Advocator, Jester, Spokesperson, Provocateur or
Boundary spanner? Exploring different
communication styles at Twitter
Sanna Ketonen-Oksi *
Department of Information Management and Logistics
Tampere University of Technology
Korkeakoulunkatu 10, 33720 Tampere
Department of Business, ICT and Chemical Engineering
Turku University of Applied Sciences
Lemminkäisenkatu 30, 20520 Turku
* Corresponding author
Purpose – The recent development of digital communication technologies, and of social
media in particular, have enhanced more direct communications between companies and
their customers. Among many other things, the use of social media has become
considerably popular in customer services. However, communicating with different types
of customers is not easy. More profound understanding is needed about how to succeed in
communicating with the customers in the increasingly impersonal, yet often emotionally
sensitive online environments.
Design/methodology/approach – Based on an extensive empirical data from Twitter
discussions on climate change and energy industry, the analysis will follow the ideas and
concepts of research on personalities and motivation in the context of social media.
Originality/value – By theorising the impacts of human personality traits to a person´s
communication style in social media, in accordance with the person´s own choices of roles
and motivations to communicate in social media, this study will provide companies new
insight on how to approach their customers in online environments.
Practical implications – This study offers significant information for any company that
wants to improve their customer service through social media. That is, by presenting the
early phase taxonomy for different social media communication styles used in Twitter, this
study will provide companies with both new insight and practical advice on how to better
share information and manage discussions on their social media channels, considering the
different communications styles of their customers.
Keywords – Social Media, Twitter, Big Five, Uses and Gratifications Theory
Paper type – Academic research paper
In his book review “Invisible Manipulators of Your Mind”, from April 2017, Tamsin
Shawn declares how modern behavioral scientists have attained the capacity to manipulate
people’s emotions, that is, their fundamental preferences, values, and desires (Shaw, 2017).
Yet, it was already in 1980 when Dervin advocated that those working in media or
conducting information campaigns should begin by examining the potential information
users and their specific needs for the information. Since then, a massive number of studies
have been conducted in order to increase the understanding of customer needs and the ways
to engage them through information sharing in different media. With a growing trend
towards stressing the individual use and choice of different media (e.g. Rubin, 1994) and
the cultural impacts (e.g. Lull, 1995) from the 1990´s onwards, the more recent research
has been more or less dominated by the emerging need to understand the shift from
analogue to digital technologies (e.g. Hargittai & Walejko, 2008).
With various studies indicating that the use of Internet, and of social media in particular,
has lead to profound changes in the media users’ personal and social habits and roles
(Amichai-Hamburger & Vinitzky, 2010; Bakshy et al., 2015; Baym, 2015), the use of
Internet and social media has now become a part of our everyday lives, and the ways how
we use it is strongly influenced by our personalities (Correa et al., 2010). In doing so, the
use of Internet and social media is now more and more linked to individuals’ motivations
for self-gratification (Ruggiero, 2000) and of value creation in multi-stakeholder service
systems (Jaakkola et al., 2015; Singaraju et al., 2016; Wieland et al., 2016).
Besides the prolonged dominance of trait theories which seek for understanding the
consistent, enduring ways of thinking, feeling or behaving according to personality traits,
it has been argued that some of the individual behavioural patterns may turn out rather
opposite once online and compared to their behaviour offline (Amichai-Hamburger &
Vinitzky, 2010; Kraut et al., 1998). That is, according to some research, individual
variations do exist within the structures of personality states and emotions, depending on
time, context and generation (Revelle, 2009; Correa et al., 2010) of the person in question.
On the other hand, it is to be considered that many of the external influences affecting our
behaviour do so through cognitive processes only and not directly. It means that observing
environmental events is at least partially impacted by the cognitive factors, thus defining
the emotional impact and the power of motivation. With that in mind, the cognitive
processes will also define how the information they convey in different forms of
experience-based symbols will be organised for future use. (Bandura, 2001.)
An extensive research has been carried out related to the development of Uses and
Gratifications Theory (Ruggiero, 2000), with a focus on reasoning people´s motivations to
use social media, and increasing the understanding of the impacts of personality, age and
gender to social media use and motivation (e.g. Correa et al., 2010; Seidman, 2013). Or, on
studies like the Social Cognitive Theory of Mass Communication (Bandura, 2001), which
offers a conceptual framework for analysing the determinants and psychosocial
mechanisms for using symbols in communication. While a lot of research has focused on
understanding the motivations of mass behaviour, the knowledge and understanding about
how to customise information sharing according to specific personalities or communication
styles is still very limited. That is, besides constructing social media architectures that guide
people towards specific behaviour, quite little is known about how to adjust to different
communication styles in customer service situations for instance.
Hence, the purpose of this study is to explore the potentially different communication
styles in social media. That is, a sample of Twitter discussions related to energy industry
and climate change was being collected, and then analysed by the means of Grounded
Theory. In doing so, the analysis was inspired by probably the most common personality
trait theory of our modern time, the “Big Five”, as well as by the Uses and Gratifications
theory. Hence, an early attempt of taxonomy of analysis for different communication styles
for social media will be presented as a conclusion of this paper.
2 Theorising social behaviour and communication in social media
Developing a theory or a new framework for studying human interactions is a complex
activity. So, understanding that this study only represents our first attempt towards creating
the taxonomy of analysis for social media communication, the term “theorising” will thus
be used instead of speaking of theory making. By theorising we also mean that this work
will entail conceiving or intuiting ideas and concepts which then will be formulated into a
logical, systematic, and explanatory scheme by grasping on the meanings of events or
happenings that might seem obscure in the first glance. In doing so, this paper aims at
constructing an explanatory scheme, providing novel and versatile understanding of the
Given that Twitter is seen as a channel for gratifying the needs to connect with others
(Chen, 2011), the motivational aspects of social media communication - hereby presented
through the lenses of the Uses and Gratifications Theory - will be considered as at least
somewhat important to the analysis of communication styles. In addition, the Five-Factor-
Model taxonomy - often called as ”the Big Five” or the OCEAN - a commonly accepted
framework that classifies the various and diverse systems of personality to five main
dimensions, will serve as a good starting point for theorising the communication styles.
(Matthews et al., 2003). Hence, both human personality traits and inner motivations may
have significant input to a person´s communication style.
2.1 Motivations for social media communication
Some recent studies indicate that the more active people are on Twitter, the more they
are expected to gratify their needs for an informal sense of friendship and connection with
other users (Chen, 2011). With a number of studies having applied the uses and
gratifications theory for understanding the motivations underpinning consumers’ media
choice and usage, it differs from many previous mass media theories by assuming that,
instead of passively receiving media, people tend to actively prefer certain media according
to the media´s ability to satisfy their specific needs (Katz et al., 1973). Considering this,
many Twitter users look for weak ties with the purpose of getting connected. As people
also tend to look for those alike, the more they discuss, the more they become alike with
those they are connected with. (Chen, 2011.)
The uses and gratifications theory represents a synthesis of many studies including
gratifications typologies of traditional mass media (Katz et al., 1973). The five categories
are: information seeking, entertainment, social interaction, self-expression, and impression
management (see table 1).
Table 1. Uses categories explaining social media motivation (adapting Katz et al., 1973).
To learn about news and events
To entertain oneself
To exchange social support, to maintain existing
relationships, and to meet new friends
To share information about themselves and to show
who they are and what they like
To give others a positive impression of oneself
The information seeking type of communication aims at lowering the barriers to
information share through networks of trust and shared interests, it refers to people who
mostly use social media to learn about news and events (Kwak et al., 2010),
recommendations about products and content (ibid.), or about any subject in their interest.
The entertainment then refers to the use of social media for entertaining oneself, that is, for
browsing interesting content shared by others and for sharing others’ life experiences, thus
often using negative expressions while trying to escape from problems or dullness. In
comparison to entertainment, the social interaction refers more to a need to exchange social
support, to maintain existing relationships, and to meet new friends. That is, social media
is used as a tool to enhance connectedness and to develop common ground (Wright, 2016).
When self-expression takes place, people are using social media to show who they are
and what they like. This behavioural characteristic is associated with a motive for self-
verification, that is, to presenting one’s true self to the outside world, to confirm an
established self-concept, and to maintain consistency in self-knowledge (Escalas &
Bettman, 2003). It is also associated with identity creation, i.e. for obtaining peer
acceptance and exchanging social support. As a difference to self-expression, the
impression management refers to social media use which aims at deliberately creating a
positive impression of the user or, at some cases, to develop social relationships and
promote self-status (Wright, 2016).
2.2 “Big Five”
What then comes to understanding different personalities, the research has been going
on for decades now. One of the pioneering taxonomies was made by Cattell (1946a) in the
1940´s. First, he subdivided all the traits under two main categories – surface and source
traits. With the surface he then referred to personality traits such as shyness of talkativeness,
easily observed in interaction with other people. With source traits he referred to qualities
that underlie beneath, such as being introvert or extrovert. Later, the same year, Cattell
(1946b) defined three main categories, referring to traits that reflect abilities, traits that are
dynamic and traits that are more stable.
With a vast number of surveys and statistical analyses conducted in the 1980´s and
1990´s, Cattell´s early work still forms the basis of modern studies of personality traits.
Along the way, the combination of the five main personality types, “the Big Five”, was also
being initialised. (John & Srivastava, 1999; Revelle, 2009.) Often referred to as the
OCEAN model, the five personality dimensions (table 2) are considered to cover hundreds
of more specific personality characteristics, as well as adjustable to different cultural or
linguistic settings (McCare & John, 1992). More importantly, considering the aim and
scope of this study, some researchers have claimed that as people do not act consistently
from one situation to another, and since people are strongly influenced by situational forces,
these traits might not be as stable as predicted (Diener, 2009).
Table 2. “Big Five” personality traits according to McCare & John (1992).
Characterisation of the dimension
New ideas, values, feelings and behaviours
Orderly, responsible, dependable
Talkative, assertive, energetic
Good-natured, cooperative, trustful
Anger, worry, sadness
3 Case study
3.1 Case description
Applying a special tool created for social media researchers (see futusome.com),
providing an easy access to all publicly open social media data in Finland, a list of most
employed keywords referred to in social media discussions connected with the energy
industry was being generated. Considering Twitter as an important social media platform
for sharing ideas and information between experts – and for both shaping and forecasting
the more popular discussions and interests on the field of renewable energy - the analysis
for limited to Twitter only.
3.2 Data collection
All in all, 58.194 tweets were thus collected in between February 2016 and February
2017. From this data, about 47,5 % of the users were civilians, that is, individuals with no
reference to companies, political parties or any other organisations in their profile. With
about 11 % of the users representing companies with no particular links to environmental
issues, less than 5 % of the users had profiles indicating of their expertise in environmental
organisations or in companies working in the energy industry. Again, about 9 % of the users
were citizens with high interests on environmental issues.
From the 30.313 original tweets and 27.881 retweets, a sample of 10.130 tweets,
representing 357 individuals having posted a minimum of 10 tweets, was selected for a
closer examination. Then, every fifth tweet was categorically selected and coded for further
analysis. Eventually, all the 2.101 tweets were individually coded according to the
following six (6) categories: 1) The function of the tweets was classified as either
informing, connecting, collaborating, confrontational or with an aim to position the tweeter
at some way. 2) The style of the tweets was classified as descriptive, editorialised, critical,
supportive or entertaining. 3) The argumentation of the tweets was judged to be reasoned
either emotionally or through logical reasoning. 4) The tweets were categorised based on
their targets: individual, organisation, and both or no target. 5) The number of used
keywords with # were counted. 6) Each tweet having a) a picture or b) a link to some other
content was marked with yes or no.
3.3 Method of analysis
The socio-psychological communication tradition as the starting point, applying the
methodology of grounded theory, intuitive ideas and concepts were conceived and then
formulated them into a logical, systematic, and explanatory scheme for further analysis. In
practice, by coding the qualitative data, the focus of analysis did not hang on single
responses or phrases, but assured a wider perspective for the analysis. In doing so, the
researchers were forced to examine their basic assumptions, their biases, and their
perspectives. As such, they were more alert than usual to notice possible properties and
dimensions or to grasp on the meanings of rather obscure events or causalities. Yet, in order
to include the impact of roles (categorised into politician, journalist, specialist,
environmentalist and citizen) on the social media communication styles, an analysis of the
Twitter profiles was also being made.
As moving from the level of description to that of abstraction will become more
effective, observing both variation and general patterns will also become more appearent.
Thereby, the analysis did not focus on finding the most relevant questions and answers, but
on properties and dimensions enhancing better understanding of the data. By consequent,
grounded theory was significantly different from the traditional research models with
researchers choosing an existing theoretical framework – with data being collecting to
demonstrate its potential implications to the phenomenon under study. (Strauss & Corbin,
The results of the data will be discussed in the light of the five personality characteristics
described in the Five-Factor-Model (FFM) (McCare & John, 1992), presented in section 2.
The theorising will consist of both making inductions (i.e. of deriving concepts, their
properties, and dimensions from data) and deductions (i.e. of hypothesising about the
relationships derived from data and found between different concepts). That is, the aim of
this paper remains in creating an understanding of the case data, not in forming an
overarching explanatory scheme through interpreting events or explaining why certain
events occurred and not others. As a result of the evolving theoretical analysis, further
research questions will thus also be suggested for continuing the theory (taxonomy)
4.1 General overview
The aims, methods and data of this research being taken into account (see section 3), all
2.101 tweets were first coded according to their contents (see table 3). The classification
into different communication styles was then initiated rather intuitively, based on
discussing the observations of both authors. Hence, a logical, systematic, and explanatory
scheme for analysis was being formulated in an excel format. The following five
communication styles were found: advocator, jerker, spokesperson, provocateur, and
The categorisations were largely based on the functions of communication, i.e. the
attitudes and/or motivations to tweet. While the advocators (41 %) and spokespersons (82
%) are considered as those mostly interested in sharing information, the jesters (68 %) and
provocateurs (52 %) seem to focus on challenging others with rather negative or at least
provocative tweet contents. Hence, the boundary spanners are active both in information
sharing (36 %) and in social interaction (37 %). As we then take a look at genders, it is
interesting to see that only 8% of jesters are females and that most of the females (30 %)
fall in the category of boundary spanners. Or, that the number of specialists, including e.g.
company representatives, researchers and public servants, is highest among the
spokespersons (38 %), whereas the number of politicians (5 %) is smallest among the
On the other hand, only 10 % of the jesters base their arguments on rational reasoning,
meaning they are very emotional (90 %) in their expressions and argumentation. The most
rationally argumented tweets are those of spokespersons (95 %), advocators (77 %) and
boundary spanners (with 63 %). More variation appears among the provocateurs as their
tweets are categorised as 41 % rationally argumented and 59 % emotionally argumented.
Table 3. Overview of the results.
While about half of the tweets were not directed to third parties at all, the spokespersons
often address their tweets to both individual(s) and organition(s) (22 %), or at least to either
of them (see 17 % for individual(s) and 11 % for organisation(s)). The advocators (39 %),
jesters (29 %) and provocateurs (38 %) often target their tweets to individual(s) only.
In addition, about half of the advocators (43 %), jesters (42 %) and provocateurs (41
%) included a (web) link to their tweets, while for the spokesperson that was a lot more
common (74 %) and the boundary spanners (6 %) rarely shared any links. Less variation
(34-55 %) appeared when observing whether the tweets had a hashtag included or not.
Furthermore, what table 3 does not tell, is that tweets by jesters went 2-3 times more viral
than the others and that they got, on average, about 4 times more likes than other tweets.
4.2 Taxonomy building – first attempt
When reviewing the data each communication style at a time, the following descriptions
were initiated (table 4).
Table 4. The five social media communication styles.
Advocator: Likes sharing information with rational arguments. Is socially active, and
mostly targets tweets to individuals, not organisations.
Jester: With strongly emotional tweets, the jester often targets individuals with his
(her) tweets. Likes challenging other people, but not necessarily in a negative way.
Spokesperson: Very rational, focus on information sharing with additional links. The
profile of a specialist or expert.
Provocateur: Likes challenging others in a mostly socially and emotionally positive
way. Targets his (her) tweets mostly to individuals.
Boundary spanner: Specialists who communicate in a rational way. Strong drive to
reconcile opposite point-of-views. Active in information sharing, but the tweets rarely
Although rather compatible with the big five personality traits or the motivational
categories of the uses and gratifications theory, it is important to note that this
categorisation by styles does refer to individuals as such. That is, the communication styles
are not stable and linked to a tweeting person, but on the context of the discussions and on
the role the person takes regarding the matter being discussed. In fact, the same person may
appear in several categories depending on the role and context.
5 Discussions and conclusions
As a result of this study, the following communication styles were identified: advocator,
jerker, spokesperson, provocateur, and boundary spanner. By identifying these five
archetypes, this study gives valuable insight on the importance of creating profound
understanding of different motives and styles for communicating in social media. As we
can see from figure 1, visualising the double dichotomy between a) connection and
disconnection and b) affective and cognitive, these five communication styles quite closely
mirror the different human personalities presented by the Big Five theory (McCare & John,
1992). Yet, following the characteristics of any living system (Maula, 2006), these different
communication styles should not be seen as stable dimensions, but as roles and functions
that continually self-produce themselves according to their context.
Figure 1. Visualisation of the five different communication styles.
Several studies have suggested that companies should use robust techniques to monitor
social media discussion (e.g. Fan & Gordon, 2014; Lee, 2017). Sentiment analysis and
opinion mining techniques have been seen as solutions for improving companies’
responsiveness to customers’ (existing and potential) needs. Without questioning the
usefulness of monitoring techniques, this paper encourages companies to analyse their
customers’ communication styles. Understanding why the customers communicate as they
do will significantly increase an organisation´s ability to interact with their customers and
to react on customer feedback in ways which best respond to the motives and
communication styles of different customers. Realising that these roles and functions are
not bound to a personality, and that some people might change their communication style
contextually, will also add the companies´ knowledge and understanding about their
possibilities to influence their customers through social media. In doing so, it will be crucial
for the companies to recognise that the existing roles and functions will not predict certain
behaviour, but that they are continuously reproducing themselves (Bolton et al., 2013) and
thus need regular updating.
By limiting this study in one theme specific Twitter data, with only a shallow analysis
on user profiles, and with no longitudinal data for observing variations in the
communication styles over time, the generalisation of these results are debatable. More than
that, a sample of Twitter users does not stand for a valid sample of all social media users
and differences can appear according to different media.
Regardless some limitations, this study offers a great stepping stone for testing the
results with other social media data, including the analysis of both individual and
organisational user profiles. Conducting discourse analysis or visual analytics of these
discussions – from a single communication style perspective or between different
communication styles – might reveal additional information worth studying for. Moreover,
considering that young adults between 18 to 34 years old are more likely than any of the
older age groups to prefer social media use for social networking, and that they are more
influenced by others' opinions in social media (Bolton et al., 2013), comparing the
communication styles between different age groups would be an interesting subject for
Amichai-Hamburger, Y., & Vinitzky, G. (2010). "Social network use and personality". Computer in
Human Behaviour, Vol. 26, pp. 1289–1295.
Bakshy, E., Messing, S., & Adamic, L. A. (2015). "Exposure to ideologically diverse news and
opinion on Facebook". Political Science, Vol. 348, No. 6239, pp.1130–1132.
Bandura, A. (2001). "Social Cognitive Theory of Mass Communication". Media Psychology, Vol. 3,
No. 3, pp. 265–299.
Baym, N. K. (2015). Digital Technology and Social Change. Maiden, USA: Polity Press.
Bolton, R. N., Parasuraman, A., Hoefnagels, A., Migchels, N., Kabadayi, S., Gruber, T., Loureiro Y.
K. & Solnet, D. (2013). "Understanding Generation Y and their use of social media: a review
and research agenda". Journal of Service Management, Vol. 24, No. 3, pp. 245–267.
Cattell, R. B. (1946a). "Personality structure and measurement. I. The operational determination of
trait unitics". British Journal of Psychology, Vol. 36, pp. 159–174.
Cattell, R. B. (1946b). "Personality structure and measurement. II. The determination and utility of
trait modality". British Journal of Psychology, Vol. 36, pp. 159–174.
Chen, G. M. (2011). "Tweet this: A uses and gratifications perspective on how active Twitter use
gratifies a need to connect with others". Computers in Human Behavior, Vol. 27, No. 2, pp. 755–
Correa, T., Hinsley, A. W., & Zuniga, H. G. de. (2010). "Who interacts on the Web?: The intersection
of users’ personality and social media use". Computer in Human Behaviour, Vol. 26, pp. 247–
Dervin, B. (1980). "Communication gaps and inequities: Moving toward a re-conceptualization".
Progress in communication sciences. Vol. 2, pp. 73–112.
Diener, R. E. (2009). "Personality, and subjective well-being", The Science of Well-Being, Vol. 37,
Escalas, J. E., & Bettman, J. R. (2003). "You are what they eat: The Influence of Reference Groups
on Consumer Connections to Brands". Journal of Consumer Psychology, Vol. 13, No. 3, pp.
Fan, W., & Gordon, M. D. (2014). "The power of social media analytics". Communications of the
ACM, Vol. 57, No. 6, pp. 74—81.
Hargittai, E., & Walejko, G. (2008). "The participation divide: Content creation and sharing in the
digital age." Information, Communication & Society, Vol. 11, No. 2, pp. 239–256.
Jaakkola, E., Helkkula, A., & Aaarikka-Stenroos, L. (2015). "Service experience co-creation:
conceptualization, implications, and future research directions". Journal of Service
Management, Vol. 26, No. 2, pp. 182–205.
John, O. P., & Srivastava, S. (1999). "The Big-Five Trait Taxonomy: History, Measurement, and
Theoretical Perspectives". Handbook of Personality: Theory and Research, Vol. 2, pp. 102–138.
Katz, E., Gurevitch, M., & Haas, H. (1973). "On the Use of the Mass Media for Important Things".
American Sociological Review, Vol. 38, No. 2, pp. 164–181.
Kraut, R., Lundmark, V., Patterson, M., Kiesler, S., Mukopadhyay, T., & Scherlis, W. (1998).
"Internet paradox: A social technology that reduces social involvement and psychological well-
being?" American Psychologist, Vol. 53, pp. 1017–1031.
Kwak, H., Lee, C., Park, H., & Moon, S. (2010). "What is Twitter, a social network or a news media?"
In WWW ’10 Proceedings of the 19th international conference on World wide web, pp. 591–
600. Raleigh, NC, USA.
Lee, I. (2017). "Big data: Dimensions, evolution, impacts, and challenges". Business Horizons, in
Lull, J. (1995). Media, communication, culture: A global approach. New York, NY, USA: Columbia
Matthews, G., Deary, I. J., & Whiteman, M. C. (2003). Personality Traits (2nd ed.). Cambridge,
USA: Campridge University Press.
Maula, M. (2006). Organizations as Learning Systems. “Living Composition” as an Enabling
Infrastructure. Bingley, UK: Emerald Group Publishing Limited.
McCare, R. R., & John, O. P. (1992). An introduction to the Five-Factor Model and its applications.
Journal of Personality, Vol. 60, No. 2, pp. 175–215.
Revelle, W. (2009). "Personality structure and measurement: The contributions of Raymond Cattell."
British Journal of Psychology, Vol. 100, pp. 253–257.
Rubin, A. M. (1994). Media uses and effects: A uses and gratifications perspective. In Media effects:
Advances in theory and research, pp. 417–436. Hillsdale, NJ, USA: Lawrence Erlbaum
Ruggiero, T. E. (2000). "Uses and Gratifications Theory in the 21st Century". Mass Communication
and Society, Vol. 3, No. 1, pp. 3–37.
Seidman, G. (2013). "Self-presentation and belonging on Facebook: How personality influences
social media use and motivations". Personality and Individual Differences, Vol. 54, No. 3, 402–
Shaw, T. (2017). Invisible Manipulators of Your Mind. Retrieved from
Singaraju, S. P., Nguyen, Q. A., Niininen, O., & Sullivan-Mort, G. (2016). "Social media and value
co-creation in multi-stakeholder systems: A resource integration approach". Industrial
Marketing Management, Vol. 54, pp. 44-55.
Strauss, A., & Corbin, J. M. (1998). Basics of Qualitative Research : Techniques and Procedures for
Developing Grounded Theory (2nd ed.). New Delhi, India: SAGE Publication Inc.
Wieland, H., Koskela-Huotari, K., & Vargo, S. L. (2016). "Extending actor participation in value
creation: an institutional view". Journal of Strategic Marketing, Vol. 24, No. 3-4, pp. 210–226.
Wright, M. (2016). Identity, Sexuality, and Relationships among Emerging Adults in the Digital Age.
(e-book): IGI Global.