Project topics and materials
Get Complete Project Material File(s) Now! »
Who: The crowds
Man is by nature a social animal […]. Society is something that precedes the individual. Anyone who either cannot lead the common life or is so self-sufficient as not to need to, and therefore does not partake of society, is either a beast or a god. (Aristotle, 1944)
The crowd before the Internet
At the dawn of the previous century, Le Bon (1895) portrays a negative pic-ture of the crowd:
Isolé, c’était peut-être un individu cultivé, en foule c’est un barbare, c’est-à-dire un instinctif. (Le Bon, 1895, p.22) 2
For Le Bon (1895), the crowd, to be understand here as a unified organism (with a biological meaning), does not have the capacity to reason but only to act, and in particular to destroy. Even if Le Bon does not trust the crowd to reason, he underlines the fact that the characteristics of the crowd is, if not better, at least different from the characteristics of its individual parts. In any case, the final aim of a crowd is to act.
A few years later, Tarde (1901) develops a theory to frame a crowd and the opinion which might emerge from it. First of all, he draws more precisely what must be understood by the crowd. He distinguishes an « audience » from a « crowd »:
Le public, en effet, est une foule dispersée. (Tarde, 1901,p.7) 3
The concept of crowd implies an organization of its structure. An audience is a spiritual group, whose members are physically separated and only linked mentally. On the contrary, according to Tarde, a crowd is more primitive, its members acting on the brain of each other, also through a physical contact. Indeed, a crowd acts through a communication method, allowing its mem-bers to coordinate themselves, in order to reach the common goal, the crowd goal.
Even if they differ on several points, Le Bon and Tarde agree on the fact that an individual belonging to a crowd loses his reason for a collective one, more primitive. Thus, the vision of political philosophers is pessimistic over the capacity of a crowd to produce positive outputs. Contemporary researchers develop a different conception of the crowd, focusing not anymore on polit-ical aspect, but also on an economic one. The following sections focus on smaller crowds, in a particular context: communities. Similar to a crowd, a community is the aggregation of individuals, sharing a common goal. Un-der the right circumstances, detailed below, such a community is capable of producing an output profitable for each individual inside the community.
Communities of Practices
A whole part of research on human behaviors analyzes how we interact one(s) to another(s). Section 2.1.1 presents the vision of intellectual class about how individuals merge into an entity: a group, a public or a crowd. This section focuses on a group within a particular organization: firms. A company, seen as an entity, is an organization which produces a product or a service in order to sell it to customers. The classical economics define that a company manages resources, under several constraints — the cost function of the firm — and aims to maximize its profits. Those resources are the work-force, on the one hand, and capital in the other (Smith et al., 1859). Becker (1962) developed his famous theory of human capital to better understand the work resource. This concept underlines the fact that each employee of a firm has his own capabilities, selling his knowledge and skills to his employer.
With the growth of companies, and then their internationalization, firms have developed new needs. Having subsidiaries all over the world increases the quantity of knowledge accumulated by a firm, but make more and more difficult to disseminate it inside the whole company (Guerineau, 2018). Thus a new management framework arises: knowledge management. We develop the theory later in the literature review (Section 2.4.1). But this framework implies an organization among employees in order to share knowledge: the Community of Practice (CoP).
This concept accepts different definitions (Johnson, 2001). Wenger (1998) sets the basis of conceptualizing a CoP: it is an evolving process for learning inside a group. Such group might exist within defined organizations, but outside and between as well. Moreover, the creation of such community, and its consolidation is a longtime process.
Wenger (1998) develops a definition around groups of professionals, shar-ing common tasks and responsibilities. For the authors and other (Winsor, 2001; Bielaczyc and Collins, 1999), one key feature of CoP is the dissemina-tion of knowledge through communication.
Thus, those CoPs are composed by employees sharing a same practice.
The development of IT improved the transfer of information inside a team, a firm, an international company. Nevertheless, the exchanges stay among members of a same organization. Internet changes this fact allowing orga-nizations to outsource information: seeking for information outside the or-ganization. We develop the notion later in this chapter (section 2.4.2). The following part addresses the concept of online communities: who are these people interacting one with another without belonging to the same organi-zation?
Online Community
Benghozi et al. (2001) and Benghozi (2006) develops a typology of commu-nities, and tries to understand what is an online community especially. The development of the Internet allowed the production of softwares which al-lows individuals to act collectively without being in the same space. In par-ticular, in firms, Enterprise Resources Planning arose and solved a part of the geographical issue. But, they also imply the need for cooperation among coworkers (Benghozi, 2006). Indeed, continuous improvement of softwares removes the issue of technological mastering (software becomes more and more user-friendly). Still, they imply important consequences on other levels: management, organizational, work practices (Benghozi, 2006).
Besides, we observe online communities outside companies. As defined by Kraut et al. (2011), an online community (OC) is a virtual space where individuals come together to interact with others (converse, exchange re-sources, play). Similar definitions are developed in different disciplines (Rhein-gold, 1993; Hagel, 1999; Andrews, 2001; Lee, Vogel, and Limayem, 2003; Iriberri and Leroy, 2009). The creation and development of a new OC faces several challenges. First of all, designers and managers are faced with a criti-cal problem: in order to attract new members, they need an important quan-tity of content but do not yet have the sufficient number of members. A sec-ond challenge is that once the OC is established, it still needs to attract new-comers, in order to replace those who leave. Attracting and socializing those new members is a challenge because their first interaction with the OC will have an important impact on their commitment, and at the same time, they will disturb for a period, the activity of previous members. Besides, man-agers need to enhance commitment. Commitment is a feeling of attachment and connection to the community. And if members are committed to the community, they tend to both be more satisfied and perform better (Mathieu and Zajac, 1990).
Finally, as highlighted by Faraj, Jarvenpaa, and Majchrzak (2011), some members can behave in an uncivilized way. Thus, a community needs a framework to regulate behaviors. The difficulty is even higher for an OC due to the anonymity of its members, an ease to enter and exit the community, and the textual communication.
Iriberri and Leroy (2009) develop precise characteristics of OC. The very core of an OC is its activity and the creation of contents. As suggested by the core-periphery model (Borgatti and Everett, 2000), almost all the content is created by a small number of members. But the issue is not the inequality of contributions, but the possible under-contribution. Hagel (1999) defines on-line communities as “computer-mediated spaces where there is potential for the integration of content and communication with an emphasis on member-generated content.” Lee, Vogel, and Limayem (2003) supports this definition. Furthermore, they ascertain that the content created in online communities brings value to business organizations.
Besides, Millen, Fontaine, and Muller (2002) highlight benefits for orga-nizations which gather such communities. From a customer’s point of view, it increases loyalty. Furthermore, it allows the organization to gather feed-back and information on customer needs and requirements, directly from customers, improving organization customer service. Alongside, it increases the organization visibility and reputation. From an employee’s point of view, it increases his trust, increases internal communication allowing everybody to follow all company projects. Then, OC has a direct impact on the pro-ductivity of the company, increasing the quality of knowledge, idea creation, product innovation and enhancing problem solving process.
Why: The motivations
Theoretically, the increase in information sharing should improve global ef-ficiency of organizations. But are individuals willing to share their infor-mation? How to make employees develop internal mechanisms in order to share their knowledge? Constant, Kiesler, and Sproull (1994) shine lights on those research questions. The theory of interdependence (Kelley and Thibaut, 1978) demonstrates how the (social) environment might put pres-sure on individuals, producing negative behaviors. An organizational envi-ronment might make an employee share his knowledge with another, even if he does not want to but it would have negative impact on his global pro-duction. In day-to-day life (cf. Section 2.3.2.3), an individual will easily act, share informations, whereas in an organization, he might not capture this in-formation sharing as a social good (an act, a behavior that may be personally costly but would be beneficial to the organization in the long run). How to understand this fact?
There is a distinction between tangible information (seen as a product – such as a document) and intangible information (considered as expertise). Authors assume that, in response to a coworker who had failed to help in the past, people would be more likely to share expertise than a document. The meaning to people of intangible information such as expertise is different than is the meaning of tangible information such as a computer program. The former reflects on its possessor’s identity and inner qualities, and that sharing it can have direct personal benefits.
Chiu et al. (2007) highlight factors that increase or reduce individuals’ satisfaction in knowledge sharing in open virtual professional communities. Organizations have understood that they do not have at their disposal all the required knowledge within their formal boundaries., some of them devel-oped professional virtual communities in order to fill that gap. But individ-uals do not necessarily want to share their knowledge, because of the fear of losing their comparative advantages.
A professional virtual community is defined by three dimensions: its members, the social network they develop and the knowledge they share. Individuals might be motivated to share knowledge because they expect fu-ture rewards, intangible and tangible benefits. Authors base their analysis upon an enhanced model of the expectancy disconfirmation theory devel-oped by Oliver (1980), finding core motivations for the continuance intention in knowledge sharing.
Faraj, Jarvenpaa, and Majchrzak (2011) theorize on how Online Com-munities (OC) engage in knowledge collaboration. An OC might be seen as a typical organizational structure, but characterized by constant changes (members, contents). Aside from the classical behaviors of knowledge ex-changes, an online environment adds possibilities. Members can also re-combine, modify, and integrate knowledge that others have contributed to. One can witness knowledge collaboration in OCs despite the lack of direct social relationships. Thus, for Faraj, Jarvenpaa, and Majchrzak (2011), the access to resources cannot explain solely the collaboration. It is due to the unique characteristic of OC, distinguishing it from traditional organization structures: its fluidity. Authors divide this fluidity into five tensions, associ-ated with five resources that have an impact on knowledge collaboration in OCs. The passion of its members (the more passionate will invest more — time, effort — in OC; can be a barrier to collaboration); an important amount of time is required from members (but if few members spend too much time, they may impact the knowledge collaboration process by rejecting newcom-ers); anonymity (encouraging participation focusing more on the merit rather than the status, but can imply bully behaviors, and even decrease participa-tion if members have the fear to not get any credit for their work); conver-gence toward a single direction (temporary and incomplete, situated among a subset of actors rather than the entire community). To counter uncivilized behaviors, the OC needs structural mechanisms (such as formal roles and participation rules).
In order to motivate members, OC managers use rewards, and especially rewards for contributions (Andrews, 2001). Member recognition, settled on psychology, is a wide area for researchers (Andrews, Preece, and Turoff, 2001; Ginsburg and Weisband, 2004; Beenen et al., 2004; Hall and Graham, 2004; Tedjamulia et al., 2005; Butler et al., 2007). Providing rewards for contribu-tions seems to increase the number of messages posted by community mem-bers, making it more active and more successful (Iriberri and Leroy, 2009).
Table of contents :
1 Introduction
2 Wisdom of the Crowd
2.1 Who: The crowds
2.1.1 The crowd before the Internet
2.1.2 Communities of Practices
2.1.3 Online Community
2.2 Why: The motivations
2.3 How: The framework
2.3.1 Swarm intelligence, Wisdom of the Crowd, or Collective intelligence
2.3.1.1 Wisdom of the Crowds
2.3.1.2 Limits
2.3.2 Wisdom of the Crowds and Democracy
2.3.2.1 Epistemic Democracy
2.3.2.2 Procedural Democracy
2.3.2.3 Deliberative Democracy
2.3.2.4 Reconciliation of the approaches
2.4 What: Adding-values to society
2.4.1 Knowledge sharing
2.4.2 Open Innovation
2.4.3 Crowdsourcing
2.5 Conclusion
2 Contents
3 Reddit – Change My View
3.1 Who: Usecase
3.1.1 Reddit
3.1.2 Change My View
3.2 Why: The will to change
3.3 How: Graph models
3.4 What: CMV descriptive analysis
3.4.1 Life-cycle of Online Communities
3.4.2 Forum dynamics
3.4.3 Threads characteristics
4 Different types of Debates?
4.1 Who: Hypothesis — discussion categories
4.2 Why: Discussion levels
4.2.1 Reddit— Change My View (CMV)
4.3 How: Clustering
4.3.1 Author Networks and Motifs
4.3.2 Clustering Algorithm
4.3.2.1 k-means Clustering
4.3.2.2 Spectral Clustering
4.3.2.3 Hierarchical and Ward Clusterings
4.4 What: Results
4.4.1 Descriptive Statistics
4.4.1.1 Motif features
4.4.1.2 Roles in the motifs
4.4.2 Clustering analysis
4.4.2.1 Optimal clustering
4.4.2.2 Clustering results
4.5 Conclusion
5 The Consent of the Crowd
5.1 Who: Rewarded discussions
5.2 Why: Individual vs. crowd
5.3 How: Econometrics
5.3.1 Reddit – Change My View (CMV)
5.3.2 First delta context
5.4 What: The Consent of the Crowd
5.4.1 Descriptive statistics
5.4.2 Regression modelling
5.4.3 Result discussion
5.4.3.1 Results
5.4.3.2 Discussion
5.4.4 Conclusion
6 Conclusion
Bibliography
