Analysing political engagement in a multimedia, multi-elections environment: predictive factors in an unpredictable political reality

ABSTRACT

This research delves into the intricate interplay between political trust, political efficacy, and engagement in the political process. Analysing survey data collected from 530 participants during the 2022 elections in Israel, at a time of heightened political instability, the study aims to uncover the nuanced relationships underlying citizens’ engagement behaviours. Grounded in political efficacy and political trust theories, the investigation explores the roles of trust in politicians and in the political system as predictors of offline and online political engagement. The mediating influence of political efficacy on these relationships is analysed. The findings reveal distinct patterns of influence for trust in politicians and trust in the political system in different modes of political engagement. Political efficacy emerges as a robust predictor, mediating online and offline engagement. This mediating role elucidates the intricate dynamics between trust variables and engagement pathways, offering insights into the complexities of civic participation in times of political turbulence. The findings contribute to a broader understanding of how trust and efficacy shape citizens’ interactions with the political realm. They underscore the need for continued investigation into the multifaceted dynamics of political engagement.

KEYWORDS:

• Online political engagement
• offline political engagement
• political efficacy
• trust in politicians
• trust in the political system

Introduction

Social media have emerged as key platforms for political communication and participation, particularly during election campaigns (Enli and Skogerbø 2013; Gilardi et al. 2022; Peeters et al., 2023; Stier et al. 2018). Their interactive and open nature empowers candidates and parties to initiate and coordinate activities, amplifying voter engagement throughout and between campaigns (Dvir-Gvirsman et al. 2022; Lilleker and Koc-Michalska 2017; Vaccari and Valeriani 2021). However, offline political engagement remains a significant aspect of the political arena (Boulianne 2009; Gil de Zúñiga and Liu 2017; Lamb and Steinberg 2021).

Political engagement is often intertwined with political trust and efficacy. Political efficacy is crucial for understanding political behaviour, which refers to peoples’ beliefs in their abilities to understand the political system and in becoming active in the political realm. it also referes to peoples’ beliefs in the responsiveness of political institutions and players to meet their expectations (Bernardi et al. 2023). Political trust is a pivotal facet of democracy and refers to confidence in political institutions such as the government, parliament, and political parties (Wu and Wilkes 2018).

Numerous studies have established that conviction in one’s capacity to participate in and influence the political process is strongly tied to online and offline political participation (Acar and Uluğ 2022; Gil de Zúñiga et al. 2019; Kiess 2022). Oser et al. (2022) noted a positive correlation between the degree of political efficacy and online and offline participation, without significant disparities between the two modalities. Beyond political trust and efficacy, voters’ motivation to engage in online and offline political activities may be stimulated by the belief that political change is required.

Nevertheless, political engagement using social media, in particular, can have detrimental effects. The profusion of online data can precipitate information overload, impeding individuals’ ability to discern between credible and dubious sources (Metzger, Flanagin, and Medders 2010). Social media algorithms may also cultivate ‘echo chambers’, permitting individuals to access exclusively that information which concurs with their preexisting beliefs (Bakshy, Messing and Adamic 2015). This can reduce political engagement and exacerbate polarisation (Flaxman, Goel, and Rao 2016).

Moreover, disseminating misinformation on social media can undermine political engagement and decision-making. Misinformation, or ‘fake news’, constitutes deliberately misleading information disseminated to manipulate public opinion (Del Vicario et al. 2016). The swift propagation of misinformation on social media can sway political beliefs and actions (Allcott and Gentzkow 2017; Elishar-Malka, Ariel, and Weimann 2020).

The current study expands our understanding of the different preconditions and factors determining voters’ preferences for online and offline political engagement. Specifically, it examines political trust as a predictor of political engagement in two different aspects: trust in the political system, and trust in candidates and elected officials (hereinafter collectively, ‘politicians’), as these two variables might not necessarily be aligned, especially under the turbulent political conditions.

From April 2019 to November 2022, the Israeli political landscape experienced significant instability, as evidenced by five general elections within three years, deviating markedly from typical democratic elections routines. This period of frequent elections, driven by political, personal, and ideological conflicts, exaggerated Israel's political landscape and fragmented political system. These challenges posed by proportional representation, and deep societal divisions hindered the formation of a stable, consensus-based government. Benjamin Netanyahu played a central role in this instability, with his long tenure, legal issues, and divisive persona deepening political rifts. The coalition formed after the 2021 elections, led by Naftali Bennett and Yair Lapid, attempted to form an anti-Netanyahu bloc. However, the November 2022 elections marked its end and further polarised party divisions.

This extended period of instability risks eroded public trust in the political system and obstructed effective policymaking, particularly in Israel’s turbulent political context (Ariel et al. 2021; Dvir-Gvirsman et al. 2022; Shamir and Rahat 2022; Yarchi, Wolfsfeld, and Samuel-Azran 2021). This study examines voter behaviour during Israel's 2022 general election, the fifth in this series, to explore the effects of such political irregularities on Online and offline political engagement patterns.

Theoretical background

Online and offline political engagement

Social media platforms, including X (formerly known as Twitter), Facebook, and YouTube, have revolutionised the landscape of democracy. These platforms manifest a surge of technological optimism and promise, a rekindled faith in the future of democratic systems (Chadwick 2007). Loader and Mercea (2011) argue that such platforms empower citizen users, highlighting the potential for democratic innovation.

Elishar-Malka, Ariel, and Weimann (2020) delineate four successive historical phases in digital political communication. The latest is marked by President Trump's extensive reliance on social media to further his political interests in the 2016 US elections. Currently, social media platforms serve as primary channels for politicians to establish direct dialogue with multiple followers. Furthermore, the potential influence of big data and artificial intelligence combined with ubiquitous digital devices and platforms is significant in moulding public opinion and political campaigns (Elishar-Malka, Ariel, and Weimann 2020; Musiał-Karg and Luengo 2023).

Peeters et al. (2023) offer a nuanced perspective, acknowledging that while social media might blur the lines between election periods and routine phases, discernible patterns emerge in politicians’ online behaviours. They note that election campaigns see a surge in online political activity. However, the essence of these interactions remains consistent, with election campaigns intensifying but not altering the fundamental dynamics between politicians and citizens.

Political engagement goes beyond the act of voting. It embodies a range of activities, such as rallying, petitions, and lobbying (Koc-Michalska and Lilleker 2017; Resnick 2001; Theocharis et al. 2022). Dahlgren and Hill (2022) depict political engagement as the multitude of ways citizens participate in political processes, including voting, protesting, and civil society memberships. The form and intensity of this engagement depend on societal dynamics, economic circumstances and access to media technologies.

Online and offline political participation appear intrinsically linked (Koivula, Malinen, and Saarinen 2021; Theocharis et al. 2022). While online activism is often criticised for its superficiality compared to offline engagement, Koc-Michalska and Lilleker (2017) counter that given the right circumstances, it can be equally influential and even replace offline involvement. Intriguingly, intrinsic motivations are more likely to drive offline than online participation. However, extrinsic motivations correlate significantly with both forms. A study by Gil de Zúñiga and Liu (2017) reinforces this interconnectedness, asserting that online-acquired social capital robustly predicts both online and offline political engagement.

The concept of ‘slacktivism’ or ‘clicktivism’ pertains to shallow involvement in politics through internet platforms (Karpf 2010). Despite criticism by Morozov (2009; 2011), Rotman et al. (2011) and Halupka (2018) view slacktivism as a kind of political involvement that is easily accessible. Matthes (2022) examines the intricate relationship between social media platforms and the political engagement of young individuals. While there has been optimism about social media's potential to enhance democratic participation among young adults, Matthes contends that such views overlook this group's essentially nonpolitical and entertainment-oriented use of social media. The research uncovers a ‘social media political participation paradox’: despite a marked rise in social media usage among young adults and substantial evidence supporting its positive influence on offline political activities, the gap in voter turnout between younger and older generations has remained unchanged in countries such as Germany, Japan, Switzerland, and the US.

If not the frequency or scope of engagement, online platforms still play a key role in enhancing the breadth of engagement (Mallipeddi et al. 2021). Engagement implies not merely information exchange but also the creation of shared meanings and relationships (Hassanein and Head 2007). Online facilitate the creation and maintenance of virtual communities, fostering idea exchanges and collective meaning-making (Chadwick and Howard 2010). However, the efficacy of such online engagement in driving offline participation remains context-dependent (Koc-Michalska and Lilleker 2017).

Although replete with opportunities, the digital realm has its share of challenges. One significant concern is the digital divide – the individual disparity in technology access (Van Dijk 2006). This inequality can jeopardize representation, information access, and political participation opportunities. Moreover, by creating echo chambers, social media algorithms risk limiting non-savvy users to self-affirming information (Bakshy, Messing and Adamic 2015). Another significant concern is political interest, which signifies an individual's political attentiveness and involvement (Strömbäck and Shehata 2010). It is pivotal in functional democracies, linking to political knowledge acquisition and civic participation (Prior 2010). The literature showcases debates surrounding the democratic virtues of political interest. Some posit that a politically engaged populace is more informed and engages in democracy-enriching behaviours (Prior 2010). Conversely, concerns arise, again, with regard to echo chambers and the potential for misinformation among the politically engaged (Chadwick and Vaccari 2019; Knobloch-Westerwick and Kleinman 2012).

Understanding the nuances of online and offline political engagement is crucial to the current digital age. Despite the inherent challenges, online platforms’ continued growth and evolution are inevitable. As such, the symbiotic relationship between online and offline political participation requires further exploration to ensure more inclusive and meaningful civic involvement.

Political efficacy

It can be bifurcated into two dimensions: internal efficacy, which denotes an individual's belief in their political competence, and external efficacy, reflecting the belief in political institutions’ responsiveness to citizen demands (Craig, Niemi, and Silver 1990).

Studies reveal a positive association between political efficacy and both online and offline political participation (Bode 2017; Karpf 2010; Oser et al. 2022). As citizen trust in the government and political institutions increases, so does their confidence in their ability to influence political outcomes, thereby enhancing their internal efficacy. Participation in political processes, through voting or other forms of activism, also bolsters political efficacy, as it offers citizens tangible experiences of interacting with the political system and influencing its outcomes (Craig, Niemi, and Silver 1990). Discrepancies in the results may be due to varying definitions and measurements of political efficacy and online political participation and the differing contexts of these studies (Oser et al. 2022). Velasquez and LaRose (2015) further propose that efficacy measures are not universally applicable, suggesting that distinct forms of political interaction may necessitate different measures.

Both political trust and efficacy are often viewed as precursors to political participation (Acar and Uluğ 2022; Gil de Zúñiga et al. 2019; Kiess 2022). Acar and Uluğ (2022) suggested that individuals might engage in online and offline political activities if they believe the country's political climate requires change. Some studies have also examined the interplay between political efficacy and trust, finding that political efficacy has a positive relationship with government trust (Brehm and Rahn 1997). Moreover, individuals with elevated internal efficacy levels are more likely to trust the government (Niemi, Craig, and Mattei 1991). Conversely, low levels of external efficacy correlate with reduced government trust (Craig, Niemi, and Silver 1990).

The antecedents of political efficacy encompass factors such as political interest and knowledge (Gil de Zúñiga and Valenzuela 2011; Prior 2010), trust in government and political institutions (Boulianne, Oser, and Hoffmann 2023; Brehm and Rahn 1997); personal encounters with the political system (Craig, Niemi, and Silver 1990); and socio-demographic elements like age, education, and income (Boulianne 2009). Oser, Feitosa, and Dassonneville (2023) investigate the complex relationships between socio-demographic characteristics and political efficacy, understood as individuals’ confidence in their ability to comprehend and influence political processes. Utilising data from 46 countries from 1996-2016, their study reveals that education and income positively correlate with both external and internal political efficacy, while being female is linked to lower internal efficacy but not external efficacy. Intriguingly, the research also shows that increased female representation in parliament corresponds with heightened external efficacy among women. Barriers to political participation, including feelings of powerlessness, cynicism towards official politics, and economic disempowerment, have also been noted by Dahlgren and Hill (2022).

As we have seen, the literature has widely accepted that political efficacy plays a pivotal role in political engagement (Morrell 2003). This concept has been further reinforced by studies demonstrating that individuals who feel more efficacious are more likely to be involved in both offline and online political activities (Kahne, Lee, and Feezell 2013). Given these insights, we hypothesise a robust positive correlation between political efficacy and offline and online engagement. Thus, our first hypothesis is that:

H1: There will be a significant positive correlation between political efficacy and both (a) offline political engagement and (b) online political engagement.

Trust in politicians and the political system

Political trust can be divided into two categories: diffused trust in the political system (Gil de Zúñiga et al. 2019), and specific trust targeted at the government, incumbents, and candidates. Generally, diffused trust persistently surpasses specific trust across most nations (Gil de Zúñiga et al. 2019).

Over recent decades, diffused political trust has declined in Western democracies (Gil de Zúñiga and Liu 2017; 2019). Bøggild (2020) highlights the pervasive public distrust of politicians in Western democracies. This pattern extends beyond dissatisfaction with specific government performance to encompass broader concerns about the conduct of democratic politics. Observational and experimental data from three countries (United Kingdom, United States, and Denmark) reveal that citizens often perceive politicians as prioritising party loyalty, known as partisan representation. However, they prefer politicians to act by following their conscience (trustee representation) and the interests of their constituency (delegate representation) rather than strictly adhering to party policy. This mismatch between expectation and perception fuels distrust in politicians.

Higher levels of political trust relate to formal political participation, such as voting and interacting with elected officials. In contrast, lower trust levels correlate with noninstitutional participation, such as protesting and boycotting (Braun and Hutter 2016; Grönlund and Setälä 2007; Hooghe and Marien 2013). Badman et al. (2022) challenge the prevailing notion that trust in political leaders is essential for public health compliance, especially during crises like the COVID-19 pandemic. Their study, encompassing multiscale trust measurements across twelve countries and territories, distinguishes between trust in institutions and interpersonal trust in political leaders. The findings reveal that trust in national and local public health institutions consistently predicts compliance, whereas trust in political leaders does not.

Relatedly, Schlipphak, Isani, and Back (2022) argue that belief in conspiracy theories may reduce distrust in the government and political system, especially when government actors share or endorse such theories. However, their findings show that a stronger belief in generic conspiracy theories is associated with increased political distrust. This relationship is less pronounced in countries where these theories are part of governmental communication. Importantly, believers in conspiracy theories consistently distrust their government more than do non-believers, irrespective of the political context.

Social media possess the potential to disrupt the political status quo and present political alternatives for those disinclined to partake in formal political processes (Koivula, Malinen, and Saarinen 2021; Sormanen and Dutton 2015). Nonetheless, online political news may further deteriorate trust in political institutions and democracy (Koivula, Malinen, and Saarinen 2021). Bennett and Pfetsch (2018) highlighted the link between diminished trust in democratic institutions and increased reliance on social media. Thus, understanding the potential influence of social networks on political trust and the subsequent repercussions for democratic societies becomes paramount. Trust in the political system, or ‘system trust’, is assumed to directly affect political engagement directly (Hetherington 1998). The underlying notion suggests that citizens who perceive their political system as reliable and effective tend to demonstrate greater political participation, offline or online (Zmerli, Newton, and Montero 2007). Consequently, we postulate that system trust will significantly predict both offline and online engagement. Thus, our second hypothesis asserts that.

H2: System trust will significantly predict (a) offline political engagement and (b) online political engagement.

In recent years, a growing number of studies have focused on another, still less-studied aspect of political trust – specific political trust in elected officials, candidates, and other political leaders (Cole et al. 2022; Druckman et al. 2021; Martini and Torcal 2019). This kind of trust goes beyond the macro level to how people perceive the political players themselves. Still, according to scholars, most people will determine their level of trust in political players according to their party affiliation. People tend to trust politicians from a party they identify with while distrusting politicians from parties that belong to the outgroup, even without any additional information about them (Martini and Torcal 2019). Following Iyengar et al. (2019) and others, Druckman et al. (2021) refer to this phenomenon as ‘affective polarization’ and describe it as ‘one of the most striking developments of twenty-first-century US politics’ (28). This finding is particularly relevant to politically polarised societies such as Israel.

Citizens’ trust in their politicians, referred to as ‘politicians trust’, can also influence their political engagement. Politicians are perceived as representatives of the political system, and citizens’ trust in them will likely influence their willingness to participate in political processes (Hetherington 2005). Therefore, we hypothesise that politicians trust will significantly predict offline and online Engagement. Thus, according to our third hypothesis,

H3: Politicians trust will significantly predict (a) offline political engagement and (b) online political engagement.

Between trust and engagement

The theoretical model of political trust and engagement further suggests a mediating role of political efficacy (Craig, Niemi, and Silver 1990; Niemi, Craig, and Mattei 1991; Oh and Lim 2017). It postulates that a higher level of system trust may increase an individual's political efficacy, leading to higher political engagement (Balch 1974). Thus, we hypothesise that political efficacy will mediate the relationship between system trust and offline and online engagement. Hence our fourth hypothesis:

H4: Political efficacy will mediate the relationship between system trust and (a) offline political engagement and (b) online political engagement. Specifically, higher levels of system trust will relate to higher levels of political efficacy, which will in turn relate to higher levels of engagement.

Extending the political trust model, we also propose that political efficacy may mediate the relationship between politicians trust and political engagement (Dalton 2004; Putnam 2000). If citizens trust their politicians, they are more likely to feel efficacious about influencing political outcomes, which can subsequently lead to increased engagement (Boulianne, 2011). Thus, our final hypothesis is that political efficacy will mediate the relationship between politicians trust and offline and online engagement. Accordingly, our fifth hypothesis states that.

H5: Political efficacy will mediate the relationship between politicians trust and (a) offline political engagement and (b) online political engagement. Higher levels of politicians trust will relate to higher levels of political efficacy, which will in turn relate to higher levels of engagement.

Methodology

Participants

This study included 530 Israeli participants recruited from the Midgam Project Web Panel, a company that provides infrastructure services for internet research and uses stratified sampling based on data published by the Israeli Central Bureau of Statistics. The sample was determined by quotas for age, gender, and education, with a maximal standard error of 4.5%. The sample size was determined using G*Power (Faul et al. 2009), to detect significant differences with a medium effect size and 90% power.

The participants were distributed nearly equally between males (49.2%) and females. They were 18–75 years old, with a mean of 42.10 (SD = 12.38). Most self-identified as Jews (80.6%), followed by Christians (10.3%), Druze (6.0%), and Muslims (3.1%); most participants regarded themselves as secular (55.8%), followed by traditional (21.7%), Orthodox (11.9%), and religious (10.6%). In terms of marital status, most (60.8%) were married (60.8%), followed by single (27.0%), separated (11.9%), and widowed (0.4%). Regarding income, slightly more than half had an above-average income (51%), with 23.2% having an average income, and 18.5% having a below-average income. Finally, in terms of education, the largest group in the sample had higher education (48.7%), followed by post-secondary (51.3%), and secondary education (26.9%).

Procedure

An online survey was administered during the two weeks prior to the 2022 elections, held in November. Participants were asked to complete a short anonymous survey (response time = 10 min) that measured the research variables and elicited demographic information. The final sample included only questionnaires completed correctly after verifying completion time (we only included questionnaires completed in more than five or less than thirty minutes; 17 participants were removed). The institutional ethics committee approved the present study.

Instrument

The research questions were examined through a structured questionnaire that included multiple-choice questions, in addition to demographic questions (see Table 1).

Table 1. Descriptive statistics and reliability coefficients for research variables.

	Min	Max	Mean	Std	Cronbach’s α
Political efficacy	1	5	3.17	.91	.77
Politicians trust	1	4.33	2.02	.85	.70
System trust	1	5	2.19	.84	.80
Offline engagement	1	5	2.25	.65	.72
Online engagement	1	5	2.01	.82	.83

Independent variables

Trust in politicians was measured using a three-item scale (Cronbach’s alpha = .70) developed specifically for this study, rated from 1 (‘strongly disagree’) to 5 (‘strongly agree’). Participants were instructed to indicate the extent to which they agreed with each statement (e.g., ‘You can trust politicians in my country’; ‘Foreign considerations taint the decisions of the politicians in my country’).

Trust in the political system was similarly assessed using a three-item scale (Cronbach’s alpha = .80) developed specifically for this study, rated from 1 (‘strongly disagree’) to 5 (‘strongly agree’) (e.g., ‘The political system in my country is reliable’; ‘You can trust state authorities’).

Dependent variables

Offline political engagement was measured using a six-item scale (Cronbach’s alpha = .72) also specific to this study, rated from 1 (‘strongly disagree’) to 5 (‘strongly agree’) (e.g., ‘I participate in demonstrations and political protests’; ‘I am active in one of the political parties or organizations’).

Online political engagement was assessed using a seven-item scale (Cronbach’s alpha = .83) developed for this study, rated from 1 (‘strongly disagree’) to 5 (‘strongly agree’) (e.g., ‘I often share political content in social networks’; ‘I usually take part in political debates on social networks’).

Mediator: political efficacy

To measure political efficacy, we used a 5-item scale (Cronbach’s alpha = .77) based on Craig and Maggiotto’s (1982) political efficacy measure. This study's scale included statements rated from 1 (‘strongly disagree’) to 5 (‘strongly agree’). Participants were instructed to indicate the extent to which they agreed with each statement (e.g., ‘People like me have skills to participate in political activities and decision making’; ‘I feel I have a pretty good understanding of the important political issues on the agenda’).

Results

In examining the platforms used to consume information about elections, participants reported their frequency of daily usage for each platform as follows: news websites (54%), Facebook (47%), television news (42%), radio news (36%), and WhatsApp groups (36%). Instagram was accessed by 22% of the respondents, followed by Telegram (16%), X (12%), and newspapers (10.5%). Moreover, 41% of participants indicated that they followed Israeli politicians on social networks. Among these, 58% followed Benjamin Netanyahu, the former Prime Minister who was re-elected in these elections, and 37% followed Yair Lapid, who served as Prime Minister during the November 2022 election campaign. The frequency of following other politicians was significantly lower. Of those who followed politicians on social media, the distribution of platform usage was as follows: Facebook (81%), Instagram (27%), X (22%), Telegram (16%), and TikTok (14%).

To examine the relations between the variables, Pearson correlation tests were calculated. As shown in Table 2, all variables were significantly correlated with online engagement, and all except for politicians trust were significantly correlated with offline engagement. This finding supports H1, which proposed that there would be a significant relationship between political efficacy and both types of engagement.

Table 2. Correlation matrix between the research variables.

	Political efficacy	Politicians trust	System trust	Offline engagement
Politicians trust	−.17***	–
System trust	−.01	.70***	–
Offline engagement	.40***	.06	.16***	–
Online engagement	.37***	.10**	.15***	.67***

Note. One-tailed significance: * p < .05, ** p < .01, *** p < .001.

Multivariate linear regression was used to test whether political efficacy, politicians trust, and system trust were significant predictors of offline and online engagement. The first model, which predicted the overall regression of offline engagement, was statistically significant (R² = .19, F(3, 502) = 38.3, p < .001). It was found that political efficacy (β = .29, p < .001) and system trust (β = .12, p < .01) significantly predicted offline engagement. Politicians trust did not significantly predict offline engagement (see Table 3).

Table 3. Model coefficients to predict offline engagement.

Predictor	Estimate	SE	T	p	Stand. estimate
Political efficacy	0.29	0.03	9.77	<.001	0.40
Politicians trust	0.02	0.05	0.33	0.742	0.02
System trust	0.12	0.04	2.61	0.009	0.15

The second model of multivariate linear regression used to predict online engagement was also statistically significant (R² = .17, F(3, 502) = 33.3, p < .001). In addition, it was found that political efficacy (β = .29, p < .001) and politicians trust (β = .14, p < .05) significantly predicted online engagement. On the other hand, system trust did not significantly predict online engagement (see Table 4).

Table 4. Model coefficients to predict online engagement.

Predictor	Estimate	SE	t	p	Stand. Estimate
Political efficacy	0.29	0.03	9.74	<.001	0.40
Politicians trust	0.14	0.07	2.23	0.03	0.13
System trust	0.06	0.06	1.05	0.29	0.06

Trust in the political system significantly predicted only offline and not online engagement, so H2 was only partly supported. The opposite results were found for politicians trust. While this variable did not significantly predict offline engagement, it did significantly predict online engagement, thus partly supporting H3.

Following the findings of the regression models, we examined the mediating role of political efficacy using Hayes’ (2018) PROCESS bootstrapping command with 5,000 iterations. The analysis treated politicians trust and system trust as predictors, political efficacy as mediator, and offline and online engagement as dependent variables. The 95% confidence interval (CI) for the direct effect between politicians trust and offline engagement did not include zero. Moreover, the CI for the indirect effect of politicians trust on offline engagement through political efficacy also did not include zero. Thus, the overall model was significant (F[2, 503] = 53.37; p < .001), indicating a direct effect (95% CI, .040 – .180) of politicians trust on offline engagement and an indirect effect (95% CI, −0.098 to −0.030) of politicians trust on offline engagement through political efficacy.

As for the correlations between system trust and online engagement, the overall model was significant (F[2, 503] = 47.13; p < .001), indicating a direct effect (95% CI, .70–0.227) of system trust on online engagement, but not an indirect effect (95% CI, 0.37 – .028) through political efficacy.

Thus, politicians trust was directly related to online engagement, and indirectly so through political efficacy, while system trust was only directly related to offline engagement. In other words, it was found that political efficacy only mediated the correlation between politician trust and online engagement and not between system trust and offline engagement (See Figure 1).

Figure 1. The mediating model of political efficacy in the correlation between system trust and offline engagement, and between politicians trust and online engagement. Note. *** p < .001.

The analysis showed that politicians trust had a direct and indirect relationship with online engagement through political efficacy, confirming H4, which proposed that political efficacy mediated the correlation between those variables. On the other hand, system trust was found to be directly related to offline engagement but not indirectly through political efficacy, refuting H5, which proposed that political efficacy would mediate the correlation between those variables.

Conclusions

This study examined the relationships between political efficacy, politicians trust, system trust, and offline and online engagement. All variables were significantly correlated with online engagement, while except for politicians trust, all were also correlated with offline engagement. The multivariate linear regression analyses demonstrated that political efficacy significantly predicted offline and online engagement. At the same time, as far as trust was concerned, the findings were somewhat more complex: trust in the political system significantly predicted offline engagement only, and trust in politicians significantly predicted online engagement.

These findings suggest that despite the key role of political efficacy as a predictor of both online and offline engagement, these are two different variables: predicted by different factors and correlated with them along different paths. The role of political efficacy as a predictor of both kinds of political engagement, as found in this study, echoes findings of previous studies, among them Oser et al. (2022), who have meta-analysed findings showed that political efficacy has a strong positive relationship with these two forms of political engagement. This replication highlights the role of political efficacy as one of the most crucial factors in any democracy wishing for active civilians and voters.

Our results also highlight the distinct roles of trust in politicians and trust in the political system in predicting voters’ engagement in the political process, especially during an election campaign. Although the correlation between these two variables is relatively strong, the findings suggest they are not identical. Voters may trust the political system in their country as a whole, but still find certain political leaders, mostly outgroup ones, as unworthy of their trust. At the same time, voters may express cynicism and distrust towards their country's political system while still trusting a few of their ingroup political players.

However, the major question remains: Why would system trust be strongly associated with offline engagement only, and politicians trust with online engagement? One possible explanation is that offline engagement is a much more demanding activity. One should deeply trust the political system and its democratic institutions and believe that elections are free and open as promised to get out of their comfort zone and even risk their freedom and other valuable resources. Online engagement, on the other hand, is not only easier and less demanding – the ‘echo chamber’ phenomenon occurs in small public sphericules (Garrett 2009; Gitlin 1998; Knobloch-Westerwick and Kleinman 2012), usually within the ingroup. Therefore, trusting the political players involved is an elementary precondition for such engagement.

Regarding the mediating role of political efficacy, the findings point to complex processes in determining engagement patterns, showing that political efficacy mediates the effect of trust in poilticians on online engagement. However, efficacy does not mediate the link between system trust and offline participation, which is direct. These mixed results provide insight into the nuanced relationships between trust, efficacy, and engagement modes.

The mediating effect in the case of trust in politicians suggests that efficacy helps convert trust in individual leaders into motivated online actions. Greater trust in specific politicians boosts one's sense of political empowerment and voice, facilitating digital participation. In contrast, the lack of mediation for system trust implies that this broader confidence can directly activate traditional offline behaviours such as voting, irrespective of efficacy beliefs. Overall, these findings reveal that political efficacy facilitates the translation of politician trust into online engagement. This highlights efficacy's role in motivating modern digital participation without assuming the nature of underlying feelings about politicians. However, traditional offline acts still respond directly to system trust without requiring a boost of efficacy. This elucidates distinct psychological pathways to online versus offline engagement. These patterns make sense in the context of mounting election fatigue after multiple votes, as in the recent Israeli case. Fatigue can amplify disillusionment with individual politicians, which political efficacy then helps channel into online engagement. Conversely, offline behaviours such as voting rely more directly on ingrained system support without requiring an efficacy boost in a tired electorate.

Although these findings relate to a unique case study, the revealed correlations between the two types of political trust and the two types of political engagement may also exist in other democracies, even in cases where elections take place as planned. Regardless of the elections ‘fatigue’, which such irregulareties might evoke, offline political participation is considered to be much more demanding and riskier than online engagement, requiring a rooted trust in the political system as a whole. Online political engagement, on the other hand, will often target specific players (i.e., following their personal social media accounts) and will therefore require more personalised trust in them. In this sense, the discussed phenomenon is of an international nature rather than a representation of a local case study.

This study’s results may be considered a major contribution to the existing literature on political trust, political efficacy, political engagement, and the interplay between them. By differentiating between two types of political trust – individual players on the one hand, and the political system as a whole on the other hand – as well as by revealing the unique correlations between these two types and online vs. offline political participation, this study adds another layer to our understanding of the role of trust in shaping voters’ political behaviour. Another major contribution concerns the unique role of political efficacy as a mediating variable in the relationship between trust and engagement. This study reminds us that online activities have become a major part of our lives. Social media has become a central political arena for elected officials and voters, and online political engagement is no longer a substitute for offline engagement. Therefore, they are expected to continue this coexistence rather than compete for voters’ time and attention.

The present study had several limitations. The exclusive focus on the Israeli case amid exceptional political circumstances restricts the general applicability of our findings to other situations. However, as previously explained, they are relevant for future investigations of online and offline political engagement during extreme political instability and other political circumstances. Future studies should examine the interplay between these variables during ordinary election campaigns, between elections, and in other democracies that might face such political irregularities. Another limitation lies in the potentially inaccurate reporting of attitudes and behaviours during an election campaign, as individuals may find it difficult to express their views truthfully. One last limitation might be possible difficulties among study participants in differentiating between trust in the political system and their perception of political players as trustworthy. Future studies should strive to find effective ways to overcome this potential obstacle, including the use of carefully worded questions to ensure participant understanding and screen the accordingly.

Understanding the nuances of online and offline political engagement is crucial to the current digital age. Despite the inherent challenges, online platforms’ continued growth and evolution are inevitable. As such, the symbiotic relationship between online and offline political participation requires further exploration to ensure more inclusive and meaningful civic involvement.

Disclosure statement

No potential conflict of interest was reported by the author(s).