When Commitment Is Not Enough : How Stress and Individual-Organization Interface Affect Activists ’ Persistence

In light of collective action and community development research, this study aims at testing a model of activist persistence that takes into account both individual and organizational levels. The proposed model predicted that commitment to a group/organization or its cause does affect an activists’ persistence. This relationship is mediated by two variables, namely the individual-organization interface and stress management processes. The model was empirically tested through a path analysis on a sample of 278 (N = 278; 43.9% female) participants recruited among active members in a variety of community groups/organizations. The results supported the pattern described by the model, showing that commitment is a precursor to activists’ persistence. However its direct impact is weaker than the impact exerted by stress levels and the fit between the individual and the group/organization. Applications for community development practice are discussed.


Introduction
There is a considerable amount of literature on collective action that has investigated the psychosocial mechanisms that lead individuals to join a variety of community groups, such as advocacy and social action groups, protest movements, and community service groups.Yet, far less is known about the factors that sustain such engagement over time, resulting in the phenomenon of activist retention (or activist persistence).The social movement literature has highlighted the role played by individual factors (e.g., changes due to life cycle; McAdam, 1988), interpersonal variables (social networks, Diani, 2005; collective identity, Klapp, 1969;Owens & Aronson, 2000;Johnston, Larana, & Gusfield, 1994;Gamson, 1992; commitment to the group, Klandermans, 1997), and organizational characteristics (e.g., level of centralization, routes of communication, relative influence of individuals on the organization, amount expected of members, see among others Snow, Zurcher, & Ekland-Olson, 1980).A community psychology perspective, such as the one shared by Kagan et al. (Kagan, Castile, & Stewart, 2005;Kagan, 2006Kagan, , 2007)), highlights that there is a potential of stress embedded in the active participation of citizens, which sooner or later is likely to make individuals quit.In a similar vein, Cox (2009) elaborated on the notion of emotional sustainability.
Despite the identification of a pool of variables affecting the persistence of individuals' civic and political engagement, to the best of our knowledge, no explanatory models have been proposed to account for the relationships between the above mentioned variables or to account for their influence on activists' persistence.Inspired by the findings of a previous qualitative study (Mannarini & Fedi, under review), we aimed to elaborate and test a model of activists' retention that takes into account the individual and the organizational lev-els.Indeed, the results of our qualitative investigation of a group of citizens involved in protest movements showed that, if on the one hand, engagement was underpinned by personal commitment and satisfaction for the organization's role structure, it was eroded on the other by the stress and strain of a long-term engagement.This is especially the case when such a stress was not compensated by supportive relationships among fellow members.Hence, we developed a theoretical model, according to which commitment to the group or its cause does affect activists' persistence.This relationship is mediated by two variables, namely the individualorganization interface and stress management processes.In detail, the current study focused on the following variables: personal commitment to the group/organization (Meyer & Allen, 1991); stress appraisal and coping strategies (Lazarus & Folkman, 1984), which can be regarded as a proxy for the emotional sustainability of engagement; social support (Cobb, 1976), and the individual-group/organization interface, intended as a combination of member-to-member relationships, role satisfaction, and the subjective evaluation of organizational functioning.

Affective, Continuance and Normative Commitment
Although commitment has a behavioral side (as behavioral persistence), researchers' attention has mainly been drawn to two factors: 1) the psychological state that characterizes members' relationships with their group/organization and 2) the consequences of their decisions to stay or leave.Following Meyer and Allen (1991) and Klandermans (1997), three types of commitment can be distinguished that are related to desire, need and obligation to maintain involvement, respectively.Affective commitment is the "partisan, affective attachment to the goals and values, and to the organization for its own sake, apart from its purely instrumental work" (Bu-chanan, 1974: p. 533).High levels of affective commitment make people feel that they want to continue being involved in the group/organization.Continuance commitment refers to the perceived costs associated with leaving the organization.The deeper a person's involvement is in the organization, the less visible and attractive are the alternatives to staying with the organization and the stronger is their commitment.An individual whose primary link to the group/organization is based on continuance commitment experiences a need to stay on.Normative commitment reflects a feeling of obligation to maintain engagement in a group/organization (Wiener, 1982).It is the result of both long-term socialization processes and of the internalization of normative pressures.Individuals with a high level of normative commitment continue their involvement because they feel they ought to.For each kind of commitment, Meyer and Allen (1991) identified antecedents and consequences related to both personal and organizational variables as well as to past experiences.However, Klandermans (1997) has defined commitment itself as both antecedent and consequent to ongoing participation, arguing that the more committed to a group a person is, the more likely it is that he/she will continue to participate, and the longer someone participates, the more committed he or she will become.

Stress Appraisal and Coping Strategies
Community psychologists have highlighted that intensely committed participation, such as that displayed by community activists, can be overloading and exhausting and therefore result in burnout and disruptive relations (Kagan, Castile, & Stewart, 2005;Kagan, 2006Kagan, , 2007)).They have argued that participation is not only time and energy consuming, but it is psychologically demanding and requires both internal and external resources.For these reasons, while civic or political engagement can be a source of gratification for engaged individuals, the risk of dropping out is real.Cox (2009) has proposed the concept of emotional sustainability to refer to the resources people can use to cope with the stress and strain experienced in their civic or political engagement.Some examples of these resources include a strong religious culture, class or political ethics, a supportive group culture, and emotional management skills (Cox, 2009;Nepstad, 2004).In a similar vein, Downton andWehr (1991, 1998) have pointed out that coping strategies are typical of persistent activists who have the ability to address issues that can disrupt their own participation.According to a cognitive approach to stress and coping (Lazarus & Folkman, 1984), the level of appraisal determines stress levels and coping strategies in which individuals partake to reduce the negative impacts of stress on their wellbeing.A primary appraisal is made when the individuals make an evaluation of the matter at hand.Then, a secondary appraisal occurs when they try to manage the event or the situation that they have perceived as a threat or harm by evaluating the personal and environmental coping resources to which they can turn.The most common typology of coping styles (Lazarus & Folkman 1984) includes problem-focused coping (such as information seeking and problem solving) and emotion-focused coping (such as expressing emotion and regulating emotions).Further coping styles that have been distinguished are avoidance coping (Moos & Billings, 1982) and social coping (Greenglass, 1993).

Social Support
Social support is utilized by individuals experiencing stress when they draw on their social network resources.The idea that social support is a resource that can serve to protect persons against the adverse impact of a stressful event is at the core of the so-called buffering hypothesis (Cohen & Wills, 1985).This hypothesis predicts that people who have little social support will have negative reactions when they experience high levels of difficult life events.Moreover, this hypothesis predicts that people who have high levels of social support will not have as intense of a negative reaction to difficult life events.However, Cohen and Wills (1985) suggest that the buffering effect of social support may be limited by particular individual responses to stress and peculiar features and processes associated with changing environments.Scholars such as Greenglass (2002) have also suggested that social support can be viewed as a form of proactive coping, and indeed, several scales of coping include items that measure the search for social support (e.g., Coping Strategy Indicator, Amirkhan, 1990;Coping Inventory of Stressful Situation, Endler & Parker, 1994).The role of social support in fostering and sustaining collective action has not been directly addressed.However, indirect indications of its influence on civic and political engagement are shown in the studies that have emphasized how the embeddedness in social networks not only provides a symbolic and material opportunity for mobilization (see Diani, 2005;Mannarini, Roccato, Fedi, & Rovere, 2009) but also contributes to reducing the costs associated with engagement (Benson & Rochon, 2004).With a more explicit argument, Nepstad (2004) suggested that some community groups intentionally provide cognitive and emotional support during the uncertainties of activism by implementing practices (collective rituals, for instance) that reinforce members' commitments.

Individual-Group/Organization Interface
Among the factors that sustain civic and political engagement, the relationships that individuals establish with the group/organization should also be mentioned.In general, we can agree that groups that foster the creation and maintenance of strong ties between group members through interaction have more possibilities to keep individuals participating in their group or to strengthen their commitment to the group as a whole over time (Corrigall-Brown, 2007).Although the engagement in a community group cannot be equated to an ordinary work activity, when involvement persists over a long period of time, similarities among community groups and work organizations increase.Hence, we can assume that as in a workplace, role and activity satisfaction, as well as a positive evaluation of the organizational structure and processes, result in positive feelings about one's situation (Robbins & Judge, 2007).Such a condition should reasonably make individuals more willing to stay in their group/organization.

Study Rationale
The study was driven both by the need to advance knowledge and the need to draw applicative indications for use in the field of community development.Our investigation was aimed at exploring the predictive influence of individual and organizational variables on activists' persistence in community groups.We first elaborated and then tested a pattern of relationships between the main psychosocial variables that the literature identified as contributing to activist retention.The point of departure for developing the model was the idea that commitment may not be sufficient per se to make activists stay because of the intervention of more "powerful" process variables, which concern both the specific experience of individuals as members of a group/organization and stress and resource management skills.Hence, we tested a model according to which the level of personal commitment to the group/organization has a direct influence on activists' persistence.At the same time, there is a second path through which commitment affects persistence, which includes the mediating role of stress management processes and the individual-organization interface.The individual-organization interface is defined as an integrated measure of member-to-member relationships, role satisfaction, and the subjective evaluation of the organizational functioning.
Specifically, we hypothesized the following: H 1 Higher levels of commitment would show a positive direct impact on persistence but would be weaker than the influence exerted on the dependent variable by stress management processes and the individual-organization interface.
H 2 Higher stress levels would reduce the probability of persistence, whereas a positive individual organization interface would increase the probability of being engaged.
H 3 Higher levels of commitment would reduce the levels of perceived stress and lead to a more positive evaluation of member-to-member relationships and organizational functioning as well as to a higher role satisfaction.
H 4 Coping strategies and social support from the group/organization members would help individuals to manage stressful events.

Method Participants
Participants (N = 278; 43.9% female) were recruited among active members in a variety of community groups/organizations.The average age was 40.48 years old (S.D. = 14.62).The majority of participants were high school graduates (48.5%), followed by college graduates (34.3%).As for the types of groups/organizations in which participants were involved, 28.8% were active in national political movements, 21.9% in environmentalist groups, 15.5% in civic organizations, 15.8% in community service groups, 12.2% in local protest movements, and 5.8% in cultural associations.

Procedures
Participants were contacted either via email (33.4%) or via the group/organization to which they were committed (66.6%) and asked to take part in a survey on civic engagement behaviors.The former were asked to fill out the online version of a questionnaire, whereas the latter were asked to complete a paper version of the same questionnaire.The questionnaire took about 20 minutes to complete.

Measures
The data were gathered by means of a self-report question-naire including the following measures.
To measure the strength of organizational commitment, an adapted version of the Organizational Commitment Questionnaire (OCQ) by Allen and Meyer (1990) was used.The scale included three components: affective, normative and continuance commitment.Sample items of the three components were "I would be very happy to spend the rest of my career with this organization", "I do not believe that a person must always be loyal to his or her organization", and "I feel that I have too few options to consider leaving this organization", respectively.All 24 items of the scale were rated on a scale of 1 to 7 (1 = very little, 7 = very much).
Stress was measured by an adapted version of the Perceived Stress Questionnaire (PSQ) by Levenstein, Prantera, Varvo, et al. (1993).We excluded the items that described a feeling or an emotional state without reference to a specific domain/ situation/environment and kept those items that were phrased so as to include either a reference to the organizational demands (e.g., "You feel that too many demand are being made on you") or to relationships with co-members (e.g., "You are under pressure from other people").Items were measured by a 4-point scale (1 = almost never, 4 = usually).
Coping was measured by the Coping Inventory for Stressful Situations (CISS) by Endler and Parker (1994).Items were rated 1 (not at all) to 4 (very much).CISS is a four-factor model of human coping with adversity that differentiates three types of coping: emotion-oriented, task oriented, and avoidant.The avoidant style includes two dimensions: distraction and social diversion.The 5 items of avoidant social coping (i.e., search for social support) were dropped because a separate measure for social support was used.
Social support was measured by an adapted version of the Multidimensional Scale of Perceived Social Support (MSPSS) by Zimet, Dahlem, Zimet, & Farley (1988).Items were adapted so as to refer to the support received by the group/organization or by co-members (e.g., "I get the emotional help and support I need from my group" and "My group is a real source of comfort to me").Items were rated on a range from 1 (strongly disagree) to 6 (strongly agree).
To investigate the individual/organization interface (IOI) participants were asked to respond to 9 ad hoc items rated 1 (strongly disagree) to 5 (strongly agree); 3 items were formulated to measure the quality of personal relationships (e.g., "The climate within the group/organization is collaborative"), 3 to assess the organizational functioning (e.g., "The group/ organization has an efficient structure"), and 3 to measure role satisfaction (e.g., "I am satisfied with my role within this group/organization").
Activists' persistence was operationalized as a behavior and measured through two ad hoc items: "How long have you been staying in this group/organization?"and "How many hours per week you usually devote to the activities of your group/ organization?" Finally, participants were asked to provide demographic information (age, gender, education, professional position, and place of residence).
alpha: OCQ = .72;PSQ = .83;CISS = .82;MSPSS = .93;IOI = .83).Correlational analyses of the variables considered (see Table 1) showed that gender was completely uncorrelated to the variables included in the study, while age was directly correlated with social support (r = .15).Education was positively but weakly correlated with stress levels (r = .14),while it was negatively correlated with commitment (r = -.13),social support (r = -.13) and the individual-organization interface (r = -.14).Stress, coping, social support and the individual-organization interface were moderately correlated with each other.The strongest correlation was found among social support and the individual-organization interface (r = .61).As for our dependent variable, persistence was positively correlated with commitment (r = .18),social support (r = .24)and the individual-organization interface (r = .20).
Finally, persistence in the group/organization was negatively affected by the stress perceived (β = -.42) and positively influenced, though moderately, by the individual-organization interface (β = .12).Most remarkably, the direct influence of commitment on persistence was very weak (β = .05),indicating that activist retention is better explained by the other variables included in the model.
To improve the fit of the model, it was necessary to add a few additional constraints, namely the correlations between commitment and coping (r = .64),commitment and social support (r = .66),and social support and persistence (r = .20).These variations did not substantially modify the proposed model, yet the changes suggested a more complex pattern of relationships between some of the variables considered.
The main purpose of this study was to validate a theoretical model of activist retention (see Figure 1).The model predicts that commitment to the group/organization has a direct influence on activists' persistence.However, the model also predicts that there are two variables that exert a greater influence on persistence, namely stress management processes and the individual-organization interface (intended as the result of co-member relationships, role satisfaction and evaluation of the organizational functioning), which mediate the relationship between commitment and the dependent variable.Moreover, as suggested by the literature on stress, coping and social support, the model predicts that both coping strategies and social support affect the stress appraisal process.The theoretical model was tested through a model of path analysis, shown in Figure 2 (all path coefficients are significant at p = .05).Table 2 shows the indices of fit of the empirical model.The results of the path analysis showed that commitment led to a reduction of per-

Discussion
Our findings provided a general, though partial, frame for understanding the psychological processes underlying sustained engagement.All the hypotheses were confirmed, as results showed that: (H 1 ) commitment is an antecedent of activists' persistence, but its direct impact is weaker than the impact exerted by stress management processes and a good fit between the individual and the group/organization; (H 2 ) high stress levels related to daily activities that individuals undertake as members of a community group reduce the probability that individuals keep engaging themselves as activists, whereas a positive individual-organization interface prevents activists from dropping out; (H 3 ) high levels of commitment affect stress appraisal and lead to a more favorable evaluation of member-to-member relationships and organizational functioning as well as to a higher role satisfaction; and (H 4 ) coping strategies and social support from the group/organization help individuals to manage stressful events.

Coping
Social support

Individualorganization interface
These findings enriched and integrated partial evidence that come from the fields of social movement research and community work.Indeed, these results highlighted a pattern of relationships that showed how a set of psychosocial variableshitherto identified as factors that underpin political and civic engagement (i.e., commitment, stress management processes, and individual-organization interface)-were related to each other.In particular, our study supported evidence drawn from community work (Kagan, Castile, & Stewart, 2005;Kagan, 2006,2007) that emphasized the stressful nature of participation, especially in highly committed forms of activism.This evidence emphasizes the pathogenic potential implied in civic and political engagement and, at the same time, brings to the fore the risk that active citizens turn into passive citizens.What tentative conclusions can be drawn from the study?In terms of application, there are three main indications that derive from our work and that can be beneficial for organizers, leaders and social entrepreneurs.First, potential sources of stress could be detected for those embedded in uncooperative or conflictual group relationships.This domain could be partially controlled by promoting interdependence and solidarity among the group/ organization members, thereby making social support (e.g., cognitive or emotional support) available.
Second, in addition to individual coping strategies, collective coping strategies (e.g., collective problem solving, collective breaks, and external support provided to the group) could be supported, so that resources that can be used to cope with problems become accessible to all the group/organization members.Finally, the fit between the individuals and the organization/group could be monitored so that the activity and the role structure meet the needs of members.Such a condition would enhance the positive feelings of the members about their situations, thereby making them more willing to stay and to contribute to the attainment of collective goals.Although we acknowledge that these actions may not be sufficient to prevent active citizens from withdrawing, they can possibly make their engagement more sustainable.
We are aware of the limitations of our findings.We acknowledge that our model included only some of the factors that sustain civic and political engagement and did not consider the influence of any variables external to the group/organization.Indeed, we focused our analysis mainly on the individual and organization levels, leaving in the background the relationships between the group/organization and the community and the relationship between individuals and the larger community.We are also aware that our results, on the one hand, apply to a small sample of active citizens and, on the other hand, do not distinguish among the different types of community groups to which the individuals belong.As a research perspective, we do believe that the proposed model should be tested within homogenous groups of active citizens so as to identify similar or differentiated patterns of activist retention.Moreover, it would be reasonable to test the validity of a circular relationship linking commitment to activist persistence.As suggested by Klandermans (1997), commitment may lead individuals to participate in a group, but their participation is likely to reinforce their commitment to the group, as in a virtuous circle.
In conclusion, our study was a first step toward a systematic comprehension of the factors that sustain the civic and political engagement of community members and prevent them from withdrawing into the private sphere.As community psychologists, it is our opinion that activist retention stands out as a relevant concern both for scholars investigating the dynamics of community and for professionals who work in the field of community development.

Figure 1 .
Figure 1.The explanatory model of activists' persistence.

Figure 2 .
Figure 2. The empirical model of activists' persistence.

Table 2 .
Tests of model fit.