Social Services and Twitter: Analysis of Socio-Political Discourse in Spain from 2015 to 2019

Abstract

The fundamental role of social networks in all areas of our lives and of social and political interactions is also very important in this new digital environment. The study focused on the analysis of tweets related to social-service issues published on Twitter during the different electoral campaigns in Spain from 2015 to 2019. The sample is 6728 tweets generated between 2015 and 2019 on the topic “social services” for quantitative analysis. In this analysis, we use the Gephi tool to observe how these messages flow on Twitter. The aim was to understand the socio-political discourse of different actors on social services in order to identify priority topics and networks for active Twitter profiles. The results show that users use Twitter for informal communication during the election period, focusing on messages, condemnation and positive evaluation to increase their visibility and influence.

1. Introduction

Social networks are the most widely used and expanded digital media due to the potential they offer for communication between users and participation in the public sphere, and stands out for the possibility for unmediated communication between users and participation in the public sphere [1]. In this public sphere, a “unitary digital agora” [2] is produced, in which citizens and public institutions participate on equal terms.

Political actors and entities use this new environment for social interaction, due to its dissemination, propagation of messages and instantaneousness. In the words of [3], in most cases they opt to maintain a unidirectional communication strategy, seeking only a greater reach for their messages.

A presence on social networks in the era of online communication has become more used and important during the development of electoral campaigns; in this electoral period, political parties and candidates disseminate their slogans and messages to attract potential voters and undecided voters. Social networks, especially Twitter, have become a means of disseminating political propaganda and Twitter is a benchmark social network for participating and interacting in politics, where citizens can access information and send their messages instantaneously [4].

Throughout the years, each temporary space has been related to a traditional means of communication and the political class and, in particular, candidates have taken advantage of and used the media’s influence on voters by making regular use of social networks during the development of electoral campaigns [5]. The development of marketing and political communication has been directly linked to the evolution and progress of mass media, and cannot be understood without them [6]. Every era has been marked by a media and the political class has always known how to take advantage of the media’s influence and impact on voters, making extensive use of the media in election campaigns [7]. In the 1920s, print media predominated, and in the 1950s, television was the most widely used medium [8].

Today, we are in the Internet era, which marks a milestone in communication, with digital media and social networks dominating the ways in which politicians interact with citizens [5,9,10,11].

The so-called information society involves a transformation of all areas of life, and, therefore, social services are not immune to it. On this level, social services are undergoing transformations that affect the activity, strategies and tactics of social work, making it necessary to study the influence of the Internet and social networks on their structure and social processes [12]. In this information society, we are constantly interacting with the Internet and social networks, and this interaction is both human and non-human, leaving a trace of this interaction in the form of data, which are given voluntarily. From the processing of this data by means of analysis techniques, data on population interest, emotions, and interests, etc., can be extracted quite reliably.

One of the sources of data for this Big-Data processing is the messages posted on social networks such as Twitter, where users provide useful information applicable to different fields. In this sense, [13] points out that Big Data has the capacity to reveal information that is in the subconscious, and of which we are not aware.

Currently, the application of Big Data in the business sector is wide [14] and it is beginning to be used in politics. Today’s social services do not have, among their approaches, the use of Big Data [15], although there is currently a call for full immersion in the digital era and for this discipline to be at the forefront of innovation [16]. In the 1980s, the ideology of “management” was imposed, also known as managerial ideologies, ideologies derived from economic neoliberalism, where the company and its management is the epicentre of everything [17]. According to this ideology, the system of accountability is imposed in social work [18], where the social-services professional is obliged to justify and explain the actions undertaken, with computer systems implemented as a mechanism to streamline and standardise processes [19]. However, these systems are aimed at measuring the efficiency and performance of the objectives imposed in order to be able to assess whether public funds have been used correctly [20]. “This managerialist view of social work causes a displacement of the relationship with users in favour of a managerialist approach based on data collection and bureaucratic completion” [21].

Although the managerial ideology is being progressively abandoned in the practice of social services, in favour of a more relationship-based system in which the bureaucratic burden has been lowered, it has not disappeared completely. To achieve this end, a step has been taken towards systems based on people’s digital participation, designing intelligent systems that manage to lower the bureaucracy in the practice of social services, and also allowing citizens to relate digitally [22]. This is achieved thanks to the fact that citizens are gradually becoming more capable of using digital technologies, which means that the data that used to be entered by social workers into management systems are now entered by users, freeing professionals from this process.

In the words of [15], “this bureaucratic work of the social services professional, as well as the use of social networks, generates a huge amount of data that is currently wasted”.

This use of ICT (information and communication technologies) in social services ranges from the level of intervention (implementation of social policies, social-justice issues, administration of organisations and communities, etc.), to the level of education and research (online courses and seminars, participation in online work, interacting with other social workers, etc.). In short, it can be said that ICTs are continuously redefining the boundaries of the profession, and in the face of these challenges, social work, both as a profession and as a scientific discipline, cannot remain on the sidelines of the digital transformation of society [23]. Just as ICTs are redefining society in all their fields of action, in social work they are redefining it in three dimensions: (1) in the social context in which it intervenes; (2) in the tools used for professional practice; and (3) in the competences and skills required.

Social networks and ICTs can impact or change social practice and offer new opportunities for social-service professionals in the coming years. Digital and social technologies provide older people with better social care solutions to improve their quality of life and well-being. Evolutions in web and digital technologies will continue to have a significant impact on all aspects of life and social care. However, this technological change and its impact on society is also accompanied by new ways of caring, sharing and being. This brings with it both opportunities and ethical, practical and human challenges. It is, therefore, essential that social services and social-care professionals can enhance their digital professionalism and their ability to support, guide and protect a digital citizenry of children, adults and families in an increasingly mediatised digital world where individual privacy is invaded [24].

Social networks are digital platforms where users generate content, messages and share them publicly or privately [5]. Messages can be both written and audio-visual [25]. In today’s society, in which we live on screens and social networks, interaction between political parties, candidates and users present on digital platforms is possible [26].

In this way, social networks have become the central node of political communication flows [27], a space in which, in the case of Twitter, numerous sources compete for the establishment of the political, economic and social agenda [28]. The traditional actors of political communication, media or partisan, coexist with many other emerging actors who, like citizens themselves, try to influence the digital public space from the same level [29,30].

Messages can be both written and audio-visual. In today’s society in which we live on screens and social networks, interaction between political parties, candidates and users present on digital platforms is possible [31]. In this decade, some strategies are already being developed to bring the political class closer to the citizenry using the Internet. During election campaigns, political parties have created web pages to support their candidates and have even included spaces for debate, such as forums or live chats [32,33,34].

From 2002 onwards, candidates started to use social networks, such as blogs, for their election campaigns, due to the potential they offer [35]. The Internet and social media introduced new possibilities for political communication, changed politics itself and redefined the relationship between politics and the media [36,37]. These new opportunities for interaction and communication have moved political activity online. Political professionals are increasingly aware of the need to have a presence on the networks and the importance of generating relevant and useful content that interests users [38].

This has meant that social networks have become an essential tool among the instruments that candidates use to interact with citizens through their messaging [39]. For this reason, we can no longer speak of political communication planning without the use of Web 3.0 and social networks [40]. Due to the advances of the current society in which we live, the mass media are no longer the only means of communication to reach potential voters. Today, we increasingly see political institutions informing citizens through social networks, and vice versa [41].

This highlights the growing use of Twitter by the general population to stay connected with their political leaders. As pointed out by the CIS (Sociological Research Centre) post-election studies of the 2015, 2016 and 2019 general elections, more than 50% of those surveyed acknowledged having followed information on the aforementioned elections via this social network. It can be stated that Twitter is a primordial social network for political analysis and study which has more than 200 million users around the world, and gives rise to the two-way communication of messages every day. Twitter is a microblogging network, as its messages are condensed into a maximum of characters; in a few words, you have to communicate the message briefly [42].

Twitter also stands out for speed in the emission and reception of its messages, which facilitates interaction between users. Twitter favours communication, direct contact and dialogue, and, since its creation, it has shown its growing importance in the political context [43,44]. From Obama’s US election campaign in 2008 to Trump and Clinton’s last one in 2017, Twitter has been the social network most used by candidates. As a result, studies on Twitter in campaigning have been incorporated into the broad field of study of electoral campaigns in various regional, national and extra-regional contexts [45].

Twitter users are opinion leaders in their environments. Of all the social networks that exist in the ICT world, Twitter stands out for its remarkable rise as an electoral tool [46]. Achieving a greater number of followers on Twitter or any other social network, gaining an advantage in digital polls, ensuring that any idea favourable to the electoral campaign reaches the top positions in terms of the most commented topics on the network or having the support of users who disseminate a good image of the party or candidate, quantitatively determine the probability of achieving an electoral victory [47,48,49].

This study aims to examine the discourse surrounding social services on Twitter in Spain, focusing on the perspectives and perceptions of different agents who participate on the platform. It differs from previous studies by not only analysing political campaigns, but also the periods between campaigns and over a longer time frame. The study will specifically focus on analysing discourse about community and specialized social services. The goal is to understand the ideologies and priorities surrounding social services as perceived by active Twitter users, with a special focus on political discourse and the use of social media to highlight the reality of social services in the national sphere.

2. Materials and Methods

Thus, this research takes as its starting point the socio-political discourse on social services of the different agents on Twitter, in order to define the priority themes of the profiles that actively participated during the elections held in Spain from 2015 to 2019.

During these years, the political landscape in Spain at a national level was involved in different events, which have marked a destabilisation of the traditional two-party system in favour of a multi-party system [50], which has led to repeated electoral due to a lack of clear majorities. In this research, only the messages posted on social networks during election-campaign periods were extracted.

Gephi is an open-source software for visualizing and analysing complex networks. It is a tool that allows users to explore and understand large and complex data sets, such as social networks, biological networks, and transportation networks, among others. With Gephi, users can import and visualize data, apply various layout algorithms, and perform various types of analysis, such as community detection, centrality measures, and statistical analysis. It also allows users to apply different types of visualizations and customizations to the network graph, making it a powerful tool for data exploration and understanding. Gephi is widely used in fields such as social-network analysis, data visualization, and digital humanities, among others. Gephi employs the CSV (comma-separated values) file format, which is a plain-text format which allows data to be stored in a table where each row represents a record and each column represents a variable. The data can be imported into Gephi through the menu File → Open or File → Import. Once the file is imported, the data can be visualised in Gephi using different network-analysis algorithms. These algorithms are used to calculate the relationships between nodes and to generate a visual representation of the network. The most common algorithms used in Gephi are the Force Atlas 2 algorithm, the Fruchterman–Reingold algorithm and the Yifan Hu algorithm [51].

Additionally, different research methods and tools can be used for data collection and analysis, such as text mining or sentiment analysis. It is also possible to use different algorithms or approaches, such as natural language processing or machine learning, to analyse the data. Mentioning these related studies can provide context and perspective on the current study and highlight the unique contributions of the current research [51].

The result of the search enabled the configuration of the database that will be the subject of analysis in this article, with a total sample of 6728 tweets collected. Gephi is a software tool for visualizing and analysing network data. To collect data in Gephi, you can use a variety of methods, such as:

• Importing data from a file: Gephi can import data from a variety of file formats, including CSV, GEXF, GDF, and GraphML. This is useful if you have data that you have collected previously and want to visualize in Gephi.
• Scraping data from the web: Gephi has a built-in web scraping tool which allows you to collect data from websites. This can be useful if you want to collect data from social-media platforms, such as Twitter or Facebook.
• Using APIs: Gephi can also connect to various APIs to collect data. For example, you can use the Twitter API to collect tweets related to a specific topic or hashtag.
• Manual data entry: You can also manually enter data into Gephi, which can be useful if you have a small dataset or if you are testing a visualization.

Once you have collected the data, you can use Gephi to visualize and analyse it. Gephi has a variety of tools and features which allow you to manipulate and explore the data, such as layout algorithms, filtering, and statistics.

The methodological process followed was based on a quantitative analysis using the Gephi network-analysis tool to see how the information flows in the messages that are emitted. Twitter user accounts with which there is interaction and two-way communication were analysed:

The methodological process for collecting data in Gephi depends on the type of data you want to analyse and the analysis approach you want to use. In general, the following steps can be followed:

(a)

Data collection: data is collected from the desired social network or source, either through automated extraction or manual collection.

(b)

Data cleaning and pre-processing: the data is cleaned of any noise or unwanted information and prepared for analysis.

(c)

Import data into Gephi: data is imported into Gephi in a compatible format, such as a CSV or GEXF file.

(d)

Exploratory analysis: visualisation tools are used in Gephi to explore and analyse the data, such as network-analysis tools, centrality measures, and clustering.

(e)

Analysis and interpretation of the results: the results of the analysis are interpreted and conclusions and recommendations are drawn.

(f)

Creation of visualisations: Gephi allows you to create attractive and easy-to-understand visualisations of the analysed data, allowing a better understanding and presentation of the results obtained [51].

3. Results

The results of the research using the free Gephi software 0.10.0 provided a glimpse of information from the set of tools that connect the network of actors (individuals, politicians, and social organisations, etc.) with the emerging social structures that result from the relationships that actors establish. A first observation will be made concerning the way in which the relationships of the accounts are formed, as well as the positions of their components in the social network of the territory and the intensity of the relationships.

The study of the sample made it possible to obtain an X-ray of the discourse on Twitter during the electoral periods analysed. The discourse of politicians, political parties, citizens, and organisations, etc., dealt with the dissemination of information from different perspectives, including complaints and positive assessments, but always from the perspective of improving social services in the broadest sense of the term.

In an exhaustive analysis of the networks of words used in the tweets and how they are related, we can observe in graph 3 the existence of certain words that have a high centrality in the network and that connect the different clusters, serving as a bridge; these are “reform” and “families”. Thus, there is cluster 1 (green) which is made up of terms related to budget issues, Europe, accounts, emergency. In addition, cluster 2 (red) is made up of terms such as protection, support, and participation, etc. In other words, most of the tweets on social services are about families and the reforms needed to help them achieve a better level of protection in terms of budget, as shown in Figure 1.

On the other hand, there are clusters of words that are weakly connected to the two main clusters seen above (1 and 2). Cluster 3 (orange) is made up of words such as citizenship, campaigns, elections, and issues, etc., as shown in Figure 1.

In this quantitative study, we are going to focus on the issues addressed in the candidates’ messages. The purpose of this quantitative study is to answer the question of whether the candidates’ discourse is adapted to the concerns of citizens.

-

Study variables: For this study, we took as a reference value the issues that most concern Spaniards from the CIS study no. 3121 Barómetro of December 2015, specifically in the question asked: “What do you consider to be the main problem that currently exists in Spain?” On the other hand, we considered the topics that politicians talked about on social networks, using the data in

Table 1. Quantifier of the topics discussed.

CIS	%
Unemployment	63.6
Corruption	13.3
Politics	6.8
The economy	3.8
Independence	3.2
Education and teaching	2.3
Health	2.3
Social inequality	0.5
Immigration	0.4
Drugs and alcohol	0.3
Public safety	0.3
Housing	0.2
Islamic terrorism	0.2
Other	2.8

(topics of the messages).

-

Methodological design: For the quantitative study, a bivariate correlation was used, and since the variables have a normal distribution, according to the Shapiro–Wilk statistical tests, the Pearson’s correlation coefficient was used [51].

-

Variables to be studied: The variables to be investigated are those shown below; the data are expressed as percentages of the total set of CIS interviews, with the “Don’t know” responses being omitted, coding them as missing values. In addition, the candidate variables are also shown as percentages of the total number of messages. Table 1 shows these values.

4. Discussion

From the study carried out with the open-source network-analysis software Gephi on the data extracted from Twitter, we can highlight that: in the series analysed from 2015 to 2019, there are increases in the emission of tweets coinciding with the periods of electoral campaigns, and decreases after them [4].

With regard to the interaction between Twitter users, it is observed that the accounts related to political parties and the associations and social entities that participate do not interact with other users, being mainly of a self-propagandistic nature. On the other hand, interaction and conversations between users can be observed, especially in the defence of ideological positions on social services in general, their coverage or funding.

Personalisation and negativism are important factors in election campaigning in Spain, and can influence the way voters perceive and receive information [52].

Personalisation refers to the tendency of political parties and candidates to focus on themselves and their leadership rather than on issues or policies. This can lead voters to base their decisions on the candidate’s image or personality rather than on their policy platform.

On the other hand, negativism refers to the tendency of political parties and candidates to attack or criticise their opponents rather than promote their own views. This can lead voters to receive biased or inaccurate information about different candidates and their policy proposals [53].

It is important for voters to inform themselves critically and seek reliable sources of information in order to make informed choices in elections.

5. Conclusions

In order to perform this study, one of the main problems encountered was the timeframe, as there are a large number of variables that make up an in-depth study of digital communication and, on the other hand, there is a large amount of data produced by this communication on social networks, even considering the short timeframe for data collection in the study, such as the electoral campaigns.

The social-services system has been one of the most neglected in recent years in terms of public services, both in terms of investment and in terms of social and public attention. This social crisis has made visible the importance of its services, especially to the most vulnerable groups and as a guarantor of basic needs.

On the other hand, when we process a large amount of data like we are dealing with in this research, it is necessary to have powerful computer equipment. This issue slowed down the performance of some analyses in Gephi, given the large amount of information that the programme had to process.

In the election campaign in Spain, social services are an important issue as they are essential to ensure a good quality of life for citizens. Social services include services such as health care, education, housing, social assistance and care for the elderly and disabled.

Political parties and candidates often propose policies to improve social services in the country. This may include increasing the budget for social services, improving the quality of services and increasing access to services for those most in need [54].

For example, some political parties propose increasing the budget for health care to improve the quality of, and access, to health services, while others propose improving education through increased investment in education and teacher training.

Immigration is another important issue in Spain’s election campaign, as the country has experienced a large influx of immigrants in recent decades. Political parties and candidates often propose policies to manage immigration, including asylum and integration policies [55].

Other political parties propose policies to improve housing, such as building social housing and improving accessibility to housing services, for people with fewer resources.

In terms of social assistance, some political parties propose increasing benefits and services for the elderly and disabled, while others propose measures to combat poverty and reduce economic inequality.

Each election has its own issues and dynamics; it is important to keep up with current news and information to know what the main issues are and how the parties and candidates are addressing them. In short, social services are an important issue in election campaigns in Spain, as they are essential to guarantee a good quality of life for citizens. Political parties and candidates propose policies to improve social services in the country, with the aim of improving the quality of life of the population.

One of the future lines of research, due to time constraints, would be to study all the variables that influence digital-communication models and to continue to deepen this analysis over the next few years to see how the discourse on Twitter evolves. A study could also be carried out on how social networks are used to talk about social services in the United States and how this differs from the discourse in Spain. Finally, another study could focus on how a different social-media platform, such as Facebook, is used to discuss social services and how the discourse differs from Twitter.