An Assessment of the Sustainability and Impact of Community Coalitions once Federal Funding has Expired. Modeling Sustainability as a Function of Coalition Membership, Structure, and Resources

06/01/2012

In light of the many significant bivariate relationships between sustained and not sustained coalitions in terms of their membership, structure, and resources, a logistic regression model was utilized to understand the multivariate relationships between these variables and coalition sustainability. The results of this model are summarized in Exhibit 19 and the full table of regression coefficients can be found in Appendix C. The dependent variable is whether the coalition is sustained or not. Additionally, the predicted probabilities discussed below are included in Exhibit 17.

There are several factors that were significant predicators of sustainability even controlling for the other variables. The greater the experience of the coalition leader in the field and working in the community (controlling for the coalition’s age), the more likely a coalition was to be sustained. For example, holding all other variables at their mean or modal value, the predicted probability of sustainability for a coalition whose leader fell in the minimum experience response category is 48 percent. This probability increased significantly to 84 percent for a coalition whose leader fell in the maximum experience response category.

Exhibit 17: Modeling Sustainability as a Function of Coalition Membership, Structure and Resources
Factors Predicting Sustainability Type of Association
Leader Experience Positive Association
Proportion Active Members Positive Association
High Membership Turnover No Association
Number of Sectors in Membership No Association
Steering Committee Indicator Negative Association
Board of Directors Indicator Positive Association
Executive Committee Indicator No Association
Community Advisory Board Indicator No Association
Have Formal Structure with Members Indicator No Association
Number Funding Sources No Association
Number In-kind support Positive Association
Proportion Funding for Programs and Operations Positive Association

In terms of membership variables, neither a high membership turnover rate nor the diversity of the membership impacted sustainability. However, the proportion of the membership that was considered active did significantly increase the likelihood of sustainability.

For example, holding all other variables at their mean or modal value, the predicted probability of sustainability for a coalition with the minimum proportion of active members, 13 percent, was 48 percent. This probability increased significantly to 87 percent for a coalition with the maximum proportion of active members, 100 percent. There is no association between having formal structure with members (e.g., MOAs or IAAs) and sustainability. Lastly, the joint, or combined, influence of the membership variables (including proportion of active members, diversity of membership, and membership turnover) significantly improves the fit of the model over a model of sustainability that only includes leadership experience (likelihood ratio test with 3 d.f., p<.05).

Two measures of coalition structure had a significant effect on the likelihood of coalition sustainability in the model. Coalitions with a board of directors were more likely to be sustained, although coalitions with a steering committee were less likely to be sustained. As shown in Exhibit 18, the predicted probability of sustainability increased from 61percent for coalitions without a board of directors to 84 percent for coalitions with a board of directors. The predicted probability of sustainability decreased from 84 percent for coalitions with a steering committee to 65 percent for coalitions without a steering committee. There was no association between having an executive committee or community advisory board and sustainability status. Overall, the joint, or combined, influence of the structure variables significantly improved the fit of the model over a model of sustainability that only included leadership experience (likelihood ratio test with 5 d.f., p<.01).

Exhibit 18: Predicted Probabilities of Sustainability

Exhibit 18: Predicted Probabilities of Sustainability

Exhibit 18: Predicted Probabilities of Sustainability
Type Predicted Probability at Minimum Predicted Probability at Maximum
Leadership Experience 48 84
Proportion Active Members 48 87
Steering Committee (0/1) 84 65
Board of Directors (0/1) 61 84
Number of In-Kind Services 71 92
Proportion of Flexible Funding 84 93

Two measures of coalition funding and resources exerted a significant and positive influence on coalition sustainability. First, the more kinds of in-kind support a coalition received, the more likely it was to be sustained. For example, holding all other variables at their mean or modal value, the predicted probability of sustainability for a coalition with the minimum number of in-kind resources was 71 percent. This probability increased significantly to 92 percent for a coalition with the maximum number of in-kind resources. Second, coalitions with higher proportions of their funding that were flexible for either operations or programs had a greater likelihood of being sustained. Holding all other variables at their mean or modal value, the predicted probability of sustainability for a coalition with the minimum proportion of flexible funding was 84 percent. This probability increased significantly to 93 percent for a coalition with the maximum proportion of flexible funding. Overall, the joint, or combined, influence of the funding and resource variables did not significantly improve the fit of the model over a model of sustainability that only included leadership experience.

View full report

Preview
Download

"rpt.pdf" (pdf, 942.2Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®