Government Information Quarterly 32 (2015) 52–62

Contents lists available at ScienceDirect

Government Information Quarterly

journal homepage: www.elsevier.com/locate/govinf

Citizens' engagement on local governments' Facebook sites. An empirical

analysis: The impact of different media and content types in
Western Europe
Enrique Bonsón a, Sonia Royo b, Melinda Ratkai a,⁎
a
Faculty of Business Administration, Universidad de Huelva, Plaza de la Merced 11, 21071, Huelva, Spain
b
Faculty of Economics and Business, Universidad de Zaragoza, Gran Vía 2, 50005, Zaragoza, Spain

a r t i c l e i n f o a b s t r a c t

Available online 26 December 2014 There is a very limited empirical research examining the impact of media and content types on stakeholders'
engagement on social media platforms, and basically none within the public sector. The purpose of this study
Keywords: is to measure the impact of media and content types on stakeholders' engagement on Western European local
Media and content types governments' Facebook pages. The sample includes 15-member countries of the EU via 75 local governments be-
Voluntary reporting longing to four different public administration styles, analyzing 50 posts from each municipality. The posts were
e-Government
coded into 16 content types (e.g. environment, housing) and 5 media types (e.g. video, text), respectively. Metrics
we-Government
Government 2.0
of popularity, commitment, virality and engagement were also computed for each post. Results show that the
Public administration styles most used media types are links and photos. Cultural activities, sports and marketing related topics are the
e-Participation most widely posted by Western European local governments, but these contents do not seem to be the most rel-
evant for citizens. Local governments' preferences to different media and content types seem to depend on the
institutional context. Significant differences on engagement levels by citizens have also been found among
media and content types and in different institutional contexts. This study provides practitioners and academics
with some initial research findings about the effect of different communicational strategies on citizen engage-
ment, which are very important factors in order to improve the effectiveness of e-participation offerings.
© 2014 Elsevier Inc. All rights reserved.

1. Introduction users visiting the site at least once a day and 40% doing so multiple
times throughout the day (Pew Research Center, 2013). The use of
Governments are accountable to citizens and society at large, since Facebook to enable contact among citizens can be characterized as a
they are responsible for safeguarding the interests of the general public. normal practice. So, among all the available SM, Facebook offers the
By using the Internet and social media (SM) a huge amount of informa- clearest possibilities for more sustained interaction between citizens
tion can be published for a relatively low cost. In this way, the use of ICTs and their local authority (Ellison & Hardey, 2013).
in the public sector (e-government) has become a powerful strategy for Prior studies have investigated government's use of SM (among
administrative reform at all levels of government. In regard to the use of others, Abdelsalam, Reddick, Gamal, & Al-shaar, 2013; Bertot, Jaeger, &
ICTs in local governments, Johannessen, Flak, and Sæbø (2012) found Grimes, 2010; Bonsón, Royo, & Ratkai, 2014; Bonsón, Torres, Royo, &
that SM is ranked third among the preferred modes of communication Flores, 2012; Ellison & Hardey, 2013, 2014; Ferro, Loukis, Charalabidis,
by the identified stakeholders' groups (politicians, administration and & Osella, 2013; Graham & Avery, 2013; Hofmann, Beverungen, Räckers,
civil society) after the e-mail and the municipality web site. They also & Becker, 2013; Meijer & Thaens, 2013; Mossberger, Wu, & Crawford,
concluded that effective municipal communication and e-participation 2013; Oliveira & Welch, 2013; Panagiotopoulos, Barnett, & Brooks,
require the use of different media, and that 70% of respondents 2013; Reddick & Norris, 2013; Snead, 2013) and the use of Facebook
preferred some kind of digital communication. for political engagement (Conroy, Feezell, & Guerrero, 2012; Rustad &
Growing dynamics of SM suggest that it is not a fad. This study focus- Saebo, 2013). They have also examined the engagement on business en-
es on Facebook since it is on the first place among SM sites according to tities Facebook pages (Bonsón & Ratkai, 2013; Dekay, 2012; Gummerus,
Alexa rankings (Alexa.com., 2013). Furthermore, Facebook has the Liljander, Weman, & Pihlström, 2012; Haigh, Brubaker, & Whiteside,
highest levels of engagement among SM users, with 63% of Facebook 2013; Waters, Burnett, Lamm, & Lucas, 2009). However, very limited re-
search exists about the impact of media and content types on stake-
⁎ Corresponding author.
holders' engagement. There are only a few papers measuring the real
E-mail addresses: [email protected] (E. Bonsón), [email protected] (S. Royo), interaction with stakeholders, and even less focusing on how content
[email protected] (M. Ratkai). type influences user engagement. Leston-Bandeira and Bender (2013)

http://dx.doi.org/10.1016/j.giq.2014.11.001
0740-624X/© 2014 Elsevier Inc. All rights reserved.
 E. Bonsón et al. / Government Information Quarterly 32 (2015) 52–62 53

conducted a content analysis of messages published on Facebook and sites (web and SM) for tracking government spending and making it
Twitter, but these studies only focused on Parliaments. Ramanadhan, visible to a wide public (http://www.recovery.gov). Some studies
Mendez, Rao, and Viswanath (2013) examined what types of content have found that transparency and accountability decrease corruption
are typically shared via SM and the response of the audience. However, (Kim, Kim, & Lee, 2009; Maor, 2004). But some studies also indicate
their study is limited to health promotion programs. The limited avail- that e-transparency can have a heterogeneous effect on different di-
ability of research into the subject demonstrates why the impact of mensions of trust (Grimmelikhuijsen, 2009) and that the impact of
media and content types on stakeholders' engagement is an interesting transparency on citizens' trust in government depends on the cultural
research topic. Therefore we decided to extend these studies to local context of the country (Grimmelikhuijsen, Porumbescu, Hong, & Im,
governments and their audiences on Facebook. 2013). Critics say that transparency will only result in less trust, and skep-
Fostering participation in SM platforms is an issue that continues to tics argue that it has no effect at all. As a consequence, e-transparency
present challenges for researchers and practitioners alike (Al-Debei, should be carefully managed (Bannister & Connolly, 2011).
Al-Lozi, & Papazafeiropoulou, 2013; Lee & Kwak, 2012). Investigating E-government has grown into an umbrella term covering all uses
this matter in terms of the public sector's SM platforms is an area that of ICTs in government (Torres, Pina, & Acerete, 2006) in an effort to
requires further research as the continuous usage and engagement of streamline transactions within government (G2G), between govern-
citizens on these platforms are key success factors. While this study ment and citizens (G2C) and between government and business
does not challenge the positive impact of SM on government per se, it (G2B). The adoption of communication technologies may aim to in-
posits that this relationship requires a more comprehensive examina- crease information dissemination, communication with stakeholders,
tion. In order to measure this impact, the paper is based on theories of and public input into government activities (Li & Feeney, 2014). There-
voluntary reporting, agency problem and citizen e-participation initia- fore, e-government is not only a tool for internal modernization,
tives, by means of media and content types' analyses, and differentiating reporting information and the delivery of public services, but an effec-
among the contexts of different public administration styles. tive communication channel for citizens to participate in democratic
Several authors have indicated the need that future research puts a institutions and political processes, which can create value for both
greater emphasis on the real impact of SM on government-to-citizen citizens and governments.
(G2C) relationships, reinforcing the qualitative analysis of what the Most conflicts in the public sector can be explained based on the
local entities do by means of SM channels (Bonsón et al., 2012; agency problem (Papenfuß & Schaefer, 2010; Waterman & Meier,
Reddick & Norris, 2013). In other words, the analyses need to be ex- 1998), where the politicians/governments are the agents and voters/
panded beyond implementation to emphasize the real use of SM tools citizens/stakeholders are the principals. The agency problem arises be-
by governments and citizens and the impact on G2C relationships. In cause these two parties have different interests and asymmetric infor-
this context, the objective of this paper is to measure the impact of mation. The stakeholders cannot ensure that the government is acting
Facebook use by Western European local governments on stakeholders' in the citizens' best interests. But this conflict can be weakened by
engagement. The effect of different media and content types on citizen disclosing information voluntarily and allowing the municipalities'
engagement will be analyzed, and also the influence of the institutional activities to be monitored, for example, by using a sophisticated
context on Facebook use and citizen engagement. So, this paper exam- digital media plan. This would enable citizens to reduce information
ines the effort of Western European local governments to promote asymmetries and uncover inefficiencies, thus changing the power rela-
transparency and citizen engagement through voluntary reporting by tionship between citizens and authorities (Katz & Halpern, 2013, p. 3).
using Facebook, with the aim of giving answer to the following research The adoption of SM in the public sector may signal how the government
questions. Does the use of different media and content types influence conforms to social expectations about the promotion of transparency,
citizen engagement? Are there any communicational differences across engaging citizens and building trust, because it applies to a context
different public administration styles? After presenting the theoretical where information asymmetry exists. SM and Web 2.0 technologies
framework, a more specific set of propositions will be presented in allow people to be integrated in administrative and policy processes
Section 2.3. Results show that there is a significant impact of media as shareholders, so they are useful tools for a successful elimination of
and content types on stakeholders' engagement and point toward the the dissociation between agents (elected politicians and assigned offi-
existence of communicational differences across public administration cials) and principals (citizens) (Karkin, 2013, p. 324). As Ellison and
styles. Hardey (2014) indicate, in an era where the scope and role of local gov-
ernment is constantly being scrutinized and where levels of citizen en-
2. Background and conceptual framework gagement are low, municipalities need to actively use SM to involve
residents properly in all aspects of local governance. Linders (2012)
This section first begins with a description of some of the key ele- called this kind of governance as “we-government”, reflecting to the
ments related to voluntary reporting, transparency, agency problem need of governmental transformation by co-operation and increased
and citizen engagement, and the role of e-government toward open citizen e-participation. Citizen participation can be detected in tradi-
government. Then, a discussion of the institutional context as regards tional forms of neighborhood watches or auxiliary policeman (Levine
citizen participation in Western Europe follows. The public administra- & Fisher, 1984), among others, but new forms of volunteerism and
tion style classification has been used to understand possible differences citizen co-production are also emerging based on the use of ICTs and
as regards to Facebook use by local governments and citizen engage- SM (Linders, 2011; Meijer, 2014).
ment. The section finishes with a proposal of conceptual framework SM relationships usually start with a like, retweet or mention, and
and a set of propositions that will be used to organize, explain and usually these relationships evolve into being followers and friends and
discuss the results of the research. engaging in online conversations. SM use started in the private sector,
although public sector entities have also started to adopt and use
2.1. Transparency and citizen engagement: the role of e-government and these tools. Panagiotopoulos et al. (2013) concluded that SM users
emerging technologies might not always represent the general public, but they do open signif-
icant opportunities to improve G2C interactions and inform policy de-
In order to gain social legitimacy (Pfeffer & Salancik, 1978) local velopment. Therefore, having a Facebook page can be considered as a
governments should voluntarily report information according to the sign of transparency and accountability. Probably it represents some-
expectations of society. This kind of disclosure can also be called as thing more than just being present where the people are, although
transparency. ICTs are powerful tools for enhancing transparency at a some authors indicate that citizen feedback via SM does not result in
very low cost. For example, the Obama Administration has several governmental change (Magro, 2012).
54 E. Bonsón et al. / Government Information Quarterly 32 (2015) 52–62

SM is more immediate and flexible than other forms of communica- Germanic cities (Royo et al., 2014). However, Bonsón et al. (2012) found
tion. Comparing to the one-to-one questions of helplines, SM are more that the public administration style is not a determining factor of the
effective and transparent, favoring the evolution from one-to-many to level of development of Web 2.0 and SM at the local level. These previous
many-to-many conversations (Panagiotopoulos et al., 2013). Linders studies have analyzed the offer of participative instruments by public
(2011) showed several examples of government initiatives toward a sector entities, not citizen engagement with these tools. In this context,
more transparent and probably less centralized kind of government this paper provides a novel approach by analyzing how citizens interact
(we-government). For example, Obama's Open Government Directive with the participative instrument (Facebook sites), in order to shed some
(http://www.whitehouse.gov/open) focuses on reducing the influence additional light on the relationship between the public administration
of special interests, creating easy-to-understand websites to track how style and Facebook use by local governments and online citizen
the government spends public money and empowering the public to in- engagement.
fluence the decisions. In order to reach this last goal, the White House
has a specialized website (http://petitions.whitehouse.gov) where peo-
ple can engage on the issues that matter to them by signing an existing 2.3. Conceptual framework and propositions
petition or starting a new one. These conversations can also be followed
on Facebook, Twitter and Google + since the administration has decid- Because of advances in technological capabilities and widespread
ed to use various SM channels. The PADGETS project (http://www. adoption, the Internet and SM have become a cost-effective medium,
padgets.eu), co-funded by the EU in the domain of ICT for e- both for disseminating information and for collecting ideas from an in-
governance and policy modeling, aims at implementing a prototype formed public that is no longer happy to be just a passive recipient of
service for policy makers that utilizes SM technologies and techniques government policy. According to Vragov and Kumar (2013), technology
to authorize cross-platform publishing, content tracking and increase can aid a peaceful transition from thin democracies (where there are
public engagement. Ferro et al. (2013) concluded that the use of SM very limited avenues of action for citizens to express their preferences)
by government has a substantial relative advantage comparing to to strong democracies (where there is a strong emphasis on engaging
previous models of e-participation, due to the first step the government the citizenry).
makes toward the citizens, rather than to wait till the citizens move Linders (2012) defines the following typology for citizen e-
their activities onto the official spaces created for e-participation participation: (1) citizen sourcing, (2) government as a platform,
(e-petitions platforms, for example). and (3) do-it-yourself government. These categories reflect to the
models for citizen co-production in the age of SM: citizen to government
2.2. The institutional context: public administration styles (C2G), government to citizen (G2C) and C2C (citizen to citizen), respec-
tively. C2G is mainly about consultation and ideation, where citizens are
Empirical studies of e-participation adoption by local govern- enabled to share their opinions with government. G2C is mainly about
ments have identified the importance of institutional characteristics informing and nudging, where citizens are equipped with data to make
(Panagiotopoulos, Moody, & Elliman, 2012; Royo, Yetano, & Acerete, informed decisions. Lastly, C2C is about self-organization.
2014). As Allen, Juillet, Paquet, and Roy (2001, 94) argue, the “necessary This paper interprets Linders (2012) classification of citizen
transformation in public sector governance and accountability is likely e-participation initiatives in the age of SM by means of C2G
to be blocked by an administrative culture that may be ill suited for a and G2C relationships and it reinforces the idea that the transition
digital world”. Bertot et al. (2010) also found that a culture of openness from e-government (citizen as customer) to we-government (citizen
is needed within the governance system to reach transparency. as partner) is “a new kind of social contract” (Long, 2002), where the
SM advocates highlight the transformational capacities of these plat- public plays a far more active role than it was traditionally. Fig. 1
forms and there is some evidence that the SM ‘logic’ indeed facilitates shows the role of Facebook on G2C and C2G communication. Online
certain practices and therefore steers communication strategies. At the communication channels make data mining possible for governments
same time, other evidence suggests that new media are adapted to the (Linders, 2011; 2012). This can help citizens and governments, as part-
specific situation and that patterns of new media use reflect underlying ners, to be more informed and make socially responsible decisions.
organizational and institutional differences (Meijer & Thaens, 2013; Crowdsourcing –or citizen sourcing– (Nam, 2012; Torres, 2007) is also
Reddick & Norris, 2013). According to Mergel (2012, p. 284), “the poten- emerging as a new and powerful problem-solving mechanism, through
tial of social media use for meaningful, multi-directional exchanges be- which citizens use their skills and expertise to solve government
tween government and its diverse audiences do not fit the highly challenges. A new level of transparent and open government has been
regulated and practiced top-down decision-making and broadcasting enabled where the datasets are available as an open-book. For example,
culture”. In this study, we will analyze whether any communicational
differences exist among local government belonging to different public
administration styles.
Kickert (1997) and Torres (2004) distinguished four public adminis-
tration styles in Western Europe: Anglo-Saxon, Nordic, Germanic and
Southern European. The public administration style has resulted as
an important element for explaining the evolution of public sector
reforms, including recent developments in transparency, account-
ability and e-participation (García-Sánchez, Rodríguez-Domínguez,
& Gallego-Álvarez, 2011; Pina, Lourdes, & Royo, 2007; 2009; 2010;
Royo et al., 2014). Anglo-Saxon and Nordic countries have a long-
standing reputation of public sector reforms, transparency, and citizen
engagement. Germanic countries have a long tradition of consultation
with social partners (OECD, 2001). On the contrary, Southern European
countries are considered laggards in introducing public sector reforms,
and in some of these countries, such as Spain, associations have tradi-
tionally been the only legal participants in most participative processes
(Allegretti & Herzberg, 2004). Therefore, a priori, a higher level of adop-
tion of e-participation tools can be expected in Anglo-Saxon, Nordic and Fig. 1. The role of Facebook on G2C and C2G communication.
 E. Bonsón et al. / Government Information Quarterly 32 (2015) 52–62 55

in forms of voluntary reporting, including financial reporting and track- Graham and Avery (2013) and Halpern and Katz (2012), this feature is
ing how public money is spent. expected to play a central role in affecting citizens' engagement.
On the basis of prior literature and theories, we explore the follow-
ing proposition through the empirical work. 3. Research design and methods

Proposition 1. Posts related to municipal management which directly The sample for this study comprises the first 15 member countries of
affect citizens' lives promote higher levels of citizen engagement. the EU represented by the five largest cities of each; thus, a total of 75
Previous research has suggested that the topic of the post has an im- Western EU local governments have been analyzed (see Appendix A).
portant influence on citizen engagement (Graham & Avery, 2013; Larger-sized local governments were chosen because they are usually
Halpern & Katz, 2012; Khan, Yoon, Kim, & Park, 2014). However, little more innovative in the adoption of new technologies, they have greater
research exists about the analysis of how different content-types affect disclosure needs, and lower relative costs for the implementation of the
to citizen engagement on public sector Facebook pages. As Ellison and new Web 2.0 tools (Bonsón et al., 2012). Besides, the municipality level
Hardey (2013) indicate, academic enquiry has largely neglected the (because of its closeness to citizens) offers a greater diversity of citizen
use of SM for open-ended conversations about local political issues, interaction possibilities. Data collection took place in March 2013, ana-
but certainly local citizens in general will engage with problems as lyzing 50 posts dated from the beginning of February backwards in
they are affected by them. The content analysis of Facebook posts in the case of each local government. The final database consists of 2950
three German local governments carried out by Hofmann et al. (2013) posts, as 16 of the analyzed local governments did not have an official
concludes that there is no topic that guarantees success in terms of Facebook presence.
discussion frequency. However, their results show that external adver- The municipalities of the sample have been classified into four groups,
tisements posted by citizens or companies are uninteresting to the according to their public administration style (see Pina et al., 2007).
followers of a government Facebook page. In contrast, information Germanic countries are Austria and Germany. Southern European coun-
about leisure activities and news about administration events seem to tries are Belgium, France, Greece, Italy, Luxembourg, Portugal and Spain.
be of interest for the followers of the pages. Nordic countries are Denmark, Finland, the Netherlands and Sweden.
Anglo-Saxon countries are Ireland and the United Kingdom.
Proposition 2. The use of multimedia content provokes citizens' The use of Facebook by each municipality was examined by applying
participation. the same method in all the cases. First of all, the official website of the
Communication in online social networks conveys improved media municipality was checked for a direct link to the Facebook platform. If
richness, since they can include rich multimedia data such as pictures, a link was found, the linked platform was accepted as the official one.
videos, or hyperlinks. This might open up new potential for organiza- For the rest of the cases, the presence of the municipality on this
tions to interact with their stakeholders. Facebook posts including platform was checked by using its official name, the link to its official
multimedia features can be used as eye-catchers drawing the users' at- website and/or the official e-mail address. Posts on the Facebook wall
tention to the published information. Previous research in German local were coded following a dual classification differentiating among
governments (Hofmann et al., 2013) has found that using pictures and media and content types, as follows.
embedded videos fosters citizen interaction, with more frequent likes For media type, the categories were the following: (1) video,
and comments on these posts than in text only posts. (2) link, (3) photo, (4) text and (5) others. Only the embedded videos
were considered in the video category. Links re-directing to a video con-
Proposition 3. The public administration style influences Facebook tent were considered as links. Even though on Facebook most of the
activity by local governments and citizens' engagement. links contain a thumbnail, they were considered as links and not as
photos. Photos with texts were counted as photos, because texts are
Proposition 4. Engagement levels by citizens are higher in more open complimentary parts of them. Only those posts where no links, no
local governments (those allowing wall posts by stakeholders). videos, and no photos were found were considered to be texts. Those
posts which did not fall in any of the four above-mentioned categories
Propositions 3 and 4 deal with the influence of the institutional con- (for example, “the events”) were counted as “others”.
text on Facebook use, by using two variables: the public administration As regards the content type analysis, sixteen categories were used
style (as a proxy of the culture of openness at the macro level) and for coding: (1) public works and town planning, (2) environment,
whether the municipality Facebook page allows for wall posts by (3) attention to the citizen, (4) citizen participation, (5) social services,
stakeholders (as a proxy of the culture of openness at the micro level). (6) citizen protection and security, (7) public transport, (8) employment
Previous research has highlighted that a combination of contextual and training schemes, (9) health, (10) education, (11) cultural activities
and path-dependency factors accounts for differences in the emerging and sports, (12) housing, (13) governance issues, (14) financial
SM strategies of public sector organizations (Meijer & Thaens, 2013; reporting, (15) marketing/city promotion/tourism, and (16) others.
Reddick & Norris, 2013). SM technologies, like other previous waves of This classification of content types is based on the lists of local services
technology innovation, have their own transformational potential, but elaborated by Torres and Pina (2001) and subsequently adapted by
organizational and contextual factors are also essential to understand Martí, Royo, and Acerete (2012), conveniently refined according to
their results in public bureaucracies (Abdelsalam et al., 2013; Meijer & the most frequent types of posts provided by local governments. For
Thaens, 2013; Oliveira & Welch, 2013). Based on our previous discus- example, “public transport”, “employment and training schemes” and
sion (Section 2.2) about the public administration styles, Proposition 3 “tourism” were all included under a common category (“economic
is presented. In addition, we also want to test whether more open activities”) in the above-mentioned studies, but they were separated
local governments (understanding by “more open” those allowing in this research because of their individual relevance. Two categories
wall post by stakeholders) are also more successful at promoting higher “governance issues” and “financial reporting” were added as they are
levels of engagement by stakeholders (Proposition 4). By allowing users not local public services, but important aspects related to transparency,
to post messages on the Facebook wall of the entity, local governments accountability and citizen engagement.
are creating a sort of virtual community with their fans in which users We also examined whether municipalities allow wall posts by
not only have the opportunity to interact directly with the posts created stakeholders. In Facebook this function is optional. So, each munici-
by the organization, but also to promote new topics themselves. That is pality has the right to decide if they allow stakeholders to post on
to say, by allowing for open communication, local governments seem to their “wall” or not, which can be considered a sign of openness on the
support democratic and participatory citizen engagement. According to part of municipalities
56 E. Bonsón et al. / Government Information Quarterly 32 (2015) 52–62

Table 1
Facebook metrics for stakeholder engagement (Bonsón & Ratkai, 2013).

Popularity P1 Number of posts liked/total posts Percentage of posts that have been liked
P2 Total likes/total number of posts Average number of likes per post
P3 (P2/number of fans) × 1000 Average number of likes per post per 1000 fans
Commitment C1 Number of posts commented/total posts Percentage of posts that have been commented
C2 Total comments/total posts Average number of comments per post
C3 (C2/number of fans) × 1000 Average number of comments per post per 1000 fans
Virality V1 Number of posts shared/total posts Percentage of posts that have been shared
V2 Total shares/total posts Average number of shares per post
V3 (V2/number of fans) × 1000 Average number of shares per post per 1000 fans
Engagement E P3 + C3 + V3 Stakeholder engagement index

In order to measure citizens' engagement on Facebook pages, the squared test, the differences among public administration styles are
number of likes, comments and shares were collected for each post in not statistically significant.
order to calculate three Facebook metrics as defined by Bonsón and As Table 4 shows, certain patterns can be found according to the
Ratkai (2013) reflecting popularity (P3), commitment (C3) and virality content type. In general terms “cultural activities and sports” is the
(V3); along with an additional aggregated index of engagement most widely used content type (23.7% of the posts), followed by
(E) (see Table 1). As can be seen in Table 1, P3, C3 and V3 have been de- “marketing/city promotion/tourism” (15.8%) and “public works and
flated by the number of fans. Since these metrics are independent of the town planning” (7.4%). The least frequent content types were “financial
size of the audience, they seem to be the most representative in terms of reporting” (0.6%); “housing” (0.7%) and health (1.6%). These patterns
measuring engagement. Therefore only these metrics and the aggregat- are quite similar among the different public administration styles.
ed level of engagement were used in this study. From now on, we will “Cultural activities and sports” is the most frequent type of content in
refer to them as popularity, commitment, virality and engagement, Anglo-Saxon, Nordic and Southern European local governments, and
respectively. the second in Germanic municipalities, where most posts (34.8%) deal
In order to examine our propositions, the engagement metrics were with “marketing/city promotion/tourism”. “Marketing/city promotion/
computed differentiating among media, content types, public adminis- tourism” is a frequent topic among all the public administrations styles,
tration styles and whether or not the local government allowed wall but the Anglo-Saxon, where only 2.4% of the posts deal with this issue.
posts by stakeholders. t-Tests, the Kruskal–Wallis test, contingency However, governance and environmental issues play a more significant
tables and the Pearson's chi-squared test were used to test for possible role in Anglo-Saxon local governments (8.4% of the posts for each of
differences, as appropriate.1 these two content types). As indicated by the Pearson's chi-squared
test at the bottom of the table, there are differences between the relative
4. Analysis and results importance of the different content types among the public administra-
tion styles.
The first part of the Analysis and results section presents an As can be seen in Table 5, regarding to media types, some basic pat-
overview of the use of Facebook among Western European local govern- terns can also be found. In general terms, the most used media type is
ments. The second part of the Analysis and results section analyzes the the link, with more than 50% of the total posts containing a URL link,
differences in citizen engagement depending on the communicational followed by photo (26.9%), text (12.6%) and video (6.1%). This ordering
strategies used (content and media types, wall posts allowed or not) in the relative importance of media types is maintained in all public ad-
and in the different public administration styles. ministration styles, but the Anglo-Saxon, where posting text was more
commonly used than the photos (20.2% versus 16%). However, some
4.1. Use of Facebook and communicational strategies differences in the percentage of the posts by media type can be found
among the different public administration styles. As indicated by the
Table 2 presents an overview of the use of Facebook among the Pearson's chi-squared test, there are differences between the relative
Western European local governments analyzed. Germanic municipali-
ties present the highest rate of adoption of Facebook (100%), followed
by Anglo-Saxon (90%), Nordic (85%) and Southern European (66%) Table 2
local governments. On average, 59 municipalities (out of 75) had a Use of Facebook at the local government level.

Facebook page in the examined period. Municipalities using Total no. of %

As regards the possibility of allowing wall posts by stakeholders, 76% Facebook municipalities
of the examined municipalities with a Facebook account allowed stake- Anglo-Saxon 9 10 90
holders to post their opinions, questions, feedback or critics on the offi- Nordic 17 20 85
cial Facebook page (see Table 3). The most “open” public administration Germanic 10 10 100
Southern European 23 35 66
style in this regard is the Germanic, where 100% of the examined
Total 59 75 79
municipalities allow posts by stakeholders. A high proportion of Nordic
municipalities (82%) also allow posts by stakeholders, but this feature is
much less common among Anglo-Saxon (67%) and Southern European Table 3
(65%) local governments. However, as evidenced by the Pearson's chi- Percentage of municipalities allowing wall posts by stakeholders.

Municipalities allowing Municipalities %

wall posts by others using Facebook
1
The t-test of difference of means is used when differences in a dependent variable
Anglo-Saxon 6 9 67
want to be tested based on an independent variable that only includes two different cate-
Nordic 14 17 82
gories (in our case, differences in citizen engagement depending on whether the local gov-
Germanic 10 10 100
ernments allow for wall posts or not). The Kruskal–Wallis test is used to test for possible
Southern European 15 23 65
differences when more than two categories exist in the independent variable (in our case,
Total 45 59 76
media and content types and the public administration style). The Pearson's chi-squared
Pearson's chi-squared test 5.470
test is used when possible differences want to be tested depending on a double codifica-
Asymptotic significance 0.140
tion of the cases, that is to say, when contingency tables are used.
 E. Bonsón et al. / Government Information Quarterly 32 (2015) 52–62 57

Table 4 Table 6
Percentage of the posts by content type and public administration style. Citizens' engagement by public administration style.

Content types Public administration style Fans Popularity Commitment Virality Engagement

Anglo-Saxon Nordic Germanic Napoleonic Total Anglo-Saxon 825 5.3803 2.0107 1.3538 8.7448
Nordic 2045 12.5128 1.3613 1.0761 14.9502
Public works and town 8.2 8.4 2.6 8.4 7.4
Germanic 182,932 4.8526 0.3478 0.5878 5.7882
planning
Southern 95,792 8.7757 0.9331 4.4607 14.1695
Environment 8.4 6.8 3 3 4.9
European
Attention to the citizen 8.2 6.4 2.8 9 7.1
Total 69,063 8.6696 1.1216 2.3551 12.1464
Citizen participation 5.6 6 1.2 2 3.6
Kruskal–Wallis 44.192 59.873 178.554 36.205
Social services 6.4 4.2 1.2 3.2 3.7
test
Citizen protection and 5.6 1.3 1 1.9 2.1
Asymptotic 0.000 0.000 0.000 0.000
security
significance
Public transport 1.3 1.8 0.6 3.9 2.3
Employment and training 1.8 3.8 0 0.9 1.7
schemes
Health 0.4 2.7 0.4 1.7 1.6
virality, which shows that clicking on the “like” button is much more
Education 4.2 4.2 1.6 3.3 3.4
Cultural activities and sports 19.1 20 29 25.8 23.7 common than commenting or sharing the post.
Housing 0.7 0.5 0.8 0.8 0.7 Interestingly, those topics more widely covered by local govern-
Governance issues 8.4 4 2.2 5.7 5.1 ments (those topics having a higher number of posts, as reported in
Financial reporting 0.9 0.2 0.4 1 0.6 Table 4) are not necessarily those that elicit higher levels of engagement
Marketing/city 2.4 14.9 34.8 13.5 15.8
promotion/tourism
on the part of citizens. “Cultural activities and sports” and “marketing/
Other 18.2 14.8 18.4 16 16.4 city promotion/tourism” are the most widely used content types by
Total 100 100 100 100 100 local governments (23.7% and 15.8% of the posts, respectively, as
Note: value of Pearson's chi-squared test = 449.607; significant at the 1% level. shown in Table 4), but these topics only obtain low (in the case of
cultural activities) or moderate (in the case of marketing) levels of en-
gagement. In contrast, the two most engaging topics (“public transport”
importance of the different media types among the public administra- and “housing”) only account for 2.3% and 0.7% (see Table 4) of the posts
tion styles. made by local governments, respectively.
The influence of the different topics on citizens' engagement
also seems to be dependent upon the public administration style (see
4.2. Differences in citizens' engagement Table 8). For example, Anglo-Saxon citizens seem to be more engaged
by issues of “employment and training schemes” and “environment”,
As regards engagement levels by citizens, the largest audiences (av- while Germanic citizens are more engaged by topics of “financial
erage number of fans) are found in Germanic countries (see Table 6). reporting”.
However, Germanic countries do not have the highest levels of engage- Table 9 analyzes the effect of the use of different media types on
ment. The aggregated index of engagement shows that, on average, the citizen engagement. As the results of the Kruskal–Wallis test show,
most engaged citizens are found in Nordic and Southern European local the media type also influences engagement levels. As can be seen,
governments, where citizens have almost twice as high engagement photos elicit the highest levels of engagement, followed by text. Consid-
levels as in the Anglo-Saxon and almost three times higher than in the ering the different types of engagement, the general pattern of “liking”
Germanic municipalities. Nevertheless, Anglo-Saxon local governments (popularity) being much more common that commenting (commit-
have the highest commitment metric, so here posts are more likely to be ment) or sharing (virality) the posts, also holds, in general terms, re-
commented on. High heterogeneity can be found in the engagement garding media types. There is just one exception, which is the “other”
metrics among the public administration styles, which is confirmed by category of media types, where virality is higher than popularity. The
the Kruskal–Wallis tests.
As Table 7 shows, the most engaging topics are “public transport” and Table 7
“housing”, and the type of content is a relevant factor to explain the Citizens' engagement by content type.
different levels of citizens' engagement (Kruskal–Wallis tests significant Content types Popularity Commitment Virality Engagement
at the 1% level). Examining the different components of engagement
Public works and town 11.0666 2.3866 2.3200 15.7732
(popularity, commitment and virality), high levels of heterogeneity can planning
be found. The most popular (most liked) topics are related to “housing” Environment 8.5096 2.7003 1.1833 12.3932
and “marketing/city promotion/tourism”. The most commented topics Attention to the citizen 8.2099 1.0831 5.2322 14.5252
deal with the “environment” and “public works and town planning”, Citizen participation 8.6935 1.0967 4.8886 14.6788
Social services 4.9343 1.5628 1.9245 8.4216
while the most viral (most shared) topics are “public transport” and
Citizen protection and 4.5072 0.3626 0.7940 5.6639
“attention to the citizen”. Also, the average levels of engagement are security
very heterogenic: 8.6 for popularity, 1.1 for commitment and 2.3 for Public Transport 9.2150 2.2933 6.9642 18.4724
Employment and training 4.6582 0.9975 2.2104 7.8661
Table 5 schemes
Percentage of the posts by media type and public administration style. Health 5.7152 0.5848 1.0149 7.3149
Education 5.1973 1.4782 1.1423 7.8178
Media types Public administration style Cultural activities and sports 6.8135 0.5657 1.7529 9.1320
Housing 13.5115 2.2155 2.2765 18.0034
Anglo-Saxon Nordic Germanic Napoleonic Total
Governance issues 6.1326 0.6503 2.5456 9.3285
Video 0 2.2 4.4 12.2 6.1 Financial reporting 6.2520 2.1802 2.4130 10.8452
Link 61.8 50.6 53.8 48.4 52 Marketing/city 11.4983 0.6439 2.3365 14.4787
Photo 16 32.9 36.4 22.7 26.9 promotion/tourism
Text 20.2 11.4 3.6 14.3 12.6 Other 11.1711 1.2448 1.8023 14.2182
Other 2 2.8 1.8 2.3 2.3 Total 8.6696 1.1216 2.3551 12.1464
Total 100 100 100 100 100 Kruskal–Wallis test 84.031 102.125 132.175 74.361
Asymptotic significance 0.000 0.000 0.000 0.000
Note: value of Pearson's chi-squared test = 245.270; significant at the 1% level.
58 E. Bonsón et al. / Government Information Quarterly 32 (2015) 52–62

Table 8 Table 9
Citizens' engagement by content type and public administration style. Citizens' engagement by media type.

Engagement Media types Popularity Commitment Virality Engagement

Content types Anglo-Saxon Nordic Germanic Napoleonic Total Video 4.3294 0.5257 2.0504 6.9055
Link 4.3967 0.6629 1.1454 6.2050
Public works and 14.3645 24.8961 7.0694 10.7994 15.7732
Photo 18.7001 1.8169 4.1420 24.6590
town planning
Text 7.6441 1.8726 3.0064 12.5231
Environment 20.2955 13.2432 4.2838 5.6889 12.3932
Other 4.9961 0.8352 5.9579 11.7892
Attention to the 6.8866 5.1834 8.5776 22.9753 14.5252
Total 8.6696 1.1216 2.3551 12.1464
citizen
Kruskal–Wallis test 450.302 199.286 231.151 414.118
Citizen participation 8.6513 5.6855 2.9146 44.2412 14.6788
Asymptotic significance 0.000 0.000 0.000 0.000
Social services 6.6652 4.6037 9.6428 13.3149 8.4216
Citizen protection and 7.4736 2.3632 4.5329 5.5147 5.6639
security
Public transport 1.1016 14.0258 0.8575 23.4451 18.4724 local governments in these public administration styles allow for wall
Employment and 29.3598 3.0688 – 6.0228 7.8661 posts by stakeholders. This research also shows that a bigger audience
training schemes
does not necessarily result in larger engagement levels. This means
Health 9.2329 8.0853 4.8826 6.4365 7.3149
Education 10.9621 8.8481 4.4306 5.9826 7.8178 that not all the fans of Western European municipalities Facebook ac-
Cultural activities and 7.1081 14.5621 3.5018 9.3587 9.1320 counts are active on these platforms. For example, the stakeholders of
sports Germanic municipalities (that had the largest number of fans, on aver-
Housing 10.1059 6.2605 1.6234 33.1351 18.0034 age) are also the least engaged. Although at first sight these findings
Governance issues 4.2677 5.2837 1.8867 15.5664 9.3285
Financial reporting 11.7390 19.5313 18.6509 7.5217 10.8452
may seem contradictory, this is not the case, as the audience can follow
Marketing/city 4.5717 19.0036 7.6973 19.0868 14.4787 the posts of the municipalities without engaging in any additional
promotion/tourism online interaction in the Facebook platform. The most engaged stake-
Other 5.0237 27.4958 6.0914 13.2868 14.2182 holders were from the Nordic and Southern European municipalities,
Total 8.7448 14.9502 5.7882 14.1695 12.1464
with very important differences as regards engagement with respect
to Anglo-Saxon and Germanic local governments.
Overall, the fact that Southern European local governments show
category of others may be considered as a mixed category, since it the lowest rate of adoption of Facebook is consistent with previous re-
collects all media types that could not be fitted to any of the four search indicating that these countries are usually laggards in introduc-
established ones covering video, link, photo and text. For example, ing public sector reforms and show lower levels of adoption of citizen
this residual category includes the “calendar events”. Obviously, for participation tools. However, the fact that Southern European local gov-
this type of media, sharing the event seems to be the most appropriate ernments have one of the highest levels of citizen engagement on their
and useful form of engagement. This explains why this media type ob- Facebook sites seems to confirm that the concepts of open government,
tains the highest value in virality. democracy and freedom of information are being transformed by new
Finally, by examining the differences in engagement levels between technologies (Picazo-Vela, Gutiérrez-Martínez, & Luna-Reyes, 2012;
municipalities allowing and not allowing wall posts by stakeholders Vragov & Kumar, 2013). As our results show, citizens in those settings
(see Table 10), we also detected some main differences. Local govern- where scarce opportunities for citizen participation have traditionally
ments allowing wall posts reached higher average levels of citizens' existed (Southern European local governments), are now making
engagement in all the metrics examined. These differences were greater use of the available technology to engage in discussions about
significant at the 1% level for popularity and the aggregated level of local issues.
engagement. The findings regarding popularity, commitment and virality are con-
sistent with prior studies in the private sector (Bonsón & Ratkai, 2013).
The most popular way of interaction is liking a post, followed by sharing
5. Discussion
it and lastly commenting on it, which is consistent with the relative
easiness of each type of interaction. So, as liking a post is much faster
Stakeholder engagement behaviors are essential part of the commu-
and easier than commenting on it, citizens are making more use of the
nity success, because without active “likers” and “commentators” there
“like” than the “comment” function, which shows a limited level on
will not be much to read about (Gummerus et al., 2012). Therefore,
engagement on the part of citizens.
improving the knowledge about how to increase stakeholders' engage-
Some patterns were found about what the local governments'
ment and which media and content types do they prefer are very
voluntarily reported on Facebook. The topics most posted were
important topics.
connected to “cultural activities and sports” and “marketing/city
Our results show that 79% of the examined European municipalities
promotion/tourism”. Previous research in German local governments
had an official Facebook page.2 Simply attaining a Facebook page is not
(Hofmann et al., 2013) has also found that most of the posts topics
enough; this SM presence should be used in a way that is mutually ben-
were concerned with leisure activities. To some extent, both topics are
eficial for both the entity and its audience. Therefore, local governments
marketing oriented, as cultural activities may also be considered as a
should be actively working on sending their stakeholders presumably
interesting and relevant content in order to promote engagement. Of
course, although not analyzed in this paper, local governments should
also commit themselves to actually take citizens' opinions and concerns Table 10
Engagement levels in local governments allowing and not allowing wall posts by
into account. stakeholders.
As expected from the literature review on public administration
styles, Anglo-Saxon, Germanic and Nordic local governments show Popularity Commitment Virality Engagement

higher levels of Facebook adoption. Meanwhile, a higher proportion of Mean LG not allowing wall 6.3697 0.9356 2.2045 9.5098
posts
2 Mean LG allowing wall posts 9.3852 1.1795 2.4019 12.9666
Probably the same local governments have other Facebook pages for more specific
Total 8.6696 1.1216 2.3551 12.1464
reasons, services or departments as well, for example, tourism, promoting a healthier life-
t-Test −3.719⁎⁎ −1.157 −0.299 −2.824⁎⁎
style, or other special purposes, like charity and fundraising, etc., but these have not been
analyzed in this research. ⁎⁎ Significant at the 1% level.
 E. Bonsón et al. / Government Information Quarterly 32 (2015) 52–62 59

kind of city and events promotion. These findings are not surprising at results show that allowing wall posts from users is an effective strategy,
all, as the use of these platforms for marketing and public relations pur- which confirms Proposition 4.
poses is well documented both in the private (Christodoulides, 2009) It seems that there is a demand from the citizens' side to a more
and in the public sectors (Graham & Avery, 2013; Karkin, 2013). effective communication about topics related to everyday life in their
But our results show that citizens do not like interacting with mar- municipalities. Therefore, local governments should encourage trans-
keting type of content as much. Therefore, these are not the contents parency via voluntary reporting and use the available technologies to
they are more interested in. In general terms, the most engaging topics promote citizen e-participation. Hopefully local governments (via
are “public transport”, “housing” and “public works and town planning”. their community managers) can realize this demand by opening effec-
So those topics related to municipal management which directly affect tive corporate dialog initiatives (Bonsón et al., 2012), based on two
citizens' lives seem to be more appreciated, which is consistent with core principles: collaboration and engagement (Eccles & Krzus, 2010).
Proposition 1 and with Johannessen et al. (2012). They found that infor- Probably this would be the starting point of a larger modernization pro-
mation about local issues and the information relevant for the individu- cess to reach the “we-government” ideal, where citizens are actively
al stakeholder are the most important communication needs. High shaping and contributing in the decisions of their municipalities.
heterogeneities were detected among the different public administra- Finally, the limitations of this study should be acknowledged and the
tion styles as regards the influence of content types on citizen engage- avenues for further research indicated. As in all SM analyses, this study
ment. As both engagement levels by citizens and the influence of the is also just a snapshot of local government practices at a specific mo-
different topics on citizens' engagement seem to be dependent upon ment in time. Internet and SM are changing and evolving continuously,
the public administration style, we can conclude that communicational so future studies are needed to update the current findings. Future stud-
differences exist among public administration styles (and perhaps ies could also broaden the number of cases studied per country, the
individual local governments), depending on the specific circumstances range in the sizes of the local governments analyzed, and include
in each jurisdiction at a given point in time, which also confirms other public administration styles within Europe (such as Eastern
Proposition 3. Therefore, the institutional context is confirmed as a European countries) and other geographical areas different from
key factor to be taken into account when analyzing citizen engagement. Europe. An issue that also deserves further attention from the academic
Not only content type preferences are largely understudied, but also literature is whether citizens' comments on SM platforms are focused
media types preferences. Common patterns have been found in on criticizing or supporting public agencies, or whether there is a bal-
Facebook media type choices by local governments. In general terms, ance between negative and positive mentions (sentiment analyses).
the resources most widely used are links, followed by photos. But Also, the C2C e-participation could be analyzed in order to gain some
again, no total correspondence exits between the media most often more insights about the third dimension of e-participation: Are non-
used by local governments and those media more engaging for citizens, official SM sites more effective at promoting citizen engagement than
as photos elicit the highest levels of engagement, followed by text. the official ones? Are citizens helping each other by providing answers
These findings are consistent with previous research (Abdelsalam on SM channels? Our results also point to a high cultural diversity
et al., 2013) in the sense that links are the most used media type, but among public administration styles (and perhaps individual local gov-
the one that promotes the lowest levels of engagement. Overall, these ernments), depending on the specific circumstances in each jurisdiction
findings only provide partial support for Proposition 2, as photos obtain at a given point in time. So, future research should provide more in-
the highest levels of engagement, but text only post are more successful depth analyses and case studies about the impact of organizational
at engaging citizens than videos or hyperlinks. In any case, the power of and institutional differences on SM use. In connection to this, future
photos to increase the intensity of citizens' reaction to posts by local studies should also analyze how citizen engagement on SM platforms
governments found in previous research (Hofmann et al., 2013) is influences the changes in public policy making.
confirmed.
In spite of the fact that citizens have been widely recognized are 6. Conclusions and implications
sources of ideas and initiatives that provide a mutual enrichment for
both parties – government and citizens – (Michel, 2005), citizens are ac- This investigation contributes meaningfully to the citizen e-
customed to being treated as clients and not as partners (Linders, 2011). participation and government 2.0 literature by demonstrating that the
This tendency seems to be changing, slowly, by recent movements to- content and media types have an impact on stakeholders' engagement
ward open government and increased attention to citizen participation. on Facebook. Our results show that marketing related contents are pre-
But as regards Facebook use by Western European municipalities, our ferred by local governments in Western Europe, but citizens do not
results show that some room for improvement still exists, especially seem to show much interest toward these contents and they prefer
in Anglo-Saxon and Southern European local governments, where topics related to municipal management more closely related to their
only around 65% of the local governments with an official Facebook everyday lives. By disclosing relevant information voluntarily, the agen-
page allow stakeholders to leave posts on their Facebook pages. The rea- cy conflict – between local governments and citizens – could be weak-
son for this can be the fear from being criticized, as prior studies in the ened. So municipalities should identify the most relevant topics for
private (Dekay, 2012) and public (Hofmann et al., 2013) sectors have citizens in their jurisdictions in order to meet citizens' needs, provide
found, but some researchers argue that this fear from criticism is not useful information for them and collect their opinions on these sensitive
well established (Shu & Chuang, 2011). Nowadays, many citizens are topics. Photo usage was also detected to provoke higher levels of citizen
discussing local policy online (Bonsón et al., 2012) and local govern- participation. So, in order to capture the attention of citizens and pro-
ments should not miss the opinions expressed there. For local govern- mote their participation, a good strategy can be to use pictures and
ments, not engaging now involves a greater risk than engaging: photos in the posts made in SM platforms.
citizens will use these networks to talk about them, whether local gov- Results also show that engagement levels by citizens are higher in
ernments utter a word to the conversation or not (Bonsón et al., 2012). the case of those local governments which allow wall posts by stake-
Until governmental agencies do not allow the citizens to express their holders. So, it seems that more open local governments not only provide
views publicly and freely, they cannot take full advantage from for a wider range of e-participation possibilities, but citizens are also
their e-participation initiatives. Previous research in the private sector more active and engaged on discussions initiated by the local govern-
(Byrd, 2012) has found that a corporation from the automotive industry ment. Engagement levels by citizens and the influence of the different
letting its consumer community to direct the conversation during a topics on citizens' engagement seem to be dependent upon the public
crisis situation resulted to be a fruitful strategy. If local governments re- administration style, confirming that the institutional setting has an im-
ally want to use the Facebook pages to promote citizen engagement, our portant influence on e-participation and citizen engagement. An
60 E. Bonsón et al. / Government Information Quarterly 32 (2015) 52–62

important finding of this research is that citizens in those settings where may be more effective if they ensure that any content posted is specially
scarce opportunities for citizen participation have traditionally existed focused around topics of interest and significance to local citizens, rath-
are now making greater use of the available technology to engage in dis- er than being of broader organizational interest or marketing-related.
cussions about local issues, which seems to confirm that SM are helping This is particularly important, especially at a time when local govern-
to empower citizens and expand democracy. ments need to be more responsive to local citizens' needs than ever
Local governments have started using Facebook as a communication before.
and reporting channel, although clear evidence about its impact or
whether it means any change on G2C relationships is still missing.
Council-led conversations about the conduct of local governance and Acknowledgments
the quality of local service provision could play a significant role in en-
couraging informal participation in local issues (Ellison & Hardey, This study has been carried out with the financial support of the
2013). A major implication of this paper is to highlight that if local gov- Spanish National R&D Plan through research projects ECO2010-17463
ernments wish to use SM primarily for citizen engagement, their efforts and ECO2011-28267 (ECON-FEDER).

Appendix A. List of municipalities.

Country City Official name Web

1 Austria Vienna Magistrat der Stadt Wien www.wien.gv.at

2 Graz Magistrat Graz www.graz.at
3 Linz Magistrat Linz www.linz.at
4 Salzburg Stadt Salzburg www.stadt-salzburg.at
5 Innsbruck Stadt Innsbruck www.innsbruck.at
6 Belgium Antwerpen Stad Antwerpen www.antwerpen.be
7 Gent Stad Gent www.gent.be
8 Charleroi La Ville de Charleroi www.charleroi.be
9 Liège Ville de Liège www.liege.be
10 Brussels City of Brussels www.brucity.be
11 Denmark Copenhagen København kommune www.kk.dk
12 Aarhus Arhus kommune www.aarhuskommune.dk
13 Aalborg Aalborg Kommune www.aalborgkommune.dk
14 Odense Odense Kommune www.odense.dk
15 Esbjerg Esbjerg Kommune www.esbjergkommune.dk
16 Finland Helsinki Helsingin kaupunki www.hel.fi
17 Espoo Espoon kaupunki www.espoo.fi
18 Tampere Tampereen kaupunki www.tampere.fi
19 Vantaa Vantaan kaupunki www.vantaa.fi
20 Turku Turun kaupunki www.turku.fi
21 France Paris Mairie de Paris www.paris.fr
22 Marseille Mairie de Marseille www.marseille.fr
23 Lyon Mairie de Lyon www.lyon.fr
24 Toulouse Mairie de Toulouse www.toulouse.fr
25 Nice Mairie de NICE www.nice.fr
26 Germany Berlin Stadt Berlin www.berlin.de
27 Hamburg Stadt Hamburg www.hamburg.de
28 Munich Stadt Munchen www.muenchen.de
29 Köln Stadt Koln www.stadt-koeln.de
30 Frankfurt Stadt Frankfurt www.frankfurt.de
31 Greece Athens Δήμος Αθηναίων www.cityofathens.gr
32 Thessaloniki Δήμος Θεσσαλονίκης www.thessaloniki.gr
33 Patras Δήμος Πατρέων www.patras.gr
34 Heraklion Δήμος Ηρακλείου www.heraklion-city.gr
35 Volos Δήμος Βόλου www.volos-city.gr
36 Ireland Dublin Dublin City Council www.dublincity.ie
37 Cork Cork City Council www.corkcorp.ie
38 Galway Galway City Council www.galwaycity.ie
39 Limerick Limerick City Council www.limerickcorp.ie
40 Waterford Waterford City Council www.waterfordcity.ie
41 Italy Rome Comune di Roma www.comune.roma.it
42 Milan Comune di Milano www.comune.milano.it
43 Naples Comune di Napoli www.comune.napoli.it
44 Turin Comune di Torino www.comune.torino.it
45 Palermo Comune di Palermo www.comune.palermo.it
46 Luxembourg Luxembourg Ville de Luxembourg www.luxembourg-city.lu
47 Esch-sur-Alzette Adm.Communale de la Ville d'Esch-sur-Alzette www.esch.lu
48 Differdange Mairie de Differdange www.differdange.lu
49 Dudelange Administration de la Ville de Dudelange www.dudelange.lu
50 Pétange Administration Communale de Pétange www.petange.lu
51 Netherlands Amsterdam Gemeente Amsterdam www.amsterdam.nl
52 Rotterdam Gemeente Rotterdam www.rotterdam.nl
53 The Hague Gemeente Den Haag www.denhaag.nl
54 Utrecht Gemeente Utrecht www.utrecht.nl
55 Eindhoven Gemeente Eindhoven www.eindhoven.nl
56 Portugal Lisbon Cámara municipal de Lisboa www.cm-lisboa.pt
 E. Bonsón et al. / Government Information Quarterly 32 (2015) 52–62 61

Appendix
(continued)
A (continued)

Country City Official name Web

57 Oporto Cámara municipal do Porto www.cm-porto.pt

58 Vila Nova de Gaia Cámara municipal de Gaia www.cm-gaia.pt
59 Amadora Cámara municipal de Amadora www.cm-amadora.pt
60 Braga Cámara municipal de Braga www.cm-braga.pt
61 Sweden Stockholm Stockholms Stad www.stockholm.se
62 Göteborg Göteborgs Stad www.goteborg.se
63 Malmö Malmo Stad www.malmo.se
64 Uppsala Uppsala kommun www.uppsala.se
65 Linköping Lidköpings kommun www.lidkoping.se
66 Spain Madrid Ayuntamiento de Madrid www.munimadrid.es
67 Barcelona Ayuntament de Barcelona www.bcn.es
68 Valencia Ajuntament de Valéncia www.valencia.es
69 Sevilla Ayuntamiento de Sevilla www.sevilla.org
70 Zaragoza Ayuntamiento de Zaragoza www.zaragoza.es
71 UK London London City Council www.london.gov.uk
72 Birmingham Birmingham City Council www.birmingham.gov.uk
73 Glasgow Glasgow City Council www.glasgow.gov.uk
74 Leeds Leeds City Council www.leeds.gov.uk
75 Manchester Manchester City Council www.manchester.gov.uk

References
Abdelsalam, H. M., Reddick, C. G., Gamal, S., & Al-shaar, A. (2013). Social media in Grimmelikhuijsen, S. (2009). Do transparent government agencies strengthen trust?
Egyptian government websites: Presence, usage, and effectiveness. Government Information Policy, 14(3), 173–186.
Information Quarterly, 30(4), 406–416. Grimmelikhuijsen, S., Porumbescu, G., Hong, B., & Im, T. (2013). The effect of transparency on
Al-Debei, M. M., Al-Lozi, E., & Papazafeiropoulou, A. (2013). Why people keep coming back trust in government: A cross-national comparative experiment. Public Administration
to Facebook: Explaining and predicting continuance participation from an extended Review, 73(4), 575–586.
theory of planned behaviour perspective. Decision Support Systems, 55(1), 43–54. Gummerus, J., Liljander, V., Weman, E., & Pihlström, M. (2012). Customer engagement in a
Alexa.com. (2013). Retrieved from http://www.alexa.com/siteinfo/facebook.com (Accessed: Facebook brand community. Management Research Review, 35(9), 857–877.
30.12.2013) Haigh, M. M., Brubaker, P., & Whiteside, E. (2013). Facebook: Examining the information
Allegretti, G., & Herzberg, C. (2004). Participatory budgets in Europe: Between efficiency and presented and its impact on stakeholders. Corporate Communications: An International
growing local democracy. Amsterdam: Transnational Institute (Retrieved from http:// Journal, 18(1), 52–69.
www.tni.org/sites/www.tni.org/archives/reports/newpol/participatory.pdf [Accessed Halpern, D., & Katz, J. E. (2012). From e-government to social network government:
07.10.2014]). Towards a transition model. Paper presented at the Proceedings of the 3rd Annual
Allen, B. L., Juillet, G., Paquet, J., & Roy, J. (2001). E-governance and government on-line in ACM Web Science Conference, Evanston, Illinois.
Canada: Partnerships, people and prospects. Government Information Quarterly, 18(2), Hofmann, S., Beverungen, D., Räckers, M., & Becker, J. (2013). What makes local govern-
93–104. ments' online communications successful? Insights from a multi-method analysis of
Bannister, F., & Connolly, R. (2011). The trouble with transparency: A critical review of Facebook. Government Information Quarterly, 30(4), 387–396.
openness in e-government. Policy & Internet, 3(1), 1–30. Johannessen, M. R., Flak, L. S., & Sæbø, Ø. (2012). Choosing the right medium for municipal
Bertot, J. C., Jaeger, P. T., & Grimes, J. M. (2010). Using ICTs to create a culture of transpar- e-participation based on stakeholder expectations. Electronic Participation. Springer
ency: E-government and social media as openness and anti-corruption tools for Berlin Heidelberg, 25–36.
societies. Government Information Quarterly, 27(3), 264–271. Karkin, N. (2013). Web 2.0 tools for public participation through government websites.
Bonsón, E., & Ratkai, M. (2013). A set of metrics to assess stakeholder engagement and Gestión y Política Pública, Special Issue on Electronic Government, 307–331.
social legitimacy on a corporate Facebook page. Online Information Review, 37(5), Katz, J., & Halpern, D. (2013). Political and developmental correlates of social media
787–803. participation in government: A global survey of national leadership websites.
Bonsón, E., Royo, S., & Ratkai, M. (2014). Facebook practices in Western European munic- International Journal of Public Administration, 36(1), 1–15.
ipalities. An empirical analysis of activity and citizens' engagement. Administration & Khan, G. F., Yoon, H. Y., Kim, J., & Park, H. W. (2014). From e-government to social
Society. http://dx.doi.org/10.1177/0095399714544945. government: Twitter use by Korea's central government. Online Information Review,
Bonsón, E., Torres, L., Royo, S., & Flores, F. (2012). Local e-government 2.0: Social media 38(1), 95–113.
and corporate transparency in municipalities. Government Information Quarterly, Kickert, W. (1997). Public management in the United States and Europe. In W. Kickert
29(2), 123–132. (Ed.), Public Management and Administrative reform in Western Europe (pp. 15–38).
Byrd, S. (2012). Hi fans! Tell us your story! Incorporating a stewardship-based social media Cheltenham, UK: Edward Elgar.
strategy to maintain brand reputation during a crisis. Corporate Communications: An Kim, S., Kim, H. J., & Lee, H. (2009). An institutional analysis of an e-government system
International Journal, 17(3), 241–254. for anti-corruption: The case of open. Government Information Quarterly, 26(1),
Christodoulides, G. (2009). Branding in the post-Internet era. Marketing Theory, 9(1), 42–50.
141–144. Lee, G., & Kwak, Y. H. (2012). An open government maturity model for social media-based
Conroy, M., Feezell, J. T., & Guerrero, M. (2012). Facebook and political engagement: A public engagement. Government Information Quarterly, 29(4), 492–503.
study of online political group membership and offline political engagement. Leston-Bandeira, C., & Bender, D. (2013). How deeply are parliaments engaging on social
Computers in Human Behavior, 28(5), 1535–1546. media. Information Polity, 18(4), 281–297.
Dekay, S. H. (2012). How large companies react to negative Facebook comments. Levine, C., & Fisher, G. (1984). Citizenship and service delivery: The promise of coproduc-
Corporate Communications: An International Journal, 17(3), 289–299. tion. Public Administration Review, 44(Special Issue), 178–189.
Eccles, R., & Krzus, M. P. (2010). One report: Integrated reporting for a sustainable strategy. Li, M. -H., & Feeney, M. K. (2014). Adoption of electronic technologies in local U.S. govern-
Hoboken, New Jersey: Wiley. ments: Distinguishing between e-services and communication technologies. The
Ellison, N., & Hardey, M. (2013). Developing political conversations? Information. American Review of Public Administration, 44(1), 75–91.
Communication & Society, 16(6), 878–889. Linders, D. (2011). We-government: an anatomy of citizen e-participation in the Informa-
Ellison, N., & Hardey, M. (2014). Social media and local government: Citizenship, tion Age. Proceeding of the 12th Annual International Conference on Digital Government
consumption and democracy. Local Government Studies, 40(1), 21–40. Research (pp. 167–176).
Ferro, E., Loukis, E., Charalabidis, Y., & Osella, M. (2013). Analyzing the centralised use of Linders, D. (2012). From e-government to we-government: Defining a typology for
multiple social media by government from innovations diffusion theory perspective. citizen e-participation in the age of social media. Government Information Quarterly,
In Electronic Participation. Proceeding of 5th IFIP WG 8.5 International Conference, 29, 446–454.
Koblenz, Germany, September 17–19 (pp. 95–108). Long, M. (2002). Beyond traditional boundaries: Government in the information age.
García-Sánchez, I. M., Rodríguez-Domínguez, L., & Gallego-Álvarez, I. (2011). The relation- Australian Journal of Public Administration, 61(1), 3–12.
ship between political factors and the development of e-participatory government. Magro, M. J. (2012). A review of social media use in e-government. Administrative
The Information Society, 27(4), 233–251. Sciences, 2, 148–161.
Graham, M., & Avery, E. J. (2013). Government public relations and social media: An Maor, M. (2004). Feeling the heat? Anticorruption mechanisms in comparative perspec-
analysis of the perceptions and trends of social media use at the local government tive. Governance: An International Journal of Policy, Administration and Institutions,
level. Public Relations Journal, 7(4), 1–21. 17(11), 1–28.
62 E. Bonsón et al. / Government Information Quarterly 32 (2015) 52–62

Martí, C., Royo, S., & Acerete, B. (2012). The effect of new legislation on the disclosure of Reddick, C. G., & Norris, D. F. (2013). Social media adoption at the American grass roots:
performance indicators: The case of Spanish local governments. International Journal Web 2.0 or 1.5? Government Information Quarterly, 30(4), 498–507.
of Public Administration, 35(13), 873–885. Royo, S., Yetano, A., & Acerete, B. (2014). E-participation and environmental protec-
Meijer, A. J. (2014). New media and the coproduction of safety: An empirical analysis of tion: Are local governments really committed? Public Administration Review,
Dutch practices. The American Review of Public Administration, 44(1), 17–34. 74(1), 87–98.
Meijer, A., & Thaens, M. (2013). Social media strategies: Understanding the differences Rustad, E., & Saebo, O. (2013). How, why and whom do local politicians engage on
between North American police departments. Government Information Quarterly, Facebook? Lecture Notes in Computer Science, 8075, 69–79.
30(4), 343–350. Shu, W., & Chuang, Y. -H. (2011). The perceived benefits of 6-degree-separation social
Mergel, I. (2012). The social media innovation challenge in the public sector. Information networks. Internet Research, 21(1), 26–45.
Polity, 17, 281–292. Snead, J. T. (2013). Social media use in the U.S. Executive branch. Government Information
Michel, H. (2005). E-Administration, E-government, E-Governance and the learning city: Quarterly, 30(1), 56–63.
A typology of citizenship management using ICTs. The Electronic Journal of E- Torres, L. (2004). Trajectories in the modernisation of public administration in European
government, 3(4), 213–218. continental countries. Australian Journal of Public Administration, 63(3), 99–112.
Mossberger, K., Wu, Y., & Crawford, J. (2013). Connecting citizens and local governments? Torres, L. H. (2007). Citizen sourcing in the public interest. Knowledge Management for
Social media and interactivity in major U.S. cities. Government Information Quarterly, Development Journal, 3(1), 134–145.
30(4), 351–358. Torres, L., & Pina, V. (2001). Public–private partnership and private finance initiatives in
Nam, T. (2012). Suggesting frameworks of citizen-sourcing via Government 2.0. the EU and Spanish local governments. The European Accounting Review, 10(3),
Government Information Quarterly, 29(1), 12–20. 601–619.
OECD (2001). Citizens as partners. Information, consultation and public participation in Torres, L., Pina, V., & Acerete, B. (2006). E-governance developments in European
policy-making. Paris: OECD. Union cities: Reshaping government's relationship with citizens. Governance:
Oliveira, G. H. M., & Welch, E. W. (2013). Social media use in local government: Linkage of An International Journal of Policy, Administration and Institutions, 19(2),
technology, task, and organizational context. Government Information Quarterly, 277–302.
30(4), 397–405. Vragov, R., & Kumar, N. (2013). The impact of information and communication technolo-
Panagiotopoulos, P., Barnett, J., & Brooks, L. (2013). Social Media and government respon- gies on the costs of democracy. Electronic Commerce Research and Applications, 12(6),
siveness: The case of the UK Food Standards Agency. Lecture Notes in Computer 440–448.
Science, 8074, 310–321. Waterman, R. W., & Meier, K. J. (1998). Principal–agent models: An expansion? Journal of
Panagiotopoulos, P., Moody, C., & Elliman, T. (2012). Institutional diffusion of Public Administration Research and Theory, 8(2), 173–202.
eParticipation in the English local government: Is central policy the way forward? Waters, R. D., Burnett, E., Lamm, A., & Lucas, J. (2009). Engaging stakeholders through so-
Information Systems Management, 29(4), 295–304. cial networking: How nonprofit organizations are using Facebook. Public Relations
Papenfuß, U., & Schaefer, C. (2010). Improving public accountability by aligning reporting Review, 35(2), 102–106.
to organizational changes in public service provision — an empirical Internet study of
all Austrian, German and Swiss towns and states from an agency-theory perspective.
International Review of Administrative Sciences, 76(3), 555–576.
Enrique Bonsón Ph.D. Professor of Finance and Accounting at the University of Huelva
Pew Research Center (2013). Social media update 2013. Retrieved from www.pewinternet.
(Spain). His main research interests are in the field of information technologies in
org (Last accessed: 10 January 2014)
business: digital reporting, continuous auditing and monitoring, XBRL and social media.
Pfeffer, J., & Salancik, G. (1978). The external control of organizations: A resource dependence
Lead researcher of the research team in new technologies in accounting and business
perspective. New York, NY: Harper & Row.
administration (SEJ290).
Picazo-Vela, S., Gutiérrez-Martínez, I., & Luna-Reyes, L. F. (2012). Understanding risks,
benefits, and strategic alternatives of social media applications in the public sector.
Government Information Quarterly, 29, 504–511. Sonia Royo Ph.D. Senior Lecturer of Department of Accounting and Finance at the Univer-
Pina, V., Lourdes, T., & Royo, S. (2007). Are ICTs improving transparency and account- sity of Zaragoza (Spain). Her research focuses on e-government, local government
ability in the EU regional and local governments? An empirical study. Public accountability and citizen participation. She participates in the research team led by
Administration, 85(2), 449–472. Lourdes Torres in accounting, management, and auditing of public sector reforms
Pina, V., Lourdes, T., & Royo, S. (2009). E-government evolution in EU local governments: (http://gespublica.unizar.es).
A comparative perspective. Online Information Review, 36(6), 1137–1168.
Pina, V., Lourdes, T., & Royo, S. (2010). Is E-government promoting convergence towards Melinda Ratkai Ph.D. Her main research interests are digital reporting, corporate
more accountable local governments? International Public Management Journal, 13(4), transparency, social media, business ethics, stakeholder engagement and business
350–380. communication. She participates in the research team led by Prof. Enrique Bonsón in
Ramanadhan, S., Mendez, S. R., Rao, M., & Viswanath, K. (2013). Social media use by new technologies in accounting and business administration (SEJ290).
community-based organizations conducting health promotion: A content analysis.
BMC Public Health, 13(1).