Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Nov 21.
Published in final edited form as: Am J Community Psychol. 2008 Sep;42(1-2):94–104. doi: 10.1007/s10464-008-9182-z

What Motivates People to Participate More in Community-based Coalitions?

Rebecca Wells 1,✉, Ann J Ward 2, Mark Feinberg 3, Jeffrey A Alexander 4
PMCID: PMC5696652  NIHMSID: NIHMS919230  PMID: 18594964

Abstract

The purpose of this study was to identify potential opportunities for improving member participation in community-based coalitions. We hypothesized that opportunities for influence and process competence would each foster higher levels of individual member participation. We tested these hypotheses in a sample of 818 members within 79 youth-oriented coalitions. Opportunities for influence were measured as members’ perceptions of an inclusive board leadership style and members’ reported committee roles. Coalition process competence was measured through member perceptions of strategic board directedness and meeting effectiveness. Members reported three types of participation within meetings as well as how much time they devoted to coalition business beyond meetings. Generalized linear models accommodated clustering of individuals within coalitions. Opportunities for influence were associated with individuals’ participation both within and beyond meetings. Coalition process competence was not associated with participation. These results suggest that leadership inclusivity rather than process competence may best facilitate member participation.

Keywords: Community-based coalitions, Participation, Inclusion, Empowerment, Shared leadership, Competence

Introduction

Throughout the United States, community-based coalitions have become a prominent mechanism for addressing issues as diverse as heart disease, substance abuse, AIDS, and violence (Alexander et al. 2003; Butterfoss et al. 1996; Butterfoss and Kegler 2002; Kumpfer et al. 1993; Mayer et al. 1998). Community-based coalitions are collaborative organizations whose members represent multiple sectors. Together they address common goals, typically related to health promotion, broadly defined (Butterfoss and Kegler 2002). Coalitions often have ambitious agendas for improving public health, including health behavioral changes and reduced disease burden. They address these goals through outreach and media campaigns and services such as health screening, healthy lifestyle classes, and support groups. Another major function of coalitions is enhancing coordination among existing services provided by member organizations (Fawcett et al. 1997; Francisco et al. 1993; Knoke 1990; Mitchell and Shortell 2000).

Community-based coalitions’ primary asset is their membership (Wandersman et al. 1987), which frequently includes representatives of nonprofits, business, schools, government, and health care, as well as private citizens. Members of coalitions do not cede authority over any of their own operations to a common governing body. One of the central leadership challenges coalitions face is thus to engage and retain such diverse constituents (Alexander et al. 2003; Butterfoss et al. 1996; Goodman et al. 1998). This challenge motivated the current study, which focused specifically on what coalition characteristics were associated with higher levels of individual member participation.

Previous theory has portrayed the motivations for member participation in coalitions in terms of three dimensions (Clark and Wilson 1961; Knoke 1990). The first is interpersonal. Examples include an enhanced sense of group identification (when “you” becomes “we”), status within the group (Clark and Wilson 1961), and enjoyment of leading and organizing (Perlman 1976; Rich 1980). The second type of member motivation is instrumental, relating to private benefits only achievable through participation in the coalition (Knoke 1990). Theoretically, instrumental benefits have monetary value (Clark and Wilson 1961), such as could be ascribed directly to additional external funding, for instance, or indirectly to increased referrals. In practice, however, instrumental goals may also include such vital intangibles as better information about the local community (Prestby et al. 1990) and increased agency legitimacy. Third, members may have normative goals such as population well-being. These public goods are collective and typically mirror the goals of the coalition (Chinman and Wandersman 1999). When members speak in terms of duty, responsibility, and values, they are discussing normative incentives for participation (Clary et al. 1998).

Coalition leaders may potentially influence a range of incentives for member participation, including helping people make new contacts, facilitating agencies’ goal achievement through coalition activities, and demonstrating community impact. In this study, we examine two incentives, each of which is foundational in that it relates to coalition capacity to achieve other member goals. These incentives are the opportunities people experience for influence within the coalition and how competent they perceive coalition processes to be. In terms of individual decisions about how much to participate in coalitions, these two factors might be framed as ‘Can I influence what this coalition does?’ and ‘How capable is this group of achieving those goals?’

What Affects Participation?

As noted previously, this study drew on both the collective action organization and more recent coalition literatures. Collective action organizations are goal-directed, boundary-maintaining activity systems that seek non-market solutions to individual or group problems; maintain formal criteria for membership on a voluntary basis; sometimes employ people as leaders; and provide formal democratic procedures to involve members in policy decisions (Aldrich 1979; Knoke 1990). This family of organizations, which includes community-based coalitions as well as national and international associations, is distinguished from other organizations by having members who are committed to pursue a public good and very few paid participants.

The collective action organizations literature frames member participation in terms of incentives (Prestby and Wandersman 1985; Rogers et al. 1993; Roussos and Fawcett 2000). Previous research has found that coalition members engage according to their opportunities to thereby meet their own agendas (Barkan et al. 1993; Butterfoss et al. 1996; Chinman and Wandersman 1999; Omoto and Snyder 1995).

Knoke (1990) builds on exchange theory (Wilson 2000) to argue that collective action organizations with more inclusive governance structures foster participation by enhancing the return on member time investment. Other frameworks of collaboration have also treated inclusivity as essential to translating member capabilities into coalition capacity, using the language of empowerment and shared leadership. A quarter century ago, authors in this journal noted the heuristic potential of an empowerment perspective on community psychology (Rappaport 1981). More recently, Lasker and Weiss (2003) have argued that individual empowerment is an essential precondition of collaborative problem solving and enhanced community health. Similarly, based on a national study of community health promotion partnerships, Alexander et al. (2003) identified power sharing as essential to fostering collective action.

As these authors put it (p. 168):

In many respects, the collaborative community health partnership operates as a virtual organization. It often lacks a formal legal status; occupies no physical space of its own; relies heavily on financial contributions from partnering organizations; and accomplishes the bulk of its work through the donated time and effort of partnering organization employees, community groups, and concerned citizens. By sharing power to set priorities, allocate resources, and evaluate performance, partnership leaders foster a sense of joint ownership and collective responsibility, from which collateral leadership emerges.

Characterizing Member Participation in Coalitions

There are essentially two ways people contribute their time and energy to coalitions: within coalition meetings and through effort devoted to coalition activities beyond meetings. During regularly scheduled coalition meetings, members decide on their collective mission and strategies, share information among member agencies (often lobbying for their respective agendas), plan interventions (Chinman and Wandersman 1999), and design related materials and tools. These interactions facilitate ties between individuals and agencies within coalitions as well as formalize and refine collective plans (Butterfoss et al. 1996; Kegler et al. 1998).

The second dimension of coalition participation is the effort that takes place outside meetings. Given very limited paid staff, coalitions typically rely heavily on members for such contributions. Between meetings coalition members often recruit new members, draft and distribute meeting agendas and minutes, design and implement needs assessments, and plan, implement and evaluate outreach activities (Butterfoss et al. 1993; Butterfoss and Kegler 2002; Goodman et al. 1998; Granner and Sharpe 2004). Thus, whereas participation in some types of groups might be adequately measured within meetings, for coalitions time outside meetings is also vital.

The current study contributes to the coalition literature in five key respects. First, we explicitly build on previous theory by identifying and testing common predictions from literatures that have not been generally linked, that is, a framework of “collective action organizations” as well as more recent work on coalitions (Knoke 1990; Lasker and Weiss 2003). This offers the opportunity to draw more effectively on all potentially relevant previous work on factors affecting participation in coalitions. Second, we examine potential ways to improve participation in operational terms. The results are practical implications for coalition leaders in terms of actions they can take. Third, because coalitions need multiple forms of member engagement to succeed, we measure participation within meetings in terms of attendance, time spent in meetings, and talking, as well as the time members devote to coalition efforts beyond meetings. Fourth, we draw on data from coalitions that address a range of related youth risky behaviors, including violence, sex, and delinquency, in addition to the substance use which has been the focus of most previous coalitions studied (Zakocs and Edwards 2006). This offers the possibility of extending generality to other health promotion coalitions that are addressing interrelated sets of health behaviors. Finally, we control for individual member attributes that may also affect participation, such as coalition tenure, education, and agency affiliation versus status as a private citizen, as well as demographics such as age, race/ethnicity, and sex (Perkins et al. 1990; Prestby and Wandersman 1985). This improves our ability to isolate the unique effects of factors that coalition leaders can shape.

Opportunities for Influence

One way leaders may share power is by explicitly asking for member input and recognizing people for the contributions they make. Path analysis of survey data from a national sample of health promotion coalition members revealed that an empowering leadership style, including member perceptions that leaders sought and recognized member talents, predicted consensus on coalition vision and in turn greater perceived participation benefits and self reported participation levels (Metzger et al. 2005). Previous evidence also generally suggests that individuals participate more when they receive personal recognition from coalition leaders (Butterfoss et al. 1996; Butterfoss and Kegler 2002; Christensen et al. 1999; Fisher and Ackerman 1998; Zweigenhaft et al. 1996).

Shared leadership may also facilitate member participation by increasing member commitment and opportunities to affect collective goals (Knoke 1990). Previous evidence suggests that opportunities to influence decision making encourage member participation (Wandersman et al. 1987) For instance, Butterfoss et al. (1996) found that opportunities to influence decisions were positively associated with the numbers of hours individual members reported devoting to coalition activities outside meetings. A recent comparative case study found that the more active coalition had a much more inclusive pattern of information seeking than did its less active counterpart (Wells et al. 2007).

Together, these studies support the prediction that:

  • H1 Coalition members will participate more when they perceive more opportunities for influence.

Coalition Process Competence

Another precondition of active member participation in coalitions is arguably members’ perception that coalition processes are sufficiently competent to facilitate goal achievement, a construct we refer to as ‘coalition process competence.’ This may matter at both strategic and tactical levels. At the strategic level, coalition leaders may develop their overarching goals and decision making processes with varying levels of clarity and realism. Such “big picture” direction, if provided effectively, may make these virtual organizations real enough to inspire active member engagement. Tactically, the day-to-day processes through which coalitions pursue strategies may also foster participation. An organization whose members only meet for a few hours a month may be particularly reliant on the efficiency and focus with which that time is used to meet member goals.

Previous theory has tended to assert the importance of process competence in fairly global terms. Knoke (1990, p. 15), for instance, argues that competence is “critical to generating support for collective actions…” In a similar vein, Reininger et al. (1999) argue that coalitions can reduce member frustrations and increase commitment by clearly defining their scope and intended efforts. Lasker and Weiss (2003) posit that collective “synergy,” which they define as the successful combination of knowledge, skills, and other resources, is a necessary precondition of effective collaborative problem solving. They note that this is an inherently collective dynamic, although we further observe that its motivational effects on participation are filtered through members’ individual perceptions. Finally, previous analyses on a subset of the coalitions examined in the current study revealed a significant correlation between board directedness and later sustainability (Gomez et al. 2005).

The modest body of empirical evidence to date about coalition process competence and member participation has been framed in terms of formalization of rules and procedures. Whereas such structure could imply rigidity in bureaucratic contexts, given coalitions’ fluid boundaries, more structure is likely essential to focusing member engagement. An early study reported that block associations’ ‘order and organization’ were significantly correlated with member reports of becoming increasingly involved over time (Giamartino and Wandersman 1983), but a later reanalysis found nonsignificant associations at both the individual and group levels when controlling for the effects of the other level (Florin et al. 1990). Another study also conducted at the organizational level revealed that mean perceived competence was higher in active block associations than in inactive associations, although not significantly so (Prestby and Wandersman 1985). Butterfoss et al. (1996) found that both perceived leader competence and ‘order and organization’ (Moos 1986) were positively associated with the number of hours individuals reported spending outside meetings on coalition activities.

Overall, we may predict based on admittedly mixed prior evidence that:

  • H2 Members will participate more when they perceive greater coalition process competence.

Methods

Sample

Communities That Care is a model for involving community leaders in coordinated strategies to reduce adolescent problem behaviors such as violence, drug and alcohol use, sex, and delinquency, and promote positive youth development (Hawkins et al. 2002). Each community’s leaders form a “prevention board” that undergoes training and then systematically assesses local risk and protective factors related to youth. They are then supposed to prioritize problems, select one or more empirically based prevention programs, and evaluate impact over time. In the United States, a randomized trial funded by four National Institutes of Health and the Center for Substance Abuse and Prevention (CSAP) is currently measuring delinquency, violence, and sexual behavior as well as tobacco, alcohol and other drug use of adolescents in intervention and control communities. Previous work has demonstrated the utility of the Communities That Care model for addressing other problem behaviors, such as bullying (e.g., the Elizabethtown Area Communities That Care). Communities That Care initiatives are currently being implemented throughout New York State and in the Seattle public schools. In the United Kingdom, the Rowntree Foundation currently funds over 30 Communities That Care coalitions. Other initiatives are underway in Australia and the Netherlands.

In Pennsylvania, four state agencies supported implementation of Communities That Care coalitions by pooling funds with federal Title V funds in the mid-1990s. A state steering committee has overseen over $15 million in funding for a total of 115 coalitions throughout the state. Coalition catchment areas have ranged from neighborhoods to counties. External support has included 1 year planning grants, 3 year implementation grants that have underwritten ongoing technical assistance and evaluation, and subsequent continued technical assistance (Feinberg et al. 2004).

Data

The unit of analysis for this study was the individual member. All but one measure (coalition founding dates, from Prevention Research Center records) were from 2004 web questionnaires of members. The web questionnaires were sent to all active members of smaller coalitions and to the most active 25 members of larger coalitions, as identified by coalition leaders. Two and six week reminders were emailed to members, who also had the option of completing pen-and-paper surveys (Feinberg et al. 2008). This procedure was followed for 1,502 individuals in 100 coalitions; 867 within 79 coalitions responded, 818 of whom provided information about their participation. Thus the final response rates were 54% at the individual level and 79% at the coalition level (Feinberg et al. 2008). Researchers at the Penn State Prevention Research Center then produced a report with each coalition’s average score on each scale compared either to the average for other sites or to the coalition’s scores the previous year and a summary of the coalition’s strengths and weaknesses. Technical assistants presented these results on site to the prevention boards, and facilitated discussions about how to build on strengths and address areas of concern.

Item missingness for seven variables exceeded 5% (the highest being 19%, for member age). In addition, although comparable in some other respects, members who were missing information on covariates tended to be less active: the mean response about the percentage of meetings attended was 3.20 on the 1–4 scale for individuals with complete information on all variables, versus 2.95 for omitted cases (t-value −3.08, 816 df, p < 0.01). Multiple imputation in SAS PROC MI reduced the bias due to this pattern of missingness by using all available information for each case to insert plausible values for missing data (Schafer 1997). This also conserved statistical power by retaining all 818 cases in each final regression model. We generated five imputed files. Very low variance in imputation parameter estimates indicated that this number of data sets was sufficient to yield stable estimates of imputed values.

The Institutional Review Board at Penn State approved the data collection process and coalition members signified informed consent by completing the on-line questionnaire.

Measures

Dependent Variables

One measure corresponded to each aspect of member participation, each based on member recollections relative to the prior year: meeting attendance (framed in the survey as 1 = less than 25%, 2 = 25–50%, 3 = 50–75%, 4 = 75– 100%), whether or not the member remembered talking in meetings (1 = yes, 0 = no), the number of hours per month the individual spent in meetings, and the number of hours per month spent on coalition activities beyond meetings. When a member provided the number of hours/month spent in meetings but left time beyond meetings blank we treated the time outside meetings as =0 (running the model without those cases led to the same pattern of results).

Independent Variables

To test hypothesis 1, that opportunities for influence would be associated with member participation, we used one perceptual scale and two members self reports of their roles in the coalition. The perceptual scale was calculated as the mean of responses to three items, concerning how coalition leadership “gives praise and recognition at meetings,” “intentionally seeks out your views,” and “asks you to assist with specific tasks” (called simply “board leadership style” in Feinberg et al. 2008). The Chronbach’s alpha coefficient of 0.80 indicated acceptable reliability. The two self-reports indicated belonging to and chairing committees, respectively (each coded as 1 when true and 0 when not).

Two additional scales were used to test hypothesis 2, that members would participate more when they perceived greater coalition process competence. The first scale addressed the coalition’s board directedness at the strategic level (Feinberg et al. 2008). This was the mean of four perceptual items: “The [coalition] Prevention has … agreed on how it will govern itself, make decisions, and clarify the roles of members; developed clear goals and objectives; identified, and is building upon, individual and community strengths; explored financing and resource development strategies to support new efforts” (α = 0.85) (Feinberg et al. 2008). The second scale used to test hypothesis 2 characterized meeting effectiveness through three items: “There’s a lot of time wasted because of inefficiencies (reverse coded)”; “This is a highly efficient, work-oriented team”; and “Team members work very hard” (α = 0.77).

In addition to the theoretical predictors, we included as controls one coalition-level attribute, the age in years, and several individual level attributes, all from the member survey: member age, sex, race/ethnicity (which the survey had framed in terms of black, Asian, Native American, or Hispanic, with the referent group being non-Hispanic white), coalition tenure in years, whether or not the member had a bachelor’s degree or higher, and whether or not participating as a “concerned citizen” rather than an agency representative. Thus the only coalition level measure was coalition age. The Prevention Research Center did not ask members about their sexual orientation, disability status, or income.

Data Analyses

Although there was fairly high agreement among members about leader style and coalition process competence (mean RWG index indicating within-coalition agreement =0.81 for leader style, 0.78 for board directedness, and 0.71 for perceived meeting effectiveness, on a 0–1 scale (James et al. 1984)) coalition-level factors only explained 1–6% of the variance in study outcomes (Bryk and Raudenbush 1992). We therefore ran the regression models at the individual level, using generalized linear models to accommodate the clustering of individuals within coalitions. The link function for each model reflected the nature of the dependent variable: an ordered logit for meeting attendance (Agresti 2002), regular logit for whether or not the member recalled talking in recent meetings, and identity links for the models predicting the two continuous measures, time in and beyond meetings, respectively. After imputing five data sets, we used SAS PROC MI ANALYZE to combine the results.

Results

Table 1 lists descriptive statistics for study measures. On average, respondents indicated having attended at least 75% of coalition meetings in the previous year (3.07 on a 1–4 scale). Over nine out of ten members (93%) reported having spoken in coalition meetings in the previous year. The mean time spent in meetings per month was 4.15 h and the mean time per month spent outside meetings on coalition activities was 6.23 h. There was much less variation in time spent within meetings (standard deviation = 4.14 h) than on time spent beyond meetings (standard deviation = 13.00 h). The mean perceived level of board directedness was 5.77 on a 1–7 scale. Member appraisals of meeting effectiveness were slightly lower, at 5.47, also on a 1–7 scale. The mean member perception of how encouraging their leaders’ style was 5.73 out of 7. Three quarters (74%) of respondents had served as committee members during the past year and over a third (35%) reported having chaired committees.

Table 1.

Descriptive statistics (Original data, prior to multiple imputation)

Variable Mean STD Range
Meeting attendance (1–4 scale) 3.07 1.15 1–4
Talking in meetings (1 = yes) 0.93 0.26 0–1
Time spent in meetings per month 4.15 h 4.14 h 0–20
Time beyond meetings per month on coalition activities 6.23 h 13.00 h 0–80
Inclusive board leadership style 5.73 1.14 1–7
Committee member 0.74 0.44 0–1
Committee chair 0.35 0.48 0–1
Board directedness 5.77 1.21 1–7
Meeting effectiveness 5.47 1.24 1–7
Coalition age (n = 79) 4.29 years 1.75 years 2.50–8.75
Member age 46.38 years 10.41 years 14–85 years
Male 0.33 0.47 0–1
Hispanic or nonwhite 0.07 0.25 0–1
Tenure in coalition 3.12 years 2.25 years 1.08–6.44
Member formal education bachelors or above 0.84 0.37 0–1
Private citizen 0.11 0.32 0–1
Open in a new tab

The mean coalition age at the beginning of 2004 was 4.29 years, reflecting the relative recency of the Communities That Care rollout from its initial cohort of 21 coalitions to 115. However, this may understate how long some individuals and agencies within coalitions had worked together, given the tendency for community organizations to cooperate under multiple auspices over time.

The average coalition member was 46 years old, female (67% of members), and white (only 7% of members reported race/ethnicity as Hispanic or nonwhite). The mean reported coalition tenure was 3.12 years. The vast majority (84%) had a bachelor’s degree or higher formal education. Only 11% were participating as private citizens rather than representing organizations.

Table 2 shows final model results. There was partial support for the first hypothesis, that coalition members would participate more when they had more opportunities for influence. Members who perceived more inclusive styles of board leadership were significantly more likely to report having attended a higher percentage of meetings in the previous year (OR = 1.361, p < 0.001) and to have spent more time beyond meetings on coalition business (with a 1 point difference on the 7 point scale assessing leaders’ style being associated with 1.725 more reported hours per month spent, p < 0.01).

Table 2.

Generalized linear models of coalition attributes and participation levels

Parameter % meetings attended (Ordinal Talked in meetings (Binary) Time in meetings (Continuous) Time beyond meetings (Continuous)
Odds ratio 95% CI Odds ratio 95% CI Coefficient 95% CI Coefficient 95% CI
Intercept1 0.072*** 0.024 0.210 0.762 0.099 5.851 1.769 −0.373 3.910 0.324 −7.508 8.155
Intercept2 0.173** 0.059 0.503
Intercept3 0.416 0.144 1.205
H1: Greater opportunities for influence will be associated with higher levels of participation
  Inclusive board leadership style 1.361*** 1.162 1.593 1.294 0.923 1.813 0.326 −0.086 0.738 1.725** 0.748 2.702
  Committee member 2.646*** 1.873 3.739 3.661*** 1.904 7.037 1.102** 0.351 1.853 0.331 −1.992 2.653
  Committee chair 2.685*** 1.844 3.909 6.047* 1.185 30.847 1.856*** 1.184 2.527 4.693*** 2.590 6.796
H2: Coalition process competence will be associated with higher levels of participation
  Board directedness 0.966 0.830 1.124 1.009 0.754 1.352 0.001 −0.314 0.315 −0.488 −1.455 0.480
  Meeting effectiveness 1.004 0.849 1.187 0.824 0.572 1.186 −0.023 −0.324 0.277 −0.047 −1.010 0.915
Controls
  Coalition age 0.889* 0.819 0.964 0.842* 0.725 0.977 −0.151 −0.317 0.015 −0.128 −0.631 0.375
  Member age 1.007 0.992 1.023 1.033* 1.004 1.064 0.005 −0.030 0.040 −0.024 −0.141 0.092
  Male 0.825 0.582 1.171 0.851 0.451 1.603 −0.627* −1.244 −0.010 −2.319* −4.307 −0.331
  Hispanic or nonwhite 0.559* 0.327 0.954 1.175 0.400 3.445 1.367* 0.232 2.502 2.618 −0.990 6.227
  Tenure in coalition 1.084 0.997 1.178 1.047 0.868 1.262 0.095 −0.068 0.259 0.183 −0.282 0.648
  Bachelors or above 0.990 0.614 1.594 1.416 0.635 3.162 −1.068 −2.226 0.090 −1.677 −4.707 1.354
  Private citizen 0.671 0.428 1.052 1.149 0.444 2.970 0.730 −0.175 1.635 0.050 −2.909 3.008
Open in a new tab
*

p < 0.05

**

p < 0.01

***

p < 0.001

Both belonging to and chairing committees or other subgroups were also positively associated with members’ reported participation. Members who belonged to coalition committees were more likely to attend a higher percentage of meetings than were non-committee members (OR = 2.646, p < 0.001), to talk in meetings (OR = 3.661), and spend more time in those meetings (an additional hour per month (1.102), all else being equal, p < 0.01). There was no association, however, between committee membership and amount of time on coalition activities beyond meetings. Committee chairs were more likely to report attending a higher percentage of meetings than were non-chairs (OR = 2.685, p < 0.001), were more likely to talk (OR = 6.047, p < 0.05), spent almost two more hours per month in meetings than non-chairs (1.856, p < 0.001), and also devoted almost five more hours per month beyond meetings to coalition business (4.693, p < 0.001).

There was no support for the second hypothesis, that coalition members would participate more when they perceived greater process competence. Neither board directedness nor meeting effectiveness was associated with members’ self-reported participation in or beyond coalition meetings.

There were scattered associations between member attributes included as controls and participation. Members of older coalitions were less likely to report having attended a higher percentage of meetings (OR = 0.889, p < 0.05) or having talked in those meetings (OR = 0.842, p < 0.05). Conversely older members were more likely than younger members to report having talked in meetings (OR = 1.033, p < 0.05). In keeping with previous research on volunteering (Obradovic and Masten 2007), male coalition members reported spending about a half an hour less per month in meetings (−0.627, p < 0.05) and over 2 h less per month outside meetings (−2.319, p < 0.05). Hispanic and nonwhite coalition members were less likely to attend a high percentage of meetings (OR = 0.559, p < 0.05) but also spent over an hour more per month in meetings than did Non-Hispanic white members (1.367, p < 0.05). Individuals’ coalition tenure, possession of a college degree or higher, and status as private citizens versus agency representatives were all unrelated to all forms of coalition participation in this sample.

Discussion

Inferences from any non-experimental study are inherently speculative. Nonetheless, based on our interpretation of results from the current sample, we offer below what we believe are some useful implications for coalition leaders and funders.

Findings from this study are congruent with the intuitive notions that people do more when they believe they can personally influence events and when they feel appreciated for doing so. Both perceptions of leader style and committee roles were associated with higher participation levels within and beyond meetings. The leadership style finding suggests that coalition leaders can foster higher participation by showing a general appreciation for member contributions and by asking people individually for that help. Being on committees may also enable members to build interpersonal ties and learn more about coalitions in the relative safety of smaller groups. This may improve socialization by providing opportunities to ask questions that people would hesitate to ask in larger group contexts, thus supporting more active (and potentially more effective) participation.

From a policy perspective, this study’s central finding raises the issues of how to select leaders who are actively inclusive as well as how to cultivate these skills and attitudes in existing leaders. There is some previous evidence that public health departments are particularly good at practicing “the politics of inclusion” (Fleishman et al. 1992, p. 554; Wells et al. 2004), although they can also be hindered by their governmental context and an attendant rule-bound culture (Kramer et al. 2005). A recent coalition study found that community-based organizational leadership was associated with lower member reports of some aspects of effectiveness, which the authors speculated might be due to the fact that such organizations may not have had sufficiently broad networks (Kramer et al. 2005). Coalition leaders and sponsors might best identify lead agencies in terms of how extensive their networks are relative to the coalition’s mission. For instance, a community-based organization focusing on violence prevention might have better networks for a violence prevention coalition than the public health department. On the other hand, a public health department might be the best lead agency for a coalition emphasizing early disease screening.

Most consultants probably believe that they already train toward an inclusive leadership style, and most coalition leaders undoubtedly share this norm. In a previous study, however, we found that coalition leaders were not always perceived by rank-and-file members to be as inclusive as they perceived themselves to be (Wells et al. 2004). It may therefore be useful to survey all members about how much opportunity they perceive to influence the coalition’s work. Depending on the dynamics within a coalition, this might best be accomplished through a group discussion, small group or one-to-one discussions, and/or an anonymous survey. It will be particularly important to solicit the views of less active members.

Although it is not surprising that people with committee member roles were generally more active than other members, only the people chairing committees spent above-average time beyond meetings on coalition activities. This may in part be an artifact of how active members in this sample were, who reported spending on average almost a day a month outside coalitions meetings on coalition business. When coalitions need more member time investment, however, forming temporary work groups to accomplish specific tasks might be a way to increase the participation of some additional members. Individuals who cannot make multi-year commitments may be willing to chair groups that have such limited time horizons.

Empirically the current study provides useful nuance to the empowerment perspective on coalitions by finding that opportunities for influence rather than process competence may be key to encouraging participation. These findings suggest the importance of distinguishing among facets of leadership. Metzger et al. (2005), for instance, measured coalition member perceptions of leadership through a 14-item scale including items reflecting how actively they included members as well as strategic and tactical process competence. Although this combined scale was associated with participation, the authors may have found differential results across subscales if they had separately measured distinct aspects of leadership behaviors.

At the same time, this study’s findings may have contributed to the identification of commonalities in what a recent review has criticized as a conceptually fragmented literature (Zakocs and Edwards 2006), despite admittedly partial measures of the constructs of interest. Those authors noted that leadership style had been measured in five different ways across empirical studies: incentive management, empowerment, shared leadership, task-focus, and multiple characteristics. Tracing our conceptual foundation back to Knoke’s (1988) framework of collective action organizations, with its basis in exchange theory, through previous coalition research (Prestby et al. 1990), we argue that incentive management is an appropriate overarching construct for understanding why people participate in coalitions. In turn, empowerment and shared leadership are two facets of inclusivity that provide members with incentives to participate actively. Greater task focus is likely to better align coalition activities with member goals, thus enhancing their incentive to participate.

The lack of associations between coalition members’ perceptions of board directedness and meeting effectiveness with their participation does not mean that process competence does not matter. An early model of team effectiveness offers another perspective on the potential role of process competence in fostering coalition effectiveness. Hackman and Morris (1975) posited that group synergies could increase the positive effects of group incentives to participate. However, unlike Lasker and Weiss (2003) and the current study’s second hypothesis, Hackman and Morris suggested that process competence might have a moderating rather than a direct effect on member participation. Such exploration is beyond the scope of the current study but illustrates another potential way that process competence may relate to member participation and coalition effectiveness.

Limitations

This study had some limitations worth noting. Contacting only the most active 25 individuals in larger coalitions yielded a sample that over-represented active members. The 54% response rate also makes it likely that there was substantial self selection bias, with more active members being more likely than others to complete the questionnaire. Previous studies suggest that active members may differ from less active members in both background attributes and perceptions of benefits and costs of participation (Norton et al. 1993; Obradovic and Masten 2007; Perkins et al. 1990; Prestby et al. 1990). Thus, inferences from study findings about how leaders may involve less active members remain speculative until further research tests associations for all coalition members. However, the study sample did include the members of the most concern to leaders, that is, those who have already shown the most interest in contributing to coalition activities. The fact that there was variation in the level of participation in this sample comprised largely of active members makes the study a conservative test of our hypotheses. In other words, we would likely find more variation in a broader sample and potentially greater effect sizes.

We also did not examine what affected whether or not people joined coalitions in the first place. This is a critical issue, given that coalitions are supposed to be grassroots, voluntary organizations that broadly represent their communities but in reality are often comprised primarily of health and social service agency employees who participate as additional duties. Another important issue we did not have the data to address was that of participation costs to members (Chinman and Wandersman 1999). Finally, all the coalitions in the current study sample were in Pennsylvania and most were fairly young. Although these coalitions were located in a range of rural, suburban, and urban locations, it is possible that some dynamics affecting their participation may not generalize nationally or to more mature coalitions.

Conclusion

Despite a growing empirical literature on coalition success factors (Giamartino and Wandersman 1983; Hays et al. 2000; Kegler et al. 1998; Prestby and Wandersman 1985; Rogers et al. 1993), there has been very little evidence about exactly how coalitions can foster greater member participation. The current study has addressed at least part of this gap, indicating that opportunities for influence may affect participation more than how competent leaders are at either strategic or tactical levels. More actively soliciting and rewarding member participation will take time and energy from very busy coalition leaders. The good news is that they may thereby better share the hard work of translating often ambitious public health goals into reality.

Acknowledgments

This research is supported by R03 CA113141-01 National Cancer Institute.

Contributor Information

Rebecca Wells, Department of Health Policy and Administration, School of Public Health, University of North Carolina, Campus Box 7411, 1104F McGavran-Greenberg Hall, Chapel Hill, NC 27599-7411, USA.

Ann J. Ward, WHO Iraq Local Area Development, Queen Margaret University College, Edinburgh, Scotland

Mark Feinberg, Prevention Research Center, The Pennsylvania State University, State College, PA, USA.

Jeffrey A. Alexander, Department of Health Management and Policy, School of Public Health, University of Michigan, Ann Arbor, MI, USA

References

  1. Agresti A. Categorical data analysis. 2. New York: Wiley Interscience; 2002. [Google Scholar]
  2. Aldrich HE. Organizations and environments. Englewood Cliffs, NJ: Prentice-Hall; 1979. [Google Scholar]
  3. Alexander JA, Weiner BJ, Metzger M, Shortell SM, Bazzoli GJ, Hasnain-Wynia R, et al. Sustainability of collaborative capacity in community health partnerships: A proposed model and empirical test. Medical Care Research & Review. 2003;60(4):130S–160S. doi: 10.1177/1077558703259069. [DOI] [PubMed] [Google Scholar]
  4. Barkan SE, Cohn SF, Whitaker WH. Commitment across the miles: Ideological and microstructural sources of membership support in a national antihunger organization. Social Problems. 1993;40(3):362–373. [Google Scholar]
  5. Bryk AS, Raudenbush SW. Hierarchical linear models: Applications and data analysis methods. Newbury Park: Sage Publications; 1992. [Google Scholar]
  6. Butterfoss FD, Goodman RM, Wandersman A. Community coalitions for prevention and health promotion. Health Education Research. 1993;8:315–330. doi: 10.1093/her/8.3.315. [DOI] [PubMed] [Google Scholar]
  7. Butterfoss FD, Goodman RM, Wandersman A. Community coalitions for prevention and health promotion: Factors predicting satisfaction, participation, and planning. Health Education Quarterly. 1996;23(1):65–79. doi: 10.1177/109019819602300105. [DOI] [PubMed] [Google Scholar]
  8. Butterfoss FD, Kegler MC. Toward a comprehensive understanding of community coalitions: Moving from practice to theory. In: DiClemente RJ, Crosby RS, Kegler M, editors. Emerging theories in health promotion practices and research. San Francisco: Jossey-Bass; 2002. p. 414. [Google Scholar]
  9. Chinman MJ, Wandersman A. The benefits and costs of volunteering in community organizations: Review and practical implications. Nonprofit and Voluntary Sector Quarterly. 1999;28(1):46–64. [Google Scholar]
  10. Christensen LA, Reininger BM, Richter DL, McKeown RE, Jones A. Aspects of motivation of a volunteer AIDS care team program. AIDS Education and Prevention. 1999;11(5):427. [PubMed] [Google Scholar]
  11. Clark P, Wilson J. Incentive systems: A theory of organizations. Administrative Science Quarterly. 1961;6:129–166. [Google Scholar]
  12. Clary EG, Snyder M, Ridge RD, Copeland J, Stukas AA, Haugen J, et al. Understanding and assessing the motivation of volunteers: A functional approach. Journal of Personality and Social Pscychology. 1998;74:1516–1530. doi: 10.1037//0022-3514.74.6.1516. [DOI] [PubMed] [Google Scholar]
  13. Elizabethtown Area Communities That Care. Retrieved July, 20, 2007, from http://www.etownctc.org/olweus_bully.php.
  14. Fawcett SB, Lewis RK, Paine-Andrews A, Francisco VT, Richter KP, Williams EL, et al. Evaluating community coalitions for prevention of substance abuse: The case of project freedom. Health Education & Behavior. 1997;24:812–828. doi: 10.1177/109019819702400614. [DOI] [PubMed] [Google Scholar]
  15. Feinberg ME, Gomez BJ, Puddy RW, Greenberg MT. Evaluation and community prevention coalitions: Validation of an integrated web-based/technical assistance consultant model. Health Education & Behavior. 2008;35:9–21. doi: 10.1177/1090198105282413. [DOI] [PubMed] [Google Scholar]
  16. Feinberg ME, Greenberg MT, Osgood DW. Readiness, functioning, and perceived effectiveness in community prevention coalitions: A study of Communities That Care. American Journal of Community Psychology. 2004;33(3/4):163–177. doi: 10.1023/b:ajcp.0000027003.75394.2b. [DOI] [PubMed] [Google Scholar]
  17. Fisher RJ, Ackerman D. The effects of recognition and group need on volunteerism: A social norm perspective. Journal of Consumer Research. 1998;25(3):262. [Google Scholar]
  18. Fleishman JA, Mor V, Piette JD, Allen SM. Organizing AIDS service consortia: Lead agency identity and consortium cohesion. Social Service Review. 1992;66(4):547–570. [Google Scholar]
  19. Florin P, Giamartino GA, Kenny DA, Wandersman A. Levels of analysis and effects: Clarifying group influence and climate by separating individual and group effects. Journal of Applied Social Psychology. 1990;11:881–900. [Google Scholar]
  20. Francisco VT, Paine AL, Fawcett SB. A methodology for monitoring and evaluating community health coalitions. Health Education Research: Theory and Practice. 1993;8(3):403–416. doi: 10.1093/her/8.3.403. [DOI] [PubMed] [Google Scholar]
  21. Giamartino GA, Wandersman A. Organizational climate correlates of viable urban block organizations. American Journal of Community Psychology. 1983;11(5):529–541. [Google Scholar]
  22. Gomez BJ, Greenberg MT, Feinberg ME. Sustainability of community coalitions: An evaluation of communities that care. Prevention Science. 2005;6(3):199–202. doi: 10.1007/s11121-005-0003-4. [DOI] [PubMed] [Google Scholar]
  23. Goodman RM, Speers M, McLeroy K, Fawcett S, Kegler M, Parker E, et al. Identifying and defining the dimensions of community capacity to provide a basis for measurement. Health Education & Behavior. 1998;25(3):258–278. doi: 10.1177/109019819802500303. [DOI] [PubMed] [Google Scholar]
  24. Granner ML, Sharpe PA. Evaluating community coalition characteristics and functioning: A summary of measurement tools. Health Education Research. 2004;19(5):514–532. doi: 10.1093/her/cyg056. [DOI] [PubMed] [Google Scholar]
  25. Hackman JR, Morris CG. Group tasks, group interaction process, and group performance effectiveness. In: Blumberg HH, Hare AP, Kent V, Davies M, editors. Small Groups and social interaction. Vol. 1. Chichester, UK: John Wiley and Sons; 1975. [Google Scholar]
  26. Hawkins JD, Catalano RF, Arthur MW. Promoting science-based prevention in communities. Addictive Behaviors. 2002;27(6):951. doi: 10.1016/s0306-4603(02)00298-8. [DOI] [PubMed] [Google Scholar]
  27. Hays CE, Hays SP, DeVille JO, Mulhall PF. Capacity for effectiveness: The relationship between coalition structure and community impact. Evaluation and Program Planning. 2000;23(3):373–379. [Google Scholar]
  28. James LR, Demare RG, Wolf G. Estimating within-group interrater reliability with and without response bias. Journal of Applied Psychology. 1984;69(1):85–98. [Google Scholar]
  29. Kegler MC, Steckler A, McLeroy K, Malek SH. Factors that contribute to effective community health promotion coalitions: A study of 10 project ASSIST coalitions in North Carolina. Health Education & Behavior. 1998;25(3):338–353. doi: 10.1177/109019819802500308. [DOI] [PubMed] [Google Scholar]
  30. Knoke D. Incentives in collective action organizations. American Sociological Association. 1988;53(3):311–329. [Google Scholar]
  31. Knoke D. Organizing for collective action: The political economies of associations. New York: Aldine de Gruyter; 1990. [Google Scholar]
  32. Kramer JS, Philliber S, Brindis CD, Kamin SL, Chadwick AE, Revels ML, et al. Coalition models: Lessons learned from the CDC’s community coalition partnership programs for the prevention of teen pregnancy. Journal of Adolescent Health. 2005;37(3 Suppl):S20–S30. doi: 10.1016/j.jadohealth.2005.05.010. [DOI] [PubMed] [Google Scholar]
  33. Kumpfer KL, Turner C, Hopkins R, Librett J. Leadership and team effectiveness in community coalitions for the prevention of alcohol and other drug abuse. Health Education Research: Theory & Practice. 1993;8(3):359–374. [Google Scholar]
  34. Lasker RD, Weiss ES. Broadening participation in community problem solving: A multidisciplinary model to support collaborative practice and research. Journal of Urban Health. 2003;80(1):14–47. doi: 10.1093/jurban/jtg014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Mayer JP, Soweid R, Dabney S, Brownson C, Goodman RM, Brownson RC. Practices of successful community coalitions: A multiple case study. American Journal of Health Behavior. 1998;22:368–377. [Google Scholar]
  36. Metzger ME, Alexander JA, Weiner BJ. The effects of leadership and governance processes on member participation in community health coalitions. Health Education & Behavior. 2005;32(4):455–473. doi: 10.1177/1090198104271967. [DOI] [PubMed] [Google Scholar]
  37. Mitchell SM, Shortell SM. The governance and management of effective community health partnerships: A typology for research, policy, and practice. Milbank Quarterly. 2000;78(2):241–289. doi: 10.1111/1468-0009.00170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Moos RH. Group environmental scale manual. Palo Alto, CA: Consulting Psychologists Press; 1986. [Google Scholar]
  39. Norton S, Wandersman A, Goldman CR. Perceived costs and benefits of membership in a self-help group: Comparisons of members and nonmembers of the Alliance for the Mentally Ill. Community Mental Health Journal. 1993;29(2):143–160. doi: 10.1007/BF00756340. [DOI] [PubMed] [Google Scholar]
  40. Obradovic J, Masten AA. Developmental antecedents of young adult civic engagement. Applied developmental science. 2007;11(1):2–19. [Google Scholar]
  41. Omoto AM, Snyder M. Sustained helping without obligation: Motivation, longevity of service, and perceived attitude change among AIDS volunteers. Journal of Personality and Social Psychology. 1995;68(4):671. doi: 10.1037//0022-3514.68.4.671. [DOI] [PubMed] [Google Scholar]
  42. Perkins DD, Florin P, Rich RC, Wandersman A, Chavis DM. Participation and the social and physical environment of residential blocks: Crime and community context. American Journal of Community Psychology. 1990;18(1):83–115. [Google Scholar]
  43. Perlman J. Grassrooting the system. Social Policy. 1976;7:4–20. [Google Scholar]
  44. Prestby JE, Wandersman A. An empirical exploration of a framework of organizational viability: Maintaining block organizations. The Journal of Applied Behavioral Science. 1985;21(3):287–305. [Google Scholar]
  45. Prestby JE, Wandersman A, Florin P, Rich R, Chavis D. Benefits, costs, incentive management and participation in voluntary organizations: A means to understanding and promoting empowerment. American Journal of Community Psychology. 1990;18(1):117–149. [Google Scholar]
  46. Rappaport J. In praise of paradox: A social policy of empowerment over prevention. American Journal of Community Psychology. 1981;9(1):1–25. doi: 10.1007/BF00896357. [DOI] [PubMed] [Google Scholar]
  47. Reininger B, Dinh-Zarr T, Sinicrope PS, Martin DW. Dimensions of participation and leadership: Implications for community-based health promotion for youth. Family and Community Health. 1999;22(2):72. [Google Scholar]
  48. Rich R. Dynamics of leadership in neighborhood organizations. Social Science Quarterly. 1980;60:570–587. [Google Scholar]
  49. Rogers T, Howard-Pitney B, Feighery EC, Altman DG, Endres JM, Roeseler AG. Characteristics and participant perceptions of tobacco control coalitions in California. Health Education Research: Theory & Practice. 1993;8(3):345–357. [Google Scholar]
  50. Roussos ST, Fawcett SB. A review of collaborative partnerships as a strategy for improving community health. Annual Review of Public Health. 2000;21:369–402. doi: 10.1146/annurev.publhealth.21.1.369. [DOI] [PubMed] [Google Scholar]
  51. Schafer JL. Analysis of incomplete multivariate data. New York: CRC Press; 1997. [Google Scholar]
  52. Wandersman A, Florin P, Friedmann R, Meier R. Who participates, who does not, and why? An analysis of voluntary neighborhood organizations in the United States and Israel. Sociological Forum. 1987;2(3):534–555. [Google Scholar]
  53. Wells R, Ford E, Holt M, McClure J, Ward A. Tracing the evolution of pluralism in two community-based coalitions. Health Care Management Review. 2004;29(4):329–343. doi: 10.1097/00004010-200410000-00009. [DOI] [PubMed] [Google Scholar]
  54. Wells R, Ford E, Holt M, McClure J, Ward A. Community-based coalitions’ capacity for sustainable action: The role of relationships. Health Education & Behavior. 2007;34:124–139. doi: 10.1177/1090198105277851. [DOI] [PubMed] [Google Scholar]
  55. Wilson J. Volunteering. Annual Review of Sociology. 2000;26:215–240. [Google Scholar]
  56. Zakocs RC, Edwards EM. What explains community coalition effectiveness? A review of the literature. American Journal of Preventive Medicine. 2006;30(4):351–361. doi: 10.1016/j.amepre.2005.12.004. [DOI] [PubMed] [Google Scholar]
  57. Zweigenhaft RL, Armstrong J, Quintis F, Riddick A. The motivations and effectiveness of hospital volunteers. The Journal of Social Psychology. 1996;136(1):25. doi: 10.1080/00224545.1996.9923026. [DOI] [PubMed] [Google Scholar]

RESOURCES