The reflection of public values in twitter use of metropolitan municipalities: A content analysis employing the analytic hierarchy process

Abstract

The reflection of public values (PVs) could be monitored through some indicators, like institutional social media tools. The measurement of how PVs are reflected over local social media accounts is of great importance. Studies in the relevant literature are mainly focusing on the benefits of social media presence, drivers, or barriers or attitudes and expectations on social media presence. We, in this study, focus on the evaluation of whether and how selected PVs are reflected through social media accounts of metropolitan municipalities (MMs) in Turkey. By taking participation, collaboration, and transparency as the main PVs, we have formed an evaluation metric composed of five sub-criteria for each determinant. With the help of seven public administration experts holding a Ph.D. degree to form an expert graded scheme by using the analytic hierarchy process (AHP), we have employed a content analysis over Twitter accounts of 30 Turkish MMs using this expert graded scheme. We have found that Turkish MMs were found not to grasp the valuation of the selected PVs since most seem to fail to reflect them over their social media accounts to the full extent possible. Moreover, many Turkish MMs seem to have violated privacy and personal data.

Keywords

Public values twitter social media local governments metropolitan municipalities

1. Introduction

Due to the ubiquitous presence of developments in information and communication technologies (ICTs), there is a profound transformation in communication between the governments and their respective shareholders, including ordinary citizens. The transformation is well defined by the introduction of web 2.0 tools, and their fast spread into the levels of governments, including local governments. Social media tools allow people and all respective stakeholders to reach the governments, irrespective of elected or appointed public officials through a real-time scenario (Mergel, 2015a).

Public institutions can employ social media tools for several communication purposes, ranging from disclosing insensitive data to dispersing the content that is generally unreachable through offline means, particularly for the public value of transparency (Goncalves et al., 2015; Mergel, 2015a). Social media tools, as per social interaction, are collaborative and participative by default. As also for interaction among people, social media tools provide the governments with connection to shareholders, being more open to socialization, disclosing the public information, and necessary abilities to form a venue to reach common public values. Arguably, social media empowers ordinary citizens by producing a space for public discourse and discussions. Besides, as a method of democratization, social media tools support the users to release real-time information (Bertot et al., 2010). With these very characters, social media tools enable both the central government and the local governments to develop their interaction with citizens, to increase transparency, public participation, and accountability (Bonsón et al., 2012).

To extend the functionalities of social media tools in government and citizen interaction, there are several applications as;

a)
integration of information, opinions, wishes, and denouncements coming from the shareholders, including ordinary citizens (Bonson et al., 2012, Sobaci & Karkin, 2013, Sobaci, 2016),
b)
disclosing information through social media channels to promote participative and collaborative democratic practices (Noveck, 2009),
c)
increasing the transparency, providing solutions for decision-making and public policy processes, increasing the identity and trust to administration (Lazer et al., 2009),
d)
collaboration with the shareholders (Mergel, 2013b: 123; Picazo-Vela et al., 2018).

Extension of the respective use of social media tools also includes the easiness of reaching public values (PVs) (Ertaş, 2015) in addition to amelioration and the betterment of public service provision (Osimo, 2008; Sobaci & Karkin, 2013). In short, social media tools could well serve the governments, particularly local governments for the actualization of PVs. Among these PVs, some are of particular interest (Mergel, 2012; Harrison et al., 2012; Mergel, 2013a) in the context of local government social media use metrics. These PVs are public participation (Jansen, 2012), transparency (Karkin & Janssen, 2014; Mergel, 2015b), and collaboration and co-creation in public policies and services (Harrison et al., 2012).

There is a void regarding the effects of social media tools on PVs creation, though studies are dealing with PVs and e-government applications and projects (Twizeyimana & Andersson, 2019) in general terms, including social media use in broad (Aladalah et al., 2018; Althaqafi et al., 2018; Barry, 2013; Criado & Gil-Garcia, 2019; Grube, 2017; Harrison et al., 2012; Hui & Hayllar, 2010; Parker & Bozeman, 2018; Omar et al., 2013; Osmani et al., 2014; Zavattaro & Brainard, 2019).

This paper aims to contribute to the extant literature by analyzing PVs in institutional use of social media tools. As asserted by Hui and Hayllar (2010), Twitter provides the local governments, metropolitan municipalities (MMs) in our case, with opening new channels for collaboration and co-creation. Thus this paper prefers to focus on analyzing PVs through institutional Twitter accounts. In determining the PVs, we picked participation, collaboration, and transparency and their respective sub-criteria up as main research dimensions.

In this paper, in the first part, we give the conceptual and theoretical backgrounds. Thereafter we present a concise literature review. Then, we elaborate on our methodological approach on how to analyze the generation and reflection of selected PVs through a two-parted analysis in the methodology section. Finally, we make a discussion on the findings and draw some conclusions.
2. Conceptual and theoretical backgrounds

In this part, we aim to present conceptual and theoretical backgrounds including social media tools as a part of broader ICTs and public values (PVs) before getting into the literature analysis. After defining these interconnected conceptualizations, we also aim to introduce and explain the importance of selected PV elements of the proposed analysis as participation, collaboration, and transparency before getting detailed through the literature review.

2.1 Public values (PVs)

As is known, due to the reflexive nature of the discipline, public administration (PA) is no indifferent to the theoretical changes and transformations within the discipline and close scholarships and policy implementations around. As there is a continuous change, particularly due to the developing nature of ICTs, there is a need for fresh perspectives by traditional theories or novel theoretical frameworks as alternative lenses. Public value management (PVM), novel interpretations of governance, and new public service are among these theoretical lenses to interpret and drive the changes and transformations occurring.

As is known, classical public administration theory and the new public management (NPM) approach as a reaction to classical PA theory has dominated the discipline over decades. Since the 1980s, NPM has been recommended as a theoretical lens to capture and drive the changes and transformations so far. Nowadays it is clear that NPM has proven itself as having internal inconsistencies (Dunn & Miller, 2007; Diefenbach, 2009) and is generally incompatible with the realities of the public sector (Karkın, 2015: 249). Lately, it seems that there are alternative paradigms and approaches compatible to capture and drive the changes and developments in the public sphere. Among them, PVM is taken for granted as aligned with the notion of publicness, the term that was asserted by Bozeman (1984). Starting from the seminal works of Moore (1994; 1995), there is an increasing interest in the notion of PVM and PVs more generally. Moore (1994; 1995) argued that managers in the public sector are supposed to create PV to tackle their strategic challenges faced as similar to private value created by private sector managers. Moore, although coined the public value in the PA context, never laid down a clear and definite conceptualization on PV, instead, he left the defining job to the discretion of the public managers on purpose (1994: 297; 1995: 17–23).

In coming years later, there are many variations and various interpretations concerning PV. At the forefront, there are studies of Bozeman (2007: 132) and Jørgensen and Bozeman (2007: 13–14). While Bozeman (2007) discusses the notion of public values, contrary to the singular conceptualization argued by Moore (1994; 1995), Jørgensen and Bozeman (2007) make their seminal universe of PVs inventory enlists more than seventy values under seven nested dimensions. Jørgensen and Bozeman (2007) argue that the public sector should place the provision of public interest and serving the common good at its focal concern, thus positing an implication that the NPM paradigm fails to meet the publicness dimension. For this very purpose, Nabatchi (2012) argues that “to put the “public” back in public values research by theorizing about the potential of direct citizen participation to assist with identifying and understanding public values”. Thus, public managers are supposed to focus their efforts to act strategically (Benington & Moore, 2011: 1) and direct public employees for the generation of PVs.

One of the main characteristics that the PV framework provides with the stakeholders is engagement (Nabatchi, 2012), collaboration (Picazo-Vela et al., 2018), and satisfaction of the relevant stakeholders (Karkin, 2015: 259–260). It is because, as promptly asserted by Benington (2011: 46), the PVs are supposed to be created in a multi-actor environment, including “private sector, voluntary sector, and informal community organizations as well as governments”, thus public sector does not constitute sole or primary locus or focus. In their study, Picazo-Vela et al. (2018) have proven that collaboration with public and private stakeholders has conditional mediating effects on the generation of PVs.

According to Stoker (2006: 42), PVM has driven a change in comprehension and conceptualization of public service to redefine or to deal with “the challenges of efficiency, accountability, and equity” in a broader term. Thus, PVM seems to be functional to provide legitimacy and co-operated/collaborated management as a derivative of networked governance. Bozeman (2019), in a similar venue, asserts by citing modern PA theorists that “governance emanates from citizens and their values” (p. 817). As per PVM, the bottom-up approach, not a top-down one, and complex multi-policy networks are at stake concerning the decision-making process in the public sphere (Stoker, 2006: 41–56) in a way to enhance stakeholder participation.

In this context, public managers should consider stakeholders, all relevant parties including citizens as stakeholders in public service design and policy implementation processes according to PVM. Thus, in defining the policy priorities, elected public sector officials and appointed public managers should promote collaboration, participation, and co-creation with stakeholders and follow primarily the interests of citizens (Algan, 2015: 5). Therefore, the definition of PV infers the participation of various stakeholders, including the relevant parties by default (McBride et al., 2019). For this very purpose, Karkin and Janssen (2014) argue that utilizing ICTs in all kinds of public service provision should aim to PVs generation, where and when possible. For example, the inclusion of ICTs in public sector communication with stakeholders through social media tools is supposed to foster or enhance the generation of PVs. Bannister and Connolly (2014: 119) assert that “ICTs can and do have transformational impacts on public values” but they also argue that these so-called impacts are not executable by default.

PV could be seen not just as the yields generated but as a reflection created by multi-dimensional and collaborated choices to reach trust or justice (O’Flynn, 2007). Bannister and Connolly (2014) assert that the inventory of PVs developed by Jørgensen and Bozeman (2007) is insightful but eclectic, thus having some lacks concerning practical dimensions. In a later study, Jørgensen and Vrangbæk (2011) have laid the dynamics of PVs over time. In that study, the authors have identified three types of change mechanisms concerning the PVs. In a similar venue, Spicer (2009) argued for the reconciliation among values, particularly conflicting ones, by taking attractions to “particular relevance of legal reasoning to public administration when dealing with the type of conflicts among rival values” (p. 538). Bannister and Connolly (2014) also argue that it is possible to develop various lists regarding the PVs. As it is previously asserted, the PV concept is about to focus on meeting the stakeholders’ needs (Cordella & Bonina, 2012). Nonetheless, according to Cordella and Bonina (2012), it is a complex duty to design performance criteria, from the perspective of governments, to meet the stakeholders’ needs, including those of citizens. Therefore, they argue that ICTs would have effects on PV but varieties are possible, which would result in various sets of values. For this very reason, various sets of values would require various public expectations. For this very purpose, we want to see whether and how MMs in Turkey are eager to take the input from their stakeholders, including citizens in a collaborative and participative manner through their Twitter accounts. As asserted by Clark (2021), in collaborative governance what is important is “which stakeholders are invited to participate (and how), but at who designs the participation agenda in the first place” (p. 199). PVs are not supposed to denote just the substantive issues, but they also pertain to procedural issues for governance criteria like transparency, participation (Alford et al., 2017), and collaboration. Pang et al. (2014) enlist public engagement capability, co-production capability, and resource-building capability among “five organizational capabilities mediate the relationship between IT resources and public value” (p. 187).

Taking a social media perspective, as this study focuses on, it is feasible to argue that social media tools have proven to change the structure of public administration and their capacity to generate PVs (Bannister & Connolly, 2014; Criado & Gil-Garcia, 2019). Smart technologies have the potentiality to generate PVs, besides, provide public services through collaboration (Criado & Gil-Garcia, 2019). For this very reason, governments have developed innovative strategies based on sophisticated and novel means to generate PVs when facing complicated socio-technical issues (Gil-Garcia, 2015). It is possible to argue that an innovative government is by default supposed to achieve better means and to endeavor the generation of PVs (Bryson et al., 2014; Crosby et al., 2017). Twizeyimana and Andersson (2019), who has focused on the generation of PVs in e-government, assert that values generated through e-government applications could have affected governments to employ the social media tools. Douglas and Meijer (2016) argue that perspectives and approaches to transparency, collaboration, and participation would not be meaningful if any of these PVs persist alone, rather yield a significant value if they are aimed together. For this very purpose, we have aimed to examine these three values and their sub-criteria after making a literature review.

The specific reasons why we have picked participation, collaboration and transparency among other possible PVs could be justified by citing first country-specific reasons. We are focusing on MMs in a particular country case, where there is lately a focus on citizen participation, a need to foster collaboration, and a widely argued self-promotion through social media accounts of local governments (Sobaci & Karkin, 2013). Moreover, social media tools are best defined by the Web 2.0 concept, which is widely fed by stakeholders’ input and collaboration. Transparency is still a very important social value, not for developing countries but also the developed ones. Thinking with the corruption, transparency is taken for granted at the forefront by default. Therefore, this study focuses on whether and how the generation of these values is made through Twitter accounts of MMs in a developing country case.

2.1.1 Participation

With the help of ICTs, local governments can get into communication with the citizens by allowing and promoting the stakeholders into interaction, including the citizens, to send direct messages (DMs) (Sevin, 2016). This could be done by allowing DM to send or mentioning the interlocutors or stakeholders in general by using “@”. By mentioning “@”, the interlocutors could be taken into interaction, thus promoting the stakeholders to get into interaction with the institution at stake (Medina & Diaz, 2016). Governments at national or local levels could get the opinions of the citizens concerning prospective projects or activities through social media tools, for both feeding them and sharing the results with them (Bhagawati, 2020; DePaula & Dincelli, 2016; Mergel, 2012). Thus, governments or local governments should be in two-way communication with the citizens to boost participation and interaction (Bonsón et al., 2019).

Two-way communication could be realized through replying to questions and wishes coming from the citizens (Moon, 2002). A fellow citizen who has sent a DM to the local government or mentioning it is in an expectation to be replied to in time (Medina & Diaz, 2016). For this very purpose, local governments should respond to the queries and messages in about 2–24 hours, depending on the content and requiring solutions (Digital.gov, 2013). If the queries or messages are not responded to in time may cause the interlocutor to get into a feeling of time-wasting or cause a decrease in eagerness (Generalitat de Catalunya, 2017). It is crucial that governments or local governments should give a detailed or effective response to the queries sent by the citizens, or advised to contribute to the discussions under postings made by the government or local government (Generalitat de Catalunya, 2017; Mietzner et al., 2015). In essence, local governments should be eager in interacting with the stakeholders and promote them (Aladalah et al., 2018; Bonsón et al., 2019) getting into participation.

2.1.2 Collaboration

Among other means, it is possible to measure the efficiency of local government via Twitter by evaluating the number of followers or population in general (Agostino, 2013; Haro-de-Rosario et al., 2016). Although followers or users by mentioning can get into communication with local governments through institutional accounts, theoretically, every tweet is just delivered to every follower but it is not plausible that every follower read the tweet in real-time, which affects the participation and collaboration measurements (Haro-de-Rosario et al., 2016). Thus, local governments are supposed to innovate means to develop collaboration with stakeholders, including citizens. Citizens could be included in decision-making processes, participatory budgeting, and designing strategies and services through social media tools, by which government-citizen interaction or collaboration in co-creation could be achieved (Althaqafi et al., 2018).

Any postings or shares of the local governments should be encouraged to be posted in various languages or dialects. Thus, it is possible to reach the mass, increase the interaction, and feed the stakeholders back. This could promote the residents, not only citizens, to get into communication with the local government (Gülaslan, 2018). Local governments could also develop collaboration with private sector firms, NGOs, or other central or local public institutions by using social media tools (Sandoval-Almazan & Gil-Garcia, 2012). If local governments could include other actors or stakeholders into their decision-making processes and public policy design and implementation processes, then it would help to achieve the generation of PVs as transparency, collaboration, and participation (Chun et al., 2010; Sivarajah et al., 2015).

2.1.3 Transparency

Local governments should share the necessary information on council meetings through various means, including institutional social media accounts. This information should include time, place, and the council meeting agenda (Generalitat de Catalunya, 2017). Local governments should also develop alternative channels or mediums to inform the citizens who could not attend to watch the council meetings for various reasons (Omar et al., 2013). It is because the disclosure of information on municipal activities or projects helps to increase the quality of the government’s business in addition to boost trust. Through that, elected or attained officials could measure the collective public opinion on municipal performance (Althaqafi et al., 2018).

It is asserted in the relevant literature that central government and local governments might make budgetary information reachable and accessible through online means if they aim to combat corruption and degeneration (Criado & Gil-Garcia, 2019). That governments and local governments inform the citizenry on legal changes and amendments over social media tools may help to provide the administration with additional transparency (Pirannejad, 2011). Table 1 presents the main evaluation criteria in the Twitter analysis of Turkish MMs gathered through analyzing the literature.

Table 1
Studies used in literature review for picking the main criteria used in developing the evaluation metrics

Participation	Sevin, 2016; Medina & Diaz, 2016; Bhagawati, 2020; DePaula & Dincelli, 2016; Mergel, 2012; Bonsón, Perea & Bednárová, 2019; Moon, 2002; Digital.gov, 2013; Generalitat de Catalunya, 2017; Mietzner, Vandenhouten & Hartmann, 2015; Aladalah, Cheung, & Lee, 2018; Bonsón, Perea, & Bednárová, 2019
Collaboration	Agostino, 2013; Haro-de-Rosario, Sáez-Martín, & del Carmen Caba-Pérez, 2016; Althaqafi, Rahim, & Foster, 2018; Gülaslan, 2018; Sandoval-Almazan & Gil-Garcia, 2012; Chun, Shulman, Sandoval, & Hovy, 2010; Sivarajah, Irani, & Weerakkody, 2015; Santoso, Rinjany, & Bafadhal, 2020
Transparency	Generalitat de Catalunya, 2017; Omar, Scheepers & Stockdale, 2013; Althaqafi, Rahim, & Foster, 2018; Criado & Gil-Garcia, 2019; Pirannejad, 2011

3. Literature review

As we argue in the previous sections that there is a causal relationship between the generation of PVs, the use of ICTs in general, and the use of social media tools in particular. Thus in here, we aim to provide the reader with a specialized causal interaction among ICTs, social media, and the generation of selected PVs.

Studies in the relevant literature argue that the relationship between governments and all relative stakeholders, particularly concerning the design and implementation of public services and policies, is changing for some time (Kekez et al., 2018; Brandsen et al., 2018; Lember, 2018; Aschhoff & Vogel, 2018). Thus, here we have some arguments.

First, it is possible to argue that we are about to leave behind a steady/stable government comprehension for a dynamic one on both conceptual and theoretical terms (Karkin et al., 2018; Karkin & Gurses, 2021). Governments as public service designers or mediators and government agencies as service producers are supposed to co-create and collaborate with all regarded shareholders, including citizens. Second, governments are seemingly forced to innovate feasible and reasonable solutions to challenges and transformations they experience and to meet the expectations arising from stakeholders. Among all other available options, social media means and co-created and collaborated generation of PVs may serve well for meeting these ends. Third, possible inputs from all possible shareholders, including citizens, should be seen as constructive, though conflicting demands or inputs might be provided. Among others, governments need increasing legitimacy and trust, thus, shareholder inputs both help to raise legitimacy and trust in government business and to generate PVs. Thus, we expect the governments including local governments to be constructive and promoters for stakeholders to get into interaction through various means, particularly through accessible, feasible and common means like social media tools. In this context, fourth, we argue that ICTs and social media tools may serve well here since it augments stakeholders’ participation as such networked governance might smoothly operate. There is a growing body of scholarship that argues ICTs-driven means like social media tools force governments to provide stakeholders with more efficient, interactive, responsive, and trustworthy services (Kelly et al., 2002; Misuraca & Viscusi, 2015; Ravishankar, 2013).

It is eloquently emphasized that the use of ICTs helps with “improving efficiency and introducing innovation in the delivery of government services, enriching governments’ relationships with citizens, and strengthening trust in, support for and legitimacy of government organizations” (Yıldız & Saylam, 2013: 143). In such a context, augmented channels of public participation enabled by social media tools have the potential to transform the existent unequal relationship between governments and their stakeholders through collaborative processes and networked governance, thus help to generate PVs. Similarly, Misuraca and Viscusi (2015) underscore the potentials of ICTs to create PVs. They emphasize that ICTs-related reforms should be evaluated based on their ability to convey the expected value to the public at large and meeting the collective expectations at stakeholders’ sides. Thus, by use of ICTs in general, governments could have an opportunity to improve processes of public service design and provision effectively but also interact through more collaborative means with stakeholders (Pereira et al., 2017). In this vein, Liu and Yuan (2015) assert that the citizens have proven to have more information. Thus they tend to behave more interactive and are eager to add to public policymaking processes through online means, including collective and real-time discussions on social media.

Some previous studies have focused on particular PVs to be created by ICTs. ICTs-led initiatives are effective in promoting citizens for collaboration and co-creation through crowdsourcing (Lember et al., 2019; Paletti, 2016; Cordella & Paletti, 2017). Thus, the use of ICTs should be considered functional in collaboration models (Ravishankar, 2013). Twinomurinzi et al. (2012) find that employing ICTs for the generation of PVs by improving public engagement has proven successful, particularly by developing a shared common space fostering community engagement with government. Twinomurinzi et al. (2012) demonstrate that in the communities of South Africa, ICTs have proven themselves functional in organizing the community members to gather in the same area, sharing their ideas and opinions with the government, and providing the local community with a feeling that they are appreciated. In another related study, Karkin et al. (2018) assert that their systematic review indicates the promotion of PVs through ICTs use in a variety of ways and manners. Thanks to improved public engagement in the design and implementation of public policy-making, Pang et al. (2014) argue that the generation of PV is strongly supported by ICTs in such a way that social networking sites and other relative tools provide citizens with an array of possibilities to quickly, effectively, and timely engage in public decision-making processes. Besides, ICTs present impressive tools and manners concerning the interaction of people of similar interests. Accordingly, ICTs-led information exchange and intra- and inter-collaboration across different organizations foster the progress of some competing PVs, including responsiveness and cost-savings (Pang et al., 2014: 199).

The relationship between social media and pre-defined PVs has been examined in a systematic literature review study consisting of 23 papers by Irfan et al. (2019). Among others, the paper finds that most of the papers examined focus on the developed Western countries while relatively fewer papers aim at examining the developing countries thus asserts the paper that there is a void regarding social media use and generation of PVs in developing countries. The studies on social media use in local governments put some PVs forward as transparency, participation, and collaboration (Haro-de-Rosario et al., 2016; Gruzd & Roy, 2016; Sobaci, 2016). In line with these PVs, studies are focusing either on the reflection of these values on institutional social media accounts, or how these PVs are reflected on the interaction between local governments and citizens or co-creation processes of public service provision (Khasawneh & Abu-Shanab, 2013; Nica et al., 2014; Picazo-Vela et al., 2012; Sevin, 2016).

It is asserted that local governments could employ two-way communication that would be provided by social media tools in their interaction with all the stakeholders, including citizens, to develop relations with their environments (Macintosh, 2004; Bonson et al., 2012; Mergel, 2013c; Lee et al., 2019).

Although there are studies on PVs and social media use, there is a void regarding the studies that particularly focus on measuring some important PVs as transparency, participation, and collaboration. There is also a need to develop some kind of metrics to evaluate the PVs embedded in social media use, particularly over institutional accounts. Thus, in this study, we determined transparency, participation, and collaboration as the main evaluation criteria, of whose sub-criteria have been identified through the analytic hierarchy process (AHP) by a group of experts holding a Ph.D. in the public administration field. In addition to that, by employing a content analysis, institutional Twitter accounts of 30 Turkish MMs have been inspected by two experts holding an MSc. degree in the public administration field. We have graded the MMs according to data gathered by content analysis and the AHP process. This paper contributed to the extant literature as twofold: first, we have determined some crucial points for social media use and interaction with citizens concerning the generation of some PVs. Second, we have evaluated and produced metrics through an expert panel for some PVs for institutional Twitter accounts of local governments, particularly metropolitan or greater municipalities.

4. Methodology

We have chosen to analyze the institutional Twitter accounts of MMs in Turkey. This is due to some practical and useful reasons. First, Twitter is mainly designed as a blogging site, though any tweet being short including 128 characters when compared to other commonly preferred social media tools in Turkey like Facebook, Instagram, or YouTube which are based mainly on visual elements. Second, Twitter has proven itself to make the business models of local governments better in addition to its shown capability to provide local governments with collaboration concerning stakeholders by opening new channels of interaction (Hui & Hayllar, 2010). Third, as asserted by Mossberger et al. (2013), local governments, particularly metropolitan municipalities, are increasingly preferring to use Twitter. As a final cause for selection, Twitter is functional in providing the researchers to gather and analyze data. As previously stated, we have chosen to focus on MMs rather than municipalities or special provincial administrations since MMs are financially powerful to adapt to technological changes and transformations, have proven technical expertise in addition to their innovativeness (Moon, 2002; Bonsón et al., 2012; Bonsón et al., 2015).

In the study, we have calculated the scores of institutional Twitter accounts of Turkish MMs concerning participation, collaboration, and transparency by analyzing the main criteria and their sub-criteria employing two different methods. The two different methods that we have picked in implementing and finalizing the research are the AHP and the content analysis. The AHP method, first coined by Myers and Alpert in 1968, was modeled by Saaty in 1977 (Özgüven, 2011). As a combination of qualitative and quantitative method (Chen & Wang, 2010: 697), the AHP is mostly preferred method, particularly for issues as planning, picking the best alternative, resource allocation, and multi-criteria decision-making (Haşıloğlu & Süer, 2008; Raharjo et al., 2009: 834; Süer, 1993: 86–87; Vaidya & Kumar, 2006: 1–2). Additionally, the AHP is functional to decide on conditions that are multi-faceted and include multi-criteria due to its potential variety regarding decision-making processes. Its main strength lies in combining quantitative and qualitative dimensions (Byun, 2001; Kuruüzüm & Atsan, 2001; Saaty, 2016). Decisions in the AHP method could be reached by an evaluation after comparing the pre-determined criteria and sub-criteria via a paired comparison (Byun, 2001). The nominal values from 1 to 9 in paired comparisons are used to finalize the values for decisions in the AHP (Saaty, 2016).

As the subject of the research includes subjective elements, in other words, the main criteria and their sub-criteria could have differentiated functionalities for various communal stakeholders (Douglas & Meijer, 2016), we have preferred to use the AHP method in measuring the weights of Web 2.0 criteria employed in analyzing the institutional Twitter accounts of the Turkish MMs. As required by the AHP method, data used in the research should have been gathered from expert views (Hasiloglu, 2012). For this purpose, we have employed 7 experts in the field. Two of them are full professors, one is associate professor, and the rest have assistant professor titles. Five of them are from the branch of administrative sciences while two of them are from the branch of local governments, urbanization, and environmental policies in the department of political sciences and public administration.

The data, aligned with the scale and hierarchical structure in Fig. 1, have been gathered from each of the experts and integrated into one metrics by having their geometrical average.

Figure 1.

The hierarchical structure of the AHP regarding the criteria and sub-criteria.

4.1 Calculation of weights for the evaluation criteria

4.1.1 The weight of the main criteria

Data regarding the paired comparison of main criteria are given in Table 2. The main evaluation criteria, as asserted in previous sections, are participation, collaboration, and transparency. Data in the matrix has been acquired by calculating their geometrical average of opinions gathered through the expert panel. As we asserted earlier in the previous parts, we asked the expert panel to compare the main criteria at a dual base (i.e. to compare the relative importance between participation and collaboration, participation and transparency, and between collaboration and transparency). We have graded the expert evaluations according to a dual comparison scale ranging from 1/9 (0.111) to 9 through AHP. For example, the grade of 3.173 at the line of participation means that the experts have evaluated the participation criterion relatively more important than that of the collaboration criterion. The grade of 1.287 means the comparison between participation and transparency is more or less equal but participation is slightly important than that of transparency (i.e. if the grade is 1.00, then it means the dual comparison between is equal). In the dual comparison, the cells above and below the diagonal line, that are represented 1.00, are symmetrical as a matrix, thus symmetrical cells are reciprocally equal (i.e. 1/3.173 is equal to 0.315, thus if 3.173 is multiplied with 0.315, it is equal to 1.000). The explanation above is valid for all other similar comparison tables (Tables 4, 6, and 8 respectively).

Table 2
The dual comparison matrix of the main evaluation criteria

	Participation	Collaboration	Transparency
Participation	1.000	3.173	1.287
Collaboration	0.315	1.000	0.397
Transparency	0.777	2.521	1.000

Table 3

The normalization matrix for the main criteria and their relative weights

	Participation	Collaboration	Transparency	W
Participation	0.478	0.474	0.479	0.477
Collaboration	0.151	0.149	0.148	0.149
Transparency	0.371	0.377	0.373	0.374

The normalization matrix and the weights for the main criteria are given in Table 3. As is seen from the weightiness vector in the last column, the participation criterion has the most weighted indicator with an effect factor of 47.7%, the second comes transparency indicator with 37.4%, and the third is the collaboration criterion with 14.9%. Table 3 is a follow-up table of Table 2. This table aims to acquire and present the W vector represented in the last column. The shown grades in each column are normally equal to 1.000, thus it is called the normalization table. We have gained the weight vector (W) from the arithmetic mean of the grades represented in the regarded lines. Thus, we have shown that all the relative weights of inner factors of the main criteria are equal to 1.000. The above explanation is valid for all other normalization tables (Tables 5, 7, and 9 respectively).

We have also checked the CR ratio for the consistency of data given in the paired comparison of the main criteria. Presenting the evaluated consistency, the CR ratio is under 10% (CR $=$ 0.000), thus it is proven that the data given in the paired comparison of the main criteria is consistent.

4.1.2 The weight for sub-criteria of participation

Table 4 presents the data on the paired comparison for sub-criteria of participation. The sub-criteria are determined by the expert panel as permission to send DMs, conducting questionnaires, a two-way interaction, feeding the messages back, and satisfying and inspiring replies. Data on the matrix has been gathered through a calculation of the geometrical average of replies taken from the expert panel.

Table 4
The matrix for the paired comparison for sub-criteria of participation

	Permission to send DMs	Conducting questionnaires	A two-way interaction	Feeding the messages back	Satisfying and inspiring replies
Permission to send DMs	1.000	3.267	0.404	0.237	0.242
Conducting questionnaires	0.306	1.000	0.378	0.243	0.227
A two-way interaction	2.474	2.646	1.000	0.741	1.274
Feeding the messages back	4.220	4.111	1.350	1.000	0.349
Satisfying and inspiring replies	4.128	4.397	0.785	2.863	1.000

Table 5

The normalization matrix and weights for sub-criteria of participation

	Permission to send DMs	Conducting questionnaires	A two-way interaction	Feeding the messages back	Satisfying and inspiring replies	W
Permission to send DMs	0.082	0.212	0.103	0.047	0.078	0.104
Conducting questionnaires	0.025	0.065	0.096	0.048	0.074	0.062
A two-way interaction	0.204	0.172	0.255	0.146	0.412	0.238
Feeding the messages back	0.348	0.267	0.345	0.197	0.113	0.254
Satisfying and inspiring replies	0.340	0.285	0.200	0.563	0.323	0.342

Table 5 gives the normalization matrix and their relative weights for the sub-criteria of participation. As is seen from the last column about weight vector in Table 5, the sub-criterion of “satisfying and encouraging replies to the questions and queries sent by the citizens” has the most weight (34.2%). In the second place comes the sub-criterion of “replying to the questions and queries in a reasonable time” (25.4%), and in the third place is the sub-criterion of “whether or not metropolitan municipality is in a two-way communication and interaction with the citizens” (23.8%).

We have also checked the CR ratio for the consistency of data given in the paired comparison of sub-criteria of participation. Presenting the evaluated consistency, the CR ratio is under 10% (CR $=$ 0.091), thus it is proven that the data given in the paired comparison of participation is consistent.

4.1.3 The weight for sub-criteria of collaboration

Table 6 presents the data on the paired comparison for sub-criteria of collaboration. These sub-criteria are determined by the expert panel as the number of followers, continuity in following the citizen requests and concerns, interaction and communication with the shareholders, posts in different languages and dialects, collaboration with stakeholders. The data on the matrix has also been gathered through a calculation of the geometrical average of replies taken from the expert panel.

Table 6
The matrix for the paired comparison for sub-criteria of collaboration

	Number of followers	Continuity in following the citizen requests and concerns	Interaction and communication with the shareholders	Posts in different languages and dialects	Collaboration with stakeholders
Number of followers	1.000	0.187	0.129	1.138	0.168
Continuity in following the citizen	5.338	1.000	0.188	3.873	0.590
requests and concerns
Interaction and communication	7.733	5.332	1.000	5.046	2.246
with the shareholders
Posts in different languages and	0.879	0.258	0.198	1.000	0.171
dialects
Collaboration with stakeholders	5.966	1.694	0.445	5.836	1.000

Table 7

The normalization matrix and weights for sub-criteria of collaboration

	Number of followers	Continuity in following the citizen requests and concerns	Interaction and communication with the shareholders	Posts in different languages and dialects	Collaboration with stakeholders	W
Number of followers	0.048	0.022	0.066	0.067	0.040	0.049
Continuity in following the citizen	0.255	0.118	0.096	0.229	0.141	0.168
requests and concerns
Interaction and communication	0.370	0.629	0.510	0.299	0.538	0.469
with the shareholders
Posts in different languages and	0.042	0.030	0.101	0.059	0.041	0.055
dialects
Collaboration with stakeholders	0.285	0.200	0.227	0.345	0.240	0.259

Table 7 gives the normalization matrix and their relative weights for the sub-criteria of collaboration. As is seen from the last column about weight vector in Table 7, the sub-criterion of “that the metropolitan municipality is getting into interaction with the citizens” has the most weight (46.9%). In the second place lies the sub-criterion of “that the metropolitan municipality is developing the collaboration with the stakeholder (i.e. NGOs, private sector and other public institutions)” (25.9%), and the third place is for the sub-criterion of “whether or not citizens follow the postings of the metropolitan municipality through Twitter account” (16.8%).

We have also checked the CR ratio for the consistency of data given in the paired comparison of collaboration’s sub-criteria. Presenting the evaluated consistency, the CR ratio is under 10% (CR $=$ 0.053), thus the data given in the paired comparison of collaboration is consistent.

4.1.4 The weight for sub-criteria of transparency

Table 8 presents the data on the paired comparison for the sub-criteria of transparency. The sub-criteria of the transparency criterion is determined by seven experts in the evaluation panel as the information about municipality council meetings, the publication of municipality council resolutions, information about municipal activities, respective postings on budget expenditures, and announcements about changes in regulations. The data on the matrix has also been gathered through a calculation of the geometrical average of replies taken from the expert panel.

Table 8
The matrix for the paired comparison for sub-criteria of collaboration

	Information about municipality council
meetings	The publication of municipality council resolutions	Information about municipal activities	Postings of budget expenditures	Announcements about changes in regulations
Information about municipality	1.000	0.342	0.510	0.466	0.426
council meetings
The publication of municipality	2.921	1.000	0.882	1.402	0.805
council resolutions
Information about municipal activities	1.962	1.133	1.000	1.346	2.026
Postings of budget expenditures	2.147	0.713	0.743	1.000	1.445
Announcements about changes in	2.347	1.242	0.494	0.692	1.000
regulations

Table 9

The normalization matrix and weights for sub-criteria of transparency

	Information about municipality council
meetings	The publication of municipality council resolutions	Information about municipal activities	Postings of budget expenditures	Announcements about changes in regulations	W
Information about municipality	0.096	0.077	0.140	0.095	0.075	0.097
council meetings
The publication of municipality	0.282	0.226	0.243	0.286	0.141	0.235
council resolutions
Information about municipal activities	0.189	0.256	0.276	0.274	0.355	0.270
Postings of budget expenditures	0.207	0.161	0.205	0.204	0.253	0.206
Announcements about changes in	0.226	0.280	0.136	0.141	0.175	0.192
regulations

Table 9 gives the normalization matrix and their relative weights for the sub-criteria of transparency. As is seen from the last column about weight vector in Table 9, the sub-criterion of “that the metropolitan municipality shares the information about municipality council meetings on a continuous” has the most weight (27.0%). In the second place, there is the sub-criterion of “that the metropolitan municipality informs the stakeholders on the publication of municipality council resolutions” (23.5%), and the third place is for the sub-criterion of “that the metropolitan municipality shares respective postings on budget expenditures” (20.6%).

We have also checked the CR ratio for the consistency of data given in the paired comparison for sub-criteria of transparency. Presenting the evaluated consistency, the CR ratio is under 10% (CR $=$ 0.028), thus the data given in the paired comparison of transparency is consistent.

4.1.5 Global and local weights

Table 10 gives the global and local weights for the criteria set in the hierarchical evaluation scale. As is seen from the global weight column of Table 10, the most weighted sub-criterion of all is “that the metropolitan municipality has satisfying and inspiring replies that promote the citizens to keep asking questions” (16.3). Among all the sub-criteria, the second most weighted sub-criterion is about the responsiveness of the metropolitan municipality “that the municipality responses to the queries in a reasonable time” (12.1%). The third most weighted is the sub-criterion of “that the metropolitan municipality is in a two-way communication with the citizens” (11.3%). The least weighted among all the sub-criteria is the “number of followers for the institutional Twitter account of the metropolitan municipality” (0.7%). Another one that lies among the least weights is the sub-criterion of “Posts in different languages and dialects” (0.8%). To note, all three sub-criteria that have the most weights are the sub-criteria of participation. Table 10 is comprised of Table 3 that contains weighted vectors of the main criteria and Tables 5, 7, and 9 that contain weighted vectors of the sub-criteria. All this combination through Table 10 helped us to find the global weights for all the main criteria. The global weight for each sub-criterion is obtained by multiplying its local weight with each of the main criteria (i.e. global weight of permission to send DMs is 0.050, which is found multiplying by 0.104 with 0.477). The global weights tell us the relative importance of each sub-criteria. For example, among all the sub-criteria, the number of followers is the least important factor while satisfying and inspiring replies is the most important factor when PVs are at stake. The relative importance of all other sub-criteria could be commented on concerning its sub-category and all other main criteria categories. As is seen from the last column, the vector for all global weights is equal to 1.000.

Table 10
Local and global weights

		Local weights	Global weights
Participation	0.477
	Permission to send DMs	0.104	0.050
	Conducting questionnaires	0.062	0.029
	A two-way interaction	0.238	0.113
	Feeding the messages back	0.254	0.121
	Satisfying and inspiring replies	0.342	0.163
Collaboration	0.149
	Number of followers	0.049	0.007
	Continuity in following the citizen requests and concerns	0.168	0.025
	Interaction and communication with the shareholders	0.469	0.070
	Posts in different languages and dialects	0.055	0.008
	Collaboration with stakeholders	0.259	0.039
Transparency	0.374
	Information about municipality council meetings	0.097	0.036
	The publication of municipality council resolutions	0.235	0.088
	Information about municipal activities	0.270	0.101
	Postings of budget expenditures	0.206	0.077
	Announcements about changes in regulations	0.192	0.072

4.2 Content analysis

The second method preferred in the research is content analysis (CA), which is one of the qualitative research methods. Content analysis is among the most preferred research method in social sciences, particularly in communication sciences and behavioral sciences like journalism and psychology (Lombard et al., 2012; Gondim & Bendassolli, 2014). Lately, CA is widely used for studies in public opinion research, monitoring the markets, or comprehension and evaluation of political tendencies in wide spectrums (Krippendorff, 2004: xiii–xiv; Neuendorf; 2017: 19). CA is a qualitative method for codifying, categorizing, and comparing the statements regarding documents, images, discourses for generating reproducible conclusions (Krippendorf, 2004; Cohen et al., 2007).

In this section of the study, we focus on the evaluation of each institutional Twitter accounts of 30 Turkish MMs according to the criteria that are determined via the hierarchical structure of the AHP. In this context, two experts coded three main criteria and their relative sub-criteria. The evaluation of institutional Twitter accounts of MMs was performed in four months between December 31, 2019, and March 31, 2020. During the evaluation process, we have examined 10353 tweets and 4453 retweets. Two experts have filled the regarding forms that contain the population and number of followers of each MMs, after making a content analysis, according to the criteria given in Table 1. Moreover, the experts have checked the responsiveness of the MMs by sending a DM or mentioning them with @regarded MM. Thus, we have cross-checked the replies, the duration for a reply, and whether a two-way interaction is provided.

We have employed Scott’s Pi for measuring reliability. As is known, the value for Pi is between 0 and 1. 0 denotes there is no consistency among the coders, 1 denotes that there is a perfect consistency (Scott, 1955: 323). The co-efficiency levels of reliability for Pi’s consistency is as follows: (Hussain et al., 2010: 105):

if Pi $>$ 0.80, then, reliability is very good, if Pi $>$ 0.70–0.79, then, reliability is good, if Pi $>$ 0.60–0.69, then reliability is satisfactory, and if Pi $<$ 0.59, then, reliability is weak.

4.3 Coefficients of the main criteria for MMs

The coders have coded the tweets and retweets in institutional Twitter accounts of MMs as “Yes”, “No”, and “Maybe/partially” according to the main criteria and their relative sub-criteria. Regarding the calculation of the coefficients, we have taken 2 as the multiplier of “Yes” while the multiplier for “Maybe/partially” is taken as 1. After deciding on procedural necessities, we have calculated the respective averages of points given by two coders according to the criteria to have the data presented in Table 11.

Table 11
Coefficients of MMs according to the main criteria

MMs	Participation	Collaboration	Transparency
Adana	4.5	2	6
Ankara	2.5	4	4
Antalya	3.5	0.5	2
Aydın	2	2	2
Balıkesir	5.5	1	3
Bursa	1.5	2	4
Denizli	1	1	3
Diyarbakır	0	1	2
Erzurum	2	2	2
Eskişehir	8	2	2
Gaziantep	2	0	2
Hatay	4	0	2
Istanbul	4.5	2	3
Izmir	6.5	2	5
Kahramanmaraş	3	1	3
Kayseri	2	1	3
Kocaeli	5.5	3	2
Konya	0	2	2
Malatya	1	1	2
Manisa	0	0	2
Mardin	0	2	2
Mersin	5.5	2	4
Muğla	7.5	2	2
Ordu	2	2	2
Sakarya	2	0	2
Samsun	2	0	1
Şaè½nlıurfa	5	3	4
Tekirdağ	4	0	4
Trabzon	2	2	6
Van	2	1	4

As is seen from the table the metropolitan municipality of Eskişehir has placed as the first with a grade of 8 among 30 Turkish MMs. When evaluated from collaboration and transparency dimensions, however, Eskişehir MM has lower grades as 2 points for collaboration (3 ${}^{\text{rd}}$ place) and 2 points for transparency (5 ${}^{\text{th}}$ place). Muğla MM has 2 ${}^{\text{nd}}$ place when evaluated from the participation dimension with 7, 5 total points. The same grades are also for Muğla MM when compared from collaboration and transparency dimensions as 2 points for collaboration (3 ${}^{\text{rd}}$ place) and 2 points for transparency (5 ${}^{\text{th}}$ place). 3 ${}^{\text{rd}}$ place in participation criterion is for Izmir MM with 6, 5 points. Izmir MM has placed as 3 ${}^{\text{rd}}$ in collaboration and placed as 2 ${}^{\text{nd}}$ in transparency. In the participation criterion, Mardin MM, Diyarbakır MM, and Manisa MM take the last place with 0 points.

When evaluated from the collaboration criterion, Ankara MM has 1 ${}^{\text{st}}$ place with 4 points among all 30 MMs. However, Ankara MM has 14 ${}^{\text{th}}$ place in participation with 2, 5 points and 4 ${}^{\text{th}}$ place in transparency with 4 points. Kocaeli MM and Şanlıurfa MM have shared 2 ${}^{\text{nd}}$ place with 3 points in the collaboration dimension. Yet Tekirdağ MM, Hatay MM, Gaziantep MM, Sakarya MM, Samsun MM, and Manisa MM have placed at the end of the collaboration list with 0 points.

In the transparency list, Adana MM and Trabzon MM share 1 ${}^{\text{st}}$ place with 6 points. Adana has been placed as 6 ${}^{\text{th}}$ with 4, 5 points while it has been 3 ${}^{\text{rd}}$ in the collaboration list with 2 points. On the other hand, Trabzon MM has found a place as 11 ${}^{\text{th}}$ with 2 points in the participation list and is in 3 ${}^{\text{rd}}$ place concerning collaboration criterion. As noted in the previous part, Izmir MM has 2 ${}^{\text{nd}}$ place concerning transparency criterion. 3 ${}^{\text{rd}}$ place has been shared by Mersin MM, Şanlıurfa MM, Tekirdağ MM, Ankara MM, Van MM, and Bursa MM with 4 points. The last place in this listing belongs to Samsun MM with 1 point.

4.4 Weights of MMs

In calculating the weights of MMs we have employed the multipliers (Participation: 0.477, Collaboration: 0.149, and Transparency: 0.374) derived from the measurement of the main criteria coefficients in the AHP. These multipliers have been used in calculating the weights by multiplying the coefficient of each MM presented in Table 12 and adding the results of all three main criteria to each other. For example, the weight of Adana MM has been calculated as;

$\displaystyle 4.6885=0.477\cdot 4.5+0.149\cdot 2+0.374\cdot 6$

Table 12
Weights of MMs

MMs	Weights
Adana	4.6885
Ankara	3.2845
Antalya	2.492
Aydın	2
Balıkesir	3.8945
Bursa	2.5095
Denizli	1.748
Diyarbakır	0.897
Erzurum	2
Eskişehir	4.862
Gaziantep	1.702
Hatay	2.656
Istanbul	3.5665
Izmir	5.2685
Kahramanmaraş	2.702
Kayseri	2.225
Kocaeli	3.8185
Konya	1.046
Malatya	1.374
Manisa	0.748
Mardin	1.046
Mersin	4.4175
Muģla	4.6235
Ordu	2
Sakarya	1.702
Samsun	1.328
Şanlıurfa	4.328
Tekirdağ	3.404
Trabzon	3.496
Van	2.599

These final values are presented in Table 12.

As is seen from Table 18, Izmir MM has acquired 1 ${}^{\text{st}}$ place with a total point of 5.268 weight gathered from all three main criteria. In second place comes Eskişehir MM with a total point of 4.862 points. Adana MM is in third place with a total weight of 4.688. Manisa MM has been placed at the end of the placement in the MM weight with a total weight of 0.748 through the CA process.

5. Discussion and conclusion

The public administration expert panel showed that participation is the most important criterion when compared to the other main criteria in the AHP evaluation. While the transparency criterion comes as the second, the criterion of collaboration is determined as the last concerning the main evaluation criteria. The expert panel has also determined that three of the sub-criteria of participation are relatively more important than those of the other two main criteria. These sub-criteria are as follow: “that the metropolitan municipality has satisfying and inspiring replies that promote the citizens to keep asking questions” (34.2%), “that the municipality responses to the queries in a reasonable time” (25.4%), and “that the metropolitan municipality is in a two-way communication with the citizens” (23.8%). There are studies in the relevant literature that also give importance to these dimensions (Aladalah et al., 2018; Bonsón et al., 2019; Generalitat de Catalunya, 2017; Medina & Diaz, 2016; Mietzner et al., 2015; Moon, 2002). Thus our findings of the AHP analysis support the importance of findings that the previous studies have found concerning the importance of participation as a PV in institutional social media use. According to the results of CA, Eskişehir MM, Muğla MM, and Izmir MM take the first three places concerning participation. These three MMs are also placed in the first four places in the placement concerning the weights of MMs, as presented in Table 17. We have found that Eskişehir MM, Muğla MM, and Izmir MM have a general inclination to reply in time to the queries and messages coming to their institutional Twitter accounts. For this very reason, we have evaluated whether these MMs are responsive to the stakeholders through two-way communication. Yet not all the queries and messages sent by the stakeholders, including citizens, have been replied to in time, or later. On the other hand, we have also observed that Izmir MM has directed some of the queries to its relevant departments, published the queries and their prospective solutions through an institutional website, and recommended all the interlocutors follow the respective website. It is concluded that MMs in Turkey have not fully captured the two-way communication over the social media accounts, though particularly Twitter allows and supports this form of interaction with the stakeholders. Nonetheless, it is hard to follow the queries up through the institutional social media accounts if the volume and number of queries increase as there are several social media accounts of MMs in use. In any case, MMs are supposed to handle the queries in time, or later on, particularly in these times as characterized by the pandemics, or under other severe natural and man-made conditions. We have also observed some situations where MMs require the interlocutors to specifically state their addresses or ID numbers via Twitter that incite the privacy issue. As is generally known, any shares that include particular information about the appealing parties are under strict and specific personal data regulations.

The expert panel evaluated transparency as the second important criterion after participation. In this context, Adana MM, Trabzon MM, and Izmir MM have the most weighted scores among all 30 MMs. The relevant literature has attributed importance to transparency as an important PV (Althaqafi et al., 2018; Criado & Gil-Garcia, 2019; Omar et al., 2013; Pirannejad, 2011). For example, as a sub-criterion, sharing information about metropolitan municipality council meetings via Twitter is mostly exercised by the Turkish MMs, some of which are broadcasting through institutional channels, or over institutional Twitter accounts. As a legal rule, municipal council meetings are open to the public if not decided to keep it closed as a council resolution. Thus, online broadcasting is a method by which council meetings are conveyed to the public at large. However, council meetings resolutions or amendments in legal provisions are not published through social media accounts, including Twitter. However, among other mediums, opening a channel via institutional social media accounts, including Twitter would increase accessibility and responsiveness as important PVs in addition to transparency. As a supporting factor for being a transparent institution, some of the MMs broadcast the public procurements through online channels, including their Twitter accounts. This could be evaluated as a positive development in direction of transparency. In general, we have observed that the MMs exploit Twitter as a medium of one-way communication and propaganda in addition to their social media tools. This observation complies with the findings of the study performed by Sobaci and Karkin (2013).

We have also observed that there is information published over Twitter on projected and prospective activities, there is no budgetary information about the announced and advertised activities. Sharing technical and budgetary information about prospective projects and activities, presenting aggregate amounts if not sharing specific and detailed amounts, would ease the measurement of MM performance in the eyes of constituents. Additionally, sharing budgetary information through alternative channels including institutional Twitter accounts would help to combat corruption and degeneracy. As is the case with participation, Turkish MMs have not fully grasped the potentiality of their social media accounts, including Twitter, in increasing the transparency of their institution.

The criterion of collaboration has been placed at the end of the AHP evaluation by the expert panel. Ankara MM, Kocaeli MM, and Şanlıurfa MM have higher scores in relative weight when compared to the other 27 MMs. Arguably, the number of followers and the increase in its volume could have some positive repercussions concerning information dispersion and collaboration. Yet it is not feasible that all of the followers see or grasp the shared content simultaneously or homogeneously, thus entering into interaction (Haro-de-Rosario et al., 2016). For this very reason, mentioning the interlocutors or respective stakeholders with “@” would take the attractions, which would trigger a snowball effect concerning participation and collaboration in a very short time interval.

We have also observed, through the CA process, that there is citizen interest in local public service provision and there is a powerful inclination to reach the MM in question through Twitter. Yet we have also found that the citizens have behaved irresponsibly in some tweets (i.e. replying to the MM citing unrelated issues, mainly ideologically driven) or having an online discussion and keeping mutual allegations with other citizens over the institutional Twitter accounts.

In this context, citizens reflect their cultural and political habits over the online mediums and Twitter is no exception to this behavior. For this very purpose, as also asserted by Picazo-Vela et al. (2012), MMs could avoid replying to the citizens in case of a potential losing control in these situations. Additionally, several factors are affecting the behaviors of the MMs over their institutional Twitter accounts as the population density of the MM (Bolívar & Muñoz, 2018), financial difficulties (Charalabidis & Theocharopoulou, 2019), technical and discursive capabilities of the regarded personnel, attitude towards the digital transformation, adaptation to technical challenges (Ogonek & Hofmann, 2018; Parente & Prescott, 1994; Rehouma, 2019).

As the result, we have analyzed the institutional Twitter accounts of 30 Turkish MMs through metrics developed to measure the PVs of participation, transparency, and collaboration with the help of a PA expert panel. This paper has a solid contribution to the extant literature as developing evaluation metrics to measure the PVs as there is an ongoing discussion in the relevant literature on whether and how to measure the PVs (Jain Gupta & Suri, 2017; Karunasena & Deng, 2012; Faulkner & Kaufman, 2018; Irfan et al., 2019).

After conducting the analysis, we have found that Turkish MMs mainly and mostly use social media tools, particularly Twitter, to promote propaganda rather than using them to promote regarded PVs for the sake of betterment of local public service provision. Thus, we recommend the prospective authors and studies to analyze the cause of this inclination both politically and culturally through a perspective of PA. Future studies are also advised to analyze why the local governments, including MMs, could not grasp the potentiality and functionality of social media tools in enriching their interaction with the citizens. Another direction for future studies should have a motivational base. It would be contributive to study whether and why citizens are would get into communication with the public institutions in general over social media tools. Under the light of these studies would the local elected and appointed people grasp and move to prepare their social media policies and guidelines and or update the present ones.

As a result, we have examined the institutional Twitter accounts of thirty Turkish MMs in a framework created by the criteria derived from the relevant literature. For this purpose, we have digitalized the PVs of participation, collaboration, and transparency besides their relevant sub-criteria. As asserted by Douglas and Meijer (2016), there could be varieties of participation, collaboration, and transparency and their sub-criteria concerning various communities. Thus, after analyzing the literature, we have consulted the experts and gathered their scores on selected PVs. We have analyzed the data provided by experts, MMs, and their interacted shareholders in this study by aiming that the similarities and differentiations of perceptions among the various actors concerning selected PVs are unveiled. By this means, we hope to help the policymakers to develop differentiated strategies toward the selected PVs through social media tools.

We have some limitations as well. First of all, we have focused on just one country case, a non-Western and developing country. Second, we have focused on MMs, not on central government institutions nor other types of local governments. We have preferred to employ just one social media tool, Twitter, thus it would be functional to focus on a blended manner with a composition of various social media tools. Finally, we have just employed two types of analytical methods on data presented through Twitter. Therefore, prospective studies could prefer to make the mixed method of analyses, including local managers and elected officials, to focus on the reasons and causes of failures that we have presented.

Nonetheless, by this analysis, we contribute to the extant literature and show some potential directions to future studies concerning the hierarchical analysis of PVs through social media by employing content analysis and the AHP method for local governments in both developed and developing countries. We plea that the AHP method could be employed to digitalize other PVs other than those preferred in this study. Finally, we argue that there is still a need for prospective studies to unveil how local governments evaluate participation, collaboration, and transparency in their business. We have proven in this study that local governments are not fully aware of the significance of PVs, thus are not so successful to reflect the PVs in their business, particularly over social media accounts. Thus, prospective studies are recommended to give some thought to reasons of whether and how local governments unveil the PVs in their business. Future studies are also recommended to deal with the motivations of local stakeholders to get into interaction with their regarded local governments. By these studies, we argue that national/local policymakers would be provided with updating their social media policies and guides for implementation.

Footnotes

Acknowledgments

The completion of the research part of the study would not have been possible without the support of the expert panel consisting of seven public administration scholars (Dr. Huseyin Gul, Dr. Nilay Yavuz, Dr. Emre Zeren, Dr. Mete Yildiz, Dr. Ozan Zengin, Dr. Volkan Gocoglu, and Dr. Levent Memis) and a coder fellow (Yasin Erkan). We, therefore, appreciate their contributions to the study, either by determining and sorting the main criteria or analyzing the tweets and retweets as a coder.

References

Agostino

(2013). Using social media to engage citizens: A study of Italian municipalities. Public Relations Review, 39(3), 232-234.

Aladalah

Cheung

, & Lee

V.C.

(2018). Towards a model for engaging citizens via gov2.0 to meet evolving public value. International Journal of Public Administration in the Digital Age, 5(1), 1-17.

Alford

Douglas

Geuijen

, & ’t Hart

(2017). Ventures in public value management: Introduction to the symposium. Public Management Review, 19(5), 589-604.

Algan

F.M.

(2015). Kamu hizmetlerinin standartlaştırılması ile verimliliğin artırılması [Increasing Efficiency by Standardizing Public Services]. Verimlilik Dergisi, 0(1), 87-108.

Althaqafi

Foster

, & Rahim

(2018). Public value creation using social media applications for the local government context: A pilot case study. Australasian Conference on Information Systems, ACIS 2018 Proceedings, pp. 1-12. Sydney, Australia.

Aschhoff

, & Vogel

(2018). Value conflicts in co-production: Governing public values in multi-actor settings. International Journal of Public Sector Management, 31(7), 775-793.

Bannister

, & Connolly

(2014). ICT, public values and transformative government: A framework and programme for research. Government Information Quarterly, 31(1), 119-128.

Barry

S.J.

(2013). Using social media to discover public values, interests, and perceptions about cattle grazing on park lands. Environmental Management, 53(2), 454-464.

Benington

(2011). From private choice to public value. In Benington

, & Moore

M.H.

(Eds.), Public Value Theory & Practice, pp. 31-51. Basingstoke, UK: Palgrave Macmillan.

10.

Benington

, & Moore

M.H.

(2011). Public Value in Complex and Changing Times. In John Benington & Mark H. Moore (Eds.), Public Value Theory & Practice, pp. 1-30. Basingstoke, UK: Palgrave Macmillan.

11.

Bertot

J.C.

Jaeger

P.T.

, & Grimes

J.M.

(2010). Using ICTs to create a culture of transparency: E-government and social media as openness and anti-corruption tools for societies. Government Information Quarterly, 27(3), 264-271.

12.

Bhagawati

(2020). Web-sites and social media technologies as implements of e-governance: A study of North East India. Indian Journal of Science and Technology, 13(31), 3188-3197.

13.

Bolívar

M.P.R.

, & Muñoz

L.A.

(2018). Political ideology and municipal size as incentives for the implementation and governance models of web 2.0 in providing public services. International Journal of Public Administration in the Digital Age, 5(1), 36-62.

14.

Bonsón

Torres

Royo

, & Flores

(2012). Local e-government 2.0: Social media and corporate transparency in municipalities. Government Information Quarterly, 29(2), 123-132.

15.

Bonsón

Royo

, & Ratkai

(2015). Citizens’ engagement on local governments’ Facebook sites. An empirical analysis: The impact of different media and content types in Western Europe. Government Information Quarterly, 32(1), 52-62.

16.

Bonsón

Perea

, & Bednárová

(2019). Twitter as a tool for citizen engagement: An empirical study of the Andalusian municipalities. Government Information Quarterly, 36(3), 480-489.

17.

Bozeman

(1984). Dimensions of ‘publicness’: An approach to public organization theory. In Bozeman, B., & Straussman, J. (Eds.), New Directions in Public Administration, pp. 46-62. Monterey, CA: Brooks/Cole.

18.

Bozeman

(2007). Public Values and Public Interest: Counterbalancing Economic Individualism. Washington D.C.: Georgetown University Press.

19.

Bozeman

(2019). Public values: Citizens’ perspective. Public Management Review, 21(6), 817-838.

20.

Brandsen

Steen

, & Verschuere

(2018). Co-creation and co-production in public services: Urgent issues in practice and research. In Brandsen, T., Steen, T., & Verschuere, B. (Eds.), Co-production and Co-creation: Engaging Citizens in Public Service Delivery, pp. 3-8. New York: Routledge.

21.

Bryson

J.M.

Crosby

B.C.

, & Bloomberg

(2014). Public value governance: Moving beyond traditional public administration and the new public management. Public Administration Review, 74(4), 445-456.

22.

Byun

D.H.

(2001). The AHP approach for selecting an automobile purchase model. Information & Management, 38(5), 289-297.

23.

Clark

J.K.

(2021). Public values and public participation: A case of collaborative governance of a planning process. The American Review of Public Administration, 51(3), 199-212.

24.

Charalabidis

, & Theocharopoulou

(2019). A participative method for prioritizing smart city interventions in medium-sized municipalities. International Journal of Public Administration in the Digital Age, 6(1), 41-63.

25.

Chen

M.K.

, & Wang

S.C.

(2010). The critical factors of success for information service industry in developing international market: Using analytic hierarchy process (AHP) approach. Expert Systems with Applications, 37(1), 694-704.

26.

Chun

Shulman

Sandoval

, & Hovy

(2010). Government 2.0: Making connections between citizens, data and government. Information Polity, 15(1, 2), 1-9.

27.

Cohen

Manion

, & Morrison

(2007). Research Methods in Education, 6

{}^{\text{th}}

Edition. London: Routledge.

28.

Cordella

, & Bonina

C.M.

(2012). A public value perspective for ICT enabled public sector reforms: A theoretical reflection. Government Information Quarterly, 29(4), 512-520.

29.

Cordella

, & Paletti

(2017). Value creation, ICT, and co-production in public sector: bureaucracy, opensourcing and crowdsourcing. In Proceedings of the 18th Annual International Conference on Digital Government Research (dg.o ’17), pp. 185-194. New York: Association for Computing Machinery. doi: 10.1145/3085228.3085305.

30.

Criado

J.I.

, & Gil-Garcia

J.R.

(2019). Creating public value through smart technologies and strategies: From digital services to artificial intelligence and beyond. International Journal of Public Sector Management, 32(5), 438-450.

31.

Crosby

B.C.

Hart

, & Torfing

(2017). Public value creation through collaborative innovation. Public Management Review, 19(5), 655-699.

32.

DePaula

, & Dincelli

(2016). An empirical analysis of local government social media communication: Models of e-government interactivity and public relations. In Kim, Y., & Liu, M. (Eds.), Proceedings of the 17th International Digital Government Research Conference on Digital Government Research, pp. 348-356. New York: Association for Computing Machinery.

33.

Diefenbach

(2009). New public management in public sector organizations: The dark sides of managerialistic ‘enlightenment’. Public Administration, 87(4), 892-909.

34.

Digital.gov. (2013). “Social Media Metrics for Federal Agencies”. Retrieved March 4, 2018, from https://digital.gov/2013/04/19/social-media-metrics-for-federal-agencies/.

35.

Douglas

, & Meijer

(2016). Transparency and public value-Analyzing the transparency practices and value creation of public utilities. International Journal of Public Administration, 39(12), 940-951.

36.

Dunn

W.N.

, & Miller

D.Y.

(2007). A critique of the new public management and the neo-Weberian state: Advancing a critical theory of administrative reform. Public Organization Review, 7(4), 345-358.

37.

Ertaş

(2015). Kamu yönetiminde sosyal medyanın kullanımı [The Use of social media in public administration]. In Ertaş, H. (Ed.), Kamu Yönetiminde Sosyal Medya Kuramdan Uygulamaya, pp. 15-42. Konya: Palet Yayınları.

38.

Faulkner

, & Kaufman

(2018). Avoiding theoretical stagnation: A systematic review and framework for measuring public value. Australian Journal of Public Administration, 77(1), 69-86.

39.

Generalitat de Catalunya. (2017). Twitter Manual for Governments. Guidelines for Public Institutions Based on the Experience of the Government of Catalonia. Barcelona: Generalitat de Catalunya.

40.

Gil-Garcia

J.R.

(2015). “Foreword”, In Bolívar, M.P.R. (Ed.), Transforming City Governments for Successful Smart Cities, pp. v-vii. Switzerland, Cham: Springer.

41.

Goncalves

Liu

Xiao

Chaudhry

Hosio

, & Kostakos

(2015). Increasing the reach of government social media: A case study in modeling government-citizen interaction on Facebook. Policy & Internet, 7(1), 80-102.

42.

Gondim

S.M.G.

, & Bendassolli

P.F.

(2014). The use of the qualitative content analysis in psychology: A critical review. Psicologia em Estudo, 19(2), 191-199.

43.

Grube

D.C.

(2017). Rules, prudence and public value: Public servants and social media in comparative perspective. Government and Opposition, 52(1), 75-99.

44.

Gruzd

, & Roy

(2016). Social media and local government in Canada: An examination of presence and purpose. In Sobaci, M. (Ed.), Social Media and Local Governments, pp. 79-94. Switzerland: Springer.

45.

Gülaslan

(2018). Kamu Yönetiminde Sosyal Medya Kullanımı ve Yönetimi [Social Media Use and Its Management in Public Administration]. Ankara: Nobel Bilimsel Eserler.

46.

Haro-de-Rosario

Sáez-Martín

, & del Carmen Caba-Pérez

(2016). Using social media to enhance citizen engagement with local government: Twitter or Facebook. New Media & Society, 20(1), 29-49.

47.

Harrison

T.M.

Guerrero

Burke

G.B.

Cook

Cresswell

Helbig

Hrdinova

, & Pardo

(2012). Open government and e-government: Democratic challenges from a public value perspective. Information Polity, 17(2), 83-97.

48.

Hasiloglu

S.B.

(2012). Evaluation of Turkish origin textile products image with fuzzy logic. Journal of Textile & Apparel, 22(3), 169-176.

49.

Haşıloğlu

S.B.

, & Süer

İ.

(2008). Beyaz eşya sanayi işletmelerinin internet üzerindeki hizmetlerinin analitik hiyerarşi süreci ile değerlendirilmesi [An evaluation of white goods manufacturers’ internet services by analytic hierarchy process]. Finans Politik & Ekonomik Yorumlar, 45(520), 87-96.

50.

Hui

, & Hayllar

M.R.

(2010). Creating public value in e-Government: A public-private-citizen collaboration framework in Web 2.0. Australian Journal of Public Administration, 69, 120-131.

51.

Hussain

Sweeney

, & Mort

G.S.

(2010). Typologies of banner advertisements’ attributes: A content analysis. Journal of Promotion Management, 16(1-2), 96-113.

52.

Irfan

Rasli

Sulaiman

Sami

, & Qureshi

M.I.

(2019). The influence of social media on public value: A systematic review of past decade. Journal of Public Value and Administrative Insights, 2(1), 1-6.

53.

Jain Gupta

, & Suri

(2017). Measuring public value of e-governance projects in India: Citizens’ perspective. Transforming Government: People, Process and Policy, 11(2), 236-261.

54.

Jansen

(2012). The understanding of ICTs in public sector and its impact on governance. 11𝑡ℎ International Conference on Electronic Government (EGOV), pp. 174-186. Kristiansand, Norway.

55.

Jørgensen

T.B.

, & Bozeman

(2007). Public values: An inventory. Administration & Society, 39(3), 354-381.

56.

Jørgensen

T.B.

, & Vrangbæk

(2011). Value dynamics: Towards a framework for analyzing public value changes. International Journal of Public Administration, 34(8), 486-496.

57.

Karkin

, & Janssen

(2014). Evaluating websites from a public value perspective: A review of Turkish local government websites. International Journal of Information Management, 34(3), 351-363.

58.

Karkın

(2015). Kamu değeri kavramı ve kamu değeri yönetimi anlayışının kamu yönetimi kuramı bağlamındaki yeri [The concept of public value and place of public value management in public administration theory]. In Koseoglu, O., & Sobaci, M.Z. (Eds.), Kamu Yönetiminde Paradigma Arayışları: Yeni Kamu İşletmeciliği ve Ötesi, pp. 249-272. Bursa: Dora Yayıncılık.

59.

Karkin

Yavuz

Sevinc Cubuk

E.B.

, & Golukcetin

(2018). The impact of ICTs-related innovation on public values in public sector: A meta-analysis. In Zuiderwijk, A., & Hinnant, C.C. (Eds.), Proceedings of the 19th Annual International Conference on Digital Government Research: Governance in the Data Age (dg.o ’18). ACM, New York, NY, USA. doi: 10.1145/3209281.3209351.

60.

Karkin

, & Gurses

(2021). An analysis of the public administration curricula in Turkey: Realities, expectations, and attitudes toward novel courses. Teaching Public Administration. doi: 10.1177/0144739421997514.

61.

Karunasena

, & Deng

(2012). Critical factors for evaluating the public value of e-government in Sri Lanka. Government Information Quarterly, 29(1), 76-84.

62.

Kekez

Howlett

, & Ramesh

(2018). Varieties of collaboration in public service delivery. Policy Design and Practice, 1(4), 243-252.

63.

Kelly

Mulgan

, & Muers

(2002). Creating Public Value: An Analytical Framework for Public Service Reform. Discussion paper prepared by the Cabinet Office Strategy Unit, UK.

64.

Khasawneh

, & Abu-Shanab

(2013). E-government and social media sites: The role and impact. World Journal of Computer Application and Technology, 1(1), 10-17.

65.

Krippendorff

K.A.

(2004). Content Analysis: An Introduction to Its Methodology. USA: Sage Publications.

66.

Kuruüzüm

, & Atsan

(2001). Analitik hiyerarşi yöntemi ve işletmecilik alanındaki uygulamaları [The analytic hierarchy process approach and its applications in business]. Akdeniz İİBF Dergisi, 1(1), 83-105.

67.

Lazer

Nebolo

Esterling

, & Goldschmidt

(2009). Online Town Hall Meetings. Washington, D.C.: Congressional Management Foundation.

68.

Lember

(2018). The increasing role of digital technologies in co-production and co-creation. In Brandsen, T., Steen, T., & Verschuere, B. (Eds.), Co-production and Co-creation: Engaging Citizens In Public Service Delivery, pp. 115-127. New York: Routledge.

69.

Lember

Brandsen

, & Tõnurist

(2019). The potential impacts of digital technologies on co-production and co-creation. Public Management Review, 21(11), 1665-1686.

70.

Liu

S.M.

, & Yuan

(2015). The evolution of information and communication technology in public administration. Public Administration and Development, 35(2), 140-151.

71.

Lombard

Snyder-Duch

, & Bracken

C.C.

(2002). Content analysis in mass communication: Assessment and reporting of intercoder reliability. Human Communication Research, 28(4), 587-604.

72.

Macintosh

(2004). Characterizing e-participation in policy-making. Proceedings of the 37

{}^{\text{th}}

Annual Hawaii International Conference on System Sciences, January 5–8 Hawaii.

73.

McBride

Aavik

Toots

Kalvet

, & Krimmer

(2019). How does open government data driven co-creation occur? Six factors and a ‘perfect storm’; insights from Chicago’s food inspection forecasting model. Government Information Quarterly, 36(1), 88-97.

74.

Medina

R.Z.

, & Diaz

J.C.L.

(2016). Social media use in crisis communication management: an opportunity for local communities? In Sobaci, M.Z. (Ed.), Social Media and Local Governments, pp. 321-335. Switzerland: Springer.

75.

Mergel

(2012). The social media innovation challenge in the public sector. Information Polity, 17(3, 4), 281-292.

76.

Mergel

(2013a). Social Media in the Public Sector: A Guide to Participation, Collaboration, and Transparency in the Networked World. San Francisco: Jossey-Bass.

77.

Mergel

(2013b). Social media adoption and resulting tactics in the U.S. federal government. Government Information Quarterly, 30(2), 123–130.

78.

Mergel

(2013c). A framework for interpreting social media interactions in the public sector. Government Information Quarterly, 30(4), 327-334.

79.

Mergel

(2015a). Social media institutionalization in the U.S. federal government. Government Information Quarterly, 33(1), 142-148.

80.

Mergel

(2015b). Opening government: Designing open innovation processes to collaborate with external problem solvers. Social Science Computer Review, 33(5), 599-612.

81.

Mietzner

Vandenhouten

, & Hartmann

(2015). Collaborative problem solving for regional development? – a new approach of public participation. In Tambouris, E., Scholl, H. J., Janssen, M., Wimmer, M.A., Tarabanis, K., Gascó, M., Klievink, B., Lindgren, I., Milano, M., Panagiotopoulos, P., Pardo, T.A., Parycek, P., Sæbø, Ø. (Eds.), Electronic Government and Electronic Participation: Joint Proceedings of Ongoing Research, Ph.D. Papers, Posters and Workshops of IFIP EGOV and EPart 2015, pp. 19-26. Amsterdam: IOS Press.

82.

Misuraca

, & Viscusi

(2015). Shaping public sector innovation theory: An interpretative framework for ICT-enabled governance innovation. Electronic Commerce Research, 15(3), 303-322.

83.

Moon

M.J.

(2002). The evolution of e-government among municipalities: Rhetoric or reality? Public Administration Review, 62(4), 424-433.

84.

Moore

M.H.

(1994). Public value as the focus of strategy. Australian Journal of Public Administration, 53(3), 296-303.

85.

Moore

M.H.

(1995). Creating Public Value: Strategic Management in Government. Cambridge, MA: Harvard University Press.

86.

Mossberger

, & Crawford

(2013). Connecting citizens and local governments? Social media and interactivity in major US cities. Government Information Quarterly, 30(4), 351-358.

87.

Nabatchi

(2012). Putting the “public” back in public values research: Designing participation to identify and respond to values. Public Administration Review, 72(5), 699-708.

88.

Neuendorf

(2017). The Content Analysis Guidebook, 2

{}^{\text{nd}}

Edition. USA: Sage Publications.

89.

Nica

Popescu

Nicolăescu

, & Constantin

(2014). The effectiveness of social media implementation at local government levels. Transylvanian Review of Administrative Sciences, 10, 152-166.

90.

Noveck

(2009). Wiki Government: How Technology Can Make Government Better, Democracy Stronger, and Citizens More Powerful. Washington, DC: Brookings Institution Press.

91.

O’Flynn

(2007). From new public management to public value: Paradigmatic change and managerial implications. The Australian Journal of Public Administration, 66(3), 353-366.

92.

Ogonek

, & Hofmann

(2018). Governments’ need for digitization skills. International Journal of Public Administration in the Digital Age, 5(4), 61-75.

93.

Omar

Scheepers

, & Stockdale

(2013). The use of social media in government 2.0 Assessed through the public value lens. 24𝑡ℎ Australasian Conference on Information Systems (ACIS), pp. 1-12. RMIT University, Melbourne, Australia.

94.

Osimo

(2008). “Web 2.0 in Government: Why and How?”. European Commission. Joint Research Centre. Institute for Prospective Technological Studies. Retrieved August 15, 2020, from https://joinup.ec.europa.eu/sites/default/files/document/2014-12/media2023.pdf.

95.

Osmani

M.W.

Weerakkody

Sivarajah

, & El-Haddadeh

(2014). The public value of social media in the UK public sector. In Janssen, M., Bannister, F., Glassey, O., Scholl, H.J., Tambouris, E., Wimmer, M.A., & Macintosh, A. (Eds.), Electronic Government and Electronic Participation: Joint Proceedings of Ongoing Research, Posters, Workshop and Projects of IFIP EGOV 2014 and ePart 2014, pp. 276-286. Amsterdam: IOS Press.

96.

Özgüven

(2011). Vakıf üniversitesi tercihinin analitik hiyerarşi süreci ile belirlenmesi [Determining the Preferences for foundation universities with analytic hierarchy process]. Dumlupınar Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 30, 279-290.

97.

Paletti

(2016). Co-production through ICT in the public sector: When citizens reframe the production of public services. In Caporarello, L., Cesaroni, F., Giesecke, R., & Missikoff, M. (Eds.), Digitally Supported Innovation. Lecture Notes in Information Systems and Organisation, pp. 141-152. Cham: Springer International Publishing.

98.

Pang

M.S.

Lee

, & DeLone

W.H.

(2014). IT resources, organizational capabilities, and value creation in public-sector organizations: A public-value management perspective. Journal of Information Technology, 29(3), 187-205.

99.

Parente

S.L.

, & Prescott

E.C.

(1994). Barriers to technology adoption and development. Journal of Political Economy, 102(2), 298-321.

100.

Parker

M.A.

, & Bozeman

(2018). Social media as a public values sphere. Public Integrity, 20(4), 386-400.

101.

Pereira

G.V.

Macadar

M.A.

Luciano

E.M.

, & Testa

M.G.

(2017). Delivering public value through open government data initiatives in a smart city context. Information Systems Frontiers, 19(2), 213-229.

102.

Picazo-Vela

Gutiérrez-Martínez

, & Luna-Reyes

L.F.

(2012). Understanding risks, benefits, and strategic alternatives of social media applications in the public sector. Government Information Quarterly, 29(4), 504-511.

103.

Picazo-Vela

Gutiérrez-Martínez

Duhamel

Luna

D.E.

, & Luna-Reyes

L.F.

(2018). Value of inter-organizational collaboration in digital government projects. Public Management Review, 20(5), 691-708.

104.

Pirannejad

(2011). The effect of ICT on political development: A qualitative study of Iran. Information Development, 27(3), 186-195.

105.

Raharjo

Xie

, & Brombacher

A.C.

(2009). On modeling dynamic priorities in the analytic hierarchy process using compositional data analysis. European Journal of Operational Research, 194(3), 834-846.

106.

Ravishankar

M.N.

(2013). Public ICT innovations: A strategic ambiguity perspective. Journal of Information Technology, 28(4), 316-332.

107.

Rehouma

B.M.

(2019). Employees’ participation in IT-Projects in the public sector. International Journal of Public Administration in the Digital Age, 6(2), 20-34.

108.

Saaty

T.L.

(2016). The analytic hierarchy and analytic network processes for the measurement of intangible criteria and for decision-making. In Greco, S., Ehrgott, M., & Figueira, J. (Eds.), Multiple Criteria Decision Analysis, pp. 363-419. New York: Springer.

109.

Saaty

T.L.

(1977). A scaling method for priorities in hierarchical structures. Journal of Mathematical Psychology, 15(3), 234-28l.

110.

Sandoval-Almazan

, & Gil-Garcia

J.R.

(2012). Are government internet portals evolving towards more interaction, participation, and collaboration? Revisiting the rhetoric of e-government among municipalities. Government Information Quarterly, 29(1), 72-81.

111.

Santoso

A.D.

Rinjany

, & Bafadhal

O.M.

(2020). Social media and local government in Indonesia: Adoption, use and stakeholder engagement. Romanian Journal of Communication and Public Relations, 22(3), 21-35.

112.

Scott

W.A.

(1955). Reliability of content analysis: The case of nominal scale coding. Public Opinion Quarterly, 19(3), 321-325.

113.

Sevin

(2016). Branding cities in the age of social media: A comparative assessment of local government performance. In Sobaci, M.Z. (Ed.), Social Media and Local Governments, pp. 301-320. Switzerland: Springer.

114.

Sivarajah

Irani

, & Weerakkody

(2015). Evaluating the use and impact of Web 2.0 technologies in local government. Government Information Quarterly, 32(4), 473-487.

115.

Sobaci

M.Z.

(2016). Social media and local governments: An overview. In Sobaci, M.Z. (Ed.), Social Media and Local Governments, pp. 3-21. Switzerland: Springer.

116.

Sobaci

M.Z.

, & Karkin

(2013). The use of twitter by mayors in Turkey: Tweets for better public services? Government Information Quarterly, 30(4), 417-425.

117.

Spicer

M.W.

(2009). Value conflict and legal reasoning in public administration. Administrative Theory and Praxis, 31(4), 537-555.

118.

Stoker

(2006). Public value management: A new narrative for networked governance? American Review of Public Administration, 36(1), 41-57.

119.

Süer

İ.

(1993). Pazarlama karması optimizasyonunda analitik hiyerarşi yöntemi uygulaması [The application of analytic hierarchy approach in optimization of marketing mix]. Verimlilik Dergisi, 3, 85-108.

120.

Twinomurinzi

Phahlamohlaka

, & Byrne

(2012). The small group subtlety of using ICT for participatory governance: A South African experience. Government Information Quarterly, 29(2), 203-211.

121.

Twizeyimana

J.D.

, & Andersson

(2019). The public value of e-Government – a literature review. Government Information Quarterly, 36(2), 167-78.

122.

Vaidya

O.S.

, & Kumar

(2006). Analytic hierarchy process: An overview of applications. European Journal of Operational Research, 169(1), 1-29.

123.

Yıldız

, & Saylam

(2013). E-government discourses: An inductive analysis. Government Information Quarterly, 30(2), 141-153.

124.

Zavattaro

S.M.

, & Brainard

L.A.

(2019). Social media as micro-encounters: Millennial preferences as moderators of digital public value creation. International Journal of Public Sector Management, 32(5), 562-580.

The reflection of public values in twitter use of metropolitan municipalities: A content analysis employing the analytic hierarchy process

Abstract

Keywords

1. Introduction

2.1 Public values (PVs)

2.1.1 Participation

2.1.2 Collaboration

2.1.3 Transparency

Table 1 Studies used in literature review for picking the main criteria used in developing the evaluation metrics

4. Methodology

4.1.1 The weight of the main criteria

Table 2 The dual comparison matrix of the main evaluation criteria

Table 4 The matrix for the paired comparison for sub-criteria of participation

Table 6 The matrix for the paired comparison for sub-criteria of collaboration

Table 8 The matrix for the paired comparison for sub-criteria of collaboration

Table 10 Local and global weights

4.3 Coefficients of the main criteria for MMs

Table 11 Coefficients of MMs according to the main criteria

Table 12 Weights of MMs

Footnotes

Acknowledgments

References

Table 1
Studies used in literature review for picking the main criteria used in developing the evaluation metrics

Table 2
The dual comparison matrix of the main evaluation criteria

Table 4
The matrix for the paired comparison for sub-criteria of participation

Table 6
The matrix for the paired comparison for sub-criteria of collaboration

Table 8
The matrix for the paired comparison for sub-criteria of collaboration

Table 10
Local and global weights

Table 11
Coefficients of MMs according to the main criteria

Table 12
Weights of MMs