The Matrix Outcomes Model of Collaborative Building among Community-Based Family Resource Centers in California ()
1. Introduction
There is a strong consensus in Child Welfare literature regarding the vital role of community-based organizations (CBOs) to prevent child maltreatment. Building strong community partnerships that involve public Child Protective Services with local family resource centers is generally seen as a crucial component of successful strategies for the prevention of child maltreatment [1]-[4]. The evidence shows that connecting families with service providers in their neighborhoods fosters enduring relationships between families and organizations even after the formal service relationships have ended creating strong safety nets of support for both birth and foster parents [5]. While the literature on the positive impact of partnerships between public child welfare agencies and non-profit CBOs on family outcomes has grown considerably during the past 2 decades, there is still a need for research on successful models for building, growing, and maintaining these partnerships. Specifically, not much is known about the process of building successful and enduring collaboratives of nonprofit CBOs in general and within family and mental health services fields [1] [4] [6] [7].
This paper presents the Matrix Outcomes Model LLC. (MOM) experience and lessons learned from building non-profit agency collaboratives since 2002. The purpose of this study is to use a mixed-methods approach to explore the extent and ways in which the MOM’s collaborative building process succeeded in building collaboratives of CBOs for the prevention of child abuse and neglect. The second part of the paper reviews the literature on non-profit agency collaboration, the third part presents the collaborative building process established by the MOM, while the fourth part presents data on agency collaboration in more than 20 collaboratives that adopted the MOM system and qualitative data (interviews) with 3 collaborative coordinators on their experiences using the MOM. The concluding section discusses lessons learned for future organizations seeking to build collaboratives in the human services field.
2. Collaboration among Community-Based Agencies
It is common for organizations whether they are nonprofit, public or private agencies, to work together for a common cause. It is important to distinguish between the depth of the coalitions they form. Himmelman proposes 4 levels of coalition depths ranging from networking to collaboration [8]. At the most basic partnership level agencies engage in networking were exchange information for mutual benefit. At the second level, agencies work together in the coordination of activities, while on the third level they also engage in cooperating by sharing resources. The final level of an interagency coalition is that of collaboration which entails not only the sharing of information, activities, and resources, but of risks, rewards, and responsibilities. While the benefits of inter-agency collaboration for service delivery in health and human services are well established, there is a recognition of the challenges of collaboration. After all, non-profit service providers usually “compete with each other for clients, status and reputation within the community, and scarce resources” [9]. Thus, successful inter-agency collaborative models are those where the benefits of collaboration for individual agencies outweigh the challenges.
In their comprehensive review of nonprofit collaboration studies, Gazzley and Guo recognize that the literature offers mixed results and gaps when explaining successful collaborations including the role of resources and government, theoretical underpinnings, and empirical definitions of successful collaborations [6]. Nevertheless, the literature points to agency, leadership, workers’, contextual, and technical resource characteristics that seem to be correlated to success in collaborative endeavors.
2.1. Agency and Contextual Characteristics
Murray explains the building of collaboratives as a multi-stage process mediated by four aspects that determine a successful process: 1) the type of collaboration being sought, 2) the characteristics of the organizations entering the collaborative, 3) the process of developing and implementing the collaborative, and 4) the environmental and contextual factors that impinge on the collaboration process [10]. Osborne and Murray use this framework to describe a successful collaborative for delivering social services in Canada in which the type of collaboration being sought recognized agency autonomy (coordination) in the beginning before trust was achieved and the relationship became more collaborative [11]. Additionally, the agencies had compatible organizational cultures that allowed them to establish a common process of building a collaborative, all the agencies faced moderate external pressure from funders for them to work collaboratively.
2.2. Leadership Characteristics
In terms of the required leadership to form successful collaboratives, Libby and Austin find that the main contributors to a successful collaborative for the provision of mental health services in Napa, CA was the directors’ buy-in, their authority to make decisions without their respective boards’ approval, and the clear independence and leadership of the coalition facilitator [9]. Similarly, O’keefe finds that agency leadership was crucial for achieving clearly defined goals and purpose which, in turn, fostered the success in building collaboratives for the provision of child welfare services in San Mateo County [12]. In addition to individual agencies’ leadership, Tong et al. recognize the vital importance of a well-funded collaborative hub coordinator in their study of successful adult services collaboratives in Canada, while Cooper et al. describe the significance of such a coordinator in their work on children’s mental health services collaboratives in the UK [4] [13].
2.3. Workers Characteristics
Individuals in organizations are ultimately the ones who will carry out a collaborative effort. Using multi-level modeling analyses of survey data, Welsh et al. find that agencies with higher levels of worker-perceived organizational support and individual adaptability were more likely to rate interagency collaborations as effective for clients and beneficial to them as a resource to serve their goals [14]. Thus, agencies in which workers felt more secure and recognized in their roles were more likely to seek opportunities outside their roles and agencies and to perform the extra tasks demanded by collaborative building process. These findings are supported by Bai et al. in their study of successful collaboratives in the child welfare system in The United States [15].
2.4. Common Data Systems
Kaasbøll et al. find that besides some of the agency and worker characteristics discussed above, having a standardized data (screening) system that was well perceived and accepted by all agencies helped facilitate an interdisciplinary collaborative for screening patients’ mental health in Norway [16]. In the United States context, the presence of a shared information system, along with clear roles (jurisdictions) was also associated with better inter-agency collaboration among mental health and child welfare [7]. Takahashi and Smutny (2001) also refer to a common information data system to track clients as a requirement for a successful collaborative, but also a result of a successful collaboration process [17]. Their study of an HIV wellness collaborative comprised of small agencies revealed that the implementation of a common data system required a high degree of integration and dependence between participating agencies as it encouraged the agencies to share skills and knowledge across the collaborative.
While the literature has a growing number of case studies of collaboratives, there is little data on the sustainability of these relationships over time (Gazzley & Guo, 2020) [6]. In addition, there is limited information on formal processes or models utilized by funders to build service provision collaboratives. The next section describes the Matrix Outcomes Model of collaborative building for the provision of child services.
3. The Matrix Outcomes Model
The MOM was created in 1998 as an assessment and case management tool that could enhance the capacity of community-based organizations to provide child and family services. The model was created during a period of changes in the practice of child welfare services and funding regulations at the federal and state levels that required a model focused on the family, and with an emphasis on outcomes. The Family Development Matrix is based on a strengths model rather than a “deficit” model. Documenting where a family is thriving as well as where it needs support allows those using it to easily identify strengths from which to start addressing needs. The process facilitates family ownership of their efforts. The caseworker becomes the assistant in helping them set and work toward short and long-term goals. This aids the family in taking both credit and responsibility for their decisions and actions. The Family Development Matrix is frequently used in a three-step process as shown in Figure 1:
![]()
Figure 1. Matrix Outcomes Model case management process.
As depicted in Figure 1, the Matrix Outcome Model of family support begins with the case manager meeting with family members to determine baseline scores for each of the Outcome Categories on which they will be working in a case management plan. Each collaborative decides on the indicators they will focus on depending on the population they serve and the needs they address. Each indicator has four status levels:
In-Crisis: Family cannot meet its needs. Family is unwilling or unable to work toward positive change. Family systems have collapsed or are in immediate danger of collapse. Strong outside intervention is needed to move the family to at least “At-Risk” level.
At-Risk or Vulnerable: Family is secure from immediate threats to health and safety but has not yet developed or committed to plans for long-term growth and change. Continuing safety-net interventions provide a platform on which the family can build its plans for improving its circumstances.
Stable: Family is no longer in danger, is ready and willing to change and is planning for its future. Supportive services are provided to assist family members in implementing their plans.
Safe/Self-Sufficient: Family is strong and has made significant progress in improving its circumstances; it is generally secure because of its own efforts. The family is economically self-sufficient and has a clear vision of its ultimate goals. Intervention is resource-oriented.
After the baseline assessment, specific subsequent meetings take place where the “scores” established at the previous meeting are revised as appropriate, and a new “action plan” is devised for implementation until the next meeting.
Services end when the family is successful at achieving its goals, after the family voluntarily leaves services, or when time-limited services conclude.
Outcomes for families using the MOM have been documented in several studies that explored family engagement in voluntary services, changes in family outcomes from first to second assessment, and parenting education outcomes [18]-[21]. The focus of this study is to explore the MOM’s impact on agencies’ collaborative building process.
3.1. Collaborative Building Process
Customizing the MOM was structured as a team-building process centered around a case management model described above and a common outcomes data system. As depicted in Figure 2, the collaborative building process begins with members of a local design team coming together to select family outcome indicators that are tested with families for validity and reliability, create an assessment protocol, and agree on case planning practices.
3.2. Trainings and Technical Assistance
In 2009, the MOM moved services to a hybrid online model, providing collaboratives with online trainings and technical assistance with collaborative and agency coordinators, design team members, family workers and data analysts. MOM training plays a crucial role in helping newly formed collaboratives use the MOM in conjunction with other systems being used and ensuring the product they create is valid and beneficial for all agencies participating. It also enforces the idea that agency leads must work together to develop one method for service delivery tracking. MOM data entry training. are provided as the design team is being established to demonstrate how the system works and their practice protocol to ensure data entry accuracy, deadlines, or timeframes for completing forms, and identifying codes they will use to track families or individuals. From time to time, modifications are made to the online system to better match and synchronize data with other databases as well.
![]()
Figure 2. Collaborative building process.
After an initial training activity, at least one more training is offered. The Matrix database trainer uses collaborative-approved documents (i.e. practice protocol and client coding sheet) to guide users in collecting data for each case management form. These training activities provide staff with the opportunity to review each step of the case management process, request clarification on the protocol, make connections between the MOM and other databases being used, test for indicator and intervention relevance and accuracy, discuss how appropriate timelines are across all collaborative agencies, and if changes should be made to the written protocol and code list. A final training may be scheduled to confirm guidelines that all agencies within that collaborative will follow also serving as a “refresher training” before all agencies “go live” in using the system with families.
We hypothesize that the MOM tool facilitates the building of community-based organization collaboratives by: 1) establishing a common goal in the form of engaging participating agencies in the developing of common outcome indicators, 2) Supporting the collaborative leadership with trainings and data reports, and 3) by providing a common data system that standardizes case management practices across agencies and allowed workers from different agencies to communicate and aid families using the same practices and family outcomes.
The next section describes the MOM collaborative outcomes and the results of a qualitative study that tested hypotheses presented in this section.
4. The MOM Collaboratives 2002-2022
During the 20 years of implementation of the MOM, the model has been used by 212 agencies. Most of these agencies (95%) operated in collaboratives of 2 or more agencies and 88% in collaboratives with 3 or more agencies. As presented in Table 1, the agencies provided services to 55,255 families. Each family served received a first assessment and an empowerment plan. About 59% of the families served returned to a second assessment. In total, 101,125 client visits were recorded by all collaboratives between 2002 and 2022. In addition, the families served included 73,020 children. 47% of the children were younger than 6 years old, 34% were between 16 and 12 years old, and 19% were between 13 and 18 years old. Most of the families served were of Latino/Hispanic descent (56%), followed by white (19%), and African American (14%).
Table 1. Assessments and Families served by MOM agencies 2002-2022.
| Statistic |
# |
% |
Total |
| Number of First Assessments |
55,255 |
|
|
| Number of Second Assessments |
32,780 |
|
|
| Total number of assessments |
101,125 |
|
|
| Children 0 - 5 |
34,224 |
46.9 |
|
| Children 6 - 12 |
25,104 |
34.4 |
|
| Children 13 - 18 |
13,692 |
18.8 |
73,020 |
| Hispanic/Latino |
27,678 |
56.5 |
|
| White |
9519 |
19.4 |
|
| African American |
6672 |
13.6 |
|
| Mixed/Other |
2155 |
4.4 |
|
| Asian/Pacific Islander |
1582 |
3.2 |
|
| Other |
749 |
1.5 |
|
| Native American |
665 |
1.4 |
49,020 |
Collaborative Characteristics
Table 2 presents the 23 collaboratives that had at least three agencies. As the table shows, collaborative’s ages at the time of analysis ranged from .9 years (Los Angeles) to 16.2 years (Sacramento). On average collaboratives remained in the MOM for 7.3 years. The collaboratives’ size was measured by the number of agencies comprising the collaborative. On average, each collaborative had about 8 agencies, but the size varied substantially. The collaborative with the highest number of agencies was San Francisco, which had 20 agencies participating in the collaborative at some point during the collaborative’s work with the MOM. The smallest collaborative was in Alpine County, which had 3 agencies, and served 29 families in their 5 years of operation. In terms of clients served, the largest collaboratives were San Bernardino and Stanislaus, which served 6694 and 7851 families respectively.
Table 2. MOM Collaboratives’ age, size, and stability.
| Collaborative |
Years in operation |
Agencies |
Families Served |
Number of years in collaborative |
| Mean |
Standard Deviation |
Coefficient of Variation(SD/Mean * 100) |
| Lake |
15.3 |
7 |
559 |
4.1 |
5.7 |
139.9 |
| Del Norte |
5.2 |
4 |
220 |
1.3 |
1.5 |
113.5 |
| Ventura |
9.1 |
8 |
2459 |
3.0 |
3.1 |
104.4 |
| San Joaquin |
14.2 |
3 |
3842 |
7.9 |
6.9 |
88.0 |
| Placer |
1.1 |
3 |
79 |
0.5 |
0.4 |
79.3 |
| Sacramento F5 |
16.2 |
14 |
5000 |
8.1 |
6.4 |
78.9 |
| Yolo |
8.8 |
5 |
1133 |
4.0 |
2.7 |
67.6 |
| SFO |
12.2 |
20 |
3776 |
5.5 |
3.7 |
66.4 |
| Fresno |
11.1 |
4 |
3037 |
5.3 |
2.9 |
54.5 |
| Alpine |
3.0 |
3 |
29 |
1.9 |
1.0 |
54.4 |
| Orange |
10.5 |
13 |
3049 |
4.8 |
2.5 |
52.9 |
| SLO |
5.4 |
6 |
341 |
3.3 |
1.7 |
52.7 |
| Yurok |
2.6 |
3 |
50 |
1.8 |
0.7 |
37.8 |
| San Bernardino |
7.8 |
13 |
6694 |
5.9 |
1.9 |
32.4 |
| Santa Barbara |
13.3 |
13 |
5498 |
10.4 |
3.2 |
30.5 |
| Siskiyou |
7.3 |
11 |
328 |
4.8 |
1.5 |
30.4 |
| New Jersey |
1.4 |
6 |
393 |
0.9 |
0.3 |
27.7 |
| Humboldt |
6.6 |
13 |
418 |
5.2 |
1.4 |
26.2 |
| Los Angeles |
0.9 |
18 |
981 |
0.7 |
0.1 |
21.4 |
| Santa Clara |
2.5 |
3 |
635 |
2.2 |
0.4 |
16.9 |
| Washington |
1.1 |
8 |
903 |
0.9 |
0.1 |
9.2 |
| Tulare |
2.9 |
5 |
1039 |
2.6 |
0.2 |
7.4 |
| Stanislaus |
10.6 |
4 |
7851 |
10.5 |
0.1 |
0.9 |
The last 3 columns in Table 2 present the average and standard deviation of the number of years agencies stayed in the collaborative and each collaborative’s coefficient of variation as a measure of agency turnover. Higher coefficients of variation reflect higher agency turnover, and lower coefficient of variation reflects lower agency turnover. As the table shows, the Lake collaborative experienced the highest agency turnover (coefficient of variation = 139.9). In its 15 years of operation using the MOM, the collaborative had 7 agencies that remained in the collaborative for 4 years each on average. On the other hand, the Stanislaus collaborative had the lowest turnover (coefficient of variation = 0.9). This collaborative was comprised of 4 agencies that stayed in the collaborative for the entire 10 years of the collaborative operation with the MOM.
While the coefficient of variation provides a relative measure of agency turnover that may be used as a proxy for collaborative stability or success, it does not explain how or why each collaborative received its respective scores. For this reason, we conducted interviews with the collaboratives of San Luis Obispo (SLO), San Bernardino, and Santa Barbara. SLO was selected because it scored a coefficient of variation of 52.7, which falls in the middle of the distribution. Half of the agencies scored higher, and the other half scored lower than the SLO collaborative. In addition, the collaboratives of San Bernardino, and Santa Barabara, (with coefficients of variation of 32 and 30 respectively) scored relatively lower than SLO. These collaboratives were selected because of their availability, size (in agencies and clients served), and because they had higher than average years of operation.
The interviews were conducted between March and May of 2023. Agency coordinators from the 3 selected collaboratives participated in interviews that were conducted by ZOOM and lasted 1 hour each. During the interviews, the coordinators were asked to share 1) their experience during the creation of their respective collaboratives, 2) their perceptions on the factors that contributed or hindered their collaborative, and 3) their opinion on how successful they perceived their collaborative was. Their responses were coded and organized around 4 common themes: the role of leadership, the contextual characteristics surrounding the implementation of their collaboratives, worker buy-in, and the common set of outcomes and data system.
5. Results and Discussion
5.1. Leadership Characteristics
As explained in section 3, the MOM requires a coordinator for the entire collaborative and each participant organization within a collaborative designates a Coordinator that represents the agency in the collaborative. The three collaboratives interviewed followed this protocol, but their responses revealed marked differences in how the agency and collaborative coordinators were selected and the specificity of their goals.
SLOs Collaborative Coordinator position was assigned to San Luis Obispo Child Abuse Prevention Council (SLOCAPC). This decision was based on their perceived level of experience in serving the community and willingness to take on the responsibility of developing a new collaborative (L. Fraser, personal communication, May 5, 2023). The main goal was “to use this leadership role to build a collaborative that could make a real impact in the community”. While implementing the MOM, SLO collaborative chose to share the responsibility across agencies and their staff. It focused on everyone’s individual and unique role, including experience and knowledge of servicing families. There was an emphasis placed on cross-generational training and transfer of knowledge to younger case managers as well.
The Santa Barbara collaborative, on the other hand, made the decision to assign the Collaborative Coordinator position based on agencies’ level of authority, funding power and standing professional partnerships (T. Johnes, personal communication, March 23, 2023). First 5 Santa Barbara had at that time funded some of the agencies recruited for the MOM collaborative and it had scheduled meetings that agency participants would attend on a quarterly basis. It had the authority to make executive decisions and build on the relationships it had with grantees. The overall goal of participating in the MOM was to build a collaborative that would synchronously use a data system that would collect information to help tell a story about families and the work being conducted. During the implementation of the MOM, efforts were made to collect data, share outcomes, and strengthen networking. Data was used to identify trends for the communities that were being served and connect those achievements to the commitment agencies had agreed upon by the collaborative. Doing so encouraged agencies to follow protocol and see the worth in data collection which in return promoted buy-in to the collaborative goal.
In contrast to the SLO and Santa Barbara collaboratives, San Bernardino, built their collaborative around First5 San Bernardino, which recruited agencies to build a collaborative based on their previous experiences with them in past projects. First 5 San Bernardino, was a funding source for participating agencies and had knowledge of the data systems being used including the program curriculums used by sites (S. McGrath, personal communication, April 7, 2023). This posed an advantage in First 5 San Bernardino’s leadership role as it minimized the amount of “figuring out” agency assets and deficits. First 5 San Bernardino took initiative early on to develop a practice protocol for agencies to follow while using the MOM. First 5 San Bernardino identified strength in numbers and sought to grow this collaborative into a team that would move together as one in seeking funding. The collaborative’s leadership strategically imposed guidelines to allow the use of various data systems. Interestingly, the collaborative coordinator role for First 5 San Bernardino did not rest on one individual, but a group within First 5 San Bernardino.
5.2. Agency and Contextual Characteristics
All three collaboratives interviewed for this study recruited agencies for MOM participation based on funding source and/or the communities and families being served. First 5 Santa Barbara and First 5 San Bernardino served as the funding source for most agencies in their collaborative and had existing partnerships with these agencies. SLOCAPC had not yet established a formal partnership with agencies, however, it was aware of the work being conducted community-wide. All agencies implementing the MOM in these collaboratives were non-profit organizations serving families within their county with similar needs although each agency served in different capacities. Agencies recognized the need to come together to make a greater impact in the communities they were serving. Their goals were similar, however, because First 5 agencies focus on family services for children ages 0 - 5, participating agencies had to accommodate their service in the collaborative with the funders’ goals. SLO, on the other hand, served families with children of any age and therefore had agencies that did not have to change their population of focus to fit the collaborative funding goals.
5.3. Workers’ Characteristics
Each of the collaboratives interviewed took on a different approach to support staff in achieving their collaborative goal. This was SLOs first time working as a collaborative, therefore guidelines were being developed and refined from scratch, thus, agencies’ staff had to add a considerable amount of time working outside of their usual responsibilities. Working as a team “forced staff to come together” to meet agency commitments and family needs (L. Fraser, personal communication, May 5, 2023). Santa Barbara, on the other hand, had already contracted agencies before and they invited agencies to the collaborative under their established protocols. Their focus was on increasing worker compliance and training them in the data systems and protocols (T. Johnes, personal communication, March 23, 2023). Agency competition was recognized during meetings but addressed openly and collaboratively. This collaborative used the MOM to share cases across different agencies. It also focused on highlighting staff’s hard work during quarterly meetings. Staff were encouraged to share insights into the work being conducted and share how collaboration could be improved. Interestingly, the San Bernardino collaborative had worked with some of the agencies before establishing their collaborative so their onboarding process into the collaborative took less time and resources of individual agencies entering the collaborative (S. McGrath, personal communication, April 7, 2023).
5.4. Common Data Systems
All interviewed collaboratives acknowledged the crucial role a common data system played in standardizing agencies’ goals and protocols. The MOM proved to be helpful for these collaboratives in coming together as a unit to create tangible and concrete collaboration across agencies to provide case management services (L. Fraser, personal communication, May 5, 2023). The database helped create a “shelf ready” product that could be adopted within the collaborative and future funding opportunities. Using the MOM database was an advantage, as it was already approved as an evidence-based tool (S. McGrath, personal communication, April 7, 2023). The MOM also helped agencies use common language to communicate family needs within the collaborative, establish a universal protocol that could be used across agencies servicing the same communities, and legitimize the work of case management at an agency level (T. Johnes, personal communication, March 23, 2023). It also helped collaboratives move towards strength-based outcomes and bring agencies together to see “what could be” of using the online system (L. Fraser, personal communication, May 5, 2023).
5.5. Measures of Success
The main goal for each collaborative was to provide services to families in need. In that sense, the measure of success for collaboratives was centered around services provided and families’ outcomes. This facilitated an alignment of measures of success from individual agencies to the collaborative as an aggregation of agencies’ success. While SLO and Santa Barbara did not mention any other measures of success for their collaboratives, San Bernardino established a sustainability goal that was to be measured as agencies’ capacity to sustain the collaborative with less reliance on First 5 funding. In their assessment at the time of the interview San Bernardino stated that Agencies in the collaborative implemented the MOM seamlessly and shared resources and insight but lacked the capacity to unite and share the rewards and risks as a singular team without A strong collaborative coordinator acting as a funding source.
6. Conclusions and Future Research Avenues
This paper describes the MOM’s experience supporting Community-Based Organization collaboratives serving children and families. During the 2002-2022 period the MOM provided services to 41 collaboratives comprised of 212 individual agencies. We found great variability in collaboratives’ stability in terms of agency turnover as measured by the coefficient of variation of average years in the collaborative.
Interviews with three collaboratives explained important commonalities and differences that may explain the differences in collaboratives’ stability. Interview responses were organized around themes found in the literature to contribute to collaborative success.
The three collaboratives that participated in the interviews faced somewhat similar contexts and recognized the crucial role of a common protocol, data, and evaluation system in the formation of the collaborative. These three aspects served as the common DNA in the collaborative. It allowed them to communicate using the same “language” and common family outcomes to work as a team. In that sense, the interviews confirmed our hypotheses regarding the value of a standardized data system based on a common set of outcome indicators and an established case-management protocol.
An unexpected, but important finding in this study was the role of leadership and funding in a collaborative. The collaborative coordinator, who was established as the leader, was recognized in each interview as the single most important aspect in bringing the collaborative together. As explained in the results section, each collaborative had a different approach to leadership, yet we found that agencies that had the first 5 collaborative coordinators had less turnover. This difference may be explained by the funding mechanism. First 5 works with a secure funding source (tobacco tax) and they use this leverage to align agencies in the collaborative with their goals. On the other hand, SLO had a more participatory relationship with its agencies given that their funding depended on the collaborative’s success. The role of a collaborative leader which is also the funder also had an impact on the type of agencies participating in the collaborative. Agencies participating in First 5 collaboratives needed to adequate their services to the funder’s population of focus and protocols. On the other hand, agencies participating in the SLO collaborative did not have to change but they had to negotiate the collaborative goals with other agencies. Thus, the role of a collaborative coordinator who is also the funder agency seems to play a mixed role in collaborative success in terms of stability. When the funder is the leader participating agencies align to the funder’s goals, protocols, and population of focus. Agencies in these types of collaboratives tend to have a lower agency turnover than collaboratives that have an external source of funding. These collaboratives tend to have a higher agency turnover rate as they negotiate the collaborative’s goals, protocols and population of focus. However, while having a funding agency as the collaborative coordinator reduces agency turnover, it seems to reduce the capacity of these collaboratives to share risks and benefits and thus may not be as sustainable as collaboratives that are dependent on an external funder agency. The implications of these findings suggest that collaboratives that have a coordinating agency that is also the funder should strive for flexibility with other participating agencies if their goal is to reduce agency turnover within the collaborative.
Finally, we found that stability and sustainability are important goals that are often overlooked by collaboratives. All collaboratives had the purpose of serving families in need and their main goals were related to their purpose, as expected. Yet only one of the collaboratives had funding sustainability as a stated formal goal. Undoubtedly, more research is needed to identify determinants of collaborative stability and sustainability. The literature is not robust in this area. We proposed agency turnover as a relative measure of sustainability which we consider to represent a crude but effective measure of collaborative stability.
Tracking collaboratives over a 20-year period provided a great opportunity to study their stability and sustainability, yet the contextual factors that changed over time may have varied differently across collaboratives. Unfortunately, this is an important limitation of this study that should be considered while interpreting the results. In addition, the three collaboratives interviewed for the qualitative component of the study were established in California and chosen because they were close to the media turnover scores. Thus, they may not be “representative” of all collaboratives. Therefore, caution should be exercised while making generalizations with these results.
Acknowledgements
This study was funded by First 5 San Bernardino. The authors would like to thank Scott McGrath, Teressa Johnes, and Lisa Fraser for participating in the interviews that were crucial for the completion of this research. All errors and omissions are the authors’ own responsibility.
Conflicts of Interest
The authors declare no conflicts of interest.