SlideShare a Scribd company logo
Social Influence and
Information evaluation
SELECTIVE EXPOSURE TO INFORMATION AND USERS INTERACTION
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
1
Context
The combination of Web-based self-publication and social media
requires new skills to evaluate information
This is the main challenge of information literacy
Relevance, cognitive authority can’t be the main criteria in an
environment defined by social interactions
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
2
“Social media”: what does it mean?
It is not social networking
It is much more to do with what people are doing with the technology than the
technology itself, for rather than merely retrieving information, users are now
creating and consuming it, and hence adding value to the websites that permit
them to do so
(Campbell et al., 2011, p. 87)
A group of internet-based applications that build on the ideological and the
technological foundations of Web 2.0, and that allow the creation and exchange
of User Generated Content
(Kaplan and Haenlein, 2010, p. 61)
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
3
Information evaluation allows
To decide where to begin
To predict which source/system would give me the best information
To select
To accept or reject
To determine whether to use or share the information
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
4
Making judgments of information
Traditional approach to information evaluation: Identifying a set of criteria
people employ when making judgments of information
Information Quality: Evaluating the values of information in terms of
excellence or truthfulness (Taylor)
Credibility: People’s assessment of whether information is trustworthy
based on their own expertise and knowledge (Rieh)
Cognitive Authority: Influence on one’s thoughts that one would recognize
as proper (Wilson). A credible source even though it did not have any
influence on our thoughts
Trust: Belief about the reliability of, dependability of, and confidence in
person, object, or process (Fogg)
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
5
Social influence
Hillman and Trier (2013, p. 3) state that social influence “provides a
broad range of concepts to explain how people’s individual actions
are affected by other people as a result of interaction”. This implies
that social influence is a natural process, but can be used by people
or businesses to change a person’s attitude or behavior.
Social influence can be used for positive actions (e.g. creating
awareness for societal problems, promoting new products) and
negative actions (e.g. social hacking, social pressure)
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
6
Normative social influence
Kelman (1985) is often cited as a fundamental analysis of normative social
influence. This type of social influence explains how individuals are
influenced, based on norms.
Kelman distinguishes three sub-types of normative social influence:
compliance, identification and internalization
Compliance occurs when an individual accepts the opinion of others,
hoping that this would lead to a favorable reaction of others.
Identification means that an individual accepts the opinion of others to
maintain a desired relationship.
Internalization represents the strongest influence and occurs when an
individual accepts and beliefs the opinion of others both in public and
private
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
7
Informational social influence
Informational social influence is explained by Lee, Shi, Cheung, Lim & Sia
(2011). This type of social influence involves accepting information or
advice from a person who may not have previously been known as a
friend or colleague.
Informational social influence is especially relevant in the context of social
media, in which user-generated content is an important type of
information.
An example of this type of social influence in social media could be a
change in purchasing behavior as a consequence of online customer
reviews of a product. These reviews change the attitudes and beliefs of
customers and thereby influence behavior.
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
8
Rationale of the study
To better understand social influence effects on information behaviour, we
compare studies on misinformation in social networks to another one,
conducted by one of our doctoral student, Albaraa Altourah, related to
agenda setting in Twitter (can a mass media theory like agenda setting be
applied to Twitter?).
Our question is:
can selective exposure to information both create division (divergence)
between users into groups who follow the same interests without being
interested to the others and also generate convergence in the form of
public opinion or common agenda setting ?
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
9
Misinformation studies
SOCIAL INFLUENCE AND FRAGMENTATION OF DIGITAL
ENVIRONMENT
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
10
Background 1
World Economic Forum (WEF Report 2013)
Massive digital misinformation is becoming pervasive in online social media: it
has been listed by the WEF as one of main threats of our society
A french resolution this year
The parliament (The French National Assemby) adopted, a few month ago, a
resolution entitled « Resolution on sciences and progess in the Republic).
It « invites the government to think about pedagogical practices based on
sensible (smart) use of digital technologies, especially information selection
learning that would make easier the difference between knowledge and
opinion without scientific basis
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
11
Original version
TEXTE ADOPTÉ n° 926
SESSION ORDINAIRE DE 2016-2017 21 février 2017
RÉSOLUTION sur les sciences et le progrès dans la République.
L’Assemblée nationale a adopté la résolution dont la teneur suit :
(…)
9. Invite le Gouvernement à réfléchir à des pratiques pédagogiques fondées sur
l’usage raisonné des technologies numériques, en particulier à l’apprentissage
du tri de l’information qui faciliterait la distinction entre des savoirs établis et
des opinions sans fondement scientifique ;
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
12
Background 2
Algorithmic-driven solutions have been proposed (Qazvinian V and al.
2011, Ciampaglia GL et al. 2015, Resnick P. 2014, Gupta A. et al. 2014, Dong XL, et al. 2015, ...)
Google tries to develop a trustworthiness score to rank the results of queries
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
13
Example: Bessi et al. 2016
« we analyze the users behavior exposed to the same contents on different
platforms—i.e. Youtube and Facebook. We focus on Facebook posts linking
Youtube videos reported on Science and Conspiracy pages. We then
compare the users interaction with these videos on both platforms »
« We limit our analysis to Science and Conspiracy for two main reasons:
a) scientific news and conspiracy-like news are two very distinct and
conflicting narratives;
b) scientific pages share the main mission to diffuse scientific knowledge
and rational thinking, while the alternative ones resort to unsubstantiated
rumors. »
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
14
Data collected
The first category (conspiracy theories) includes the pages that
disseminate alternative, controversial information, often lacking
supporting evidence and frequently advancing conspiracy theories.
The second category (science news) includes the pages that
disseminate scientific information. The third category (trolls)
includes those pages that intentionally disseminate sarcastic false
information on the Web with the aim of mocking the collective
credulity online.
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
15
Results
« Focusing on the consumptions patterns of YouTube videos posted
on Facebook pages, we compute the Spearman’s rank correlation
coefficients between users’ actions on Facebook posts and the
related YouTube videos. We find strong correlations on how users
like, comment and share videos on Facebook and Youtube. Despite
the different algorithm for content promotion, information
reverberate in a similar way »
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
16
Similar polarization of Science and Conspiracy
users in Facebook and Youtube according Bessi
et al. 2016
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
17
« Echo chambers »
“We observe sharply peaked bimodal distributions. Users concentrate
their activity on one of the two narratives. To quantify the degree of
polarization we use the Bimodality Coefficient (BC), and we find that
the BC is very high for both Facebook and YouTube. (…) Content has a
polarizing effect, indeed, users focus on specific types of content and
aggregate in separated groups—echo chambers—independently of
the platform and content promotion algorithm”.
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
18
Users Polarization on Facebook and Youtube
according Bessi et al. 2014
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
19
Core concepts
Echo chamber
Filter bubble (Eli Pariser 2011)
Collective intelligence is a myth
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
20
Issues for Information evaluation
Identification of cognitive biais including:
◦ Confirmation biaises
◦ Representativity biais (Bronner 2017)
◦ And many other biaises that economists and psychologists have already
studied to understand behaviours in financial markets (especially
anomalies)
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
21
Agenda setting study
SOCIAL INFLUENCE AND PUBLIC OPINION IN SOCIAL NETWORK
ENVIRONMENT
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
22
Proposed questions
•Are social media be able to build convergence between users and
contribute to shape a public opinion despite the fact of its high
potentiality of causing echo chamber effects?
•Is there an inter-user influence while using social media
platforms?
•Presentation of research findings of one of my doctoral students,
Alabraa Altourah who conducted a study at the end of 2016 about
Twitter
A. Altourath research questions
•How can the agenda setting be understood in the context of Twitter?
◦ What limits or enables issue salience creation and transfer to subsequently lead to the
manifestation of agenda setting effect on Twitter?
◦ What are the practices that instigate issue salience in an extremely diverse media
platform such as Twitter?
◦ How may the agenda setting be defined in a platform that enables audiences to be an
active part in the communication cycle?
•How can the newly constructed meanings of agenda setting help
understanding agenda building within Twitter?
◦ Is the agenda building process on Twitter strongly attached to the socio-cultural
setting where the study took place?
◦ Are there any external factors that contaminate the agenda building process within
Twitter?
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
24
A. Altourah research design (1)
•Platform study
• Using Twitter analytics tool to identify the manifestation of
agenda setting
• Isolating the agenda setting effects and analyze the conditions
that led to its establishment
• Examining the possible existence of homogenous information
on Twitter that leads to the augmentation of the appearance of
certain news
• Examining the role of cultural setting in hindering or
promoting the agenda setting. Comparison between France
and Koweit
Why Twitter? why France and Kuwait?
•Data may be accessed easily
•Data are available publically
•Data may be researched based on different research settings
•The platform is very popular
•France is one of the top 10 countries in number of Twitter users
(Forbes.com, 2016)
•Kuwait has the highest number of Twitter users per capita globally (one in
three people in Kuwait have a Twitter account) (Forbes.com, 2016)
•The two countries have a fundamentally different socio-political setting
Research design (2)
•Users study
• Identify the elements that leads to the establishment of issue
salience
• Identify the role of users as an active part in the
communication cycle in promoting, hindering or avoiding the
effects of agenda setting
• Examine the socially constructed understanding with respect
to information available on Twitter
• Identify the perceived reliability of information found on
Twitter
Platform study: methodology 1
The first step is executed by analyzing 5 selected Twitter accounts from France
and Kuwait to identify if an issue salience has been created and transferred in
the corresponding cyberspace region.
A period of seven days has been set to collect the tweets of the selected Twitter
account to identify the influencer’s agenda. During the seven days period, two
days were selected as the most applicable days for content analysis, which will
be referred to as the first and the second day of analysis.
The first day of analysis represents the day where the selected users posted the
highest number of tweets in relevance to the week’s average. The second day of
analysis on the other hand refers to the day where the selected users posted the
lowest number of tweets in relevance to the week’s average.
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
28
Platform study: methodology 2
In both the days of analysis, a content analysis method is applied to
identify the most recurrent keywords in the posted tweets. Once the
keywords are identified, they are researched within the corresponding
cyberspace region three days before and three days after their
introduction to assess whether there is an increment or a decrement in
number of tweets mentioning those keywords. This process aims to
identify the Twitter agenda.
The second step is devoted to investigate the level of homogeneity of
media content between selected users, in each study location. This is
conducted by using the socialbearing.com as to collect three-days worth
of tweets of the selected users then analyzing the data using content
analysis.
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
29
The selected Twitter accounts
France :
Nicolas Sarkozy (@NicolasSarkozy)
Marine Le Pen (@MLP_officiel)
Jean-Luc Mélenchon (@JLMelenchon)
Najat Belkacem (@najatvb)
Le Monde (@lemondefr)
Kuwait :
Waleed Altabtabie (@altabtabie)
Nasser Alduwailah
(@nasser_duwailah)
Almajlliss (@Almajlliss)
Safaa Alhashem (@safaalhashem)
Faisal Almuslem (@faisalalmuslem)
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
30
Examples
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
31
Reuse of keywords
after their first introduction
22-24 August 2016 25-27 August 2016 Change in %
Impôt 2351 3837 +63.2%
Être francais 3129 3882 +24.1%
L’immigration 1100 3487 +217%
L’autorité de l’état 75 1668 +2124%
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
32
Agenda setting effects exist on Twitter
09-10 September
2016
11-12 September
2016
Change in %
Laïcité 2454 4624 88.40%
Frexit 717 765 6.70%
L’outre-mer 12 442 3583.30%
Droit des femmes 159 1572 888.70%
Keywords comparison before and after First Day of Analysis (Marine Le Pen)
(Number of tweets)
Keywords reuse/tweet category
(Ex:Sarkozy account)
22-24 August 2016 25-27 August 2016
Tweet Retweet Reply Mentions Tweet Retweet Reply Mention
Impôt 1018 1195 138 818 1123
+10.3%
2434
+103.7%
280
+102.9%
1513
+84.9%
Être francais 441 2276 412 1161 1123
+10.3%
1123
+10.3%
1123
+10.3%
1123
+10.3%
L’immigration 131 861 108 384 578
+341.2%
2494
+189.6%
418
+287%
1648
+329.2%
L’autorité de l’état 26 42 7 16 316
+1115.5%
1251
+2878.6%
101
+1342.9%
637
+3881.3%
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
34
Information on Twitter can be
homogenous (France)
•Twitter account with different political orientation engage in a level of
consistency with respect to tweets they post
•Between the 18th of October 2016 and the 21st of October 2016 the
selected accounts engaged in different topics. However the content
analysis identified similar topics between these accounts:
• Police
• Russie
• Immigration
• CETA (Comprehensive Economic and Trade Agreement)
• Goodyear
• Islam
• US
• Impôt
• Intérêt de la France
The process of agenda building in France
is different than Kuwait
In the Kuwaiti twitter sphere there isn’t any structural attempt to
build an agenda that coincides with pressing matters that are of
high relevancy to the public. Instead, the proposed issues on
Twitter are random, based on the personal political believes and
do not affect the majority of the public.
It is rather an attempt to appeal to a specific political or societal
group. Therefore, the establishment of an agenda and the
increment of discussions pertaining to the issue proposed are
limited (agenda setting effects are less occurring, the
twittersphere is less homogenous)
Social influence
in the Kuwaiti twitter sphere
Twitter accounts in Kuwait are less likely to be influenced by each other’s
with respect to tweets content or by exterior influence source
The Twitter usage in Kuwait may be primarily to support and promote
personal propaganda irrespective what other users are concerned about,
benefiting from their high number of followers and the fact that users in
Kuwait are more likely to use Twitter as their prime source of information
Therefore, the possibility of having an exterior influential entity that might
establish a common agenda among Twitter users in Kuwait is unlikely. The
Kuwaiti culture for long has regarded number of major news sources as
largely biased
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
37
What is similar: Users are immersed in
the illusion of news selection (user study)
What are the factors that determine your choice of Twitter accounts to
follow? (This question aims to determine the elements of the Twitter account
that encourage users to establish a connection with account i.e. number of
followers, type of content, political orientation etc.)
France: 72% of the participants chose to follow accounts on Twitter because
they are known persons
Kuwait: 54% of the participants chose to follow certain account on Twitter
because they are known persons or subject matter experts
But both France and Kuwait attributed Twitter popularity (number of
followers) as the main variable to determine the person as known or not.
Research Findings
Agenda setting effects can exist on Twitter
Information on Twitter can be homogenous
The process of agenda building in France is different than Kuwait
Users are immersed in the illusion of news selection
Some conclusions
related to political information
When Twitter is the one and primary source of information because
of the lack of trust in other sources of information, we observe that
convergence to common topics is low
Common agenda setting and public opinion shaping in Twitter
depend on the number and the credibility of external information
sources
Information overload is a problem to evaluate information but it is at
the same time a condition (when information quality is good) to
overcome selective information exposure drawbacks
LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN
THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES
40

More Related Content

PDF
Do your employees think your slogan is “fake news?” A framework for understan...
Ian McCarthy
 
PPTX
Social media in the public sector south korea twitter
Danube University Krems, Centre for E-Governance
 
PDF
Digital development and Online Gender-Based Violence
Anand Sheombar
 
PPTX
Accessing and Using Big Data to Advance Social Science Knowledge
Josh Cowls
 
DOC
The Design of an Online Social Network Site for Emergency Management: A One-S...
guest636475b
 
PDF
GlobalPulse_SAS_MethodsPaper2011
UN Global Pulse
 
PDF
Reflections On Social Media Use Along The Academic Research Life Cycle
Anand Sheombar
 
Do your employees think your slogan is “fake news?” A framework for understan...
Ian McCarthy
 
Social media in the public sector south korea twitter
Danube University Krems, Centre for E-Governance
 
Digital development and Online Gender-Based Violence
Anand Sheombar
 
Accessing and Using Big Data to Advance Social Science Knowledge
Josh Cowls
 
The Design of an Online Social Network Site for Emergency Management: A One-S...
guest636475b
 
GlobalPulse_SAS_MethodsPaper2011
UN Global Pulse
 
Reflections On Social Media Use Along The Academic Research Life Cycle
Anand Sheombar
 

What's hot (20)

PPT
Investigating Internet-based Korean politics using e-research tools Kaist Cu...
Han Woo PARK
 
PDF
2053951715611145
Firas Husseini
 
PDF
Research-Open Access-Social Media: A winning combination
Eileen Shepherd
 
PDF
A framework of Web Science
vafopoulos
 
PDF
From Informal Academic Debate To Cyber Harassment- Navigating The Minefield A...
Cape Peninsula University of Technology
 
PPTX
Research-Open Access-Social Media: a winning combination
Rhodes University Library
 
PPTX
WEBINAR: Joining the "buzz": the role of social media in raising research vi...
HELIGLIASA
 
PDF
Computational Social Science
Kiarash Kiani
 
PPT
Entradas & Dimopoulos: Public science on the web
Observatori Comunicació Científica - Universitat Pompeu Fabra
 
PDF
All a twitter reddit edition
Andy Crosby
 
PPTX
Are we all knowledgeable now
Petr Lupac
 
PPT
e- Research As Intervention (5 April 2010) J Unit
Webometrics Class
 
PDF
Integrating Behavioural Science in Government Communication
Mike Kujawski
 
PPT
Information, Science, and Society
Melanie Swan
 
PDF
Studying Cybercrime: Raising Awareness of Objectivity & Bias
gloriakt
 
PDF
Computational Social Science
journal ijrtem
 
PDF
Joining the ‘buzz’ : the role of social media in raising research visibility ...
Eileen Shepherd
 
PDF
Analysing Social Media Conversations to Understand Public Perceptions of Sani...
UN Global Pulse
 
PPTX
Gaza_Audience Gatekeeping
Kyounghee Hazel Kwon
 
Investigating Internet-based Korean politics using e-research tools Kaist Cu...
Han Woo PARK
 
2053951715611145
Firas Husseini
 
Research-Open Access-Social Media: A winning combination
Eileen Shepherd
 
A framework of Web Science
vafopoulos
 
From Informal Academic Debate To Cyber Harassment- Navigating The Minefield A...
Cape Peninsula University of Technology
 
Research-Open Access-Social Media: a winning combination
Rhodes University Library
 
WEBINAR: Joining the "buzz": the role of social media in raising research vi...
HELIGLIASA
 
Computational Social Science
Kiarash Kiani
 
Entradas & Dimopoulos: Public science on the web
Observatori Comunicació Científica - Universitat Pompeu Fabra
 
All a twitter reddit edition
Andy Crosby
 
Are we all knowledgeable now
Petr Lupac
 
e- Research As Intervention (5 April 2010) J Unit
Webometrics Class
 
Integrating Behavioural Science in Government Communication
Mike Kujawski
 
Information, Science, and Society
Melanie Swan
 
Studying Cybercrime: Raising Awareness of Objectivity & Bias
gloriakt
 
Computational Social Science
journal ijrtem
 
Joining the ‘buzz’ : the role of social media in raising research visibility ...
Eileen Shepherd
 
Analysing Social Media Conversations to Understand Public Perceptions of Sani...
UN Global Pulse
 
Gaza_Audience Gatekeeping
Kyounghee Hazel Kwon
 
Ad

Viewers also liked (17)

PDF
Erika Janiūnienė, Lina Markevičiūtė: The Quality Assessment of Information Se...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Alicja Waszkiewicz-Raviv: Visual Information and Visual Persuasion in Public ...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Samia Takhtoukh: The Practices of Historians in the Digital Age: a case study
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PPT
Nauka o informacji w XXI wieku (nowa prezentacja)
Sabina Cisek
 
PPSX
Mieczysław Muraszkiewicz, Warsaw University of Technology: Artificial Intelli...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Christopher Biedermann, EmiTel Ltd: Cybersecurity and the Internet of Things
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Tibor Koltay, Eszterházy Károly University: Beyond Literacies: The evolving l...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Zhenfei Feng: The Impact of Social Influence on Users’ Ratings of Movies
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Pablo Benalcazar: Modern Tools on Patent Thicket Identification
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Zachowania informacyjne
Sabina Cisek
 
PDF
Karolina Zawada: Toruń University’s Open Access Data Project – the new role f...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Maciej Dziubecki, Aleph Poland: Applying UX Principles to the Design of Libra...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Tan Tran: Ethical Dimension in Knowledge Organization Systems and Applicable ...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Radosław Lipiński: Information Flow Model as an Effective Tool For Supporting...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Emmanuelle Chevry Pébayle:Towards Open Innovation of Services Using Social Me...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Amel Fraisse, Ronald Jenn, Quoc-Tan Tran, Samia Takhtoukh: Merging Crowdsourc...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
PDF
Gerhard Budin, University of Vienna: Beyond Accessibility: “Operational Usabi...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Erika Janiūnienė, Lina Markevičiūtė: The Quality Assessment of Information Se...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Alicja Waszkiewicz-Raviv: Visual Information and Visual Persuasion in Public ...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Samia Takhtoukh: The Practices of Historians in the Digital Age: a case study
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Nauka o informacji w XXI wieku (nowa prezentacja)
Sabina Cisek
 
Mieczysław Muraszkiewicz, Warsaw University of Technology: Artificial Intelli...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Christopher Biedermann, EmiTel Ltd: Cybersecurity and the Internet of Things
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Tibor Koltay, Eszterházy Károly University: Beyond Literacies: The evolving l...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Zhenfei Feng: The Impact of Social Influence on Users’ Ratings of Movies
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Pablo Benalcazar: Modern Tools on Patent Thicket Identification
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Zachowania informacyjne
Sabina Cisek
 
Karolina Zawada: Toruń University’s Open Access Data Project – the new role f...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Maciej Dziubecki, Aleph Poland: Applying UX Principles to the Design of Libra...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Tan Tran: Ethical Dimension in Knowledge Organization Systems and Applicable ...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Radosław Lipiński: Information Flow Model as an Effective Tool For Supporting...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Emmanuelle Chevry Pébayle:Towards Open Innovation of Services Using Social Me...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Amel Fraisse, Ronald Jenn, Quoc-Tan Tran, Samia Takhtoukh: Merging Crowdsourc...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Gerhard Budin, University of Vienna: Beyond Accessibility: “Operational Usabi...
Katedra Informatologii. Wydział Dziennikarstwa, Informacji i Bibliologii, Uniwersytet Warszawski
 
Ad

Similar to Laurence Favier, University Charles De Gaulle – Lille 3: Social Influence and Information Evaluation (20)

PDF
Social Media Research: with focus on Twitter and misinformation
khornberger
 
PPT
European librarians theatre - Social Media Spotlight
Julien Houssiere
 
PDF
Misinformation, Disinformation, Malinformation, fake news and libraries
Dr Trivedi
 
PPTX
Social media for research impact
Aberdeen CES
 
PPTX
Fake news presentation lirt summit 2018
PamelaPfeiffer1
 
PPTX
Digital Literacy in an age 'post-truth' world (Keynote at the International M...
Queen's University Belfast
 
PDF
Group 7 ppt sts. Technology. Information.pdf
serenitygayle2325
 
PPTX
Web Science Session 2: Social Media
Stefanie Panke
 
PDF
Media literacy in the age of information overload
Gmeconline
 
PPTX
Georgia Tech Library Information Literacy Program-First Year Experience
Seth Porter, MA, MLIS
 
PPTX
quarter 2 of media and information literacy
teachercarlocortez
 
PDF
Social Networks and Well-Being in Democracy in the Age of Digital Capitalism
AJHSSR Journal
 
PDF
Evaluation of Internet Information
Ahmed-Refat Refat
 
PPTX
Info-competencies for the disruptive digital whirlwind age: The Library Role 
Jesus Lau
 
PPTX
MEDIA AND INFORMATION LITERACY FOR GRADE 12
teachercarlocortez
 
PPTX
Health (mis)information behaviour in the COVID-19 era
Diane Rasmussen Pennington
 
PPTX
Shareology and Social Media in Academia
Sue Beckingham
 
PDF
21st Century Research Profiles: Using social media to benefit your research h...
Emma Gillaspy
 
PDF
A Scientist's View of Twitter
Craig McClain
 
Social Media Research: with focus on Twitter and misinformation
khornberger
 
European librarians theatre - Social Media Spotlight
Julien Houssiere
 
Misinformation, Disinformation, Malinformation, fake news and libraries
Dr Trivedi
 
Social media for research impact
Aberdeen CES
 
Fake news presentation lirt summit 2018
PamelaPfeiffer1
 
Digital Literacy in an age 'post-truth' world (Keynote at the International M...
Queen's University Belfast
 
Group 7 ppt sts. Technology. Information.pdf
serenitygayle2325
 
Web Science Session 2: Social Media
Stefanie Panke
 
Media literacy in the age of information overload
Gmeconline
 
Georgia Tech Library Information Literacy Program-First Year Experience
Seth Porter, MA, MLIS
 
quarter 2 of media and information literacy
teachercarlocortez
 
Social Networks and Well-Being in Democracy in the Age of Digital Capitalism
AJHSSR Journal
 
Evaluation of Internet Information
Ahmed-Refat Refat
 
Info-competencies for the disruptive digital whirlwind age: The Library Role 
Jesus Lau
 
MEDIA AND INFORMATION LITERACY FOR GRADE 12
teachercarlocortez
 
Health (mis)information behaviour in the COVID-19 era
Diane Rasmussen Pennington
 
Shareology and Social Media in Academia
Sue Beckingham
 
21st Century Research Profiles: Using social media to benefit your research h...
Emma Gillaspy
 
A Scientist's View of Twitter
Craig McClain
 

Recently uploaded (20)

PDF
Electricity-Magnetic-and-Heating-Effects 4th Chapter/8th-science-curiosity.pd...
Sandeep Swamy
 
PPTX
Understanding operators in c language.pptx
auteharshil95
 
PPTX
Dakar Framework Education For All- 2000(Act)
santoshmohalik1
 
PPTX
An introduction to Prepositions for beginners.pptx
drsiddhantnagine
 
DOCX
SAROCES Action-Plan FOR ARAL PROGRAM IN DEPED
Levenmartlacuna1
 
PPTX
Tips Management in Odoo 18 POS - Odoo Slides
Celine George
 
PDF
Landforms and landscapes data surprise preview
jpinnuck
 
PPTX
Information Texts_Infographic on Forgetting Curve.pptx
Tata Sevilla
 
PPTX
family health care settings home visit - unit 6 - chn 1 - gnm 1st year.pptx
Priyanshu Anand
 
PPTX
ACUTE NASOPHARYNGITIS. pptx
AneetaSharma15
 
PDF
High Ground Student Revision Booklet Preview
jpinnuck
 
PPTX
NOI Hackathon - Summer Edition - GreenThumber.pptx
MartinaBurlando1
 
PDF
Module 3: Health Systems Tutorial Slides S2 2025
Jonathan Hallett
 
PPTX
Software Engineering BSC DS UNIT 1 .pptx
Dr. Pallawi Bulakh
 
PPTX
TEF & EA Bsc Nursing 5th sem.....BBBpptx
AneetaSharma15
 
PPTX
Introduction and Scope of Bichemistry.pptx
shantiyogi
 
PDF
Phylum Arthropoda: Characteristics and Classification, Entomology Lecture
Miraj Khan
 
PPTX
How to Manage Leads in Odoo 18 CRM - Odoo Slides
Celine George
 
PPTX
Nursing Management of Patients with Disorders of Ear, Nose, and Throat (ENT) ...
RAKESH SAJJAN
 
DOCX
Action Plan_ARAL PROGRAM_ STAND ALONE SHS.docx
Levenmartlacuna1
 
Electricity-Magnetic-and-Heating-Effects 4th Chapter/8th-science-curiosity.pd...
Sandeep Swamy
 
Understanding operators in c language.pptx
auteharshil95
 
Dakar Framework Education For All- 2000(Act)
santoshmohalik1
 
An introduction to Prepositions for beginners.pptx
drsiddhantnagine
 
SAROCES Action-Plan FOR ARAL PROGRAM IN DEPED
Levenmartlacuna1
 
Tips Management in Odoo 18 POS - Odoo Slides
Celine George
 
Landforms and landscapes data surprise preview
jpinnuck
 
Information Texts_Infographic on Forgetting Curve.pptx
Tata Sevilla
 
family health care settings home visit - unit 6 - chn 1 - gnm 1st year.pptx
Priyanshu Anand
 
ACUTE NASOPHARYNGITIS. pptx
AneetaSharma15
 
High Ground Student Revision Booklet Preview
jpinnuck
 
NOI Hackathon - Summer Edition - GreenThumber.pptx
MartinaBurlando1
 
Module 3: Health Systems Tutorial Slides S2 2025
Jonathan Hallett
 
Software Engineering BSC DS UNIT 1 .pptx
Dr. Pallawi Bulakh
 
TEF & EA Bsc Nursing 5th sem.....BBBpptx
AneetaSharma15
 
Introduction and Scope of Bichemistry.pptx
shantiyogi
 
Phylum Arthropoda: Characteristics and Classification, Entomology Lecture
Miraj Khan
 
How to Manage Leads in Odoo 18 CRM - Odoo Slides
Celine George
 
Nursing Management of Patients with Disorders of Ear, Nose, and Throat (ENT) ...
RAKESH SAJJAN
 
Action Plan_ARAL PROGRAM_ STAND ALONE SHS.docx
Levenmartlacuna1
 

Laurence Favier, University Charles De Gaulle – Lille 3: Social Influence and Information Evaluation

  • 1. Social Influence and Information evaluation SELECTIVE EXPOSURE TO INFORMATION AND USERS INTERACTION LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 1
  • 2. Context The combination of Web-based self-publication and social media requires new skills to evaluate information This is the main challenge of information literacy Relevance, cognitive authority can’t be the main criteria in an environment defined by social interactions LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 2
  • 3. “Social media”: what does it mean? It is not social networking It is much more to do with what people are doing with the technology than the technology itself, for rather than merely retrieving information, users are now creating and consuming it, and hence adding value to the websites that permit them to do so (Campbell et al., 2011, p. 87) A group of internet-based applications that build on the ideological and the technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content (Kaplan and Haenlein, 2010, p. 61) LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 3
  • 4. Information evaluation allows To decide where to begin To predict which source/system would give me the best information To select To accept or reject To determine whether to use or share the information LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 4
  • 5. Making judgments of information Traditional approach to information evaluation: Identifying a set of criteria people employ when making judgments of information Information Quality: Evaluating the values of information in terms of excellence or truthfulness (Taylor) Credibility: People’s assessment of whether information is trustworthy based on their own expertise and knowledge (Rieh) Cognitive Authority: Influence on one’s thoughts that one would recognize as proper (Wilson). A credible source even though it did not have any influence on our thoughts Trust: Belief about the reliability of, dependability of, and confidence in person, object, or process (Fogg) LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 5
  • 6. Social influence Hillman and Trier (2013, p. 3) state that social influence “provides a broad range of concepts to explain how people’s individual actions are affected by other people as a result of interaction”. This implies that social influence is a natural process, but can be used by people or businesses to change a person’s attitude or behavior. Social influence can be used for positive actions (e.g. creating awareness for societal problems, promoting new products) and negative actions (e.g. social hacking, social pressure) LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 6
  • 7. Normative social influence Kelman (1985) is often cited as a fundamental analysis of normative social influence. This type of social influence explains how individuals are influenced, based on norms. Kelman distinguishes three sub-types of normative social influence: compliance, identification and internalization Compliance occurs when an individual accepts the opinion of others, hoping that this would lead to a favorable reaction of others. Identification means that an individual accepts the opinion of others to maintain a desired relationship. Internalization represents the strongest influence and occurs when an individual accepts and beliefs the opinion of others both in public and private LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 7
  • 8. Informational social influence Informational social influence is explained by Lee, Shi, Cheung, Lim & Sia (2011). This type of social influence involves accepting information or advice from a person who may not have previously been known as a friend or colleague. Informational social influence is especially relevant in the context of social media, in which user-generated content is an important type of information. An example of this type of social influence in social media could be a change in purchasing behavior as a consequence of online customer reviews of a product. These reviews change the attitudes and beliefs of customers and thereby influence behavior. LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 8
  • 9. Rationale of the study To better understand social influence effects on information behaviour, we compare studies on misinformation in social networks to another one, conducted by one of our doctoral student, Albaraa Altourah, related to agenda setting in Twitter (can a mass media theory like agenda setting be applied to Twitter?). Our question is: can selective exposure to information both create division (divergence) between users into groups who follow the same interests without being interested to the others and also generate convergence in the form of public opinion or common agenda setting ? LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 9
  • 10. Misinformation studies SOCIAL INFLUENCE AND FRAGMENTATION OF DIGITAL ENVIRONMENT LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 10
  • 11. Background 1 World Economic Forum (WEF Report 2013) Massive digital misinformation is becoming pervasive in online social media: it has been listed by the WEF as one of main threats of our society A french resolution this year The parliament (The French National Assemby) adopted, a few month ago, a resolution entitled « Resolution on sciences and progess in the Republic). It « invites the government to think about pedagogical practices based on sensible (smart) use of digital technologies, especially information selection learning that would make easier the difference between knowledge and opinion without scientific basis LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 11
  • 12. Original version TEXTE ADOPTÉ n° 926 SESSION ORDINAIRE DE 2016-2017 21 février 2017 RÉSOLUTION sur les sciences et le progrès dans la République. L’Assemblée nationale a adopté la résolution dont la teneur suit : (…) 9. Invite le Gouvernement à réfléchir à des pratiques pédagogiques fondées sur l’usage raisonné des technologies numériques, en particulier à l’apprentissage du tri de l’information qui faciliterait la distinction entre des savoirs établis et des opinions sans fondement scientifique ; LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 12
  • 13. Background 2 Algorithmic-driven solutions have been proposed (Qazvinian V and al. 2011, Ciampaglia GL et al. 2015, Resnick P. 2014, Gupta A. et al. 2014, Dong XL, et al. 2015, ...) Google tries to develop a trustworthiness score to rank the results of queries LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 13
  • 14. Example: Bessi et al. 2016 « we analyze the users behavior exposed to the same contents on different platforms—i.e. Youtube and Facebook. We focus on Facebook posts linking Youtube videos reported on Science and Conspiracy pages. We then compare the users interaction with these videos on both platforms » « We limit our analysis to Science and Conspiracy for two main reasons: a) scientific news and conspiracy-like news are two very distinct and conflicting narratives; b) scientific pages share the main mission to diffuse scientific knowledge and rational thinking, while the alternative ones resort to unsubstantiated rumors. » LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 14
  • 15. Data collected The first category (conspiracy theories) includes the pages that disseminate alternative, controversial information, often lacking supporting evidence and frequently advancing conspiracy theories. The second category (science news) includes the pages that disseminate scientific information. The third category (trolls) includes those pages that intentionally disseminate sarcastic false information on the Web with the aim of mocking the collective credulity online. LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 15
  • 16. Results « Focusing on the consumptions patterns of YouTube videos posted on Facebook pages, we compute the Spearman’s rank correlation coefficients between users’ actions on Facebook posts and the related YouTube videos. We find strong correlations on how users like, comment and share videos on Facebook and Youtube. Despite the different algorithm for content promotion, information reverberate in a similar way » LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 16
  • 17. Similar polarization of Science and Conspiracy users in Facebook and Youtube according Bessi et al. 2016 LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 17
  • 18. « Echo chambers » “We observe sharply peaked bimodal distributions. Users concentrate their activity on one of the two narratives. To quantify the degree of polarization we use the Bimodality Coefficient (BC), and we find that the BC is very high for both Facebook and YouTube. (…) Content has a polarizing effect, indeed, users focus on specific types of content and aggregate in separated groups—echo chambers—independently of the platform and content promotion algorithm”. LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 18
  • 19. Users Polarization on Facebook and Youtube according Bessi et al. 2014 LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 19
  • 20. Core concepts Echo chamber Filter bubble (Eli Pariser 2011) Collective intelligence is a myth LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 20
  • 21. Issues for Information evaluation Identification of cognitive biais including: ◦ Confirmation biaises ◦ Representativity biais (Bronner 2017) ◦ And many other biaises that economists and psychologists have already studied to understand behaviours in financial markets (especially anomalies) LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 21
  • 22. Agenda setting study SOCIAL INFLUENCE AND PUBLIC OPINION IN SOCIAL NETWORK ENVIRONMENT LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 22
  • 23. Proposed questions •Are social media be able to build convergence between users and contribute to shape a public opinion despite the fact of its high potentiality of causing echo chamber effects? •Is there an inter-user influence while using social media platforms? •Presentation of research findings of one of my doctoral students, Alabraa Altourah who conducted a study at the end of 2016 about Twitter
  • 24. A. Altourath research questions •How can the agenda setting be understood in the context of Twitter? ◦ What limits or enables issue salience creation and transfer to subsequently lead to the manifestation of agenda setting effect on Twitter? ◦ What are the practices that instigate issue salience in an extremely diverse media platform such as Twitter? ◦ How may the agenda setting be defined in a platform that enables audiences to be an active part in the communication cycle? •How can the newly constructed meanings of agenda setting help understanding agenda building within Twitter? ◦ Is the agenda building process on Twitter strongly attached to the socio-cultural setting where the study took place? ◦ Are there any external factors that contaminate the agenda building process within Twitter? LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 24
  • 25. A. Altourah research design (1) •Platform study • Using Twitter analytics tool to identify the manifestation of agenda setting • Isolating the agenda setting effects and analyze the conditions that led to its establishment • Examining the possible existence of homogenous information on Twitter that leads to the augmentation of the appearance of certain news • Examining the role of cultural setting in hindering or promoting the agenda setting. Comparison between France and Koweit
  • 26. Why Twitter? why France and Kuwait? •Data may be accessed easily •Data are available publically •Data may be researched based on different research settings •The platform is very popular •France is one of the top 10 countries in number of Twitter users (Forbes.com, 2016) •Kuwait has the highest number of Twitter users per capita globally (one in three people in Kuwait have a Twitter account) (Forbes.com, 2016) •The two countries have a fundamentally different socio-political setting
  • 27. Research design (2) •Users study • Identify the elements that leads to the establishment of issue salience • Identify the role of users as an active part in the communication cycle in promoting, hindering or avoiding the effects of agenda setting • Examine the socially constructed understanding with respect to information available on Twitter • Identify the perceived reliability of information found on Twitter
  • 28. Platform study: methodology 1 The first step is executed by analyzing 5 selected Twitter accounts from France and Kuwait to identify if an issue salience has been created and transferred in the corresponding cyberspace region. A period of seven days has been set to collect the tweets of the selected Twitter account to identify the influencer’s agenda. During the seven days period, two days were selected as the most applicable days for content analysis, which will be referred to as the first and the second day of analysis. The first day of analysis represents the day where the selected users posted the highest number of tweets in relevance to the week’s average. The second day of analysis on the other hand refers to the day where the selected users posted the lowest number of tweets in relevance to the week’s average. LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 28
  • 29. Platform study: methodology 2 In both the days of analysis, a content analysis method is applied to identify the most recurrent keywords in the posted tweets. Once the keywords are identified, they are researched within the corresponding cyberspace region three days before and three days after their introduction to assess whether there is an increment or a decrement in number of tweets mentioning those keywords. This process aims to identify the Twitter agenda. The second step is devoted to investigate the level of homogeneity of media content between selected users, in each study location. This is conducted by using the socialbearing.com as to collect three-days worth of tweets of the selected users then analyzing the data using content analysis. LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 29
  • 30. The selected Twitter accounts France : Nicolas Sarkozy (@NicolasSarkozy) Marine Le Pen (@MLP_officiel) Jean-Luc Mélenchon (@JLMelenchon) Najat Belkacem (@najatvb) Le Monde (@lemondefr) Kuwait : Waleed Altabtabie (@altabtabie) Nasser Alduwailah (@nasser_duwailah) Almajlliss (@Almajlliss) Safaa Alhashem (@safaalhashem) Faisal Almuslem (@faisalalmuslem) LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 30
  • 32. Reuse of keywords after their first introduction 22-24 August 2016 25-27 August 2016 Change in % Impôt 2351 3837 +63.2% Être francais 3129 3882 +24.1% L’immigration 1100 3487 +217% L’autorité de l’état 75 1668 +2124% LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 32
  • 33. Agenda setting effects exist on Twitter 09-10 September 2016 11-12 September 2016 Change in % Laïcité 2454 4624 88.40% Frexit 717 765 6.70% L’outre-mer 12 442 3583.30% Droit des femmes 159 1572 888.70% Keywords comparison before and after First Day of Analysis (Marine Le Pen) (Number of tweets)
  • 34. Keywords reuse/tweet category (Ex:Sarkozy account) 22-24 August 2016 25-27 August 2016 Tweet Retweet Reply Mentions Tweet Retweet Reply Mention Impôt 1018 1195 138 818 1123 +10.3% 2434 +103.7% 280 +102.9% 1513 +84.9% Être francais 441 2276 412 1161 1123 +10.3% 1123 +10.3% 1123 +10.3% 1123 +10.3% L’immigration 131 861 108 384 578 +341.2% 2494 +189.6% 418 +287% 1648 +329.2% L’autorité de l’état 26 42 7 16 316 +1115.5% 1251 +2878.6% 101 +1342.9% 637 +3881.3% LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 34
  • 35. Information on Twitter can be homogenous (France) •Twitter account with different political orientation engage in a level of consistency with respect to tweets they post •Between the 18th of October 2016 and the 21st of October 2016 the selected accounts engaged in different topics. However the content analysis identified similar topics between these accounts: • Police • Russie • Immigration • CETA (Comprehensive Economic and Trade Agreement) • Goodyear • Islam • US • Impôt • Intérêt de la France
  • 36. The process of agenda building in France is different than Kuwait In the Kuwaiti twitter sphere there isn’t any structural attempt to build an agenda that coincides with pressing matters that are of high relevancy to the public. Instead, the proposed issues on Twitter are random, based on the personal political believes and do not affect the majority of the public. It is rather an attempt to appeal to a specific political or societal group. Therefore, the establishment of an agenda and the increment of discussions pertaining to the issue proposed are limited (agenda setting effects are less occurring, the twittersphere is less homogenous)
  • 37. Social influence in the Kuwaiti twitter sphere Twitter accounts in Kuwait are less likely to be influenced by each other’s with respect to tweets content or by exterior influence source The Twitter usage in Kuwait may be primarily to support and promote personal propaganda irrespective what other users are concerned about, benefiting from their high number of followers and the fact that users in Kuwait are more likely to use Twitter as their prime source of information Therefore, the possibility of having an exterior influential entity that might establish a common agenda among Twitter users in Kuwait is unlikely. The Kuwaiti culture for long has regarded number of major news sources as largely biased LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 37
  • 38. What is similar: Users are immersed in the illusion of news selection (user study) What are the factors that determine your choice of Twitter accounts to follow? (This question aims to determine the elements of the Twitter account that encourage users to establish a connection with account i.e. number of followers, type of content, political orientation etc.) France: 72% of the participants chose to follow accounts on Twitter because they are known persons Kuwait: 54% of the participants chose to follow certain account on Twitter because they are known persons or subject matter experts But both France and Kuwait attributed Twitter popularity (number of followers) as the main variable to determine the person as known or not.
  • 39. Research Findings Agenda setting effects can exist on Twitter Information on Twitter can be homogenous The process of agenda building in France is different than Kuwait Users are immersed in the illusion of news selection
  • 40. Some conclusions related to political information When Twitter is the one and primary source of information because of the lack of trust in other sources of information, we observe that convergence to common topics is low Common agenda setting and public opinion shaping in Twitter depend on the number and the credibility of external information sources Information overload is a problem to evaluate information but it is at the same time a condition (when information quality is good) to overcome selective information exposure drawbacks LAURENCEFAVIER-THE 4TH INTERNATIONALSCIENTIFICCONFERENCEINFORMATIONSCIENCEIN THE AGE OF CHANGEINNOVATIVEINFORMATIONSERVICES 40