Voluntary Sector Quarterly 2002, 31 (1), 62–83. Thomson, A. M., Perry, J. L., and Miller, T. K. “Linking Collabora- tion Processes and Outcomes: Foundations for Advancing Empir- ical Theory.” In L. B. Bingham and R. O’Leary (eds.), Big Ideas in Collaborative Public Management . Armonk, N.Y.: Sharpe, 2008. Van Slyke, D. M. “Agents or Stewards: Using Theory to Understand the Government-Nonprofit Social Service Contracting Relation- ship. Journal of Public

Administration Research and Theory , 2007, 17 (2), 157–187. 422 HEN , G RADDY Nonprofit Management & Leadership DOI: 10.1002/nml