Human service delivery in a multi-tier system: the subtleties of collaboration among partners.
Abstract: This article examines the nature of interorganizational relationships that are formed within a multi-tier human service delivery system. Taking into account the hierarchical structure of a statewide initiative to support early childhood education, the study investigates the differences in the relationships between organizations at the service and administrative levels of the system. Forty-nine administrative level and 146 service delivery level relationships are evaluated. Findings indicate that organizations involved in direct service delivery form more collaborative relationships. Thus, when government provides funding for human services, policymakers must seek to balance public accountability with the advantages believed to be inherent in devolved service delivery. Furthermore, practitioners who appreciate the importance and nuances of interorganizational relationships will be in a position to better manage their organizations in an environment of increased collaborative activity and joint delivery of services. Going forward, human service systems will continue to involve organizations from the public, nonprofit, and private sector. A better understanding of how these organizations work together is crucial to the effective delivery of these essential services.
Article Type: Report
Subject: Human services (Management)
Public health administration (Research)
Public-private sector cooperation (Management)
Interorganizational relations (Management)
Author: Mayhew, Fred
Pub Date: 06/22/2012
Publication: Name: Journal of Health and Human Services Administration Publisher: Southern Public Administration Education Foundation, Inc. Audience: Academic Format: Magazine/Journal Subject: Government; Health Copyright: COPYRIGHT 2012 Southern Public Administration Education Foundation, Inc. ISSN: 1079-3739
Issue: Date: Summer, 2012 Source Volume: 35 Source Issue: 1
Topic: Event Code: 200 Management dynamics; 310 Science & research Computer Subject: Company business management
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 304051553
Full Text: In the United States, the delivery of human services is a constant puzzle for policy makers, practitioners, and academics. The difficulty stems from the interrelatedness of issues being addressed and the role and scope of government involvement in the process. Since the complexity of the social problems in question is not going away, the primary area of interest for this study revolves around administration and the role of government. Currently, much of government's work is being carried out through complex and indirect administrative approaches (Kettl, 2000), leading to third-party entities playing an increasingly significant role in the design, management, and execution of policy responsibilities (Heinrich et. al., 2010). This devolution strategy is envisioned as a way to deliver services in an innovative manner that is consistent with local needs. This approach is embraced in the field of human services since policy makers and practitioners realize that no single organization is in the position to successfully address the multifaceted problems that face society. The focus on collaboration is evident in the fact that governmental funding streams often require documentation of ongoing human service collaboration in order for communities to secure funding (Sandfort, 1999). Underlying the push towards collaboration is the belief that by working together human service organizations can integrate their services, producing more effective and efficient delivery and addressing the needs of multiproblem clients in a more comprehensive manner (Konrad, 1996).

This article examines how collaboration manifests itself in a hierarchical service delivery system. By examining a statewide multi-tier system that seeks to deliver a "high quality, comprehensive, accountable system of care and education for every child beginning with a healthy birth" (North Carolina Partnership for Children), the study addresses two principal questions. First, what is the nature of the interorganizational relationships between members of the service delivery system? Second, does the degree of collaboration differ between organizations at the service delivery level of the system and those at the administrative level? Collaboration is a central component of most strategies to address social issues and the literature relating to the concept accepts that a variety of organizational relationships may arise. Yet, public policy often views collaboration as a one-size-fits-all concept, promoting collaborative activity without taking into account the complexity involved in such efforts. Increasing our understanding of how collaborative activity may, or may not, differ within a service delivery system begins to build a foundation upon which we can better appreciate the nuances of collaboration in the field of human services, and should assist in developing strategies that can address the issues of effectiveness and accountability in a devolved system of governance.

HIERARCHY IN THE HOLLOW STATE

Since the great society programs of the 1960's most social programs have been managed through a similar approach. First, ambitious goals are set by the federal government which provides state and local governments with funding to manage programs, and in turn the state and local governments contract out the delivery of the programs to local organizations. This leads to the joint production of services and, if to be done effectively and efficiently, requires collaboration between a number of dissimilar organizations. It is the nature of the relationships between these organizations that provides the "hollow" in hollow state. Milward and Provan (2000) speak of the hollow state as the degree of separation between a government and the services it funds, and identify the central task of the hollow state as arranging networks rather than managing hierarchies (p. 362). However, when looking at a multi-tier service delivery structure that receives substantial funding from government, as is common in the human services, it is difficult to view it as having completely abandoned hierarchy. Instead one sees a web of interorganizational relationships that form within a more or less hierarchical structure. Kettl (2002) notes that horizontal relationships have not displaced vertical relationships; rather they have been added to the system. In their review of over 800 research studies on governance, Hill and Lynn (2005) do not find evidence of the decline of hierarchical governance. In addition, they raise the possibility that new administrative tools and techniques are being created to facilitate governance within a hierarchical system. This suggests that the emphasis within governance theory on primarily horizontal structures may be missing a key vertical element that holds the system together (1).

When considering human service delivery, it is important to keep in mind that most systems incorporate public and private organizations. In addition, the services being provided are typically viewed as public goods and substantial funding for their provision comes from government. In the context of devolution this creates a scenario where government can choose to contract out services and then set about managing those contracts, or government can choose to partner with other organizations to realize the delivery of services. For services that lack complexity and offer straightforward solutions the former is a viable strategy. However, when issues are complex and goals are broad, government may not be able to clearly specify what a contractor needs to do to meet public needs (Whitaker et. al., 2004). When this occurs, the option of partnership and collaboration is more appealing. Evident in this choice is that the success of whatever strategy is selected will depend on the nature of the interorganizational relationships that are formed.

COLLABORATION AND INTERORGANIZATIONAL RELATIONSHIPS WITHIN A HIERARCHY

Collaboration is essential to the joint production of human services, yet what is meant by collaboration is seldom specified. While collaboration has many definitions and can result in numerous organizational forms, this article identifies collaboration as taking place within the context of an interorganizational relationship, that is, when two or more organizations work together. Within a multitier hierarchical system of service delivery, there are varying levels at which these relationships will manifest themselves. In a devolved service delivery system the management issues will differ from level to level, suggesting that the appropriate form of relationship may also differ. Moreover, the challenges and barriers to collaborative efforts will vary at different levels within the delivery system (Sandfort, 1999; Miller & Ahmad, 2000), indicating that the nature of the relationships will be important to overcoming the difficulties and that various forms of collaboration may be more effective at different levels.

A range of concepts seek to capture the collaborative nature of interorganizational relationships, including partnerships, alliances, and joint production. While the terminology may differ, what is consistent is that the nature of these relationships organizes along a continuum, and that positioning on the continuum is based on the level of organizational integration that is evident in the relationship (Bailey & Koney, 2000; Frey et al., 2006; Gajda, 2004). Bailey and Koney (2000) propose a four level continuum of interorganizational processes that moves from cooperation to coordination, collaboration, and finally coadunation. In a human service delivery system, coadunation would entail the joining of two organizations into a single entity, a merger. The scope of this study does not encompass coadunation; it does however add a point to the continuum. When a funder contracts with a private or nonprofit organization in a purely market based transaction, where organizational integration is minimal, the principal/agent relationship is evident. In such conditions, the principal (funder) is more concerned with ensuring the agent (contractor) is fulfilling its obligations and does not intimately involve itself with the delivery of the good or service.

In a multi-tier human service delivery system that is initiated by government and includes a multitude of organizations from the public and private sectors, the interorganizational relationships formed will fall along a continuum ranging from principal/agent to collaborative. The appropriate nature of the relationship will depend on the goals of the organizations entering into the relationship. Goals for state or federal level governmental organizations will differ from those of local governments or service providers. Therefore, in an effective service delivery system the nature of the relationships should differ depending on the level of the hierarchy involved. In other words, state level organizations have different objectives than do local level service providers. State level agencies are concerned with aspects of control, coordination, and equity. Their desire is to see that all services meet certain expectations, are not redundant, and are available to all members of their constituency.

Local level organizations are interested in crafting services to meet their community's needs, and integrating services to provide those in need with seamless delivery. Thus, more collaborative relationships will need to be formed at the local level, where services are actually delivered, with less need for fully collaborative efforts at the higher administrative levels.

THE NORTH CAROLINA SMART START INITIATIVE

The Smart Start initiative is a public-private venture that provides funding for early childhood education. Funds are administered by the North Carolina Partnership for Children (NCPC), a statewide nonprofit organization, through local nonprofits called partnerships. Partnerships then distribute funding and work with local agencies (nonprofit, governmental, and for-profit) who address issues associated with early childhood education. Funds are used to improve the quality of childcare, make childcare more affordable and accessible, provide access to health services and offer family support. In the 2006-07 fiscal year, local partnerships funded programs ranging from family literacy to transportation for pregnant teens and prenatal patients.

The variety of programming that is supported speaks to the intent of the originating legislation that stated, "There is a need to explore innovative approaches and strategies for aiding parents and families in the education and development of young preschool children" (North Carolina Senate Bill 27, 1993). Currently 77 partnerships cover all 100 counties in North Carolina. NCPC distributes funds that are allocated by the state legislature and provides technical assistance and oversight to the local partnerships. NCPC also conducts fundraising activities to supplement the state allocations, a 10 percent match is mandated by legislation and NCPC has met this requirement, claiming $382 million in cash and in-kind donations since 1994. The initiative began in 1993 with 12 "pioneer" partnerships and $20 million in appropriations, and now boasts appropriations of around $200 million and is used as a model by 14 other states for their early childhood programs. In addition, in 1998 the initiative was awarded the Innovations in American Government award from Harvard University and Ford Foundation.

[FIGURE 1 OMITTED]

Collaboration is the backbone of this multi-tier service delivery system, with interorganizational relationships occurring at different levels within a hierarchical structure and among a variety of organizations, both public and private, with differing missions and objectives (see Figure 1). The system is designed with the intent of delivering effective services that meet local community needs while remaining accountable to the state legislature. The most recent state legislation requires that funds be used to support services that target child care providers, children, and families. The programs to be funded are identified by local partnerships (2) and designed to be driven by local, regional, and statewide needs.

DATA & METHODS

This study examines the interorganizational relationships found within the North Carolina Smart Start initiative that includes a statewide agency (NCPC), 77 local partnerships, and over 500 service providing organizations. The policy goal of this multi-tier system of delivery is to enhance early childhood education throughout the state and assure that every child is prepared physically and mentally to enter school. The units of analysis are the relationships between service providers and the local partnerships, and the relationships between the local partnerships and NCPC. In this system, the local partnerships and affiliated service agencies provide services tailored to the needs of their community, and the NCPC provides oversight and technical assistance to the partnerships. This creates a hierarchical arrangement where NCPC works with the local partnerships to provide administration and coordination and the partnerships work with local organizations to deliver services.

A survey was used to examine the nature of these relationships. The survey instrument contained 8 Likert scale questions related to the nature of the relationship, rated from 1 to 5, with 1 indicating that the respondent strongly disagrees with the question and 5 indicating strong agreement (see Appendix I for survey items). Questions were designed to capture the degree of integration and formalization in the relationship; addressing issues of communication, shared decision-making, and understanding of each partner's goals and responsibilities. In an effort to validate the instrument prior to distribution, three steps were taken. First, questions were composed based on a thorough review of the literature and previously conducted surveys that addressed similar concepts (Frey et. al., 2006; Van de Ven & Ferry, 1980). Second, a focus group was conducted with managers from human service organizations that were involved in interorganizational relationships. They were asked to review and comment on the survey in an attempt to attain a level of face validity. Third, the survey was reviewed by individuals familiar with the North Carolina Smart Start program. In addition, principal components analysis substantiated that the relationship items held together as a unified construct with factor loadings above .5 (3).

The survey was distributed online to all 77 local partnerships and 448 direct service providers that had been identified as having received funding from one of the local partnerships in the 2006-07 fiscal year. Forty-nine of the partnerships completed the survey (64%) as did 146 of the service providers (33%). Individual respondents answered on behalf of their respective organizations with the intent to have individuals respond that were in management positions and knowledgeable about the nature of the initiative and resultant interorganizational relationships. Respondents for the 49 local partnerships included 32 executive directors, 1 deputy executive director, 11 program directors, and 5 evaluation coordinators. One hundred and twenty-six of those who responded on behalf of the service agencies (86%) identified themselves as director, supervisor, or manager. Once the surveys were compiled, an indexed variable for relationship was formed. Analysis of variance (ANOVA) was conducted using SPSS software to compare the two groups. The level at which the interorganizational relationship occurs, between service provider and local partnership or between local partnership and NCPC, served as the independent variable or factor. Responses to the eight survey items as well as the indexed relationship variable served as the dependents in one-way ANOVA analysis.

RESULTS

Findings are reported for the indexed relationship variable as well as each survey item. All variables are examined independently to assess whether differences are evident between the two groups. Prior to conducting analysis of variance the distribution of data was examined. Tests for skewness and kurtosis met the more lenient criterion for normal distribution with skewness statistics in the +2 to -2 range, and kurtosis statistics in the +3 to -3 range.

Descriptive statistics can be seen in Table 1. The data suggest that the service providers perceive their relationship with the local partnerships as more collaborative than the partnerships perceive their relationship with the statewide agency. In other words, where an organization sits within the hierarchy makes a difference in how they view the level of collaboration with the organization directly above them in the hierarchy. Service providers view their relationship with local partnerships as more integrated with more open and frequent communication, shared decision-making, and involving a greater understanding of each other's goals and responsibilities. This can be seen by examining the results to individual survey items. All items signify higher levels of collaboration between service providers and local partnerships than between partnerships and the NCPC.

To a greater degree service provider's felt that local partnerships made themselves available to discuss issues (Rel1), that there was a known channel through which to voice concerns (Rel5), and meetings were held on a customary schedule (Rel8). Likewise, service providers acknowledged a greater level of shared decision-making (Rel2) within the relationship as well as higher levels of understanding between the partners concerning each other's goals and the expectations of each organization within the relationship (Rel3, Rel4, Rel6, and Rel7).

Of greater interest are the results for the one-way ANOVA. The results of the analysis indicate that there is a true difference between the two groups. The p value of the F-test is highly significant and signifies that the variation in the two group's view of the nature of their relationship is not by chance (see Table 2).

In addition to the one-way ANOVA, Levene's test of homogeneity of variance was conducted to determine that each group of the independent has the same variance. The statistic was not significant at the .05 level for the indexed relationship variable, allowing acceptance of the null hypothesis of equal variances. However, five of the individual survey items were found to be heterogeneous in variance, violating an assumption of ANOVA.

To address this issue, as well as the unequal sample sizes between the two groups, Brown and Forsythe's and Welch's tests of the equality of means were conducted. Both statistics were significant at the .01 level for all variables in the analysis, allowing for the rejection of the null hypothesis that there is no difference between the groups.

DISCUSSION

This article addresses the nature of interorganizational relationships that are formed within a multi-tier human service delivery system. The study examines a diverse set of actors seeking to advance their own organizational missions while integrating themselves into a larger system of service. Within this arrangement, many relationships are formed between actors at different levels of the system. Findings suggest that the collaborative nature of these relationships differs between those at the service delivery level and those at the administrative level of the system's hierarchy. The following sections will address these results as they relate to the design and implementation of human service systems. First, the broad issues of accountability and effectiveness will be discussed. Next, implications for practice and future research will be addressed.

Governmental agencies often provide the bulk of financing for human service provision. In the case of the North Carolina Smart Start initiative, the NCPC receives nearly 99% of their revenues from government sources. In addition the local partnerships receive an average of 67% of their revenue from NCPC grants and twenty-six of the partnerships receive over 90% of the total revenue from NCPC (4). This suggests that despite the involvement of organizations from the private and nonprofit sectors, the driving force behind large-scale human service efforts such as this one is still often government. If governmental agencies are bankrolling these efforts, questions of public accountability must be dealt with. This study addresses this issue in that accountability can be viewed within relationships at various levels in the hierarchy. Ebrahim (2005) speaks of accountability as centering on relationships between various actors, within a multi-tier service delivery system this would indicate that there are multiple accountabilities built into the system. Understanding what each actor expects, and what is expected of them, will move the entire system towards a greater degree of accountability. Whitaker and his colleagues (2004) state that too often systems that are set up to ensure public accountability focus on catching mistakes rather than looking for ways to encourage better service (p. 115). The key to designing human service delivery systems in the era of devolution is to balance the need for public accountability with the advantages of decentralized approaches. To accomplish this, a system must de designed to identify mistakes while simultaneously promoting effective and innovative practices. Too much emphasis on catching mistakes may lead to a bureaucratic and rigid system that does not provide the flexibility to meet the differing challenges present at the local level. Too much emphasis on decentralization and local control may lead to a disorganized service delivery with no accountability ties to the public that provide funding.

A multi-tiered arrangement that understands and acknowledges the differing forms of relationships and degrees of collaboration within those relationships can begin to design mechanisms that provide accountability while promoting improvement in service delivery. Administrative entities within the system may be more inclined to view accountability as a top-down function in which the lower levels of the hierarchy are answerable to them. On the other hand, accountability at the service level may be viewed as more shared in nature, leading to greater levels of collaboration as found in the Smart Start system. Government may not become simply a contract manager as envisioned in some interpretations of the hollow state; it may in fact become an active administrator working through intermediary organizations to foster collaborative service integration. These intermediary organizations can provide levels of upstream accountability from local service providers to state or federal agencies. In addition, their role as an intermediary enables them to form more collaborative interorganizational relationships with local service providers, in turn developing a sense of mutual accountability.

In the system studied here, NCPC serves as a surrogate for the state. In fact, the organization was born from state legislation and the Secretaries of the Department of Human Resources, the Department of Environment Health and Natural Resources, and the Superintendant of Public Instruction are all ex officio members of the NCPC Board of Directors. NCPC provides a two-way conduit through which policy preferences can be delivered downstream to local partnerships, and measures of financial and programmatic accountability can be delivered upstream from the partnerships and the organizations that they fund. This creates a system in which public accountability can be maintained, but day-to-day monitoring of the many organizations operating at the local level does not fall to already taxed governmental agencies. Additionally, while the system creates a framework in which public accountability can be preserved, it also allows for flexibility at the local level where services are delivered. The local partnerships are service delivery intermediary organizations, identifying community needs and marshaling resources to meet those needs. As long as they are within the policy boundaries, local partnerships are free to pursue unique strategies that fit their community's needs. This freedom to devise local strategies for local problems is a centerpiece of arguments that advocate devolutionary policy implementation. In theory, such a system would entail more collaborative interorganizational relationships at the service delivery level, and relationships geared more towards traditional top-down accountability at the administrative level. The results of this study support this notion, indicating a greater degree of integration between organizations at the service level of the system.

Implications for Practitioners

Those engaged in the delivery of human services are the most knowledgeable regarding the complexities involved with tackling interrelated social problems. It is also those same practitioners that understand how the problems associated with human service delivery vary depending on where one sits within the system. Local service providers face many different challenges than state or federal agencies when seeking to implement programs, and the issues faced are vastly dissimilar when one organization is delivering services and another is delivering funding. Actors within a multi-tier system that accept the differences inherent in their undertakings and embrace the fact that collaboration is not a one size fits all solution, will be in a better position to understand what is expected of them and to work with their partners to effectively deliver services.

In the provision of public services a traditional view of accountability, often equated to punishment, is common. As collaborative and networked forms of service delivery become the standard, many have proposed a new understanding of accountability that involves collective responsibility rather than top-down authority. In a multi-tier system with differing levels of collaboration as an accepted operational paradigm, the understanding of accountability will be aligned with the nature of the interorganizational relationship. Those at the service delivery level, where organizational integration is greatest, may be freed to take risks and implement innovative programs that would not see the light of day under a traditional bureaucratic hierarchy. At the same time, those whose responsibility it is to ensure top-down accountability of the system can still maintain levels of command and control, monitoring program effectiveness without undue influence in the delivery of services. The findings of this study indicate that there is a significant difference between the interorganizational relationships formed at the service delivery level and those formed at the administrative level. Likewise, responses to individual items identify the way in which these relationships are managed and understood by those involved. Levels of shared decision-making, more open lines of communication and a clearer understanding of the partnering organization's goals are all present to a greater degree at the service delivery level. Collaboration between organizations varies depending on where they are in the hierarchy. Those charged with leading and managing large scale human service efforts, in both design and implementation, would be well served to understand the differences in the nature of relationships formed within the system. Efforts to cultivate the types of relationships that serve the different yet complimentary goals of a multi-tier system will facilitate human service delivery that takes advantage of local resources and knowledge while still remaining accountable to the larger public that supports it.

Future Research

While the findings of this study show that differences in the nature of interorganizational relationships exist at different levels of a multi-tier human service system, many interesting questions remain. Although this study examines the nature of the relationships it does not address the effectiveness of the system from the vantage point of the end user. Research into how the effectiveness of local service providers is impacted by such a system is warranted. While the Smart Start initiative has garnered national attention, received substantial support from outside the state (5), and been shown to produce impressive benefits when studied on a large scale (6), the initiative's influence on service providers is less well documented. Further research is needed to determine whether the existence of collaborative relationships at the service level leads to more effective or innovative programming. Also, this study examined dyadic relationships. For human service provision to be most effective it will entail the involvement of many organizations, networks, at the local and possibly administrative levels. Does a multi-tier system like the one studied help to develop networks at the local level? If so, a series of questions arises: What do these networks look like? Are they effective? How are they managed? Do the intermediary organizations serve as network administrative organizations? Additionally, questions regarding funding and efficiency need to be explored. While over 60% of the local partnerships received 50% of their revenue from NCPC grants in the 2007-08 fiscal year, that leaves a number of local partnerships that received less than 50% of their revenue and 13 where NCPC funding accounted for less than 40% (see appendix II). Examination of how funding is leveraged at the local level and how efficiently it is put to use is needed. Future research that expands our understanding of how these organizations work together will assist in the creation of systems that take advantage of devolved service delivery while maintaining public accountability.

CONCLUSION

Human service agencies often deal with the most complex issues, and while delivery of these services is essential, it is also complicated. These organizations provide services to some of the most vulnerable groups in our society, and do so under difficult and ever-changing circumstances. As the recent economic downturn has shown, overreliance on governmental funding puts smaller local organizations at risk. It also seems clear that services targeted at populations that cannot afford to pay stand no chance of competing in a market driven system, and require a level of public support that is appropriate for the public goods they provide. Devolution and collaboration are seen as strategies that can address these issues by improving the delivery of services in a manner that is consistent with local needs, and do so more efficiently by leveraging local resources (organizations and dollars). However, implementation of this approach is not without its problems. When public funding is involved issues of accountability come with it, and a system designed to ensure accountability can often impinge on adaptability and effectiveness of programming at the local level. Likewise, complete freedom to implement programming funded by the public at large goes against the tenets of good governance. Systems designed to address these issues will need to incorporate a multitude of organizations from the public, nonprofit, and private sector and balance the questions of accountability and effectiveness.

In an environment of devolved service delivery the goals of public accountability and effective service delivery can be at odds. This study adds to our understanding of how organizations work together within a large scale, hierarchical system. The increased collaboration and organizational integration suggests joint effort in the delivery of services at the local level, while less collaboration at the administrative level of the hierarchy points to relationships more geared towards top-down traditional accountability. By employing an intermediary organization, in this case a nonprofit, the system seems capable of serving local needs and providing a level of answerability to funding entities at the top of the hierarchy. Additionally, by working directly with the service organizations in their regions the intermediaries can offer a level of support and assistance that a governmental agency could not. While certainly not a silver bullet, this set of relationships seems to offer a framework for dealing with the issues of public accountability and effective locally driven service provision.

APPENDIX I

Nature of the Funder-Fundee Relationship

Rel1: A representative of the local partnership (NCPC) was typically available to discuss the relevant issues.

Rel2: There was shared decision-making between your organization and the partnership (NCPC).

Rel3: Your organization has a clear understanding of the local partnership's (NCPC's) goals.

Rel4: The local partnership (NCPC) has a clear understanding of your organization's goals and services.

Rel5: There is an agreed upon channel through which your organization can voice concerns and make recommendations.

Rel6: Your organization understands what the local partnership (NCPC) expects from you.

Rel7: Your organization has a clear understanding of what the local partnership (NCPC) will provide you.

Rel8: There are regularly scheduled meetings (bi monthly, quarterly) between your organization and the local partnership (NCPC).

REFERENCES

Bailey, D., & Koney, K. (2000). Strategic alliances among health and human services organizations. Thousand Oaks, CA: Sage Publications.

Ebrahim, A. (2005). Accountability myopia: Losing sight of organizational learning. Nonprofit and Voluntary Sector Quarterly, 34 (1), 56-87.

Frey, B., Lohmeier, J., Lee, S., & Tollefson, N. (2006). Measuring collaboration among grant partners. American Journal of Evaluation, 27 (3), 383-392.

Gajda, R. (2004). Utilizing collaboration theory to evaluate strategic alliances. American Journal of Evaluation, 25 (1), 65-77.

Heinrich, C., Lynn, L., and Milward, H. B. (2010). A state of agents? Sharpening the debate and evidence over the extent and impact of the transformation of governance. Journal of Public Administration Research and Theory, 20 (Suppl 1), 3-19.

Hill, C., & Lynn, L. (2005). Is hierarchical governance in decline? Evidence from empirical research. Journal of Public Administration Research and Theory, 15 (2), 173-195.

Kettl, D. (2000). The transformation of governance: Globalization, devolution, and the role of government. Public Administration Review, 60 (6), 488-497.

Kettl, D. (2002). The transformation of governance: Public administration for the twenty-first century. Baltimore, MD: Johns Hopkins University Press.

Konrad, E. (1996). A multidimensional framework for conceptualizing human services integration initiatives. New Directions for Evaluation, 69, 5-19.

Miller, C. & Ahmad, Y. (2000). Collaboration and partnership: An effective response to complexity and fragmentation or solution built on sand? International Journal of Sociology and Social Policy, 20 (5/6), 1-38.

Milward, H.B., and Provan, K. (2000). Governing the hollow state. Journal of Public Administration Research and Theory, 10 (2), 359-379.

North Carolina Partnership for Children. About Smart Start, Mission Statement. http://hugh.ncsmartstart.org/category/smart-startinformation/about-smart- start (accessed April 4, 2011).

North Carolina Senate Bill 27. 143B-168.10. Early childhood initiatives; findings. http://www.ncga.state.nc.us/Sessions/1993/Bills/Se nate/HTML/S27v6.html (accessed June 20, 2011)

Sandfort, J. (1999). The structural impediments to human service collaboration: Examining welfare reform at the front lines. The Social Service Review, 73 (3), 314-339.

Smith, B. (2008). The sources and uses of funds for community development financial institutions: The role of the nonprofit intermediary. Nonprofit and Voluntary Sector Quarterly, 37 (1), 19-38.

Van de Ven, A., & Ferry, D. (1980). Measuring and assessing organizations. New York: Wiley and Sons.

Whitaker, G., Altman-Sauer, L., & Henderson, M. (2004). Mutual accountability between governments and nonprofits: Moving beyond "surveillance" to "service". American Review of Public Administration, 34 (2), 115-133.

FRED MAYHEW

Northern Illinois University

(1) For an in depth discussion on the impact on governance of the use of third parties to deliver services see the 2010 special issue of the Journal of Public Administration Research and Theory, "A State of Agents".

(2) The legislation does place a number of constraints on how the funding will be used. For instance, 70% of the funds are to be used for child care related activities and no less than 30% of the local partnerships direct service allocation shall be used to expand child care subsidies.

(3) The survey was part of a larger research effort that examined the impact of interorganizational relationships on the use of funder mandated evaluation by human service organizations. This enabled the use of principal components analysis.

(4) These numbers were derived from an analysis of IRS form 990 filings for the 2007-08 fiscal year. Information was obtained through the GuideStar online database.

(5) In 2001 NCPC started the Smart Start National Technical Assistance Center with funding from the W.K. Kellogg Foundation.

(6) A recent study by the Sanford School of Public Policy at Duke University found evidence of significant benefits to children in the counties that receive funding.

Frederick Mayhew is an assistant professor in the Division of Public Administration at Northern Illinois University. His experience in the nonprofit and public sectors serves as the foundation for his current research interest in how government interacts and works with nonprofit and private sector organizations to deliver public goods and services.
APPENDIX II

                                      Total
                 NCPC grant 2007-08   revenue 2007-08   %

Partnership 1    $ 1,376,517.00       $ 4,870,611.00    28%
Partnership 2    $ 4,474,382.00       $ 15,001,684.00   30%
Partnership 3    $ 678,716.00         $ 2,157,070.00    31%
Partnership 4    $ 997,540.00         $ 3,114,771.00    32%
Partnership 5    $ 937,911.00         $ 2,835,123.00    33%
Partnership 6    $ 5,710,719.00       $ 16,754,139.00   34%
Partnership 7    $ 565,142.00         $ 1,621,883.00    35%
Partnership 8    $ 1,014,653.00       $ 2,889,784.00    35%
Partnership 9    $ 921,669.00         $ 2,604,236.00    35%
Partnership 10   $ 1,774,930.00       $ 4,959,858.00    36%
Partnership 11   $ 458,257.00         $ 1,273,839.00    36%
Partnership 12   $ 1,575,385.00       $ 4,118,191.00    38%
Partnership 13   $ 611,403.00         $ 1,557,780.00    39%
Partnership 14   $ 1,002,139.00       $ 2,509,970.00    40%
Partnership 15   $ 705,841.00         $ 1,737,670.00    41%
Partnership 16   $ 210,179.00         $ 504,331.00      42%
Partnership 17   $ 3,073,272.00       $ 7,255,032.00    42%
Partnership 18   $ 734,818.00         $ 1,717,184.00    43%
Partnership 19   $ 1,031,679.00       $ 2,358,714.00    44%
Partnership 20   $ 305,789.00         $ 696,063.00      44%
Partnership 21   $ 2,049,793.00       $ 4,574,938.00    45%
Partnership 22   $ 1,203,822.00       $ 2,685,383.00    45%
Partnership 23   $ 1,339,627.00       $ 2,889,069.00    46%
Partnership 24   $ 559,368.00         $ 1,196,518.00    47%
Partnership 25   $ 1,605,352.00       $ 3,405,084.00    47%
Partnership 26   $ 721,122.00         $ 1,518,871.00    47%
Partnership 27   $ 2,318,687.00       $ 4,853,134.00    48%
Partnership 28   $ 2,135,050.00       $ 4,442,655.00    48%
Partnership 29   $ 1,888,550.00       $ 3,923,825.00    48%
Partnership 30   $ 2,039,976.00       $ 4,043,247.00    50%
Partnership 31   $ 1,994,052.00       $ 3,727,889.00    53%
Partnership 32   $ 2,523,773.00       $ 4,672,457.00    54%
Partnership 33   $ 7,753,394.00       $ 14,312,798.00   54%
Partnership 34   $ 2,255,787.00       $ 4,071,351.00    55%
Partnership 35   $ 5,485,036.00       $ 9,580,673.00    57%
Partnership 36   $ 3,364,773.00       $ 5,582,506.00    60%
Partnership 37   $ 2,715,136.00       $ 4,450,961.00    61%
Partnership 38   $ 1,258,935.00       $ 2,046,890.00    62%
Partnership 39   $ 2,668,467.00       $ 4,236,398.00    63%
Partnership 40   $ 466,921.00         $ 732,833.00      64%
Partnership 41   $ 329,829.00         $ 500,355.00      66%
Partnership 42   $ 362,568.00         $ 546,403.00      66%
Partnership 43   $ 1,655,722.00       $ 2,490,354.00    66%
Partnership 44   $ 1,705,715.00       $ 2,350,193.00    73%
Partnership 45   $ 1,150,255.00       $ 1,488,066.00    77%
Partnership 46   $ 1,929,434.00       $ 2,355,683.00    82%
Partnership 47   $ 1,614,314.00       $ 1,893,131.00    85%
Partnership 48   $ 574,870.00         $ 666,416.00      86%
Partnership 49   $ 543,350.00         $ 623,920.00      87%
Partnership 50   $ 423,480.00         $ 481,383.00      88%
Partnership 51   $ 2,943,852.00       $ 3,317,345.00    89%
Partnership 52   $ 273,551.00         $ 299,864.00      91%
Partnership 53   $ 233,962.00         $ 256,439.00      91%
Partnership 54   $ 890,378.00         $ 973,864.00      91%
Partnership 55   $ 971,250.00         $ 1,061,032.00    92%
Partnership 56   $ 207,208.00         $ 224,534.00      92%
Partnership 57   $ 1,336,286.00       $ 1,445,169.00    92%
Partnership 58   $ 416,527.00         $ 448,293.00      93%
Partnership 59   $ 1,095,692.00       $ 1,174,095.00    93%
Partnership 60   $ 1,712,166.00       $ 1,794,581.00    95%
Partnership 61   $ 539,253.00         $ 564,927.00      95%
Partnership 62   $ 1,719,836.00       $ 1,781,792.00    97%
Partnership 63   $ 463,157.00         $ 479,670.00      97%
Partnership 64   $ 943,183.00         $ 975,251.00      97%
Partnership 65   $ 2,224,861.00       $ 2,294,280.00    97%
Partnership 66   $ 981,350.00         $ 1,011,723.00    97%
Partnership 67   $ 362,351.00         $ 370,015.00      98%
Partnership 68   $ 502,810.00         $ 510,037.00      99%
Partnership 69   $ 512,860.00         $ 519,877.00      99%
Partnership 70   $ 686,891.00         $ 695,623.00      99%
Partnership 71   $ 516,814.00         $ 522,594.00      99%
Partnership 72   $ 1,015,341.00       $ 1,026,374.00    99%
Partnership 73   $ 239,199.00         $ 241,440.00      99%
Partnership 74   $ 1,034,849.00       $ 1,044,396.00    99%
Partnership 75   $ 9,534,535.00       $ 9,613,446.00    99%
Partnership 76   $ 619,319.00         $ 621,561.00      100%
Partnership 77   $ 1,164,433.00       $ 1,166,233.00    100%


Table 1
Descriptive Statistics

                             Std.      Std.
               N    Mean   Deviation   Error
Rel Service   146   4.35     .529      .043
    Admin.    49    3.59     .644      .092
    Total     195   4.16     .647      .046

Rel Service   146   4.37     .695      .058
1   Admin.    49    3.92     .862      .123
    Total     195   4.26     .764      .055

Rel Service   146   4.14     .830      .069
2   Admin.    49    3.31     1.045     .149
    Total     195   3.93     .958      .069

Rel Service   146   4.51     .646      .053
3   Admin.    49    4.08     .886      .127
    Total     195   4.40     .735      .053

Rel Service   146   4.40     .670      .055
4   Admin.    49    3.59     .934      .133
    Total     195   4.20     .822      .059

Rel Service   146   4.36     .723      .060
5   Admin.    49    3.61     1.017     .145
    Total     195   4.17     .868      .062

Rel Service   146   4.51     .554      .046
6   Admin.    49    3.80     .979      .140
    Total     195   4.33     .750      .054

Rel Service   146   4.48     .566      .047
7   Admin.    49    3.57     .957      .137
    Total     195   4.25     .789      .056

Rel Service   146   4.03     1.063     .088
8   Admin.    49    2.92     1.152     .165
    Total     195   3.75     1.186     .085

              95% Confidence
               Interval for
                   Mean

              Lower   Upper
              Bound   Bound   Min.   Max
Rel Service   4.26    4.43    2.5    5.0
    Admin.    3.41    3.78    2.0    4.8
    Total     4.07    4.25    2.0    5.0

Rel Service   4.26    4.48     1      5
1   Admin.    3.67    4.17     1      5
    Total     4.15    4.36     1      5

Rel Service   4.01    4.28     1      5
2   Admin.    3.01    3.61     1      5
    Total     3.80    4.07     1      5

Rel Service   4.40    4.61            5
3   Admin.    3.83    4.34     1      5
    Total     4.30    4.50     1      5

Rel Service   4.29    4.51            5
4   Admin.    3.32    3.86     1      5
    Total     4.08    4.32     1      5

Rel Service   4.24    4.48     1      5
5   Admin.    3.32    3.90     1      5
    Total     4.05    4.30     1      5

Rel Service   4.42    4.60            5
6   Admin.    3.51    4.08     1      5
    Total     4.22    4.43     1      5

Rel Service   4.39    4.57            5
7   Admin.    3.30    3.85     1      5
    Total     4.14    4.36     1      5

Rel Service   3.85    4.20     1      5
8   Admin.    2.59    3.25     1      5
    Total     3.58    3.92     1      5

Table 2
                 Sum of              Mean
                 squares   df        Squaare   F          Sig.

Between Groups   20.674    1         20.674    65.898     .000

Within groups    60.550    193       .314

Total            81.224    194
Gale Copyright: Copyright 2012 Gale, Cengage Learning. All rights reserved.