Volume 54, Number 1 Winter 2021

Student Issues

Edited by Jessica S. Saucedo, Michigan State University

A Mixed Methods Study on Community Academic Partnerships in Public Health: Preliminary Findings from Phase 1

Written by Tatiana Elisa Bustos, Michigan State University


Public health needs call for greater community participation and control in processes that define community problems and design and implement interventions that are both meaningful and feasible within the community (Israel, et al., 1998; Wallerstein & Duran, 2010). To that end, a systems approach to public health challenges in underserved communities can utilize community-academic partnerships (CAPs)—partnerships extending beyond academic boundaries to translational research in real-world settings—to support and enhance the capacity of existing community-based initiatives and integration of evidence-based programs. CAPs involve community-partnered research that includes community stakeholders into the decision-making processes of interventions, programs, practices, and other health-related efforts; likewise, academic stakeholders are integrated into the decision-making processes of community-based organizations’ (CBOs) real-world application of said practices, interventions or treatments into the community (Drahota et al., 2016; Pellecchia et al., 2018).  

While CAPs are increasingly utilized, the perspectives of community partners participating in these collaborations remain understudied. Other studies on CAPs have indicated a significant gap in documenting the effectiveness of partnerships and an overall limited understanding of how partnership characteristics relate to successful or unsuccessful outcomes (Drahota et al., 2016; Lasker et al., 2001; Ortega et al., 2018). Furthermore, little attention is paid to the interactive relationships in CAPs, which can be a proxy to its effectiveness (Behringer et al., 2018; Bunger et al., 2014; Honeycutt & Strong, 2012; Ortega et al., 2018). For instance, factors such as motivation to participate, perception of success, and reasons for continuing to collaborate can strongly predict the long-term effectiveness of the network of CAPs (Carney et al., 2011; Valente et al., 2008). Furthermore, assessing contextual factors of CAPs (such as facilitators or barriers) and eliciting details on community partner perspectives have strong potential to create strategies that are more effective for real-world settings (Behringer et al., 2018; Suarez-Balcazar et al., 2005). Thus, there is a need to broaden understanding of how


Figure 1

communities can benefit from CAPs to inform strategies that sustain partnerships for health efforts over time. The current dissertation study explores CAP partners’ motivation to participate, perception of benefits, and overall CAP experiences. Further, the study explores how external factors related to the COVID-19 has changed (or amplified) CAP efforts. The report here will focus on phase 1 of the dissertation project, as phase 2 data collection remains underway. 


Study Design. The project utilized a longitudinal, sequential mixed methods design (QUAN → QUAL) to explore an instrumental case study of a CAP centered on public health efforts in Michigan. Specifically, the project applied mixed methods social network analysis (MMSNA) utilizing social network analysis in tandem with qualitative interviews. Of note, a longitudinal approach allowed for a temporal assessment of how contexts can change over time, revealing underlying processes of partnership dynamics. Guided by the Model of Research-Community Partnership (Brookman-Frazee et al., 2012), the project aimed to identify facilitating and hindering factors to the collaboration process, understand partner perspectives about the collaboration, and examine changes in the overall

network structure over time (1 year apart). The quantitative phase applied social network analysis (SNA), with the unit of analysis as the network and outcome measures related to existing network ties, degree of collaborative activities, level of trust, and frequency of interactions. Qualitative interviews then solicited community perspectives to contextualize quantitative responses, providing the breadth of collaboration experiences. Through the use of MMSNA, results are expected to provide an in-depth assessment of factors that contribute to or hinder the growth of networks in CAPs, along with the perspectives of community partners, including motivation to participate and perceived success.  

Recruitment and Sample Characteristics.  In January 2020, the CAP was comprised of 28 primary agencies within the Midwest region, including community members, academic partners, and policymakers involved in health equity efforts at the local, state, and national level. All participating agencies within the CAP were recruited to participate in Phase 1 quantitative data collection using phone calls and emails. Phase 1 eligibility criteria included: (a) represent a participating agency in the CAP; (b) read and speak in the English language; and (c) be 18 years of age or older. In this case, key representatives were members who attended meetings, completed evaluation assessments, and acted as site facilitators to communicate between the CAP and their affiliated agencies. Given the strain of the COVID-19 pandemic on community-based agencies, only CAP core leaders were interviewed for phase 1 qualitative data collection. However, all partners will be recruited for phase 2 data collection (QUAN + QUAL).

Materials. The PARTNER Tool is a social network analysis tool designed to assess the collaboration efforts among partners within a collaborative. The tool was integrated with other assessments exploring motivations to CAPs, including items from the Decision to Participate (DPQ) survey (Gomez et al., 2018) and the CAP survey, which included facilitators and barriers identified from a prior systematic review on CAPs (Drahota et al., 2016). The final adapted survey collected information on: (1) facilitators and barriers to CAPs; (2) partners’ motivations to participate; (3) demographics; (4) perceived goals; (5) perceived success; (6) trust; (7) perceived value; and (8) network metrics on interactions between partners. These details were expected to provide insight into characteristics that have contributed to the ultimate success of the community partner’s goals (Williams et al., 2018). Efforts were made to incorporate the context of the community partner in collaboration with community core members. The quantitative data analyses then informed the development of a semi-structured, individual interview. The interview protocol aimed to elicit: (a) perspectives on the collaboration process with the community partner; (b) barriers and facilitators to the CAP efforts; (c) motivations for joining the CAP; (d) expectations of community partner’s outcomes; and (e) suggestions for improving the partnership. Participants were provided a $15 Amazon gift card as an incentive for their participation in the survey and for completing the interview at time-point 1. Of note, these interviews with community representatives were halted when COVID-19 rates were rising. In response to COVID-19, the interview protocol for phase 1 qualitative data collection was modified to explore how the pandemic changed the structure of the CAP, how the partnership continued to support collective health equity efforts, and what advice would be given to another CAP experiencing these same emergencies from the perspective of the CAP leaders. 

Data Analysis Plan

Social Network Analysis. For the quantitative portion of this study, social network analysis (SNA)—a systems science methodology—was used to assess the CAP at two time-points approximately one year apart (January 2020, January 2021) to examine how partnerships grow and develop over time with the unit of analysis as the network of the CAP. Using the PARTNER Tool platform, applied network analysis was used for time-point 1 to assess the overall structure of the network and identify significant relationships between network metrics and CAP characteristics. Visual sociograms were then created to depict the nodes, representing agencies, and network ties, conveying the links between multiple pairs of nodes within the network (Bergenholtz & Waldstrom, 2011; Borgatti et al., 2013; Monge & Contractor, 2001). This sociogram provides an overview of how agencies within the CAP are interacting, if there is a clique or cluster of groups working together, where interactions are predominantly occurring, and how organizational attributes (e.g., type of agency) are related to frequency of ties. Figure 2 provides an example of a sociogram where the size of the nodes indicates the average rate of ties with other partners in the network (e.g., density).


Figure 2

SNA can integrate other characteristics with visual illustrations, such as sociograms for frequency of collaborations or separate sociograms for different levels of collaborative activities. Additional descriptive statistics, including means, standard deviations, t-tests, and correlational analyses were used to explore the quantitative data (e.g., number of months participating with the CAP, type of services contributed through the agency, job title, and other CAP characteristics). Analyses for time-point 2 are underway. A community report with preliminary social network data from time-point 1 was presented to CAP leaders for a formative discussion to inform next action steps for the collaboration as a whole (see Figure 4). 

Content Analysis. Qualitative data from time-point 1 was analyzed using directed content analysis, a widely used, flexible qualitative approach (Bernard, 2006; Hsieh & Shannon, 2005). Transcriptions of interviews with leaders were completed with Rev Transcription services. Using MAXQDA, text data was then coded to create themes that explain underlying patterns or meanings (Hsieh & Shannon, 2005; Vaismoradi & Snelgrove, 2019). Codes were quantized to present ever-coded (e.g., the number of transcripts that had the code assigned ever) and frequency (e.g., the number of times the code was assigned throughout all of the transcripts) counts, which provide additional data to support the salience of the emergent themes (Bernard, 2006; Hsieh & Shannon, 2005). Qualitative analysis for phase 2 of the project (final interviews with community partners) remains underway. 

Integration of QUAN + QUAL. Convergence refers to the process of bridging the quantitative and qualitative data strands to explain the phenomenon of interest (Palinkas et al., 2019; Plano Clark et al., 2015). Phase 1 QUAN results will be converged with the Phase 2 QUAL data at time-point 2 to allow for expansion of collaboration experiences and comparisons within and across partners’ responses. The product resulting from this mixed-method approach will yield a joint display to demonstrate the salience of themes related to barriers and facilitators, along with motivation and perspectives from key representatives regarding the CAP process.


Participants. A total of 23 quantitative survey responses were received at time-point 1, resulting in a response rate of 83%. Twenty respondents represented local, state, and national level community partners, and three respondents represented academic partners. Participants who did not complete the network survey at time-point 1 included 2 policymakers, 1 academic partner, and 2 local community partners. At time-point 1, CAP partners’ involvement with the CAP averaged at 23.95 months (SD = 12.675) with a range of 2 - 40 months (e.g., since the CAP began in 2016). Further exploration of these rates by partner type revealed that community and academic partners were both involved with the CAP around the same duration, at an average of over 24 months (M = 24.26, SD = 12.346; M = 21, SD = 21.213, respectively). 

Knowledge on relationship ties exchanged between partners in CAPs. Network scores indicated that there were 195 ties in the network of 27 nodes, with a mean average of 13 ties with other partners in the CAP. See Figure 4 for an overview of other network-level metrics. The dissertation project contributes to the application of a social network perspective to CAPs in public health, with emphasis on observational changes to a CAP’s network structure. A social network perspective can allow for the visual representation of patterns in ties over time and facilitate discussions on governance, legitimacy, trust, with collaboration in a community-based context. Qualitative findings from interviewing leaders further demonstrated how CAPs mobilize and collaborate to improve ties amongst community members representing major public health departments, particularly when faced with a public health crisis.  


Figure 3


Figure 4

Motivations to participate. Understanding motivation to participate is important to ensure efficient and successful strategies for collaboration to result in health equity outcomes (Carney et al., 2011). For time-point 1, both community and academic partners were motivated by the opportunity to network with others engaged in health equity efforts. However, many community partners were also motivated by their shared sense of values and mission in health equity; whereas, academics prioritized their motivations for a systematic process to implementing evidence-based practices. The current project will assess motivating factors again in December to explore how motivations may have changed over time, particularly in response to COVID-19. Additionally, qualitative interviews at time-point 2 will help expand on these findings by revealing why such motivational factors were particularly important to community partners and how they may have led to the CAP’s outcomes. 

Perceived goals and success. Positive perceptions of the efforts affiliated with a CAP can influence a community member’s decision to participate (Ortega et al., 2018). Interestingly, many community partners viewed the most important goal of the CAP as a means to reduce health disparities, whereas academics rated the most important goal as “increasing knowledge sharing.” This suggests that community and academic partners in collaboration are working towards practice-oriented goals rather than goals set by other commitments. Further, both community and academic partners viewed the CAP as successful in meeting its goals, with some qualitative responses indicating where improvements can be made. In this case, both partners prioritized improvements in role and participation. 


Organizations, communities, and partnerships involved in public health efforts must not be seen as static but as fluid and cumulative efforts that are dynamic by nature (Behringer et al., 2018). Studies in public health are progressing towards integrative, system-level approaches with SNA (Bright et al., 2017, 2019; Chambers et al., 2012; Franco et al., 2015; Leischow & Milstein, 2006). However, the use of SNA is underutilized in the study of CAPs, in particular (Bright et al., 2017; Franco et al., 2015). The impact of this research builds on systems-level, ecological perspectives grounded in community psychology, emphasizing how networks of CAPs in public health within larger systems of marginalized communities can function collaboratively to better understand and resolve health disparities. In utilizing SNA, the dissertation explored how network properties and partnership dynamic processes relate to CAP success and outcomes. Community stakeholder participation in the CAP process is also important to understand in order to create and design strategies that are relevant, useful, and responsive to the needs of a given community (Benoit et al., 2005). To this end, this project contributes to research on the CAP collaborative process to inform future efforts to develop and maintain successful partnerships for broader public health equity impacts. Findings ultimately contribute to community and academic perceptions of CAP collaborations that highlight dynamic processes intertwined with contexts related to community, operational, and interpersonal processes. 


If you have any questions or are interested in following up on more specific details about the study, please reach out to me via email at or follow me on Twitter (@telisa72). 


Thank you to my community partner, advisor, and the SCRA community for supporting this dissertation project.