Psychological Aspects of Civic Engagement during the Legislative Crisis in Israel ()
1. Introduction
Civic engagement is a term describing active initiation, participation or active support of behaviors associated with community life, social issues and political dynamics (Ekman & Amna, 2012) . Civic engagement in times of socio-political unrest and crisis is a major force shaping the evolution and at times-revolutions societies and nations go through (Bull, 1982; Danopoulos, 2019) . Civic engagement is widely acknowledged as a leading force leading social and political change in various settings, cultures and nations throughout history: Civic activism has been associated with pivotal events from the establishment of the USA (Skocpol, 1997) , Ghandi’s popular march facilitating India’s independence from the UK (Zachariah, 2011) to the popular peaceful uprising of Solidarity in Poland (Biela, 2013) and others (Phan & Kloos, 2023) . Such examples of civic involvement and activism at its best demonstrate how potentially powerful and influential the phenomenon can be in shaping societies in times of crisis. Understanding factors underlying civic engagement, especially in times of socio-political crisis, is therefore important for those wishing to better understand, predict and shape this behavior pattern. The present study focuses on a specific case of a society in crisis – Israel during the legislative reform of 2023.
Researchers have examined various types of factors associated with civic engagement in various populations: Social settings seem to be among the strongest factors associated with these behavior patterns, especially norms and expectations in the peer group (McIntosh & Muñoz, 2009; Rossi et al., 2016) , Community cohesiveness and collective sense of efficacy (Collins et al., 2014; McIntosh & Munoz, 2009) . Social capital, especially world knowledge associated with the target subject and social issues, but also interpersonal skills, and education were also often associated with the probability of individuals and groups taking action on community matters (Apaliyah et al., 2012) . Positive, empowering leadership has also been named as a major factor predicting civic engagement and involvement (Wheeler & Edlebeck, 2006) .
Somewhat less explored is the personal level of civic engagement. What personal factors and characteristics shape our tendency to take action on socio-political issues? Bekkers (Bekkers, 2005) showed that social activism was associated with human and social capital as well as political interest and empathic concern for others as a personal disposition. Within the concept of human capital, education, and area of living were most frequently and strongly associated with civic involvement (Ajaps & Obiagu, 2021) . Within the concept of social capital—social connections (including online), social skills and activities were associated with the tendency to be politically active (Ferrucci et al., 2020) . A few studies showed that openness and extraversion are consistently associated with civic engagement (e.g.: Omoto et al., 2010 ). Beyond personality traits, certain typical behavior patterns were also associated with civic engagement, mainly social media usage (Kim et al., 2013) . The Values-Belief-Norm model, based on the theory of planned behavior tries to propose a comprehensive model integrating social, psychological and context factors to predict civic engagement behavior (Skarmeas et al., 2020) : While values and beliefs represent personal level valence of issues at hand, the perceived norm is the individual’s representation of social, cultural and political factors within their environment—these 3 factors result in accounting for civic engagement. Studies as the ones quoted above represent individual-level factors that often represent individual pre-disposition and resources. But what dynamics connect these factors to actual behavior in times of socio-political upheaval?
Emotional and cognitive responses, often associated with both the Values-Belief-Norm and models of attitudes have long been considered ‘behavioral navigators’ linking pre-dispositions (such as personality traits, or attitudes) and actual behavior or responses (Macafee, 2013; Teper et al., 2015) . In the cases of crisis—situational anxiety is the most immediate emotional experience that many claim—drives behavior (Al Majali, 2020) . Examples of these dynamics can be gleaned from studies on recent pandemics worldwide and the roles stress and anxiety played in driving various behavior patterns, for example: leading individuals to take a more proactive role in their societies (Kang et al., 2018) . What determines how much anxiety we experience and how it may channel our responses? Authors suggest that coping resources play a pivotal role in these dynamics. We all have at our disposal a set of potentials allowing us to mobilize resources in times of crisis, and these potentials will shape our response – its magnitude and direction (for example: fight or flee?). Two leading resources often associated with better outcomes in times of crisis are resilience, and optimism.
Resilience is a concept pertaining to the process and outcome of successfully accommodating and adapting to crisis or change, by means of flexibility and mobilization of resources. This process often does just result in adaptation or “survival” but also in improved self-efficacy, and flourishing (APA, 2023) . A large body of evidence supports the associations between resilience and various positive results 9or low levels of negative ones) for people and communities coping with crisis (e.g.: Steen & Morsut, 2020 ). Optimism is somewhat a different resource, but no less important: It is defined as a personality trait that is associated with the tendency to expect positive outcomes and consequences even in dire circumstances (Carver et al., 2010) . Optimism is considered an effective coping resource for its ability to motivate perceptions and actions motivated by the expectation of doing well or reaching beneficial outcomes.
Based on all the above a multi-tiered model of individual-level factors and processes underlying civic engagement was suggested in which demographic variables associated with socio-emotional outcomes (gender, age, education, and income) political attitudes and attitudes toward the legal reforms, alongside personal coping resources (resilience and optimism) served as exogenous factors of civic engagement, mediated by two types of responses to the reforms. The model is depicted in Figure 1 herein.
Demographics. Certain demographic variables are associated in the literature with civic involvement and activism, especially education, income and to a lesser extent age (Barrett & Pachi, 2019; Mahatmya & Lohman, 2012; Oliver, 2000) . Gender has often been associated with differences in emotional responses to life
![]()
Figure 1. The conceptual model reflects two potential theoretical models of civic involvement.
events (e.g.: Boussalis et al., 2021 ). In existing studies women tend to express emotional responses in a more pronounced manner, especially negative responses (e.g.: Moran et al., 2013 ). It was therefore deemed relevant to the study model.
General and specific political attitudes. Since attitudes reflect a relatively stable reaction pattern toward an object, it is only logical that political attitudes related to the reform will predict the emotional reactions, perceptions, and behavioral intents of individuals (Weiss, 2020) . As attitudes change their effectiveness in shaping behavior by their specificity, they were assessed as both a general political orientation (location on a left-to-right semantic differential) and a specific attitude regarding the legal reforms.
Resilience. Resilience is a relatively recent psychological concept depicting the resources and processes by which individuals and groups mobilize resources to effectively cope and make the most of dealing with adverse conditions. As such it is held by many as a predominant resource in the understanding of how people weather hardships and crises (Schwarz, 2018) . Since resilience is a resource shaped and developing across the lifespan it was defined as one of the exogenous factors in our models (as people “bring it with them” to the crisis situation).
Optimism. Similarly, we treated the propensity to view life situations as ones that are likely to end up well, or yield positive results, as a resource shaped over time that helps individuals and groups withstand crisis and challenges for what they are on one hand. On the other hand—expecting them to be beneficial despite the hardship (Ruiz-Rodríguez et al., 2022) . While optimism and resilience are associated of course—they are nonetheless differentiated enough both theoretically and empirically to cover different types of resources associated with coping with crisis.
Situational Anxiety. Viewed as the “first line of response” in numerous contexts stress and anxiety serve as “mental alarm bells” to mobilize resources and set the stage for coping behaviors and reactions aimed at surviving and possibly thriving vis-à-vis a crisis. It is therefore our first mediating factor, representing a preliminary response.
Behavioral responses to crisis. Following the initial response, individuals may choose to engage in active coping with the crisis or distance themselves and avoid the challenge (i.e.: approach vs. avoidance, see: Roth & Cohen, 1986). Responses to the perceived threat or opportunity set by the legal reform could be divided into “positive” (engagement, active coping for example: donate to organizations representing my opinion on the matter, participating in demonstrations, etc.) and “negative” (Avoidance, for example—planning a move to another country). These responses go beyond the semi-automatic patterns of anxiety. These are further steps consisting of planned behavior—thus representing another tier of mediation in our model.
Civil engagement/involvement. The end behavioral outcome in our model is not only the extent to which individuals participate in activities promoting their opinions and beliefs within the socio-political controversy—but also the level of activism represented by the involvement behavior. Based on existing measures of political activism and responses of a local pilot sample a ranked checklist of activities was devised ranging from low involvement (e.g.: following the media outlets regarding this subject) to the highest level of involvement (e.g.: organizing political movements and demonstrations).
While the general model provides an insight into potential pathways from personal characteristics and resources to civic engagement, it is also of interest to look for the specific dynamic underlying the model, namely—the role of anxiety as “an engine for change” in such circumstances.
Two models of emotional reaction’s role in civil engagement
The proposed model emphasizes the role of the basic emotional response, situational anxiety, in driving civic engagement—but how does it do that? Two alternatives may suggest two somewhat different paths within the same overall theoretical model:
1) Civic involvement (activity) as a means of reducing anxiety:
The literature on stress, anxiety and sense of control suggests that one of the most powerful mechanisms for reducing stress and anxiety is taking action, thus regaining a sense of control over the anxiety-inducing situation (Schneider & Leonard, 2022) . If we follow this rationale—civic involvement and participation can serve to reduce anxiety elicited by the threat of political change. Thus, we would expect a negative association between civic involvement and anxiety (e.g.: for an example from the world of environmental conservation see Schwartz et al., ( Schwartz et al., 2023 ).
2) Civic involvement as a response to threat (anxiety):
An alternative view of the association between the two concepts goes the other way around and harkens back to the origins of our understanding of stress and anxiety—as response patterns preparing organisms for action when facing threats or opportunities—back to the “fight or flight” survival pattern (Santos & Van Daalen, 2018) . In this respect, we will expect a positive association between situational anxiety and civic involvement.
Hence, within the proposed model depicted above, two possible dynamics associating emotional responses and civic involvement are tested.
This study and the model presented herein are among the only recent studies examining a comprehensive model, as previous studies usually focused on single or limited numbers of factors. Moreover, it tests a model during, actually, at the peak of a real-life socio-political crisis, rather than at the aftermath of such events—as with most studies in this field.
2. Method
2.1. Settings and Study Design
The study took place during the summer of 2023, in Israel, at the peak of the social upheaval, following the Israeli government’s intention to pass a series of laws limiting the power of the legal system vis-à-vis the government thus raising serious doubts worldwide about the deterioration of the democratic regime in the country (Gidron, 2023) . This resulted in a series of unprecedented demonstrations and acts of civic activism in the Israeli public arena. Within this context, the study questionnaires were disseminated among diverse groups in Israeli society in an attempt to represent a broad range of communities and socio-political agendas.
2.2. Sample
Three hundred and sixty-eight participants were included in our data analysis. Of them 62% were women and the rest were men. The mean age was 49.96 (s.d.: 18.36) with participants’ age ranging 15 to 82. Table 1 summarizes demographic distributions for our sample.
While our sample is slightly older, quite more educated and financially stable than the general population in Israel (based on reports of the Israeli central bureau of Statistics 2022) At least as attitudes are concerned our sample reflects what general surveys reveal regarding public attitudes in Israel toward the reform: While left and right-wing general political attitudes are more or less balanced, current surveys show that up to 45% of the Israeli public are against the reform with about 35% - 39% supporting the reforms and the rest are undecided (Cohen, 2023) . These statistics are echoed in our sample demographics.
2.3. Measures
Demographics were collected using a brief questionnaire containing items referring to the participants’’ gender, age, education, marital status, number of children, income level, area of residence, political orientation (ranging “far left” to “Far right” on a 5 point scale), the party they voted for in the last elections, and the ranking of their support of the current legal reforms, ranging from “fully supportive” to “fully against” on a 5 point grading scale.
![]()
Table 1. Sample demographic variable distribution (n = 368).
Resilience was measured by a questionnaire assessing resource recruitment and usage to obtain resilience (Lucas, et al, 2001). This 14-item questionnaire shows evidence of good internal consistency and reliability at 0.71 for the Hebrew version, as reported by Zohar and Bar David (Zohar & Bar Davi, 2011) .
Optimism was measured by a Hebrew version of Scheier and Carvers’ Optimism questionnaire—brief version (Scheier et al., 1994) . This 9-item version demonstrated adequate reliability ranging 0.67 - 0.80 in Hebrew speaking settings (e.g.: Malka-Zemach, 2019 ).
Situational Anxiety was assessed using the State-Trait-Personality Inventory (STPI, Spielberger, 1996 ) translated into Hebrew by Ben-Zur (Ben-Zur, 1998). This 10-item self-report scale shows good construct validity and internal reliability with coefficients ranging 0.82 - 0.87.
Reactions to the legal reforms were assessed by a questionnaire based on public reactions documented in the public press in the months preceding the study. These were divided into “positive reactions” —which included seeking more involvement and activity regarding the reforms (e.g.: seeking new ways to contribute to the country’s policies and getting more politically active) and “negative reactions” which included withdrawal or avoidant behaviors (e.g.: considering leaving the country, seeking foreign citizenships, etc.). Both subscales were tested for content validity in a pilot sample of 20 Israeli citizens representing the full range of political attitudes. The internal consistency of each subscale was tested in the full sample and was 0.76 for positive and 0.89 for negative items.
Civic involvement/engagement was assessed using a checklist of civic activities ranging from superficial (listening to the news, following events through media) to deep involvement (e.g.: organizing demonstrations, leading a social movement). Items were generated by 10 Israeli citizens from both the right and left political wings who were asked to identify and rank civic involvement activities. Ranks were then amalgamated to create a cumulative scale of increasing engagement or involvement. Validation of the item’s relevance and location along the scale was checked vis-a-vid similar measures of social or political involvement (e.g.: Ha et al., 2013 ). Thus, the higher the grade, the higher the level of engagement/involvement. Due to the nature of the scale, we did not expect to find internal consistency among the items.
2.4. Procedure
The study was approved by the author’s IRB. Questionnaires were disseminated electronically via social networks during a period of 1 month during July 2023, at the height of the internal conflict in Israel around the legal reform. Participants were asked to take the questionnaires and contribute to a study about people’s perceptions and feelings about the legal reform and events surrounding it. Filling out the questionnaires took approximately 10 minutes. Potential participants were assured that there was no obligation to participate and that no identifying data was collected and were asked to forward the link to the questionnaires in their social circles. No rewards were offered for participation. Data collection stopped when response rates dropped below 1 person per day.
3. Results
3.1. Descriptive Statistics
As a first step, descriptive statistics were obtained for our main study variables. These are summarized in Table 2 herein.
The descriptive statistics showed appropriate distributions, with no ceiling or floor effects. Further examination of the distribution of personal characteristics such as resilience, optimism, anxiety and engagement/involvement showed normal distributions. We therefore continued to calculate simple correlations among the study variables. These results are summarized in Table 3 herein.
The bivariate associations reveal a few interesting patterns, in light of our proposed model.
![]()
Table 2. Descriptive statistics for the main variables (n = 368).
![]()
Table 3. Intercorrelations among the study variables.
*p < 0.05, **p < 0.01.
- Demographics were identified as coping resources: thus, for gender, women tended to be more anxious but reacted less negatively, Education was associated positively with other resources such as resilience, income, optimism and associated positively with involvement.
- Political attitudes and attitudes toward the reform showed marked association with optimism, anxiety (in opposing directions), but not resilience and with the reaction patterns.
- Resilience, optimism as well as anxiety showed associations with both reaction patterns and engagement/involvement.
These patterns lend partial support to the second study model. Thus, we next used path analysis to test a double mediation model in which anxiety and reaction patterns mediate the associations between demographic and personal resources and civic engagement.
3.2. Model Testing
Since the preliminary analyses supported the second model, it was put to the test using path analysis in AMOS 28.0 software package. The original model (identical to the conceptual model, see Figure 1) was not supported by the data. However, once non-significant paths were removed from the model and exogenous variables that did not show any significant associations in the model were excluded, a partial model was supported by the data. This model, as well as its goodness-of-fit indices are summarized in Figure 2 below.
The model does not only provide an insight into how the relevant factors may work in these seldom explored dynamics and political settings, but also reveals some additional insights drawn from understanding the factors excluded from the model. First, of the group of demographic variables included in our preliminary model, only a few showed roles in our actual model: gender (but not age or socio-economic status), resilience, optimism, attitudes toward the reforms (but not general political attitudes) were left in the model as exogenous variables. Anxiety and behavioral responses (both positive and negative) showed a partial
![]()
Figure 2. The empirical model. Goodness of fit indices: Chi Square = 119.56; df = 18; p < 0.01. NFI = 0.95, CFI = 0.96, RMSEA = 0.03.
(anxiety also had a direct path) mediation effect on civic engagement/involvement. Again, supporting this model rather than the “control hypothesis model” —anxiety showed positive associations with both types of reactions as well as with engagement (in both direct and indirect paths).
4. Discussion
Civic engagement/involvement in times of socio-political crisis is a behavior pattern that may help better understand revolutions as well as the political evolution of societies and nations. Civic engagement is also considered a pivotal goal for most industrialized countries, considering civic involvement as a tool to maintain democratic values and stabilize conflicted societies (Devesa, 2022; Hechler et al., 2023) . Understanding what individual factors contribute to civic engagement, especially in times of crisis and threat, is still lacking as research focusing on personal level factors is less prevalent than sociological, socio-political and historical approaches to this issue. The current study aimed to address this lacuna, within the real-life context of the socio-political unrest in Israeli society in light of recent legal reforms that many believe may threaten the country’s democratic nature (Shultziner, 2023) .
When asking why people choose to be more or less involved in civic activities (regardless if they are for or against the change introduced), this study considered two somewhat opposing theoretical models that may account for the dynamics of civic engagement: 1) The first model is based on the veteran sense of control hypothesis, suggesting that in times of threat and distress, involvement and active engagement in events that are often perceived as distant and beyond one’s personal control may help regain sense of control, thus reducing anxiety and distress; 2) The alternate model sees anxiety as an engine for action, especially given coping resources such as resilience, and hope. According to this approach, higher levels of anxiety will positively associated with engagement and active involvement. Both models see anxiety and its consequences as key factors in understanding engagement, but they predict different association patterns between these factors.
The data collected in this study showed support for the 2nd model while revealing interesting insights regarding the dynamics of personal resource utilization and responses leading to individual-level civic engagement. The results show the central adaptive role of situational anxiety as a preliminary response to the events. The emerging model suggests that the higher the perceived threat or anxiety response (rather than trait anxiety), the more like are people to respond in both positive and negative patterns (which is unexpected) and in turn—they are more likely (in case of positive reactions) or less likely (in case of negative reactions) to take a more active approach to their civic involvement. Anxiety seems to be associated with engagement and involvement in both direct and indirect paths—all showing positive coefficients.
Another interesting point may be derived from the fact that attitudes toward the reforms but not general political attitudes found their way into the empirical model. While these two were highly correlated (the more right-wing people reported their attitudes to be, the more support they expressed toward the reform), Attitudes toward the reform explained anxiety and reactions better than general attitudes. Another surprising finding is that age, socio-economic status and area of living did not play a significant role in our model. We expected education and economic status to be substantial predictors, as suggested by political figures associated with the reform-suggesting the protests can be seen as actions of dominant elites fighting to maintain their power (Hilay, 2023) .
It is also worthwhile to dwell on the roles of coping resources (resilience and optimism) in the final model: Resilience is associated positively with both types of reaction (positive and negative) as well as with engagement results that both fit with existing ones (Deng et al., 2023) and contradicts others. How can we account for this positive association? The results may suggest that resilience encourages a response, any kind of response, and action in the face of a threat or anxiety-inducing event. Optimism, however only associated negatively with anxiety and positively with engagement. Optimism was associated with supportive attitudes toward the reform.
These results lend support to the approach portraying situational anxiety as a cue or a preliminary response pattern preparing individuals and groups for action in the face of threats or opportunities (Gutiérrez-García & Contreras, 2013; Pantel, 2008) . Regardless of the political attitudes, situational anxiety serves as a preliminary response pattern, seemingly underlying a process of resource reallocation and preparation toward action. While anxiety was associated with both positive and negative responses (which in turn showed opposing association patterns with civic engagement) it showed a robust positive direct association with engagement. The literature on coping with threats and anxiety usually conceptualizes anxiety as a crippling experience, negatively associated with adaptive responses (e.g.: see Clayton, 2020, for an example of a recent field of the study showing the toll of anxiety regarding the environment). The results of this study emphasize the adaptive role (alongside the toll of anxiety and threat) of situational anxiety in mechanisms accounting for civic engagement in times of crisis.
Study Limitations
This study focused on a single series of events in a particular cultural and political context to test the roles and dynamics of concepts and models with a much global aim. While the results and the dynamics described here are not dissimilar to other events in other political settings (for an example see: Horwitz, 2021 for a US example of a very similar nature), the current results should be taken carefully when generalizing their meanings and implications. Our sample, while providing ample statistical power, may not be representative enough of populations such as the Arab society in Israel, and the ultra-Orthodox society which were underrepresented in this sample. The correlational nature of the study design allows discussion of associations between variables but not conclude regarding causal relationships among the study variables.
Directions for Future Studies
Given the results of the study and its limitations, further studies may wish to examine the model in various settings (socio-political events) and target populations to learn more about its applicability beyond the current context. Using a longitudinal design to better establish timelines and causal associations among the variables may also help clarify the roles of pivotal factors in the process described herein. Last but not least—while this study’s model addresses central concepts that seem to serve the clarification of the dynamics at the focus of our attention—competing models can be proposed, borrowed from the fields of social change, intergroup relations and positive psychology – just to name a few. Proposing more competing models may help better identify and optimize models that provide deeper insights and better understanding of what individual-level characteristics play critical roles in how individuals and groups change their social circumstances.