Cross-System Collaboration: A Mixed-Methods Network Analysis of Interoganizational Challenges in Child Welfare

Background

Collaboration among a wide variety of organizations is essential to child welfare practice. Though research consistently identifies challenges in these collaborative relationships, results remain mixed, and studies typically examine only two or three organizations rather than a more encompassing interorganizational landscape. Using network analysis and qualitative interviews, the purpose of this study was to clarify and broaden a systematic understanding of challenges experienced among a ‘whole’ complex network of organizations connected to child maltreatment prevention and service delivery.

Methods

As part of a larger study, data were gathered via a mixed-methods design. Snowball sampling began with a county’s Department of Children and Family Services, and included all organizations connected to child maltreatment prevention and service delivery efforts (n = 105 organizations). Key informants from each organization completed a survey (n = 80, 76%) indicating their organization’s experience with 15 pre-identified challenges derived from the literature on collaboration in human-services. Examples include “interagency politics”, “loss of organizational autonomy”, and “time”. Respondents were asked whether challenges were experienced during service delivery, during prevention efforts, or both. Semi-structured interviews were conducted with 67 of the 80 organizations (84%) to deepen understanding of challenges encountered when collaborating within the network. UCINET and NETDRAW were utilized for network data analysis and visualizations, and Atlas. Ti for qualitative data management and thematic analysis.

Results

Data revealed differences between challenges experienced during service delivery and prevention tasks, and that challenges were differentially experienced by organizations. Findings will be discussed in detail using network visualizations that map this variability. Of the 15 pre-identified challenges, some were excluded in the final analysis, and other categories were collapsed or broadened given the combined interpretation of survey and qualitative findings. For example, based on survey responses, the most prevalent challenge in collaborating was “time” (n=57), yet qualitative data yielded a broader version of this challenge in which shortages of time, resources, staff, and funding were inseparable. In total, 16 challenges were supported by the mixed-methods analysis, seven of which were not categories on the pre-identified list, but newly articulated by respondents. Threaded through all of the challenges was an identified pattern of “instability” suggesting that the challenges themselves are dynamic aspects of the complex system’s functioning, rather than static, stand-alone constructs.

Conclusions

The mixed-methods network design generated a more complex and nuanced picture of challenges experienced in interorganizational child welfare collaboration than evident in past research. Findings are derived from a diverse representation of organizations, and suggest numerous policy and practice implications, including implications specific to how challenges vary by task and by organization. Further, the observed thread of instability offers a paradigm-shifting lens through which to view and improve collaboration. Because the pattern of instability rejects the notion of challenges as constants, new opportunities for tackling challenges emerge at the intersection of multiple stakeholders, and at different points in collaborative processes. Policy and practice that anticipates, and even embraces instability, may better-equip organizations to navigate their networked environment and will be discussed as an area ripe for future research.