Understanding Factors That Support Well-Functioning Community Coalitions

Introduction

Extension plays a vital role in convening stakeholders around shared issues (Parisi et al., 2018), which is essential for tackling complex, community-level challenges (Clark et al., 2010; Smathers & Lobb, 2015). Indeed, Extension professionals are known for leading or participating in community-based collaborations addressing a variety of health initiatives, including disparities related to nutrition and physical activity (Butel et al., 2018). Extension's role in many states includes implementing Supplemental Nutrition Assistance Program Education (SNAP-Ed), the largest nutrition education outreach program of the U.S. Department of Agriculture (USDA). Federal SNAP-Ed guidance encourages coalition work as a tool for leveraging resources across programs that address similar goals (USDA Food and Nutrition Service [FNS], 2018). In Arizona, local-level coalition participation is common among SNAP-Ed agencies, including Cooperative Extension. These coalitions bring together SNAP-Ed and other stakeholders in support of collectively identified initiatives, such as the development of new farmers' markets or efforts to increase recreational facility access.

Although it is understood that coalitions seek to achieve community-level outcomes via the collective contributions of their members (Hersey et al., 2012; Zakocs & Edwards, 2006), success or failure may rely in large part on a given coalition's characteristics, including its resources, governance, and structure. The collective impact model offers a framework for implementing coalition-based initiatives, suggesting that informal relationships between members, coupled with participation requirements, can support effective coalition practices (Gillam, Counts, & Garstka, 2016). Research also has suggested that consistency of participation in a coalition, rather than number of participating organizations, may be more strongly associated with successful outcomes (Clark et al., 2010).

A recently developed national SNAP-Ed Evaluation Framework (USDA FNS, 2016) includes the recommendation that local SNAP-Ed agencies measure the functioning of their coalitions, including with regard to effective planning and capacity building to achieve their goals. To understand more about specific factors related to well-functioning coalitions, and to assist local Arizona coalitions in strengthening capacity to realize their goals, our evaluation team collected and analyzed data from coalitions that (a) included a SNAP-Ed representative and (b) were actively pursuing community-level changes that aligned with local SNAP-Ed programming goals. Here we describe our relatively low-burden method for measuring coalition characteristics that support goal achievement, which may benefit Extension professionals who participate in or lead coalitions in a variety of community and programmatic contexts.

Methods

In 2016, we assessed SNAP-Ed coalitions in Arizona that met the following criteria: (a) the coalition focused on community-level initiatives related to SNAP-Ed programming; (b) the coalition included representatives from at least five organizations in the community, including the agency implementing SNAP-Ed; and (c) the coalition had been active for at least 6 months.

We selected the Wilder Collaboration Factors Inventory (WCFI) to assess eligible coalitions (Mattessich & Johnson, 2018) because it is a research-tested questionnaire that allows members of a coalition to anonymously evaluate the strengths and areas for improvement within their collaborations. In addition, it has been identified in the National SNAP-Ed Evaluation Framework as an appropriate tool for measuring coalition readiness to engage in community-level initiatives (USDA FNS, 2018). Other reasons we selected the tool are that there is no cost for coalitions to participate in its use and it is publicly available through the Wilder Foundation's online platform (http://wilderresearch.org/tools/cfi/index.php).

The WCFI measures collaboration characteristics called "success factors" in six domains identified as essential for goal achievement: environment, member characteristics, process and structure, communication, purpose, and resources (Mattessich, Murray-Close, & Monsey, 2001). Multiple evaluators have tested the reliability of the WCFI, with Cronbach's alpha scores ranging from .58 to .92 across the factors (Derose, Beatty, & Jackson, 2008; Townsend & Shelley, 2008). Respondents provide their levels of agreement with 40 statements aligned to success factors in the six domains, and Likert scale scores range from 1 to 5 for each success factor. Scores of 4.0 or above reflect coalition strength for the factor, scores from 3.0 to 3.9 suggest the need for possible attention, and scores of 2.9 or lower indicate concerns that should be addressed. Examples of success factors linked to each domain are listed in Table 1.

Table 1.
Wilder Collaboration Factors Inventory Domains and Success Factors

Domain	Description	Example success factor
Environment	The geographic location and social context within which the coalition exists	Group seen as a legitimate leader in the community
Member characteristics	The skills, attitudes, and opinions of members as well as the cultures and capacities of the organizations that form the coalition	Appropriate cross-section of members
Process and structure	The management, decision-making, and operational systems of the coalition	Development of clear roles and policy guidelines
Communication	Channels used by members to send and receive information, keep one another informed, and convey opinions to influence the coalition's actions	Open and frequent communication
Purpose	Reasons for the collaborative effort, the result or vision the coalition seeks, and the tasks or projects the coalition defines as necessary to accomplish	Concrete, attainable objectives
Resources	The financial and human "input" necessary to develop and sustain the group	Sufficient funds, staff, materials, and time

Prior to rollout of the WCFI evaluation of Arizona SNAP-Ed coalitions, we provided a 60-min in-person training to SNAP-Ed agency staff. Our training included the protocols for informing coalition members about the WCFI and instructions for collecting completed questionnaires. Additional information collected by the SNAP-Ed coalition member or our evaluation team included the age of the coalition and organizations participating. The Arizona Department of Health Services Human Subjects Review Board approved the evaluation procedures.

After training, SNAP-Ed agency staff invited members of their eligible coalitions to complete a web- or paper-based WCFI questionnaire during coalition meetings or via email. In either case, mean scores for each of 20 success factors were calculated and a data report was generated by the Wilder Foundation's online platform for all questionnaires that were fully completed. The Wilder Foundation's reports provided descriptive statistics for each coalition, and our evaluation team conducted additional data analysis in the form of independent-samples t-tests using SPSS software to test for significant differences between urban and rural coalitions.

To encourage use of the findings, we provided the Wilder Foundation's report to each coalition after assessment. In addition, we developed a second document that (a) summarized each coalition's mean highest and lowest scoring success factors and (b) offered tailored recommendations for action based on the results.

Results

Ten coalitions in six Arizona counties were assessed in 2016, including four coalitions in primarily rural counties. SNAP-Ed staff reported 222 members across their participating coalitions, with 46% (n = 103) completing questionnaires. Extension staff completed a WCFI in eight of the coalitions. Coalition characteristics are summarized in Table 2.

Table 2.
Coalition Characteristics in Arizona, by County

County	Coalitions assessed (#)	Community type	Members (#)	Completed questionnaires (#)	Participation rate (%)
Coconino	1	Rural	10	10	100
Maricopa	4	Urban	74	43	58.1
Mohave	1	Rural	32	9	28.1
Pima	2	Urban	32	17	53.1
Pinal	1	Rural	34	7	20.6
Yavapai	1	Rural	40	17	42.5
Total	10	Rural 4, urban 2	222	103	46.4

Figure 1 illustrates that across all assessed coalitions, the strongest domains were communication (M = 3.99) and purpose (M = 3.91) and the lowest were resources (M = 3.25) and process and structure (M = 3.67). Rural coalitions (n = 43 respondents) scored higher than their urban counterparts (n = 60 respondents) in all domains, with significantly different means between the two groups in the domains of resources (p = .007) and process and structure (p = .011).

Figure 1.
Mean Coalition Scores by Domain (n = 103 Respondents)

The three highest and three lowest scoring individual success factors across all coalitions are shown in Table 3, along with their corresponding domains.

Table 3.
Highest and Lowest Rated Wilder Collaboration Factors Inventory Success Factors

Success factor	Mean score	Domain
Highest
Members see collaboration in their self-interest	4.3	Member characteristics
Skilled leadership	4.2	Resources
Unique purpose	4.1	Purpose
Lowest
Sufficient funds, staff, materials, and time	2.8	Resources
Appropriate cross-section of members	3.3	Member characteristics
Development of clear roles and policy guidelines	3.4	Process and structure

Discussion

Our evaluation team sought to understand the characteristics of community-level coalitions in which SNAP-Ed staff participated and to assist coalitions in using the findings to strengthen their capacity to realize their goals. Members of coalitions participating in the evaluation reported engaging in robust communication regarding their collective goals and maintaining a collectively shared purpose. These findings echo those of previous research, whereby robust structure, operations, and leadership were reported as essential for coalition function (Barnes, Erwin, & Moonesinghe, 2014).

However, having adequate resources to accomplish their coalitions' goals was scored lower by coalition members. Interestingly, success factors related to the resources domain appeared in both the highest and lowest scoring categories, suggesting that this area of collaborative enterprise is not clear-cut. Coalition members scored their leaders highly as a resource, which may encourage ongoing commitment from members and facilitate new member recruitment. However, the lower score on sufficient funds, staff, materials, and time suggests that some coalitions may have lacked the adequate human and/or financial capital to advance their collective aims.

Feedback from several SNAP-Ed staff who administered WCFIs for the project indicated that the process provided a deeper understanding of the collaboration characteristics that are likely to support their goals. Specifically, receiving the WCFI results provided a formal opportunity for coalition members to review their governance structures and processes. After such review, one coalition implemented a subcommittee-based organizational model in order to allow members to prioritize their coalition responsibilities. In addition, because sufficient resources was a concern identified through the WCFI, the coalition created a dedicated subcommittee to identify grants to support their work. Another coalition used their WCFI results to develop a more specific work plan, with clearer objectives and accountability. The process helped members be more structured in their approach to addressing areas for improvement. More broadly, these examples illustrate how coalitions can use their WCFI results to enhance collaborative practices to support their longer term goals.

Our differential findings between rural and urban coalitions suggest that despite previously reported challenges to collaborative work in rural areas (Moore et al., 2010; Smathers & Lobb, 2017), the rural coalitions we studied may have experienced collaboration advantages as compared to their urban counterparts. In Arizona SNAP-Ed, rural staff have reported that some coalition members represent more than one agency or role in their communities. Members with multiple types of decision-making authority could have a positive effect on coalition function by streamlining processes and communication. Indeed, the development of a collaborative working group in one rural Arizona county increased the number of agencies coordinating to address diabetes, a goal that may be easier to accomplish within smaller community and service networks (Hill et al., 2008).

Our project had several limitations. Only 46% of members across all coalitions completed a WCFI. Because questionnaires were distributed during coalition meetings and via email membership lists, it could be that those who completed a WCFI were more active in their coalitions. Consequently, less involved members were likely underrepresented in the sample, and those not participating may have had different perspectives about their coalitions.

In addition, use of the WCFI limits the scope of our analysis in some ways. The WCFI does not measure certain coalition features, such as which members and/or organizations are more active, nor does it directly address equity issues within a coalition. Finally, the WCFI does not assess whether a coalition succeeds in achieving the outcomes it sets out to accomplish.

Implications

We identified resource constraints as one barrier coalitions face, suggesting that there may be a negative impact on coalitions when collaborative work is underresourced. Discussing the role of funders in coalition-based work, Easterling (2013) noted that "one of the most valuable forms of support is to cover the costs of a network coordinator who can facilitate communication, plan meetings and other networking events, and carry out the follow up steps that come out of those meetings" (p. 70). Although funding is an ongoing challenge for many public health initiatives, the acquisition of dedicated funding for coalition expenditures, including staff time, may benefit community coalitions such as the ones we studied.

The collective impact model described earlier has been used in Extension to strengthen the conceptualization and implementation of community-based work (Parisi et al., 2018; Vines, 2018), including work with coalitions. Collective impact's four phases may be useful to coalitions seeking to strengthen their collaborative work after involvement in a WCFI. These phases are generating ideas and dialogue, initiating action, organizing for impact, and sustaining action and impact (Bradley, Chibber, Cozier, Vander Muelen, & Ayres-Griffin, 2017).

Importantly, both the collective impact model and the WCFI have been revisited in recent years to more explicitly address equity within coalitions and in the communities affected by them (Kania & Kramer, 2015; Mattessich & Johnson, 2018). An equity lens is particularly salient for SNAP-Ed, which requires programming to be centered in communities where the underrepresentation of resident voices has a historic precedent (Wolff et al., 2016).

For Extension professionals across many communities and programs, a notable benefit of the method described here is that the WCFI is a free, relatively low-burden tool for assessing the characteristics and processes within a coalition that support collective goal achievement. To further build on our project, those undertaking future studies could measure WCFI scores over time as an indicator of changing coalition capacity or examine linkages between WFCI results and coalition-level successes and/or failures.

Conclusion

Any program initiative's success in addressing complex community challenges relies in large part on collaborative efforts to organize, execute, and sustain shared goals, usually through partnerships such as coalitions. Shifting an evaluative lens to the coalitions directly involved in such efforts allows for a deeper understanding of the collaboration characteristics that are likely to support the achievement of identified goals. Extension professionals who participate in or lead coalitions, including with programs such as SNAP-Ed, may wish to consider characteristics such as resource allocation and the strengthening of governance processes and structures in order to further accelerate their progress on shared community-level goals.

Acknowledgments

This project reported here was supported by the U.S. Department of Agriculture's Supplemental Nutrition Assistance Program Education (SNAP-Ed) in collaboration with the Arizona Department of Health Services (Agreement No. 16-099661) and the Arizona Department of Economic Security. We also thank the SNAP-Ed local implementing agencies that participated in the evaluation, including University of Arizona Cooperative Extension.

References

Barnes, P. A., Erwin, P. C., & Moonesinghe, R. (2014). Measures of highly functioning health coalitions: Corollaries for an effective public health system. Frontiers in Public Health Services & Systems Research, 3(3), 1–7. https://dx.doi.org/10.2105/AJPH.2014.10412e43.1

Bradley, K., Chibber, K. S., Cozier, N., Vander Muelen P., & Ayres-Griffin, C. (2017). Building Healthy Start grantees' capacity to achieve collective impact: Lessons from the field. Maternal and Child Health Journal, 21(Suppl 1), S32–S39.

Butel, J. A., Banna, J. C., Novotny, R., Franck, K. L., Parker, S. P., & Stephenson, L. (2018). Validation of a collaboration readiness assessment tool for use by Supplemental Nutrition Assistance Program–Education (SNAP-Ed) agencies and partners. Journal of Nutrition Education and Behavior, 50(5), 501–505. doi:10.1016/j.jneb.2017.11.002

Clark, N. M., Lachance, L., Doctor, L. J., Gilmore, L., Kelly, C., Krieger, J., . . . Wilkin, M. (2010). Policy and system change and community coalitions: Outcomes from Allies Against Asthma. American Journal of Public Health, 100(5), 904–912. doi:10.1177/1090198114547507

Derose, K. P., Beatty, A., & Jackson, C. A. (2008). Evaluation of Community Voices Miami: Affecting health policy for the uninsured (Report No. TR-177-CCPP). Santa Monica, CA: Rand Corporation. Retrieved from the Rand Corporation: https://www.rand.org/pubs/technical_reports/TR177.html

Easterling, D. (2013). Getting to collective impact: How funders can contribute over the life course of the work. The Foundation Review, 5(2). doi:10.9707/1944-5660.1157

Gillam, R. J., Counts, J. M., & Garstka T. A. (2016). Collective impact facilitators: How contextual and procedural factors influence collaboration. Community Development, 47(2), 209–224. doi:10.1080/15575330.2015.1133684

Hersey, J., Kelly, B., Roussel, A., Curtis, L., Horne, J., Williams-Piehota, P., . . . Farris, R. (2012). The value of partnerships in state obesity prevention and control programs. Health Promotion Practice, 13(2), 222–229. doi:10.1177/1524839910383945

Hill, A., de Zapien, J. G., Stewart, R., Whitmer, E., Caruso, Y., Dodge, L., . . . Staten, L. (2008). Building a successful community coalition–university partnership at the Arizona–Sonora border. Progress in Community Health Partnerships: Research, Education, and Action, 2(3), 245–250. http://doi.org/10.1353/cpr.0.0029.

Kania, J., & Kramer, M. (2015, October). The equity imperative in collective impact. Stanford Social Innovation Review, 13(4). Retrieved from https://ssir.org/articles/entry/the_equity_imperative_in_collective_impact

Mattessich, P., & Johnson, K. M. (2018). Collaboration: What makes it work (3rd ed.). St. Paul, MN: Fieldstone Alliance.

Mattessich, P., Murray-Close, M., & Monsey, B. (2001). Collaboration: What makes it work (2nd ed.). St. Paul, MN: Fieldstone Alliance.

Moore, J. B., Jilcott, S. B., Shores, K. A., Evenson, K. R., Brownson, R. C., & Novick, L. F. (2010). A qualitative examination of perceived barriers and facilitators of physical activity for urban and rural youth. Health Education Research, 25(2), 355–367. doi:10.1093/her/cyq004

Parisi, M. A., Northcutt, J. K., McKendry, J. E., Sherrill, W. W., Dye, C. J., & Snow, J. Z. (2018). Extension: The backbone organization in statewide population health management. Journal of Extension, 56(2), Article 2IAW1. Available at: https://joe.org/joe/2018april/iw1.php

Smathers, C. A., & Lobb, J. M. (2015). Extension professionals and community coalitions: Professional development opportunities related to leadership and policy, systems, and environment change. Journal of Extension, 53(6), Article 6FEA1. Available at: https://joe.org/joe/2015december/a1.php

Smathers, C. A., & Lobb, J. M. (2017). Mobilizing rural communities to prevent obesity: A tool kit. Journal of Extension, 55(6), Article 6TOT1. Available at: https://joe.org/joe/2017december/tt1.php

Townsend, A., & Shelley, K. (2008). Validating an instrument for assessing workforce collaboration. Community College Journal of Research and Practice, 32(8), 101–112. doi:10.1080/10668920701707813

U.S. Department of Agriculture Food and Nutrition Service. (2016). SNAP-Ed evaluation framework: Nutrition, physical activity, and obesity prevention indicators. Retrieved from https://snaped.fns.usda.gov/snap/EvaluationFramework/SNAP-EdEvaluationFramework.pdf

U.S. Department of Agriculture Food and Nutrition Service. (2018). FY2019 SNAP-Ed plan guidance. Retrieved from https://snaped.fns.usda.gov/snap/Guidance/FY2019SNAPEdPlanGuidanceFULL.pdf

Vines, K. A. (2018). Exploration of engaged practice in Cooperative Extension and implications for higher education. Journal of Extension, 56(4), Article 4FEA1. Available at: https://joe.org/joe/2018august/a1.php

Wolff, T., Minkler, M., Wolfe, S. M., Berkowitz, B., Bowen, L., Dunn, F., . . . Lee, K. S. (2016, Winter). Collaborating for equity and justice: Moving beyond collective impact. Nonprofit Quarterly, 23(4), 42–53.

Zakocs, R. C., & Edwards, E. M. (2006). What explains community coalition effectiveness? A review of the literature. American Journal of Preventive Medicine, 30(4), 351–361. doi:10.1016/j.amepre.2005.12.004