Understanding different levels of e-participation in Spanish municipalities based on city and citizen profiles

Recent studies (Alcaide-Muñoz et al., 2022) have determined that open government (OG) Data (OGD) initiatives and transparency portals are among the most developed, primarily aimed at increasing the legitimacy of public actions rather than promoting citizen engagement in public decisions. Nevertheless, transparency represents a valuable instrument for fostering citizen participation (Caba Pérez et al., 2008), thereby reinforcing the relationship between governments and citizens (Alarabiat et al., 2021; Liu and Zowghi, 2022) which has led governments to create e-Participation platforms in recent years (Pina et al., 2022; Campo et al., 2022). However, these e-Participation projects have often recorded low participation levels (Oliveira and García, 2019) due to limitations such as the digital divide, web design and usability, citizens’ capacity, and trust in the democratic process (Rodríguez-Bolívar and Alcaide-Muñoz, 2019).

To counteract citizen apathy, governments have increased their use of social media or instant messaging tools for interaction and engagement. Additionally, governments are adopting emerging technologies (ETs) to improve communication channels (Rodríguez-Bolívar and Alcaide-Muñoz, 2024). These technologies are radically novel and relatively fast-growing technologies characterized by a certain degree of coherence persisting over time and have the potential to exert a considerable impact on the socioeconomic domain(s), primarily in the future, being somewhat uncertain and ambiguous in the emergence phase (Rotolo et al., 2015), address current participation channel limitations, fostering more fluid interactions and making citizens feel more involved and confident that their opinions are considered (Cortés-Cediel et al., 2023; Lnenicka et al., 2024). In this regard, recent research has demonstrated that chatbots in OGD collections have the potential to improve the usability, openness, accountability, trust and legitimacy of the data used in relation to e-Participation platforms (Cortés-Cediel et al., 2023).

Prior research (Wijnhoven et al., 2015) has demonstrate a range of interactions between governments and citizens in public decision-making (Choi and Song, 2020), possibly due to its continuous evolution and the increasing implementation of ETs (Rodríguez-Bolívar and Alcaide-Muñoz, 2019, 2024). This suggests that e-Participation is an emerging field that has not yet reached maturity. Additionally, there is currently no evidence on different e-Participation implementation models in urban contexts, nor the influence of contextual factors (city profile and/or citizen profile) on these models, which is crucial for understanding the various open and collaborative governance models in cities.

Thus, our study focuses on analysing the e-Participation channels offered by the large Spanish local governments (SLGs), with the intention of determining the different e-Participation models adopted, and whether these models are defined by the profile (characteristics) of both the cities and the citizens. Concretely, our study seeks to ascertain whether demographic (municipality profile) and social (citizen profile) factors motivate public managers to offer different levels of e-Participation. By this way, the findings of this research seek to offer added value to prior literature in this field. The Spanish case study is of particular interest in the context of the Open Data Maturity Report 2024 -6th position in Europe- and, especially, the country’s leadership position in the field of Open Data impact on addressing the governmental, societal, economic and environmental challenges (Page et al., 2024), which could provide citizens with a clear motivation to be actively engaged with open data as a catalyst of social innovation and a more participatory approach to public knowledge (Santos-Hermosa et al., 2023).

Although there are numerous definitions of participation, this study focuses on political participation aimed at solving collective or community problems (the targeting perspective, as proposed by Van Deth, 2016, or civic engagement, as outlined by Weiss, 2020), rather than participation focused on voting rights (the locus perspective, as outlined by Van Deth, 2016).

While new forms of participation vary according to the level of decision-making control, they can be broadly classified as online and offline participation (Lim and Oh, 2016). The existing literature on offline participation examines its advantages for enhancing both local government policy-making procedures (Walters et al., 2000) and trust in policy implementation processes. In contrast, online participation or e-Participation has been demonstrated to exhibit both different levels of citizen participation within the digital realm (Errandonea, 2023) and certain limitations and barriers to the achievement of more transparent government organizations (La Porte et al., 2002) or real openness to encouraging citizen dialogue (Pina et al., 2009).

Over the past decade, the growing demand of citizens to play an active role in policymaking and the advent of digital transformation have created an environment conducive for policymakers to the development of strategies that facilitate public engagement (Errandonea, 2023). A significant number of governments worldwide have established official websites that incorporate e-Participation instruments, including participation platforms, social media platforms (Facebook and Twitter), and instant messaging tools (Telegram and WhatsApp). Moreover, in view of the limitations encountered by governments in managing the substantial volume of data and communications generated through these channels, there is an emerging trend towards the deployment of ETs with the objective of enhancing and diversifying the exchange of information between governments and their citizens (Cortés-Cediel et al., 2023).

However, the extent of participation achieved through these various technological tools has varied, not only due to the intrinsic characteristics and capabilities of these tools (Oliveira and García, 2019), but also due to citizen perceptions regarding the degree of control exerted by governments in these collaborative spaces, which may limit their genuine opinions (Rodríguez-Bolívar and Alcaide-Muñoz, 2019).

Prior research has concentrated its analysis on citizen perceptions (Choi and Song, 2020), psychological factors (Edelmann and Cruickshank, 2012), contextual factors (Krishnan et al., 2012), political orientation and environment (Dias, 2020) or the achievement of democratic values -representative democracy, legitimacy, and government effectiveness (Stratu-Strelet et al., 2021)- to elucidate the extent of citizen participation. However, there has been a paucity of investigation into the city and citizen profiles as potential sources for explaining these discrepancies in citizen participation.

Also, although evidence suggests that individuals who are inclined to engage in political activities in an offline context are also likely to participate in online political activities (Oser et al., 2013), this is not a universal phenomenon. It is therefore imperative to examine not only the factors that predict both this phenomenon (Boulianne, 2015) and those that influence or encourage e-Participation (Feezell, 2016), but also the efficacy of the different systems for engaging citizens in the policy-making process (Lim and Oh, 2016).

This study focuses on analysing the different tools offered by SLGs, with the intention of determining the different participation models offered to citizens. In order to ascertain the various degrees of involvement proposed, this study is grounded in the Arnstein’s Participation Ladder (ALP) (Arnstein, 1969). Despite criticism, adaptation, and evolution (Salem, 2016) of the ALP model due to its linear model approach (Tritter and McCallumb, 2006) and its static and rigid vision of the participation process (Akers, 2022), it remains a widely used model due to its simplicity, coherence, and widespread application. Consequently, its work has recently been used as a framework for participatory models’ evaluation (Contreras, 2019; Akers, 2022).

This study does not to analyse the evolution of e-Participation models at the municipal level. Rather, it examines the different e-Participation models that have been implemented and assesses the influence of city and citizen profiles on these models. To achieve this objective, the eight rungs for citizen participation included in the ALP (Arnstein, 1969) are employed. The eight rungs of the ladder symbolize the manner in which local governments have approached the engagement of citizens, based on the varying degrees of redistribution of power in public decision-making processes. Arnstein (1969) grouped the eight rungs into three principal approaches in accordance with the level of power delegation between the government and citizens: (a) non-participation; (b) tokenism; and (c) citizen power.

The lowest-level rungs of the ladder (Manipulation -Rung 1- and Therapy -Rung 2-) do not imply any form of participation. Rather, they focus on the manipulation and learning from “uneducated” citizens who lack power in the decision-making processes. The middle-level rungs of the ladder (Informing -Rung 3-, Consultation -Rung 4-, and Placation -Rung 5-) primarily focus on disseminating information about government decisions and encouraging citizen participation through opinion solicitation. However, the ultimate decision-making authority remains with the government. This approach to citizen participation enables those who are marginalized to be heard and to have a voice. However, while citizens are unable to ensure that their views will be considered under Rungs 3 and 4 (Informing and Consultation), the rules of Rung 5 (Placation) permit the less powerful to offer advice and exert some influence in the decision-making processes. Nevertheless, the right to make decisions remains with the powerful.

Ultimately, the third and final approach (high-level rungs: Partnership -Rung 6-, Delegated power -Rung 7-, and Citizen control -Rung 8-) entails the delegation of power, whereby governments establish conducive conditions and processes for citizens to engage in the implementation of public policies and decisions. In conclusion, citizens have the potential to exert influence on decision-making processes at rungs 5 to 8 of the ALP, whether to a limited extent (Rung 5) or to a significant degree (Rungs 6, 7 and 8) -see Table 1. Consequently, in our research, we proceeded to consider Rungs 5 to 8, which entail the assumption of citizen participation. The aforementioned rungs were used to configure our dependent variable, which will be explained in greater detail below.

Previous studies have analysed the factors that shape citizens' e-Participation (Stratu-Strelet et al., 2021; Alarabiat et al., 2021; Choi and Song, 2020; Guillamón et al., 2016; Hao et al., 2016). While Stratu-Strelet et al. (2021) examined the institutional factors and their influence on e-Participation, highlighting that government leadership is a key driver in promoting e-Participation, beyond technological infrastructures, Alarabiat et al. (2021) and Guillamón et al. (2016) focused their research on the use of social networks for citizen participation. The former showed that citizens' attitude, participation efficacy and perceived behavioural control are the main factors influencing e-Participation through Facebook (Alarabiat et al., 2021), while the latter found that local governments use Facebook for e-Participation according to the municipality’s population size, citizens' income level and debt level (Guillamón et al., 2016).

Additionally, Hao et al. (2016) explored how social media enhances government-citizen interactivity, highlighting the impact of structural and content features, such as multimedia elements and external links, on citizen engagement. Their findings suggest that municipalities with advanced social media strategies are more likely to adopt dynamic and participatory governance models.

Finally, Choi and Song (2020) analysed the factors influencing citizen e-Participation using technology acceptance model factors, theory of planned behaviour factors, individual social capital factors and control variables. Although this study used citizens' age and education level, individual social capital factors are the most important and significant in influencing citizen participation.

Our research contributes to prior research by determining city profile influence on the participation management models adopted by municipal governments. Also, this analysis considers the different channels used by local governments to communicate with citizens (offline and online), encompassing both participation platforms and social media. Based on solid theoretical framework (stakeholder theory, institutional theory, or agency theory), this paper identifies the city and citizen profile using sociodemographic characteristics -see Table 2.

According to stakeholder theory (Jensen and Meckling, 1976), policymakers and public managers do not inherently align with the interests of citizens, necessitating mechanisms of accountability to ensure they act responsibly. Research indicates that larger municipalities face heightened pressure in political decision-making and public resource management (Alcaide-Muñoz et al., 2017, 2022). This pressure frequently results in increased innovation with regard to the adoption of new technologies, thus creating enhanced opportunities for citizen participation through a variety of online services (Bonsón et al., 2015). This is supported by studies showing a positive correlation between population size and e-Government adoption (Moon, 2002; Reddick, 2004).

However, the evidence is not entirely conclusive. Some studies have found no significant evidence that population size alone is a predictor of government financial transparency (Caba Pérez et al., 2008) or e-Participation (Novo and Vicente, 2019). Therefore, this issue is, up to now, blurred and deems a greater attention. This way, our research proposes to test the following hypothesis:

Previous research has identified age as a critical factor in understanding individuals’ behaviour towards using ICTs, mobile systems and apps (Ahmad and Kalid, 2017). Age is also a significant predictor of e-Government adoption by citizens (Tavares and Da Cruz, 2020). Studies have consistently shown that younger people tend to engage more actively with ICTs compared to older citizens, who are less likely to adopt e-Government solutions (Van Deursen and Van Dijk, 2009). This age-related digital divide is further corroborated by the findings of Novo and Vicente (2019) in the context of e-Participation.

Conversely, some research suggests that the elderly population exhibits a higher demand for online information and public participation. Lowatcharin and Menifield (2015) noted that older adults may have specific needs and motivations for engaging with e-Government services. Additionally, Ma and Zheng (2018) observed that teenagers do not typically utilize municipal e-Government services due to their specific needs and interest not being catered to by these platforms. In brief, while younger population are more likely to use e-Government services, there are also segments within the elder population that actively seek and benefit from these services. Based on these assumptions, we propose the following hypothesis:

Population density is a significant factor to be analysed within the framework of stakeholder theory (Jensen and Meckling, 1976) and institutional theory pronouncements (Powell and DiMaggio, 2012). According to institutional theory, the design of organizations is shaped not only by rational processes but also by internal and external pressures that drive the adoption of similar structures and practices over time (Powell and DiMaggio, 2012). Organizations, including municipalities, adopt structures and practices that are perceived as legitimate and socially acceptable, leading to homogeneous practices and structures (Alcaide-Muñoz et al., 2022).

Municipalities with similar population densities tend to exhibit similar characteristics, leading them to undertake comparable initiatives, provide services under similar conditions, and manage their organizations in analogous ways (Alcaide-Muñoz et al., 2022). Dense cities facilitate social interactions, which promote the flow of knowledge and innovative ideas (Glaeser and Gottlieb, 2006). This makes them potentially more inclined to introduce ICT-based initiatives (Neirotti et al., 2014).

High population density can increase stakeholder pressure on local governments, creating a greater need for e-Participation initiatives. Recent research by Silal and Saha (2021) confirms that higher population density is associated with a greater likelihood of community action, which in turn promotes governmental efforts to involve citizens in support of proposed plans and local actions.

Previous studies (Moon, 2002; Monaharan, 2012) have shown that municipalities with higher population densities tend to adopt more sophisticated e-Government practices due to the increased demand for services and the necessity to manage large, more complex urban environments. Moreover, densely populated areas are often characterized by higher level of social capital and civic engagement, which further drivers the adoption of e-Government services (Reddick, 2004). Considering this finding, we propose the following hypothesis:

The framework of Smart Cities (SCs) has significantly advanced technological development and innovation, fostering a participative and collaborative environment among governments and citizens to improve the quality of life (Rodríguez-Bolívar, 2019). Smart city initiatives aim to enhance citizen engagement in public decisions, implement collaborative governance structures, and foster co-creation and collective intelligence (Khan and Krishnan, 2021).

Recent research (He and Ma, 2021) reveals that when citizens feel their opinions are considered by the government, their satisfaction with public services, governmental decisions, and public policies increases, leading to stronger trust in government. Pereira et al. (2017) affirmed that closer interactions between citizens and government lead to the development of smart government models.

Dense urban environments characteristics of SCs facilitate social interactions and the flow of knowledge and innovative, ideas, making these municipalities more inclined to introduce ICT-based initiatives (Glaeser and Gottlieb, 2006). Based on these pieces of evidence, we propose the following hypothesis:

According to agency theory (Verbeeten, 2008), politicians are driven more by self-interest than by a commitment to the public good. As a result, using ICTs, citizens demand more information enabling them to be well-informed and to monitor governmental decisions, prompting politicians to manage government affairs in a more efficient manner (Rodríguez-Bolívar and Alcaide-Muñoz, 2019).

The use of ICTs and participation channels, such as social networks and e-Participation platforms, have empowered citizens to take a more active role in decision-making processes. Citizens no longer need to wait for municipal elections to express their opinions; they can do so frequently on various topics. Previous studies (Oliver, 2001) suggest that voter turnout levels, the frequency of contacts between citizens and local officials, and participation in political meetings are indicators of citizen involvement and commitment to political participation. Moreover, higher levels of electoral participation are often linked to the presence of robust e-Government services, which can enhance citizen satisfaction with public services and governmental decisions (He and Ma, 2021). Therefore, we propose the following hypothesis:

In contrast to their interest in government financial transparency (Caba Pérez et al., 2008), people with higher levels of education tend to be more active in using eGovernment services (Thomas and Streib, 2003). They also demonstrate a greater understanding of and ability to utilize e-Government platforms, which in turn leads to increased participation, encompassing the provision of feedback and the contribution to the formulation of public policies and services (Tolbert et al., 2008), the shaping of governmental decisions (García et al., 2017), and the monitoring of governmental activities (Monaharan, 2012). Consequently, municipalities may be compelled to adopt more sophisticated e-Government services in response to an increased demand for transparency and efficiency in government operations (Reddick, 2004).

In summary, individuals with higher qualifications tend to demonstrate greater ease in utilizing new technologies and exhibit a more active role in citizen participation and the monitoring of municipal management. Thus, the following hypothesis is derived:

Gender plays an important role in the technology acceptance model due to the different socially constructed cognitive structures between men and women in the making decisions process (Venkatesh et al., 2012). Ahmad and Kalid (2017) emphasized that women are generally more empathetic and aware of the others’ feelings compared to men. By contrast, male web users exhibit higher activity on OG platforms compared to female users (Schmidthuber et al., 2017).

Research has shown that men tend to be more active in political affairs, while women are equally or more active in community activities and grassroots movements (Enns et al., 2008). Kim and Lee (2019) found that men who perceive greater instrumental value of e-Participation are more likely to use e-Participation actively. Conversely, women are more likely to use e-Participation when they are connected to offline social groups, using them as supplementary channels. In brief, gender differences significantly influence how individuals engage with e-Government platforms. The following hypothesis is therefore derived:

The presence of an immigrant population can negatively influence the finances of local governments (Rodríguez-Bolívar et al., 2021; Choi and Song, 2020), leading to increased demands for transparency and information about the management of public resources. Moreover, Maya-Jariego et al. (2009) emphasized the pivotal function of ICTs in fostering social cohesion by facilitating the rapid adaptation of immigrants to their new surroundings, including the acquisition of the local language, culture and customs. The Internet, as a user-friendly mass medium, can help remove social, cultural, and racial barriers, fostering shared experiences and knowledge transfer in community meeting places such as libraries (Paganoni, 2012). Such venues frequently offer access to computers and online training, which can mitigate social exclusion and enhance employment opportunities for immigrants (Van der Meer and Van Winden, 2003).

In short, the presence of immigrant residents may increase the demand for information among local citizens. In this context, ICTs play a pivotal role in facilitating access to information and supporting the integration of immigrants into the community. Consequently, immigrants will also be keen to understand government decisions and public policies relevant to their interest, leading to an increased demand for participation channels from all citizens. Based on these assumptions, we propose the following hypothesis:

Civil status significantly influences offline political participation. Wolfinger and Wolfinger (2008) found that single individuals are less likely to vote and that single parents have lower offline participation rates, as they perceive their interests to be less represented in policy formulation. Additionally, participation tends to decrease for individual whose marriages have ended through death, divorce, or separation. However, in the context of online participation, Gertrudix et al. (2016) found no significant evidence linking civil status to participation levels.

Further research by Gaboy et al. (2020) and Marcial (2017) indicated that civil status is negatively related to the use of software applications, with single and younger individuals more likely to use various applications and software. Marcial (2017) also found that single people tend to have higher ICTs skills. Moreover, married couples without children or with fewer children show greater confidence in using information and communication technology (ICT) equipment and tools.

Despite these insights, there is limited evidence concerning how civil status influences participation management models within the context of e-Government. In light of the discrepancies in ICT utilization and participation proclivities contingent on civil status, it is imperative to investigate the potential impact of these variables on e-Participation models. That’s why we propose the following hypothesis:

In recent years, the great deterioration in public finances of Spanish regional and local governments has jeopardized the financial sustainability of public services and led to hard management cutbacks (Rodríguez-Bolívar et al., 2021). Also, the continuous cases of corruption have increased distrust towards institutions and political leaders, especially at the local government level (Cifuentes-Faura et al., 2022). Spain has therefore carried out a greater number of smart initiatives (Rodríguez-Bolívar and Alcaide-Muñoz, 2019), whose technological environment favours the provision of innovative online services, a greater openness of public institutions and citizen’s participation in the development of public policies. The central government of Spain launched an ambitious OG strategy, adopting the OG Partnership philosophy in 2011, following the Recommendations of Organization for Economic Co-operation and Development (OECD) on the Digital Strategies of the Government (OECD, 2014). To date, this strategy has been implemented in several phases and is constituted by four action plans (www.transparencia.gob.es).

As SLGs are the most likely to adopt ETs and innovations in the provision of public services (Alcaide-Muñoz et al., 2022; Criado and Zarate, 2022), they could fit well with the aim of our research. Indeed, Spanish SLGs have initiated and implemented OG initiatives and projects with the objective of facilitating citizen participation in public policies and enhancing citizen trust by disclosing a greater amount of information (Alcaide-Muñoz et al., 2022), which could be useful and interesting for governments in other countries with similar settings around the world.

Concretely, our study focuses on Spanish SLGs with a population exceeding 50,000 inhabitants, representing over 50% of the Spanish population (https://www.ine.es/), as these local governments assume a series of competencies and responsibilities to deliver complex and efficient public services (Law 7/1985, Regulation of the bases of local regimes) and their public policies influence on the daily lives of their citizens. Moreover, they are technologically innovative (Criado and Zarate, 2022) and are positioned as pivotal actors in the creation of a participatory environment and an interactive urban environment based on information (Rodríguez-Bolívar, 2015).

The data presented in this study were collected from the official websites of 145 municipalities during the months of May and June 2022.

The official websites of the SLGs were analysed using an item of participation rungs based on ALP (Arnstein, 1969) -see Table 1. As previously stated, our analysis is focused on the four higher levels of rungs, which include some degree of participation (included in the tokenism and citizen power levels -see section 2-).

The first sub-item is related to the Rung 5 “Placation Phase” where citizens begin to have more influence but retain for powerholders the right to judge the legitimacy of feasibility of the advice (Arnstein, 1969). This entails bidirectional information flow through consultation, encompassing surveys conducted via social networks, mobile applications (Apps), and online participation platforms, through which citizens may express their opinions and remain apprised of developments. Nevertheless, this does not directly impact governmental decision-making (Nabatchi, 2012).

The second sub-item is related to the Rung 6 “Partnership” whereby power is redistributed through negotiation between citizens and powerholders. It is agreed that planning and decision-making responsibilities will be shared through the establishment of structures such as planning committees, assemblies, and associations (Arnstein, 1969). It can thus be asserted that participation is occurring, albeit in an insufficient manner and with citizen influence on public decisions remaining indirect. This indicates a slight authority of citizens (Miller et al., 2019).

The third sub-item is related to the Rung 7 “Delegated Power” where negotiations between citizens and public officials may result in citizens attaining dominant decision-making authority over a specific plan or program (Arnstein, 1969). In this ladder, powerholders must initiate the bargaining process rather than responding to pressure from the other party. In this research, the term “participation” is used to refer to the involvement of citizens in person at municipal plenary sessions, which may be conducted online. Consequently, the participation process is initiated via online channels (by the municipalities) to enable citizens to engage in the process and present their concerns, suggestions, and issues through various online channels (participation platforms, social networks, and apps).

The final sub-item is associated with the Rung 8 “Citizen Control”, wherein citizens are seeking the degree of power that ensures their participation in the governance of a program or institution (Arnstein, 1969). At this juncture, the role of citizens in the governance process is reinforced by public policies (Routzouni et al., 2019). In this regard, it may be possible to negotiate the conditions under which external parties may propose changes. At this stage, the municipality provides a variety of online platforms for facilitating dialogue between public managers and citizens.

Finally, regarding the value assigned to each sub-item in Table 1, the four Sub-items (Sub-items 1, 2, 3 and 4) were accorded a weight of 100% each. The score was distributed equally among the different items included in each Sub-item, which represent the various technological platforms (participation platform, social networks or apps) that could facilitate citizen engagement in decision-making processes for each of the analysed rungs, with the exception of Sub-item 2, which was only offered in person among the sample municipalities and thus considered an offline method of participation.

In accordance with the extant literature (Conroy and Evans-Cowley, 2005; Evans-Cowley and Conroy, 2006; Theocharis and Van Deth, 2018), a binary dichotomic value system (0/1) has been employed in each rung, whereby the absence or presence of each technological tool (in the online method) or a way of participation in person (in the offline method) for citizen engagement is determined based on the responses to the questions posed. This approach is intended to reduce subjectivity in the evaluation of aspects that lack specific rules for assigning value (Jones and Alabaster, 1999) -see Table 1.

To achieve the objective of this study, the data analysis was conducted in two stages. The methodology was conceived with the objective of examining the influence of demographic factors on the implementation of e-Participation tools. This methodological approach aligns with extant research that has identified a correlation between citizens’ demographic profiles and their engagement in public affairs (Guillamón et al., 2016; Kim and Lee, 2019; Choi and Song, 2020).

The initial stage of the research entailed the performance of hierarchical cluster analysis, with the objective of identifying homogeneous patterns among cities based on demographic variables pertinent to e-Participation. The objective was to categorize municipalities exhibiting analogous demographic characteristics and we examined then their e-Participation levels. This stage was crucial for understanding the correlation between demographic variables and the level of citizen participation achieved.

The present study employed the Ward Method (Ward, 1963), also known as the minimum inertia loss method, to perform the aforementioned clustering. This method involves the iterative linkage of cases with analogous demographic profiles to minimize variance within each group. The results of the cluster analysis are presented in Table 3, where the municipalities are grouped according to characteristics such as population size, density, age, education level, and percentage of immigrants, among others. This classification lays the foundation for the second stage of the analysis, which focuses on identifying the variables that influence the formation of these groups and their relationship to participation levels.

In the second stage of the study, the objective was to ascertain which demographic variables function as determining factors for the level of citizen participation and to explain the rationale behind the assignment of municipalities to specific clusters. To this end, we conducted a series of statistical tests to explore the relationships between the clusters and demographic characteristics. Initially, within each cluster, we performed U Mann–Whitney and t-Student tests (depending on the distribution of the variable as determined by the Kolmogorov–Smirnov normality test in Table 4) to identify the primary factors influencing municipal e-Participation models. The goal was to determine which demographic variables (such as age, education level, or population density) are significantly associated with different levels of participation. The results of these statistical tests are shown in Table 5.

Secondly, we employed Kruskal-Wallis and ANOVA tests to compare the characteristics of the different clusters. These tests were implemented to ascertain statistically significant disparities in the demographic characteristics that define each cluster and to explore the potential contributions of these factors to the adoption of particular e-Participation models. The findings from these tests are presented in Table 6.

Contrary to the findings of prior studies (Guillamón et al., 2016; Kim and Lee, 2019; Choi and Song, 2020) that examined the direct correlation between demographic characteristics and participation levels, our approach employed clustering to categorize municipalities based on their similar demographic profiles. This methodological decision facilitates a more nuanced understanding of how demographic factors influence the development and implementation of e-Participation tools. Moreover, by identifying the variables that differentiate the clusters, we can better explain the mechanisms driving e-Participation in municipal governance.

In conducting this research, it was imperative to adhere to the established ethical guidelines to ensure the integrity of the study. The data used in the analysis was gathered from publicly available sources, and no personally identifiable or sensitive information was used. Additionally, no direct interaction with participants took place, as the study relied exclusively on secondary data from municipal reports and publicly accessible records. The research process was carried out in accordance with established ethical standards, thereby ensuring the validity and reliability of the results.

The results of cluster analysis are presented in Table 3. Municipalities in this cluster (Cluster 1–60% of the total sample-) offer citizen participation tools for consultation (R5) or citizen collaboration (R8). However, these tools are not yet fully developed, and no other forms of participation are considered. These municipalities tend to be medium-sized, with a population exceeding 75,000. They exhibit a high average population density and a high percentage of foreign citizens. As development in participation levels increases, so too does the proportion of foreign residents in these municipalities. Finally, the level of citizen participation is average, although citizens have low levels of education in both secondary and higher education, with slightly greater levels in municipalities with R8 initiatives.

Municipalities included in Cluster 2 (20% of sample municipalities) have fully accomplished R5. To reach R8, they pass through R6, although it is poorly developed. These municipalities tend to have larger populations (average population over 88,000 inhabitants), higher population density, and generally higher education levels, especially in higher education. This municipalities also have a significant percentage of immigrant population.

Municipalities included in cluster 3 (4% of sample municipalities) have made significant progress in their participation strategies, progressing through all the levels of the ALP (from R5 to R8). They achieve high levels of accomplishment in R5 and R6 but need further focus to advance in R7 and R8, where they reach only 50% accomplishment. Municipalities reaching R7 are characterized by large populations (average population over 80,000 inhabitants) but low population density, low education levels and a low percentage of immigrants. On the other hand, municipalities reaching the R8 are characterized by even large population (average population over 130,000 inhabitants), high population density and a high percentage of immigrant population.

Finally, sample municipalities grouped into Cluster 4 (16% of sample municipalities) are characterized by their concentration on either R5 or R8, with only a few initiatives in R6. Although the results may seem similar to those of Cluster 2, municipalities in Cluster 4 reach the highest level of accomplishment at R8, while their level of achievement in at R6 is minimal. These municipalities are larger, with an average population over 270,000 inhabitants, high population density (over 3,900 inhabitants per km²) and higher level of education, particularly in higher education. They are also considered SCs and have high percentages of immigrant populations.

Table 4 presents the normality tests, showing that all analysed variables have non-normal distributions, except for citizens over 64 years, electoral participation, and single and married citizens. Table 5 presents the results of the first difference test, enabling us to identify the determining factors that influence the e-Participation models implemented by municipalities. In Cluster 1, age is a significant differentiating factor, especially in municipalities with an ageing population (%AGE+64; p > 0.05). Therefore, we cannot reject H₂, confirming that municipalities with higher ageing population are less likely to offer advanced e-Participation tools due to the digital divide, as younger populations are generally more adept at using ICTs (Novo and Vicente, 2019).

Furthermore, the gender distribution of the population and the proportion of immigrants may also cause differences in the e-Participation models implemented, as both variables are statistically significant at the 5% level. Therefore, H7 and H8 cannot be rejected. This is in line with prior research that indicates participation channels vary depending on the gender of the participants (Schmidthuber et al., 2017; Ahmad and Kalid, 2017; Kim and Lee, 2019) as well as the role of immigrants in utilizing ICTs to engage in public affairs (Choi and Song, 2020).

Our results also indicate that participation models are more advanced in SCs. In Cluster 4, this SC variable is statistically significant at the 5% level, confirming that H4 cannot reject. This suggests that the technological environment in SCs promotes a more participatory and collaborative model of e-Participation.

Finally, Table 6 presents the results of the second difference test. The results indicate that all variables, except for the age of the population and the percentage of immigrants, influence the e-Participation models implemented by municipalities, leading us to reject H2 and H8. Regarding civil status, the percentage of widowed and separated citizens appears to influence the adopted e-Participation models, supporting H9, which suggests that these demographic groups may require more tailored participation models.

Furthermore, the statistical tests confirm that population size and density are correlated with e-Participation models. Politicians and public managers feel pressure from citizens to provide more advanced technologies to enhance citizen participation (Alcaide-Muñoz et al., 2022). As previously stated, the technological environment of a SC enhanced opportunities for better infrastructures, technologies, and resources to facilitate e-Participatory initiatives. Additionally, cities with politically active populations and high educational levels will require more technological tools for public participation. Finally, the gender and civil status of the population may influence municipalities to implement varying models of e-Participation.

Prior studies have revealed spatial and socioeconomic determinants of SCs, as well as the factors influencing citizen participation in the official Twitter accounts of municipalities (Stone and Can, 2021). However, this study fills an important gap by empirically identifying different e-Participation management models and explaining how city and citizen profiles could account for these variations.

The findings of this research suggest that public managers are under increasing pressure to adapt participation tools due to rapid urban population growth, particularly higher population density and the influx of immigrants. In response to these trends, local governments have adopted technological tools that empower citizens to collaborate and exert influence on decision-making processes to address urban challenges (the ALP has been scaled at levels 5 to 8 -Arnstein, 1969-). This has been primarily achieved through the proliferation of OG initiatives worldwide, which promote transparency and provide participation tools (Alcaide-Muñoz et al., 2022). Public managers must therefore implement OG projects with interactive tools that foster collaboration, improve public service delivery, and enhance citizens’ quality of life (Rodríguez-Bolívar, 2021).

However, the study reveals that most e-Participation initiatives that have been developed are at R5, seeking the advice of citizens but restricting their capacity to influence decision-making processes. By contrast, initiatives at R6 are scarce, and R7 is almost non-existent. Indeed, although citizens’ opinions are gathered through consultations or participatory budget platforms, the decision-making power remains with politicians and public managers. These findings suggest that although citizens’ opinions are gathered, decision-making power remains largely with politicians and public managers. The use of ICTs is often aimed at increasing the legitimacy of public actions and making citizens feel heard, but it does not translate into genuine power-sharing.

Large municipalities, particularly those branded as “SCs”, with high population densities and large proportions of qualified citizens, offer tools that favour R8, placing citizens in a position of authority over policy and managerial aspects, and empowering them to negotiate the conditions under which 'outsiders' may effect change (Arnstein, 1969). These findings support legitimacy theory, which posits that organizations adopt structures and practices that are socially acceptable and legitimate to maintain credibility and accountability (Weber, 1978). In the context of rapid urban growth and increasing citizen demands, policymakers face significant pressure to implement participatory governance models demonstrating their commitment to democratic principles. By adopting advanced e-Participation tools, these cities address residents’ immediate needs and enhance their legitimacy and trustworthiness.

The results of this study indicate that municipalities implemented advanced e-Participation platforms, especially those that allow citizens to influence decision-making (R6 and R8), tend to generate higher levels of trust in the government. SCs are better positioned to develop these platforms, which enhance the legitimacy and credibility of local government. In contexts where citizens demand greater transparency, adopting participation tools that go beyond consultation (R5) can be a crucial factor in boosting public trust. These findings emphasize the importance of moving beyond mere consultation and towards collaboration and empowerment in decision-making processes to increase citizens’ engagement and trust.

In cities with high population density and significant cultural diversity, public policies should be designed to facilitate active collaboration between citizens and authorities. The use of interactive e-Participation platforms can promote the co-creation of public policies, allowing citizens not only to offer their opinions but also to participate actively in decision-making. This leads to greater involvement and a sense of belonging, which fosters cooperation in addressing urban challenges.

As observed, demographic characteristics (such as age, education, and the percentage of immigrants) play a significant role in the adoption and success of e-Participation tools. Public managers must consider the demographic profile of their population to design accessible and effective e-Participation tools. For example, in cities with an ageing population, it is essential to offer intuitive platforms that overcome the digital divide, while municipalities with large immigrant populations could benefit from the use of multilingual platforms to ensure that all groups have a voice in decision-making processes.

The results of the difference tests further confirm stakeholders’ theory (Jensen and Meckling, 1976), showing a correlation between population size and density with e-Participation models and rungs of citizen participation -Table 6. Larger cities with higher population densities have more diverse populations that demand greater transparency and accountability. This pressure compels local governments to adopt more interactive and inclusive communication channels. The evidence corroborates previous research (Alcaide-Muñoz et al., 2022), indicating that larger cities, with more active citizenry, are more likely to develop advanced e-Participation models.

Moreover, our findings are consistent with those of other studies (Rodríguez-Bolívar, 2019), indicating that the technological environment in SCs serves as a catalyst for e-Participation, fostering collective intelligence (Khan and Krishnan, 2021). Similarly, e-Participation initiatives are more developed in cities where citizens have the necessary knowledge and skills to use them. Our research indicates that a higher percentage of citizens with higher educational levels correlate with a greater number of tools promoting higher rungs of participation.

Our research has significant practical implications for city governments and public managers. Firstly, the findings suggest that city governments need to enhance their e-Participation initiatives and leverage ETs to better engage citizens. However, the implementation of these initiatives is heavily influenced by the city profile. The vast majority of SLGs are currently positioned at R5 offering citizens only the capacity to advise local governments. This requires substantial improvements to foster deeper citizen participation and to delegate more decision-making power to the public.

Understanding the city profile is crucial for explaining how e-Participation models are managed by politicians and public managers. By identifying the characteristics of the city profile and the stage of development of e-Participation models, public managers can pinpoint which municipalities are most advanced in these initiatives. This knowledge allows them to learn from the experiences of these municipalities, understand their best practices, and implement similar initiatives more efficiently in their own context.

Local governments in municipalities characterized by high population density, significant immigrant populations, and high educational levels of the citizenry tend to make greater efforts to design sophisticated e-Participation platforms or offer social media as communication channels. These platforms facilitate greater citizen involvement in public affairs, promoting transparency and accountability, which in turn can lead to an increase in citizen trust in government for individuals who value information accessibility (Grimmelikhuijsen, 2009; Kettl, 2017; Mabillard, 2022). This trust in government also encourages citizens to have a sense of cooperation with government and in turn, take appropriate actions (Lee and Kim, 2014). Nonetheless, although the structural openness of the government through participatory policymaking marginally increases trust and confidence in public authorities (Cole, 1975; Schmidthuber et al., 2021; Campbell, 2023), it is imperative to pay more attention to the process and adopt a management capacity-building approach to effectively manage e-participation in government (Kim and Lee, 2012). Public managers and politicians should therefore consider all these attributes as critical factors when adopting open initiatives and policies.

Analysing these attributes can help public managers identify the specific needs and preferences of their citizenry, enabling them to tailor e-Participation tools accordingly. This approach serves to enhance the effectiveness of e-Participation initiatives, while also ensuring that these initiatives are inclusive and accessible to all segments of the population.

In brief, our findings about large SLGs indicate that they are not using the ICTs, technological tools, and ETs to enhance citizen participation in public affairs. Therefore, possible questions for future research are: (1) Do small-size municipalities have the needed resources to undertake open initiatives? (2) Does the administrative culture have influence on the low level of development of e-Participation initiatives in Spain? (3) Do citizens residing in cities with more advanced e-Participation models really consider that there is effective participation in public management? (4) Do more developed rungs of e-Participation really result in greater citizen collaboration and co-creation in public services? (5) Are there other factors, such as political, organizational, institutional, and so on, to better explain the different e-Participation models? Future studies should consider these questions to be analysed to gain a deeper understanding of e-Participation initiatives and their implementations, and to really know both the environmental and the contextual factors that facilitate the development of these initiatives and their perception as effective by the citizens.

Funding: The authors acknowledge the financial support from the Regional Government of Andalusia (PY20_00314 and B-SEJ-556-UGR-20).