The Interconnectedness of Brain Hubs

Scale free networks are networks that have few nodes with more power that dominate the network. This week in class we were assigned, Clay Shirky (2003) in “Power laws, weblogs, and inequality”. He describes how there are core groups of nodes known as “hubs” that are more connected within networks which in turn creates the power laws of a network distribution. They tend to arise in social systems where many people express their preferences among many different options (p.46). I decided for this blog to explore a different viewpoint of how power laws and scale free networks arise in different types of networks than human interaction. Below we can examine how scale free networks help scientists understand and examine the human brain and its interconnectedness.

Image result for scale free network

Brain networks in neuroscience are referred to by scientists as “connectomes”. When doing an initial search for large scale network research in health “Network hubs in the human brain” by van den Heuvel and Sporns (2014) caught my attention. There are different regions of the human brain that serve different functions such as behavioral characteristics and cognitive processes. These different regions of the brain have dynamic interactions with each other. In relation to social network analysis, “The brain’s anatomical and functional organization can be approached from the perspective of complex networks” (p. 684).  Certain parts of the brain of high degree of connectivity and are central to the neural network of the brain that can be depicted using SNA and graph theory.

Image result for connectomes

The purpose of this article was to look at the concept of network hubs in relation to neural function networks and neural connectivity. The article discusses two ways patterns of brain connectivity can be recorded: structural and functional. Structurally, studies have shown the existence of specific sets of hubs within the brain. Node hubs in the brain are highly connected with each other and the structural density of these hubs can be noted below. Functionally, describes the way these hubs are highly interactive with each other. How these hubs communicate with each other ultimately creates a network that is interrelated of different parts of the brain and its hubs. Ultimately, these hubs can help scientists detect differences among individuals, how these hubs develop over time, and the roles hubs play in brain disorders.

How do these brain hubs play a part in examining brain disorders? The next article I reviewed titled, “The hubs of the human connectome are generally implicated in the anatomy of brain disorders” the authors hypothesized that brain lesions would be concentrated in hub reasons since hubs are very important for reasons listed in the above article in relation to their structural and functional importance within the network (Crossley et al., 2014).

Image result for alzheimer disease

The researchers found that were were nine brain disorders in which lesions were significantly more likely to be located in the brain hubs. This article shows that hubs which are are nodes of high degree centrality plays a significant part in hub anatomically abnormal brain disorders. If scientists have an understanding where in the network these disorders are occurring they have narrowed down where to begin examining the “hows” and “whys” these hubs have such a high degree centrality within the network as can be seen in the picture below of the darker blue sport “hubs”.

Crossley, N. A., Mechelli, A., Scott, J., Carletti, F., Fox, P. T., McGuire, P., & Bullmore, E. T. (2014). The hubs of the human connectome are generally implicated in the anatomy of brain disorders. Brain, 137(8), 2382-2395.

Shirky, C. (2003). Power laws, weblogs, and inequality.

van den Heuvel, M. P., & Sporns, O. (2013). Network hubs in the human brain. Trends in cognitive sciences, 17(12), 683-696.

The Structure of Organized Crime: Money Laundering

As an emerging scholar who focuses on the positive value participating in college athletics provides to student-athletes, it is always unfortunate that most of the information that hits the news is negative publicity of both the NCAA and college athletic programs. At the end of this past September, the world of college basketball was rocked with the news that four college basketball assistant coaches at high profile athletic schools were indicted by federal authorities for their alleged involvement in fraud and corruption that also included personnel from Adidas sportswear company. Long-story short, the old saying in Division I athletics is “If you aren’t cheating you aren’t winning” got its press in on the front page of many news outlets. This incident had me really interested in exploring articles for this week’s SNA topic on crime data. How can money laundering and bribery be tracked and investigated using SNA as a methodological tool?

Since last week we talked about friends and family networks I thought the first article I investigated was fitting in that incorporated both friends and crime. It is titled, “Using friends for money: The positional importance of money launderers in organized crime”. Money laundering from the article is described as “involves introducing the illicit funds into the financial system, distancing the fund(s) from the criminals, and converting the money into legitimate business earnings”.

The research question under investigation were:

  • Do money-launderers occupy a distinct role separate from the commission of predicate crimes, or are most individuals involved in the drug market doing their own money laundering?
  • What structural position do different types of money launderers occupy in the drug market?

Data for the study was extracted from the Royal Canadian Mounted Police, National Threat Assessment Report which included an inventory of the social network or organized crime groups, that engaged in qualifying crimes within British Columbia from ‘04-06.  In total 129 crime groups and 2,197 individuals were identified. Information included: group members, co-offending information, demographic characteristics, nature of illicit drug trade involvement, and types of relationships that existed among individuals. Any relationship whether business or friendship was marked down as a relational tie between two individuals.

Image result for money laundering crime

Two SNA centrality measures were used to analyze the network: betweenness and eigenvector. Results of this network data demonstrated that majority of individuals are involved in money laundering are fall under the category of self-laundering criminal activity. Of the 102 drug market money launderers examined, 80% were self-launderers. Centrality metrics revealed that when comparing money-launderers with others in the drug market, launderers were significantly higher in betweenness centrality which the researchers concluded these individuals are more likely to be well positioned within the network to control the flow of information and materials. They are the brokers and most likely to handle the money. On the other-hand those not in money-laundering positions had higher eigenvector scores. This highlights that those involved in money laundering are distancing themselves from the key players who are highly connected within the network of the drug industry who are usually smuggling or handling the drugs themselves.

Overall, I think the biggest contribution this network provides is that if policy makers and law enforcement want to crack down on the drug market those who are actually smuggling drugs and supplying them are those they need to find! The picture above provides a great visual among different actors within a drug market and their interactions.

——————————————————————————————————————————————————————-

In exploring more about how this phenomenon using SNA I reviewed a second article by Drezewski, Sepielak, and Filipkowski (2014) “The application of social network analysis algorithms in a system supporting money laundering detection”. The research questions under investigation were

  • In which companies do shareholders hold shares?
  • In which other companies do members of the management board hold management positions?
  • In which other companies are certain persons present as, for example, proxies, or agents?
  • Do other business entities have the same registered office address?
  • Is the registered office located in a country that have been put on one of the “black lists” containing countries or terrorists that offer more advantageous fiscal conditions?
  • Are there companies with similar names, e.g. the one with only one or several letters, a single word, or the legal or organizational form that are different?

To answer these very specific research questions the researchers collected data from a national register operating under the name of “National Court Register” and bank statements. Data collected from such registers are: Name of each business or legal entity along with many other identifiable node attributes.  This article overall is providing an example of how SNA can be used as a visualization tool that allows the identification of interaction patterns of offenders and their roles in criminal groups.

The researchers used the SNA technique of hierarchical clustering in which separated the network into subgroups and were able to identify patterns within these subgroups. These subgroups included individuals from entities and assigned roles to these persons of interest to help detect money laundering processes and the proposal of intervals for roles which was utilized by a role finding algorithm. These algorithms were run by a “Money Laundering Detection System” which was built to facilitate the analysis of financial flows in order to fight money laundering systems. This system seems VERY complex and complicated. As you can see below the SNA Module is only one component used for the imported data of this system and is used to: Assign roles to nodes, analyze connections between nodes, look for proximity of entities, and compare roles assigned to nodes. These crime roles were labeled by the researchers as: organizers, insulators, communicators, guardians, extenders, monitors, crossovers, soldiers, recruits, outsiders, and occasional (as seen in the network above). These are all nodal attributes assigned to nodes within the crime network.

For data analysis the experts of police are used to for their knowledge of the structure and connections between these node roles. The node proximity module helped determine if different bank accounts are possessed by the same persons. Clustering and hierarchical analysis allows for detecting suspicious transactions. Since this article was more about a framework and methodology of using different modules there were no real findings, just examples of how SNA could be used for examining these different interactions among actors.

Dreżewski, R., Sepielak, J., & Filipkowski, W. (2015). The application of social network analysis algorithms in a system supporting money laundering detection. Information Sciences295, 18-32.

Malm, A., & Bichler, G. (2013). Using friends for money: the positional importance of money-launderers in organized crime. Trends in Organized Crime16(4), 365-381.

 

Whether Old or Young, All Need Social Connection

What does it mean to have a close group of friends? One of the ways social science researchers can examine the intricate dynamics of friendship is through social network analysis. A specific topic in the literature being explored is on mobile youth. This topic relates to how families move geographically or just switch the schools in which their adolescents attend. Those who switch schools or move living situations are considered “mobile”. Friendship for these adolescents can be influenced greatly by such moving and changes in their lives. The role of friendship networks as a theoretical lens has been used to understand more about how students changing schools or living locations, known in the literature as “school mobility” can affect youth in various ways. One such article by Scott and Haynie (2004) titled, “Friendship Networks of Mobile Adolescents” examines this phenomenon.  According to, Scott and Haynie (2004) “The friendship networks of mobile adolescents are thought to be less complete, less satisfying, and less conductive to prosocial behavior than are the networks of residentially stable youth”. This article was dense and packed with findings.

Image result for new kid

The research questions under investigation were “Does the effect of individual residential and school mobility on network structure vary by the percentage of newcomers in the adolescent schools. In the terms of the characteristics and interconnectedness of one’s close associates, is it better to be the “new kid in town” where many others are relative newcomers, or is it better to enter a “stable” social environment, where most of one’s schoolmates are long-term residents?” (p.317). The purpose was twofold, examining the effects of adolescent mobility on broad characteristics of their friendship network. First, by examining their size, structure, and position. Second, the degree to which parents’ know members of their children’s friendship cliques.

Data for this study was collected by the Add Health multisurvey multiwave study administered to US adolescents, their parents, and their schools. 13,000 useable adolescents data was used for analysis.  The nodes of the network structure are indicated by the adolescents who took the surveys. For the edges/links within the network, adolescents were given a roster of all the students enrolled in the school and were asked to list up to five male and five female friends.

Image result for friends in your network

Results of the networks found that compared to residential stayers, residential movers have significantly smaller networks and individuals in these networks are less popular. Residential movers are about 2/3 more likely than stayers to be isolated within the network. Networks of residentially mobile adolescents are significantly denser than stayers. Residentially mobile adolescents are less centrally located and have less prestige in comparison to stayers. Networks of school movers and stayer’s were also examined. Results indicated adolescents who move to a new school are less popular, are more likely to be isolated, and have less prestige within the network compared to students who have been at the school longer. Overall, mobile youth have fewer friends, are less popular, and are more likely to be isolated compared to their peers who have been in places (schools or communities) longer. SNA advances the literature on this topic by being able to examine different network structures with network measurements of density, centrality, and prestige. By examining these measurements researchers were able to analyze how those adolescents who were less in these measurement areas were less likely to nominate someone as a best friend, reciprocate that nomination, and were less popular. These findings are very informative for school officials and parents who have mobile students. This information can aid practitioners in being more proactive in implementing school or community adjustment programming to make more successful and thriving friendship networks for adolescents.

Image result for adjusting to new school

Another social network analysis research article that examined social structure of community relationships took a different direction by examining adults rather than adolescents. Cornwell, Laumann, and Schumm’s (2008) purpose was to examine how older adults are directly connected to each other. Having interpersonal relationships leads to successful aging in adult populations. The research question was not explicitly stated but was inferred that researchers were trying to understand what influences interpersonal connectedness among aging adult population.

Image result for aging adults

Researchers used data collected from the National Social Life, Health, and Aging Project (NSLHAP) and consisted of 4.400 respondents with ages ranging from 57-85. Individuals who answered the survey making up the sample are the nodes of the networks. The edges/links between individuals was measured on nine forms of social connectedness by having individuals to list people with whom they discuss important topics with. Network measures used were: egocentric network size, volume of contact with network measures, emotional closeness to network measures, network composition, and network density.

Image result for retirement community

Results from this study suggest age (aging adults) relate to social network connectedness among individuals. The “oldest-old” have smaller networks and have fewer in their primary group compared to “young-old” networks. The network approach utilized in this study was able to determine that compared to past literature, there is a more complex and nuanced profile of older adults’ social and interpersonal connections with others. The authors were able to utilize SNA to develop a profile of older adults’ social integration given nine dimensions of connectedness in interpersonal networks. Those individuals experiencing retirement or a loss in life actually may become more involved in their community and more connected. By utilizing SNA these researchers were able to take a different approach and have contradictory findings to past research finding those who age may become more isolated.

Cornwell, B., Laumann, E. O., & Schumm, L. P. (2008). The social connectedness of older

adults: A national profile. American sociological review73(2), 185-203.

South, S. J. & Haynie, D. L. (2004). Friendship networks of mobile adolescents. Social Forces,

83, 315-350.

Density and Centralization Measures of a Sports Teams’ Shared Leadership

In a few of my blogs I have talked a lot about leadership and how social network analysis (SNA) can be a tool to further examine shared leadership among a team. This week we are going to explore an article that examines density and centralization as a form of network analyses. An article by Fransen et al. (2015) “The Art of Athlete Leadership: Identifying High-Quality Athlete Leadership at the Individual and Team Level Through Social Network Analysis” is one article I found that looks at density within the network as a point of analysis. The researchers’ purpose was to move athlete leadership forward by using SNA as a novel tool in sports contexts to provide a deeper insight in high-quality athlete leadership, both at the individual and the team level. This “team level” the authors talk about is how they will examine density among the network. There hypotheses involving network density and centralization are as follows: 1) “Because the specific role description of the social leader focuses on the social relations with the other team members, we expect that also at the team level the social leadership quality network will be most strongly related with the social conectedness network”; 2) Teams with higher degrees of shared leadership are characterized by stronger social connectedness.

Image result for shared leadership

The authors performed two studies in writing up this one manuscript. For the purposes of this blog I will mostly be relating to study 1 of this manuscript. The first study involved 25 sports teams which included 308 athletes. Thirteen of these teams were high elite level teams playing at a national level and 12 of the teams were regional level. The researchers created a “leadership network” by having each player on the team rate each teammate with respect to their leadership quality which resulted in a directed, valued network graph. The question answered by each teammate was, “To what extent do you consider each teammate as having good leadership qualities in general?”.  The researchers also created a “social connectedness network” by having each player on the team rate each teammate with respect to how they felt connected to each player. The question answered by each teammate was, “to what extent they felt connected to this person”. Data was collected in person by the researchers who had players fill out survey questionnaires. The nodes within the networks are each player on the team and the links between them were based on the question in which each player rated the others on their general leadership qualities and social connectedness.

Data analysis used density and centralization network measurements to examine measures at the team level.  Density was computed by summing the values of all relations and dividing this result by the number of all possible relation ties. “As a result, high density scores refer to teams with an average high-quality athlete leadership, whereas low density scores characterize teams with on average low-quality athlete leaders”. The researchers also used in-degree centrality to examine centralization of shared leadership among the teams. They concluded that, “teams with high quality shared leadership are characterized the combination of a high network density (high overall leadership quality) and low centralization (i.e. leadership is spread throughout the team).

Image result for network density Image result for network centralization spread

Results revealed that the perceived quality of leadership in general was significantly related to the density of the connectedness network. So, in line with hypothesis 1, the density of the social leadership quality was strongly correlated with the density of the social connectedness network (however this came from the participants and teams in study 2). You can see this in Table 3. Below. The density of the general leadership network was also significantly related to the social connectedness network as well. I highlighted this result in red for the purposes of study one. Also, can be seen is the negative correlation between leadership centralization and general leadership network. This was also hypothesized.The high correlation is .57*** of social connectedness network and general leadership network for density. There is negative correlation of -.16 between of social connectedness network and general leadership network for leadership centralization.

Results of hypothesis 2 can be seen in Table 4. And for purposes of study one I highlighted in red. The highest social connectedness was found in teams in characterized by high leadership density. The differences between the high/low leadership centralization was negligible. . High Density/High Centralization seems to be highest for the general leadership and the density of the social connectedness network.

To summarize, the study suggests that social connectedness is not only an attribute of the perceived leadership quality at the individual level, but also a team-level attribute for teams with high quality athlete leadership. Having higher general leadership qualities was positively associated with higher levels of social connectedness. Also, results revealing the negative correlation shows leadership is shared amongst the team members. Leadership density (Table 4.) is found to be more decisive for the team’s social connectedness in leadership centralization across the different teams.

The Evolution of the Public Sphere

According to German Philosopher Jurgen Habermas, the public sphere is the middle ground between the state and the people. This neutral space according to Habermas is an integral part of democracy. Individuals within this space (primarily physical) come together as a public to openly discuss matters of public concern, which in turn forms a public opinion. This space is open to all and does not matter on characteristics such as class and ideally is universally accessible to all. This is how the public sphere thrives. According to Habermas, the public sphere is weakening due to the power of the public which now includes powerful organizations. These organizations have influence. This influence works against the public sphere by infusing political agenda and commercial power via the media that biases the public’s debate and ultimately the public’s opinion (Introduction to Sociology, 2016b).

Image result for public sphere

One of the reasons the public sphere is fighting to survive is because of the rise of technology and specifically communication technology in a digital age. Socialized communication based on internet networks is what Manuel Castells (2008) refers to as the “Network Society”. This network society is becoming the new public sphere. Societal debate according to Castells has shifted from a national to a global communication of networks and from a physical space to an internet and wireless space. The difference in a network society is that the social networks are driven by electronic communication which he considers “democracy in action”. This new society of information flows can use the internet to disrupt the power dynamics (Introduction to Sociology, 2016a).

Related image

As mentioned earlier, public discourse and shared opinions has shifted from a physical to a virtual interconnected network. This shift has created a change in the way information is shared from a linear to a non-linear hyperconnected and pervasive public sphere. This has made for a new set “rules” in which information is shared, how much an individual has access to information, and the amount of knowledge one if able to gain (Complexity Labs, 2017). The below YouTube video is an amazing insight into the evolution towards the information revolution that Castells speaks of. It really put into perspective just how much our networks and communication has shifted and how influence spreads and an individual’s prominence within a network is gained.

Since we can virtually be in several places at the same time, what implications does this new form of social organization hold? One of shifts we have already seen taken form is through the educational system. Online courses for both higher education and k-12 education especially in the United States has seen a societal shift to online learning. Exchange of information through virtual classrooms, blogs, and even social media has shifted the way educators teach and students learn. The access of remote learning has been a benefit to students who work full-time jobs or live in rural areas. However, are we sacrificing learning for convenience? It depends on the level of engagement that both the teacher and the learner bring. This can be said for both the physical meeting space and the virtual meeting space. Whether we are choosing to engage in on-ground learning or online, we still have access to our teachers 24/7 as learners. This communication happens via email, telephone, or even social media platforms. This information revolution has and will continue to shape many other forms of our society such as government, health, and national relations. These changes are neither good nor bad, they are just the “new normal” of our everyday lives as humans. These complex relationships embedded within society are creating new social institutions that will continue to shape and reform our current political economy and the world as we know it.

Image result for online learning

References

Castells, M. (2008). The New Public Sphere: Global Civil Society, Communication Networks,

and Global Governance. The Annals of the American Academy of Political and Social 

Science, 616, 78-93

Complexity Labs. (2017). Network Society Short Field. YouTube. Retrieved from: https://www.youtube.com/watch?v=UUT4B3au5h4&t=669s

Introduction to Sociology. (2016a). Castells and the Network Society. YouTube. Retrieved from:

https://www.youtube.com/watch?v=tLF5J8Y5zyg

Introduction to Sociology. (2016b). Habermas & the Public Sphere. YouTube. Retrieved from: https://www.youtube.com/watch?v=1PzTyNe4tP4

Relationship Ties Among a Division I Lacrosse Team: Who are the Leaders?

Did you know that shared leadership amongst a team has been found to be positively related to more cohesive teams? Also, that more cohesive teams have been found to have greater team performance (Mathieu, Kukenberger, D’Innocenzo, & Reilly, 2015)? Team cohesion and team dynamics has been a long interest of sport and org. theory researchers interested in investigating what is needed for teams to reach optimal performance levels. This semester for the Social Network Analysis (SNA) class, we are tasked with conducting a research project to apply the skills we have been learning throughout this semester. I am excited to begin this venture as an SNA researcher!

Before explaining my project let me explain some of the research on team cohesion and group dynamics. Carter, DeChurch, Braun, and Contractor (2015) offer a framework of why SNA can be a powerful tool to examine study leadership. The researchers provide the following reasons for using SNA to examine leadership: leadership is relational, leadership is situated I context, leadership is patterned, and leadership can be formal or informal. The researchers continued on to develop a framework of three network approaches that can be taken to further examine leadership networks. The first is “Leadership in Networks”, this would be examining relationship ties such as trust and communication to examine leadership emergence. Second, is “Leadership as a Network”, this would be actual leadership ties between members of a team who recognize and accept leadership relations. Lastly, “Leadership in and as Networks” which would be researchers who try to examine the embedded social context and model the phenomenon of leadership as a relational network. There is evidence to show that if formal leaders are more centralized in social context networks predicts follower’s perceptions of that leader (Carter, DeChurch, Braun, and Contractor, 2015). For my small project this framework fits perfectly, I will be looking at “Leadership in and as a Network” by examining social contexts such as individual’s perceptions of technical skills and leadership and where the formal leaders are within the boundary of the network in regards to these relational ties.

Coaches are always looking for athletes who are leaders. Focusing on both informal and formal leadership roles among team members is a recently new conceptualization as a team-level construct (Mehra, Smith, Dixon, & Robertson, 2006). Formal leaders and those who are emerging must recognize their shared leadership for one another for a team to have optimal performance. So, a question to ask is, “do the leaders on the team see each other as leaders”? My study will also be taking into account the informal (ex: a player with high interpersonal skills) and formal leadership (ex: captain assignment) roles of whole team rather than just examining formal leadership roles. One of my favorite motivational speakers and life coaches is Robin Sharma. Below I think he sums up what a leadership role is all about. How do individuals see and perceive other’s leadership behaviors on a team and do these ties make a difference based on formal and informal leadership roles?

Team cohesion and its effect on performance has been primarily focused on the relationships among the actors of the networks. In my last blog post I referred to an article by Warner, Bowers, and Dixon (2012) that examined relationships among players on two basketball teams. The study was longitudinal and followed the structure and cohesion of the two teams at four times over the course of a season. They specifically examined each actor and team member’s role, centrality, and position within the network as well as the group’s overall structural cohesion throughout the season. Similarly, to how I will be conducting my small project they examined social constructs of friendship, advice, efficacy, and trust. They found that teams with more structural cohesion within the network on the four social constructs had a higher winning percentage than the team with less structural cohesion. They aso found that those with informal leadership roles were central to the network, I believe this lies a finding that warrants for further investigation because it contradicts past research on formal leadership. Cotterill and Fransen (2016) also address the need that there is a gap in the literature in the examination of informal and formal leadership roles on athlete leadership. Therefore, the purpose of my project is to investigate relationships and player perception of their teammates amongst a network of a college lacrosse team, as a way to examine team cohesion.

My small project will be guided by the following research questions:

  • Do team members who perceive playing ability of their teammates to be related to having more perceived leadership qualities than those with less playing ability?
  • Do team members perceive those who hold a formal leadership position on a team to be related to having more perceived leadership qualities than those who don’t hold formal leadership positions?
  • Do team members perceive playing ability of their teammates to be related to having more perceived technical skills than those with less playing ability?
  • Do team members who perceive those who hold formal leadership positions on a team to be related to having more perceived technical skills than those who don’t hold formal leadership positions?

Operational definitions were derived from the variables under examination. Perceived playing ability will be measured by “starter” “starts sometimes” and “non-starter”. Formal leadership will be defined if they are an assigned captain or squad leader (six squads). The edge attributes of leadership quality and technical skill ability are relational ties in which each player answered about each teammate on the following questions “Our team relies on her for leadership” and “Our team relies on her for her technical lacrosse skills”.

The data I will be using was gathered during a survey in the 2016-2017 pre-season from a mid-major NCAA Division I lacrosse team. There is a defined network boundary for this team case analysis of 31 players (Nodes). Node attributes I am interested in examining for the research questions are formal/informal leadership assignment; playing ability (starter/non-starter); position played (offense/defense); tenure on the team (number of years played). The edge attribute I am interested in examining are the in-degree and out-degree from each player based on their perceptions of the teammates leadership (RQ1) and teammates technical skills (RQ2).

References

Carter, D. R., DeChurch, L. A., Braun, M. T., & Contractor, N. S. (2015). Social network approaches to leadership: An integrative conceptual review. Journal of Applied Psychology, 100(3), 597-622.

Cotterill, S. T., & Fransen, K. (2016). Athlete leadership in sport teams: Current understanding and future directions. International Review of Sport and Exercise Psychology, 9(1), 116-133.

Warner, S., Bowers, M. T., & Dixon, M. A. (2012). Team dynamics: A social network perspective. Journal of Sport Management, 26, 53-66.

Mathieu, J. E., Kukenberger, M. R., D’Innocenzo, L., & Reilly, G. (2015). Modeling reciprocal team cohesion–performance relationships, as impacted by shared leadership and members’ competence. Journal of Applied Psychology, 100(3), 713-734.

Mehra, A., Smith, B. R., Dixon, A. L., & Robertson, B. (2006). Distributed leadership in teams: The network of leadership perceptions and team performance. The Leadership Quarterly, 17(3), 232-245.

What’s at the Centrality of Team Cohesion?

In the world of sport, team cohesion is a topic that is heavily researched by scholars. What makes a team successful? What do teams need from their players to be the top performers? This all depends on the context and theory that is able to explain intended results from the reserch.  Social Network Analysis (SNA) is a methodology just beginning to explore the aspect of what makes a cohesive team. When visualizing a network, there are different ways to measure how individual actors and relationships within a network interact as a unit. A lot of these questions about what makes a successful team is answered by the amount of team cohesion is present. There is theory to explain that the more a team works together and trusts each other the more optimal their performance will be.

SNA is a tool that can help examine team cohesion at both the individual and the group level. It can be used to examine the various relationships within the group. Since we know performance and cohesion are highly correlated, SNA can be used as an explanatory tool in seeking to understand the relational embeddedness and connectedness on the individual levels of a team. When looking at the centrality of a node within a network we are trying to examine the prominence of important nodes and the number of connections a node has to other nodes. One way to measure degree centrality is by its “closeness” to other nodes. This can be measured by either the in-degree (prominence of nodes) or out-degree (influential nodes) within a network. Another measurement of node centrality is “betweenness”, this is the extent to which a node is connected to other nodes. And lastly “eigenvector” centrality of a node is its connection to other important nodes.

One article by Warner, Bowers, and Dixon (2012) used SNA to examine the concept of team cohesion among a basketball team. The researchers used in-degree centrality (prominence of nodes) to look at the density (visualize team cohesion). They were also able to use SNA longitudinally by measuring team cohesion links (efficacy, friendship, trust etc.) at four different times during a team’s season and watch the density (team cohesion) change over time. Depicted below you will notice from the off-season (top) to post-season (bottom) we can visually see the team became more cohesive by its density and indegree measures of the individual actors.

In contrast to how players connect by links of intangible factors, Trequattrini, Lombardi, and Battista (2015) looked at tangible factors of actual soccer passes between players during a soccer game in relation to team performance. Like the other article, these scholars used density to examine team relations. The picture below demonstrates as an example how relationships (passes) between players of a team which can used in a different way to examine team cohesion during an on-field performances.

Both of these articles of great examples of how theory and past literature can drive different perspectives in analyzing similar networks. Each article used different links among their players to look at how cohesive the team operated as a unit. SNA has a lot of potential in understanding how the networks of players effect cohesion and group dynamics among the sporting world!

Trequattrini, R., Lombardi, R., & Battista, M. (2015). Network analysis and football team performance: a first application. Team Performance Management, 21(1/2), 85-110.

Warner, S., Bowerts, M. T., & Dixon, M. A. (2012). Team dynamics: a social network perspective. Journal of Sport Management, 26, 53-66.

Social Capital

Social capital put simply – is the investment in social relations with expected returns (Lin, 1999). Resources within a social network are embedded to enhance outcomes of actions for the common good. According to a seminal piece by Putnam (1995) in Bowling Alone: America’s Declines in Social Capital, “social capital” refers to features of social organization such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit” (p.67). He makes the claim that social capital is eroding in which he highlights diminished civic engagement and social contentedness. It is important to note that Putnam published his thoughts in the mid 1990’s just as the internet was starting to claim its spot in our network embedded world. I think the age of the internet initially drew people away from connecting in-person (where we see the lull in social capital Putnam describes) to making a total transformation of how we connect to online and virtual meeting spaces in which we now today use to connect with others. It has totally shifted how networks and social capital is distributed amongst individuals.

 

The question is if technology and the digital age has us dealing with an increase in social capital or social isolation?

As described in “Building a Network Theory of Social Capital” by Nan Lin (1999) there are four elements to explain why social capital (embedded resources) can enhance outcomes of the overall network: information, influence, social credentials, and reinforcement of resources. Most scholars in the field of social sciences believe that social capital is a benefit to both the collective and individuals within the collective. This makes me think of my place of employment the Center for Sport Leadership at VCU. All four of these resources flow and are exchanged and social capital is increased on an individual and group level of all current master students and the alumni base across the country. This is facilitated through networking events, guest speaking of alumni, information interviews, and ultimately job searches among current and past students. The network is in person and virtual on all platforms in which resources are shared and gained.

Nan Lin. 2001. “Building a Network Theory of Social Capital.” Pp. 3-30 in Social Capital: Theory and Research, edited by Nan Lin, Karen Cook and Ronald S. Burt. Aldine de Gruyter.

Putnam, Robert D.  1995. “Bowling alone: America’s declining social capital.” Journal of Democracy 6: 65-78.

It’s a Small World Afterall

As a beginning researcher, one important aspect when reviewing literature is to recognize the major contributions of scholars and the seminal manuscripts that have contributed to a body of knowledge. Any novice interested in Social Network Analysis would be remissed if they did not read the seminal work of Dr. Mark Granovetter (1973) “The Strength of Weak Ties”. In his work he examined for one of the first times, the strength of interpersonal ties and the importance of intermediate overlap of connections within a network. For example, let’s picture a triad of of “A” “B” and “C” actors. “A” has a strong tie to both “B” and “C” but is also the bridge and “weak” tie between “B” and “C”. This could also be explained by the picture below as an “open triad”.  “A” which is a bridge assumes an important role in this relationship network because of the diffusion of information that can cross through the weak tie and information from “B” can indirectly affect “C”. This is a powerful concept to think about how influence and information can reach beyond our personal interactions.

So what is the actual definition of a “weak tie” you ask? This can be difficult to define because of the way relationships in nature are. It could be defined for example by the amount of time, emotional intensity, intimacy, and reciprocal services which characterize a tie (Granovetter, 1973). Visualizing how “strong” or “weak” this tie is like picturing the connection and the link between two actors. My imagination took me to the picture below to try to conceptually understand how we can make sense of the relationship among actors. One of the most important things to understand is that weak ties serve an extremely important role in our social lives because of the indirect flow of information and influence they have the power to connect. We may find out a piece of gossip about our best friend’s cousin which whom we have never met. However, we would never know this information without our best friend! And now we know a little secret in which might influence our lives moving forward without even recognizing it! 

In Kadushin’s (2012) “Understanding Social Networks” he gives a great example of how our weak ties can influence us. He uses Facebook to portray this idea that if we have 100 friends on Facebook and those 100 friends none of them are friends with each other we now have access to 10,000 people (who also have the access to reach us!) (p.6). Riding on Walt Disney’s Magic Kingdom ride “It’s a Small World Afterall” has a whole other meaning now as a social science researcher interested in social network analysis. We have the potential to be connected with people, organizations, and national entities from all over the WORLD from our network of weak ties. This past summer I got to travel to Astana, Kazakhstan on a grant using sport as a vehicle for social change. Now when singing or hearing this song I will forever think of the friends I met one time in Kazakhstan who I am now connected with on my social network platforms who now have connected me via weak ties probably all over South Central Asia!

What’s a SNA Philosophical Perspective?

Much of the information in this post has been a conceptual understanding from the book “Doing Social Network Research” by  Dr. Garry Robins. Social Network Analysis (SNA) as a methodology requires its own way of thinking. As a social science researcher, who uses SNA, one must be willing to step-out of their comfort zone to examine social contexts and social phenomena in a different way. This way of thinking is important to understand from a SNA philosophical perspective. A philosophical perspective is the lens in which a researcher understands reality and creates new knowledge. Specifically, a SNA perspective is applied to the way researchers examine a social system of connections.The ontology (knowledge) through this perspective examines the network relationships and also the social actors who have intentionality differs from traditional social science research in the sense that individuals are all tied together with the assumption of complex structures among the variables that are studied (Robins, 2015).

To perform systematic inquiry, the research questions asked as researchers should guide the study and the statistical procedures used for data analysis. SNA differs from traditional social science research designs from the questions asked which impacts the structure of the data during analysis. Most social science research involves linear models or the utilization of randomized control experimental settings to look for group differences. Observations in these designs are all assumed to be independent from one another. Hence, the overall conceptual framework in which is used to organize a research study using SNA  may be foreign to some social science researchers who have not been trained in performing SNA specifically. Robin (2015) encompasses this idea perfectly when stating, “Networks are based on connectivity not atomization. Networks are structured and patterned, not summed and averaged. Yet, this is more than a methodological nuisance denying us the comfort of standard statistics and classic research designs. It is the heart of a network theorization and we need to adapt to its demands, rather than try to contort network research back into a more familiar shape” (p.10).

As mentioned above, traditional social science research designs are of a linear model or even way of thinking. Data from these designs are usually statistical analyses portrayed in statistical tables carried about by a linear thinking conceptual model. When utilizing SNA social scientists must be able to abstract the observed social systems which should be heavily influenced by theory, finding the best way to explain the social elements on a social network graph is not a linear process.

Relational data explains the difference contents of social connections, which could be many things pertaining to just one relationship. The data is pretty much the variable relations being examined for, example collaboration among co-workers, information flows, or who is communicating with who are all examples of relational data. The relational data describes who the actors are, what types of tie to the study, what are the relevant outcomes and at what level which are all theoretical decisions and conclusions drawn from the data which ultimately makes traditional social science designs like predictive analysis difficult to address.