What Do We Want? Examining the Motivating Role of Goals in Social Movement Mobilization

The main purpose of any social movement organization is to achieve the goals of its followers. Little is known, however, about what type of goals disadvantaged group members strive to reach and which of those may motivate them to join a social movement organization. Using a door-to-door survey (N = 351), we investigated themobilizing effects of goals among inhabitants of the North of the Netherlands that are adversely affected by gas-extraction induced earthquakes. We distinguished between collective (e.g., reduce gas extraction) versus individual goals (e.g., financial compensation), and outcome versus means goals (e.g., influence policy-makers). Moreover, we examined how perceptions of shared opinions with other affected citizens versus with people who are not negatively affected by gas extraction motivate the inhabitants to join a movement and attach importance to different goals. Our results indicate the existence of two pathways for potential mobilization: the first one through the perceptions of shared grievances, which can motivate people to join the movement and pursue collective solutions; and a second one through the perceptions of deprivation, which can motivate people to exert influence on power holders by joining a movement. Individual outcome goals were important but did not motivate disadvantaged citizens to join a social movement organization. We discuss the role of goals as a link between individual level and meso level factors for movement mobilization and collective action.

One of the key aspects of social movement organizations is that they allow members of disadvantaged groups to exert influence through a collective endeavor, which cannot be achieved through individual action (Oberschall, 1973). In order to achieve social change, social movements try to attract followers by communicating sets of beliefs, values and goals which are supposed to resonate with the views of potential members (Benford, 1993;Benford & Snow, 2000). However, people often have multiple goals which can potentially motivate them to engage in collective action (Hornsey et al., 2006), and activists and their potential followers may have differing opinions about the same situation (Snow, Rochford, Worden, & Benford, 1986). Thus, successful mobilization requires that movements and their potential members have shared goals and work together to achieve them (e.g., Klandermans, 2014;van Zomeren, Postmes, & Spears, 2008). Little is known, however, about which type of goals disadvantaged individuals want to achieve and which ones may motivate them to act collectively.
In this paper, we start from a very basic but important question: What do people want to achieve by joining a social movement organization? Research on goal attainment (Gollwitzer, 1999) and collective action (Hornsey et al., 2006;Polletta & Jasper, 2001;Tajfel & Turner, 1979) suggests that members of disadvantaged groups can pursue various individual or collective goals that can have different functions (e.g., outcome goals or means goals). Importantly, these goals may arise from different concerns, such as perceptions of shared grievances or a sense of entitlement, and they may not necessarily match those of a social movement organization looking to mobilize this population. Previous research, however, focused either on the goals of social movement organizations (McCammon, 2012;Simmons, 2014;Snow et al., 1986) or on individual level motivations to participate in collective action (e.g., Klandermans, 1997;van Zomeren et al., 2008;Walgrave, van Laer, Verhulst, & Wouters, 2013). In this paper, we bridge the two perspectives and test the mobilizing effect of goals. We do this in the context of gas-extraction induced earthquakes in the northern Dutch province of Groningen (van der Voort & Vanclay, 2015), a slowly deteriorating situation (Postmes et al., 2018) in which the communities living in the vicinity of the gas field have been dealing with the negative consequences of gas extraction for several decades (Dost & Kraaijpoel, 2013). This situation provides a relevant context to test the mobilizing power of goals, because there are no clear and straightforward solutions, nor a powerful social movement organization that could unite the affected population.

Gas Extraction and Earthquakes in the Province of Groningen, the Netherlands
In 1959, the largest natural gas field in Europe was discovered underneath the province of Groningen and the Nederlandse Aardolie Maatschappij (NAM), a shared enterprise between the Dutch state and two big oil companies subsequently started extracting the gas in 1963 (Stäuble & Milius, 1970). The first recorded earthquake occurred in this region in 1986, a region which had not experienced any natural earth tremors before. The intensity and frequency of these earthquakes varies considerably by location, as the Groningen gas field covers around 900 km 2 with 20 different gas extraction sites. Some municipalities, such as Loppersum, have been experiencing more frequent and stronger tremors than others (van der Voort & Vanclay, 2015). Despite growing unrest among the inhabitants of the region, it took over a decade before the first official reports tentatively linked the earthquakes to gas extraction (de Crook, Dost, & Haak, 1998), and the causal connection between the two has only recently been acknowledged (Dost & Kraaijpoel, 2013). While earlier reports predicted only minor tremors resulting from gas extraction (de Crook, Dost, & Haak, 1995), subsequent research suggested a gradual increase in the expected maximum magnitude of the earthquakes to 4.1 degrees on the Richter scale (de Crook, Dost, & Haak, 1998;Dost & Kraaijpoel, 2013;Nederlandse Aardolie Maatschappij, 2013;SodM, 2016). Recent reports suggest that the earthquakes do not only pose substantive financial strain on the local population because of property damages, but they are also associated with increased stress and health problems (Postmes et al., 2018;van der Voort & Vanclay, 2015). Thus, the affected population may be motivated to join a social movement organization in order to seek short term relief and improve their individual situation, nevertheless they may also want to pursue a longterm collective solution that would secure better outcomes for the whole region.
The goals that people want to achieve through collective action are likely to be contingent on the type of grievance they want to address, but also on the social and political context (Jennings & Andersen, 2003;Simmons, 2014).
Whereas suddenly imposed grievances (Walsh, 1981) might evoke swift mobilization driven by anger or outrage, mobilization as a result of enduring disadvantage depends on a more complex set of factors involving multiple actors, organizations, and motivations (e.g., Klandermans, 1997;Simon & Klandermans, 2001;Subašić, Reynolds, & Turner, 2008). The context of our study contains elements of both; after years of smaller tremors, one relatively strong earthquake in 2012 has generated substantial fear and anger in the local community, and has finally led the government and the gas company to acknowledge the severity of the situation (Dost & Kraaijpoel, 2013). Due to continued gas extraction however, the situation has not improved and the community was pulled into a complicated power struggle between local and national government, businesses, and the general public.

Goals and Movement Mobilization
Goals are consciously intended end-states of an action (Locke, 1996;Ryan, 1970) that guide human behavior (Higgins, 1997). Research on goal setting theory has consistently shown that setting difficult goals leads to higher performance for both individuals (Locke & Latham, 1990) and groups (Kleingeld, van Mierlo, & Arends, 2011).
Among participants in collective action, Hornsey et al. (2006) demonstrated that goals reflecting different interand intragroup concerns, such as influencing policy-makers and third parties, building an oppositional movement, and expressing individual values were important predictors of participation in future protests. Activists with experience in protest have however been found to differ from those who have never participated, with the latter facing greater barriers of engaging in protest for the first time (Verhulst & Walgrave, 2009 The Mobilizing Role of Goals 36 (Tajfel & Turner, 1979). In contrast, collective coping strategies may improve the position of the group as a whole, while also (in)directly benefiting the individual (Tajfel & Turner, 1979;van Zomeren, Spears, Fischer, & Leach, 2004;Wright, Taylor, & Moghaddam, 1990). Social identity theory suggests that people are most likely to engage in individual efforts first, and only engage in collective strategies when they cannot improve their disadvantaged position by leaving the group individually (Tajfel & Turner, 1979). Therefore, the goals that could motivate people to join a social movement organization may encompass both individual and collective concerns at the same time.
In the context of earthquakes in the province of Groningen, the affected community might strive for various short and long-term solutions. With the earthquakes causing damages to houses for example, ensuring better financial compensation for damages represents an important goal. Although many people have reported damages, this goal arises mostly out of individual concerns (i.e., damage to one's household), and it represents only a shortterm and temporary improvement of one's individual status. Because the intensity and frequency of the earthquakes seems to be increasing over the years (van der Voort & Vanclay, 2015), individual financial compensation may not represent a long-term solution that changes the situation structurally and benefits all inhabitants of the affected region, including those who do not yet have property damages. One of the proposed long-term solutions that can benefit all inhabitants of the region would be to reduce or stop the gas extraction that is causing the earthquakes (SodM, 2016). In line with previous research, we expect that such a collective goal should be more predictive of intention to join a movement (Smith, Pettigrew, Pippin, & Bialosiewicz, 2012;van Zomeren et al., 2004;Walsh, 1981).
Furthermore, research on goal attainment (Gollwitzer, 1999;Gollwitzer & Sheeran, 2006;Polletta & Jasper, 2001) differentiates between final or outcome goals and means to achieve these outcome goals. In the context of collective action, the activists may seek the support of third parties or policy-makers (i.e., means) to help them achieve better status for their group (i.e., outcome goal). Simon and Klandermans (2001) conceptualized collective action not only as a power struggle between two parties with competing interest, but they also argued that gaining thirdparty support is crucial in this dynamic. Moreover, Hornsey et al. (2006) showed that goals such as influencing third parties and policy-makers play an important role in enduring engagement in collective action.
In the context of gas-extraction induced earthquakes, the inhabitants of Groningen need to gain wider support from Dutch citizens in general, but also the support of the government and policy-makers in order to ensure better financial compensation or reduce gas extraction. Although this can be done individually, for example by writing letters to politicians, exerting influence on the authority and/or third parties is more likely to be achieved through a collective endeavor (Simon & Klandermans, 2001). Thus, our expectation is that endorsing these types of goals should also be predictive of individuals' motivation to join a social movement, because they are crucial for achieving the desired individual and collective outcomes.

Shared Opinion and Severity Perceptions
The literature on environmental and technological hazards (e.g., social contamination, fracking, or nuclear plant accidents) suggests that the affected population responds to collective stress situations by engaging in an assessment of severity and the associated risks (van der Pligt & de Boer, 1991). This process involves the evaluation of available information communicated by the authorities, as well as the comparison between one's individual stance on the issue and the perceptions of relevant others (van der Pligt & de Boer, 1991). Van  tantly, through social comparison, individuals can evaluate whether the other members of their group share their opinion or not.
Previous research on environmental risks has not specified to whom people compare themselves in assessing risk, nor which effects these social comparisons might have on different collective action goals. In line with the work on politicized collective identity (Simon & Klandermans, 2001), two different types of social comparisons are relevant for members of disadvantaged groups. First, one can compare oneself to those who share the disadvantage to assess whether others who are affected see the situation as equally dangerous and severe. Comparing oneself to other members of the disadvantaged group can foster a sense of shared grievances around which a strong and unified group can be formed to fight the disadvantage (e.g., Bliuc, McGarty, Reynolds, & Muntele, 2007;Simon & Klandermans, 2001;van Zomeren & Iyer, 2009). Second, members of the disadvantaged group can also compare themselves to those who are not affected (i.e., third parties). This social comparison is important because it can increase feelings of relative deprivation and entitlement (Runciman, 1966;Smith et al., 2012), especially if the third parties are seen as underestimating the severity of the situation.
However, if an individual perceives the situation as less serious and urgent in contrast to the relevant others, he or she may be less likely to engage in a collective action. System justification theory suggests that individuals sometimes justify the current status quo in order to reduce feelings of uncertainty (e.g., Jost & Hunyady, 2005), which may lead them to underestimate the severity of the situation. Importantly, research on natural hazards and disasters (e.g., floods, natural earthquakes) shows that underestimating risks has grave consequences because people fail to adequately prepare for a potential disaster (Paton, 2003;Paton, Smith, & Johnston, 2000). We expect that people engage in comparisons of risks with both those who share their situation and those who do not. In line with system justification theory and prior work on environmental hazards, both comparisons may reduce people's motivation to engage in actions if they perceive the situation as less severe than others. However, we explore whether these two comparisons could discourage people from joining a movement by changing the importance of different goals. Perceiving the situation as less serious than one's community may signal that collective solutions to the disadvantaged situation are not necessary. In other words, those who underestimate the severity in contrast to their fellow disadvantaged group members may attach less importance to collective outcome goals.
On the other hand, those who underestimate the severity in contrast to unaffected third parties may feel les entitled to be compensated. Thus, they might attach less importance to securing better individual outcomes. Lastly, we expect both comparisons to motivate the endorsement of means to attain these goals.
In sum, in the current research we examine the mobilizing potential of different types of goals for those not previously mobilized by a social movement organization. We do so in the context of gas-extraction induced earthquakes in the province of Groningen, an environmental disaster which has only recently gained recognition by authorities (Dost & Kraaijpoel, 2013), and where the affected population did not engage in collective action on a large scale prior to our data collection. We differentiate between outcome goals that reflect desired end-states (i.e., individual financial compensation or the collective goal of reduced gas extraction) and means to achieve these goals (i.e., informing the wider public and influencing policy-makers). We expect collective goals and means towards goals to have greater motivating power for people to join a social movement organization. Finally, we examine whether and how perceptions of shared opinions with the local community or third parties motivate the endorsement of different goals and the intention to join a movement.

Method
Participants 351 inhabitants of the province of Groningen took part in a larger project on the effects of gas-induced earthquakes (for more details about the sampling procedure see Kutlaca, van Zomeren, & Epstude, 2017). The survey was carried out by six research assistants in nine towns of similar size situated in different parts of the province, in order to reach both people who were strongly and weakly affected by the earthquakes. The research assistants handed out approximately 50 surveys per town, with a maximum of one survey per household. The person who opened the door was asked to fill out the survey, except in the case of a minor younger than 16 i . The data was collected during November 2013, almost one year after the reports explicitly linking gas extraction to the earthquakes were published (Dost & Kraaijpoel, 2013). We excluded four questionnaires from the analyses because they were filled out by two people, and one survey because it arrived long after the data collection was finished. The average age of the participants was 51.17 (SD = 15.11) and 53.3% were women (5.5% of the sample did not fill out the demographic questions).

Measures Movement Membership
First, we checked whether our participants were already members of any of the social movement organizations active at the moment of the data collection (for exact items, please see the Supplementary Materials). The participants could choose between the largest social movement organization -the Groninger Bodem Beweging-and the much smaller and newly established organization Schokkend Groningen, specify another movement, or indicate that they did not belong to any movement. Next, we asked the participants how likely it was that they would join a social movement organization (1-Not at all likely to 5 -Very likely), and to specify which organization they might join.

Movement Goals
The participants were asked to think about how important different goals should be for a social movement organi-

Shared Opinion
In order to investigate how people assessed the severity of the situation, we asked them to compare their views to the views of two relevant groups: their affected community (i.e., ingroup) and the rest of the Dutch society (i.e., van Bezouw & Kutlaca equally severe or 5-They see it as more severe than I do). We also asked participants to indicate how similar the opinion of those not affected by the earthquakes (i.e., the rest of the Netherlands) is to their views. Thus, higher scores on both items indicated that people underestimate or perceive the situation as less severe in contrast to the relevant group.

Socio-Demographic Background
At the end of the survey, participants answered questions about their socio-demographics: age, gender, average monthly household income, highest obtained education level, whether they owned the house they lived in, and whether their home was damaged due to the earthquakes.

Descriptive Analyses
Out of 336 people who responded to the question about membership in a movement, a large majority (93%) indicated they were not a member of any organization. Importantly, the members and non-members did not differ in their perceptions of importance of goals, with the exception that members attached greater importance to the goal of influencing policy-makers than non-members (see Table 1). Those who were already in an organization were more eager to become a member of another organization and they thought that the rest of the country was even less aware of the severity of the earthquake situation. Below we report the analyses on those who were not members of a social movement organization, however the key findings remained the same when we analyzed the full sample (see Supplementary Materials). Non-members were not very motivated to join any organization (M = 2.41, SD = 0.85), with the average score being well below the scale mean point of 3, t(310) = -12.37, p < .001, and a large majority did not specify which movement they would consider joining. Among those who did, the Groninger Bodem Beweging as the largest social movement organization was the most popular (see Table 2). Among non-members, individual financial The Mobilizing Role of Goals 40 compensation was deemed a more important goal than reducing the gas extraction, although these two goals were positively correlated r(299) = .32, p < .001. In line with the findings in Hornsey et al. (2006), goals reflecting intergroup concerns, e.g., informing others and influencing policy-makers, were also highly endorsed. Moreover, these two goals were positively related to both individual and collective outcome goals (see Table 3), suggesting that they represent the relevant means to achieve the desired outcomes. We also checked whether the participants mentioned additional goals that we did not cover. Only 21 (6.7%) of the non-members responded to the openended question, mainly elaborating on whether reducing gas extraction is feasible or not, and how a movement can influence policy-makers or third parties. One participant mentioned investing in alternative green energy as a relevant goal. Thus, overall we are confident that our choice of goals captured the most salient goals at the time of data collection. People slightly underestimated the severity of the situation in comparison to their local community; the average response was significantly different from the scale mean point of 3, t(306) = 10.50, p < .001. Closer inspection of the variable revealed that the participants almost exclusively indicated one of two options: They felt that the others have the same perspective (64.1%) or that the others see the situation as more severe (31.7%). Less than 5% of the sample chose other options. In contrast, our participants perceived the situation as more serious in contrast to the people who are not affected by the earthquakes. The average score differed significantly from the scale midpoint t(307) = -19.90, p < .001. In other words, they felt that the rest of the country seriously underestimated the severity of the situation in the affected region.

Testing the Mobilizing Power of Goals
First, we examined whether perceptions of shared opinions with the community and third parties were related to the importance people attach to different goals (see Table 4). In line with our expectations, we found that perceptions of shared opinions with the community negatively predicted the importance attached to the collective goal of reducing gas extraction, meaning that underestimating the risk vis-à-vis other affected citizens was related to decreased endorsement of the collective goal. The perception of shared opinions with third parties negatively predicted the importance of the financial goal. In other words, underestimation of the severity compared to the unaffected citizens predicted less support for seeking financial compensation. Moreover, the two variables predicted the importance attached to informing third parties and policy-makers. Next, we ran a stepwise regression analysis to test whether different goals and perceptions of shared opinions with the affected community and third party motivate intentions to join a movement (see Table 5). Table 5 Regression Analysis: Goals and Shared Opinion as Predictors of Intention to Join a Movement.
Step 2 b Step The additional analysis controlling for socio-demographic variables can be found in the Supplementary Materials.
First, perceptions of shared opinions with the community and the rest of society predicted lower intentions to join a movement. Second, supporting our assumptions about the importance of collective goals, reducing gas extraction and influencing policy-makers predicted the intention to join a movement. The financial goal, on the other hand, did not contribute to the model, nor did informing third parties. We ran a sensitivity power analysis in G*power The Mobilizing Role of Goals 42 (Faul, Erdfelder, Buchner, & Lang, 2009) to check the effect size our study was able to detect with the following parameters: 6 predictors, alpha level set at .05, power .80, and sample size of 286. The analysis yielded an effect size f 2 = .05 (≈ to R 2 = .05), non-centrality parameter λ = 13.93, and a critical F-value = 2.13.
Finally, we ran a path analysis in Mplus (version 7.4) to examine whether the shared opinions predicted the intention to join a movement both directly and indirectly via two goals (i.e., reducing gas extraction and influencing policymakers). We allowed the two shared opinion variables as well as the two goals to co-vary with each other. The model fitted the data well (see Figure 1): The chi square test for model fit was not significant χ 2 (2) = 2.66, p = .265, and fit indices indicated a good fit CFI = 0.99, RMSEA = .03, 90% CI [.00, .12], SRMR = .03. The indirect effect of shared opinion with the community through the collective goal of reducing/stopping gas extraction was only marginally significant B = -0.03, SE = 0.02, p = .065 (see Table 6). The second indirect effect of shared opinion with the rest of the country through influencing policy-makers was significant B = -0.06, SE = 0.02, p = .001 (see Table 6). We also tested a reverse model, where the shared opinions acted as mediators and goals as predictors. This model also showed good fit χ 2 (2) = 2.04, p = .36, CFI = 1.00, RMSEA = .01, 90% CI [.00, .11], SRMR = .02. However, the initial model fits better with prior theoretical work on social change, which proposes that perceptions of shared grievances are a more distant predictor of engagement in collective action (Simon & Klandermans, 2001;van Zomeren et al., 2008). .065 (0.02) -0.03 SO Third party → Intention to join a movement (via Influence policy-makers) .001 (0.02) -0.06 Figure 1. Results of path analysis. Single-headed arrows refer to significant paths. Double-headed arrows refer to correlations between the variables. *p < .05. **p < .01. ***p < .001.

Discussion
What do people want to achieve by joining a social movement organization? In this study we integrated insights from research on goal attainment (Gollwitzer, 1999), collective action, and movement mobilization (Hornsey et al., 2006;van Zomeren et al., 2004) to examine the motivating role of goals and perceptions of shared opinions in movement mobilization. We found that different outcome-and means to achieve goals stemmed from an individual's assessment of risk in comparison to two relevant groups: their affected community and an unaffected third party. More specifically, those who underestimated the severity of the situation in contrast to their fellow ingroup members attached less importance to achieving a collective solution to the situation (i.e., reducing/stopping gas extraction) and gaining greater societal support. This fits with previous research on risk assessment and protest against technological hazards (van der Pligt & de Boer, 1991). Moreover, those who underestimated the severity of the earthquakes in contrast to those who are not affected felt less entitled to seek better personal outcomes (i.e., improving individual financial compensation) or, interestingly, push for a movement to renegotiate the situation with policy-makers. Furthermore, in line with our hypothesis, people highly endorsed both individual and collective outcome goals, but only the latter was predictive of their intentions to join a social movement organization. Finally, we found that people's underestimation of risk affected their intention to join a movement both directly and indirectly by affecting the importance attached to the collective outcome goal and means goal of influencing policymakers. Taken together, the findings suggest that there are two different routes that could motivate disadvantaged citizens to join a social movement organization. On the one hand, emphasizing the shared grievances with their close community might motivate people to strive for a collective solution, which would benefit all that are affected in the long run. On the other hand, increasing the sense of entitlement by focusing on the neglect and indifference of unaffected others might propel people to join in order to collectively exert influence on power holders.

Theoretical and Practical Implications
This study extends previous research on movement mobilization in three important ways. First, we expand previous findings by examining the goals of those who have not yet been mobilized by a movement. Protest survey data shows that first-time protesters differ from experienced ones (Verhulst & Walgrave, 2009) or those who are occasional protesters (Saunders, Grasso, Olcese, Rainsford, & Rootes, 2012) in terms of the strength of emotions, but also their biographical availability. First-time protesters face bigger obstacles to participation because they need to overcome the barrier of having sufficient time and resources to participate (Verhulst & Walgrave, 2009).
Our findings suggest that potential followers resemble inexperienced protesters to the extent that they are mostly concerned with their individual interests and improving their own situation. In other words, potential followers need to overcome the tendency to seek short-term improvements to their situation, but rather focus their strength on finding a long-term resolution. We doubt that successful mobilization can be achieved if movements would allude too much to these individual interests. Instead, focusing on articulating collective outcomes seems to be a more promising route to motivate those who are affected but not yet mobilized.
Furthermore, we showed how comparisons to relevant others play an important role in intentions to act collectively.
Comparisons with other affected group members provides affected individuals with important information about the presence (or absence) of a shared view of the potentially disastrous situation (van der Pligt & de Boer, 1991) and may foster a sense of urgency to seek a collective solution (Simon & Klandermans, 2001). A strong and unified community is also an important source of emotional and instrumental support to those affected by injustice (van Zomeren et al., 2004). Downplaying the risks or disagreeing with the collective views may help an individual cope with uncertainty (Jost & Hunyady, 2005) or provide a feeling of relative gratification compared to others who are worse off (e.g., Leach, Snider, & Iyer, 2002;Vanneman & Pettigrew, 1972). However, these types of social comparisons may also increase distance to the community, which could have negative social consequences for people and render them particularly vulnerable in case of a next disaster. In contrast, we believe that the comparison with unaffected others facilitates the awareness that one is indeed worse off than those who are unaffected. The realization that one is being deprived of important outcomes may increase feelings of entitlement and the likelihood of participating in collective action (Smith et al., 2012;van Zomeren et al., 2008). Importantly, we contribute to this literature by showing that this perception can motivate people to enter the power struggle (Simon & Klandermans, 2001) by joining a social movement organization, but mainly through the instrumental means goals of exerting influence on policy-makers. Whereas we expected that the means goal of informing others could also be a response to the lack of acknowledgement from third parties, we did not find this relation. This could signal that in the context of gas-extraction induced earthquakes, the lack of acknowledgement from third parties does not stem from a lack of knowledge about the situation, and that concrete outcomes from power holders after years of neglect are deemed more important than trying to convince the general public. In other contexts with less public awareness of certain issues, shared opinions with third parties could, however, spur movement mobilization with the goal of informing others.
Third, our study provides a way to connect the micro level -individual goals, beliefs, and values-with the aims of social movement organizations, i.e., the meso level (Ketelaars, Walgrave, & Wouters, 2014;Snow et al., 1986).
Previous research on movement mobilization assumed that movements can increase their mobilization potential by communicating messages that align with potential members' views (Benford & Snow, 2000;Snow et al., 1986), but failed to specify which psychological mechanism underlies this process (Kutlaca, van Zomeren, & Epstude, 2016;McCammon, 2012;Simmons, 2014 solutions, emphasizing collectively shared grievances, and the sense of entitlement through comparisons with those who are affected versus those who are not affected can be fruitful ways to increase the movement's mobilization potential. Furthermore, our findings may further clarify the puzzling absence of collective action in the province of Groningen that should theoretically be a fruitful ground for a social movement to gain substantial following (van Zomeren et al., 2008). One possible reason for inactivity in Groningen is that the existing movements such as the Groninger Bodem Beweging focused too strongly on securing financial compensation, something that could be achieved individually by claiming damages from the gas company. Moreover, seeking financial compensation would not provide any relief to those whose damages are not only material, but also psychological.
A recent study in the region showed that people reported continued safety and health concerns despite the improved individual financial compensation over the last three years, since the risk of a stronger earthquake with potentially more devastating consequences remains high due to continued gas extraction (Postmes et al., 2018). Thus, social movement organizations should focus on individual and especially short-term concerns only in conjunction with addressing collective needs, otherwise they run the risk of losing out on many more potential followers.

Limitations and Directions for Future Research
An important limitation of this study is that the cross-sectional design cannot capture how the goals develop and possibly change as potential followers pass through different stages of mobilization (Klandermans, 1984). Previous research suggests that social identities can change through intergroup interactions (Drury & Reicher, 2000), and when the content of these identities becomes more politicized, the likelihood of engagement in political action increases (Turner-Zwinkels, van Zomeren, & Postmes, 2015). Similarly, the goals of experienced activists, first time protesters and potential followers can change as people become more politicized. For potential followers who are in the early stages of the mobilization process, the most important thing is to crystalize which collective (and possibly individual) goals should be achieved. Second, when people come to share the same ground and identify strongly with the goals, establishing important means may become more important for movements to keep the motivation high. Lastly, for long-term participation another goal may become the main priority, namely building, sustaining and expanding the movement. Future research should use a longitudinal design and include both activists and potential followers to clarify these processes.
Furthermore, the context of gas-extraction induced earthquakes is interesting for the study of movement mobilization, however its unique characteristics make it difficult to generalize in every respect to other contexts such as climate change activism or situations that are characterized by structural disadvantage (e.g., gender or race inequalities).
Nonetheless, we believe that our differentiation between individual and collective goals on the one hand, and outcome-versus means goals on the other hand may help explain why social movements and their potential followers may not always share a common ground. For example, the movements and their followers may endorse the same outcomes but disagree about the means to achieve them (e.g., Klandermans & Oegema, 1987;Stuart, Thomas, & Donaghue, 2018). This sometimes occurs with less normative feminist movements (e.g., Becker & Barreto, 2014) where many women may not feel comfortable with tactics involving violence or explicit nudity to raise awareness about gender inequality. It would be interesting to investigate when the agreement on either the outcomes or the means is more important for participation in collective action.
To conclude, we believe that both sociological and social psychological work on social change can benefit from looking more closely at what people want to achieve by engaging in a collective struggle. We think that our classification of individual versus collective goals, and means-versus outcome goals may be a good starting point to