Now Reading
Fundamentals of Social & Organizational Network Analysis

Organizational Network Analysis (ONA)

Organizational network analysis has been an important part of my life and career for a number of years now. I wrote my first Master’s thesis on it, and continue to use the method in most if not all of my consulting engagements. It’s a powerful tool.

But so many people don’t have a foundation to help them begin studying it on their own so I thought I’d provide them with a small posting with basic explanations and a few references, to help them begin their network journey.

First let’s provide an overview of the discipline.The large majority of social sciences, including human resources, carry within them an assumption that actors (employees), or members of any organizational or social system behave without being affected by other actors or members of the same system (See Knoke & Yang, 2008). In the least, we can assume that they do not take influence into consideration.

In other words, the dominant social sciences of today concentrate heavily on analyzing the attributes of a member of a social or organizational system (like a firm) without regard to that member’s relationship to other members in the same system.

Network analysis, also known as social network analysis, and organizational network analysis, and when superimposed against process controls—Talent Sphere Mapping™, focuses on measuring and analyzing the structural and relational connections between members of any given system (see Wasserman & Faust, 1994).

But what is Network Analysis…

Network analysis is an interdisciplinary subject that draws upon a diverse field that in consequence draws on a number of subjects—subjects such as anthropology, economics, organizational studies, epidemiology, human resources, business management, information science, communication, complexity, chaos and game theory ( see Barabasi, 2003).

Social/Organizational network analysis has enjoyed a research focus within sociology, psychology, and management theory (See Rob Cross’ great introductory book called “Driving Performance Through Social Networks”), which is why it has piqued the interest of so many applied researchers from a variety of different backgrounds, including physicists (like me!)

Just so you know, in the industry we use either the term organizational network analysis (ONA), social network analysis (SNA) or more commonly just network analysis. That’s the general term will use when trying to describe the whole book of knowledge of network.

To keep things simple however, go ahead and assume a distinguishable difference between organizational network analysis and social network analysis which we will mostly ignore for now. What we’ll do is discuss this topic under the assumption that organizational network analysis is an adapted form of social network analysis, which is used mainly in business and applied contexts—in other words it’s the application of SNA.

In fact, when I attended the Network Roundtable conference in Arlington, VA earlier this year, one of the attendees asked Rob Cross, who was there throughout the entire conference why we use the term ONA (organizational network analysis) instead of SNA (social network analysis), to which he responded “well every time I would speak to corporate executives about ‘social’ network analysis, they would think I was talking about Facebook or social media in general, and I would spend a lot of time trying to re-educate them, so we came up with the term organizational network analysis instead,” he said. Now, I’m not sure if by “we” he meant “I” and that he was taking the credit for it or if he generally meant a “we”.

I’ll ask him the next time I see him—maybe. 🙂

The whole point of distinguishing the two is that we need to focus our analysis efforts to the framework of the organization alone—because that’s what I really care about. Anklam, Cross & Gulas (2005), make that distinction in their article so I think I’m good.

The study of the network perspective began with Dr. Jacob L. Moreno when he attempted to map all the relationships within some kind of institution for young girls, because they were having problems with some of the “patients” escaping. He started to look for a pattern by graphing the friendships between them and noticed some interesting things. Back then the subject was called sociometrics—I think, but it’s unclear to me right now how separate it was from other sciences in the 1930s.

Though the whole subject was truly built on hundreds of years of social research in multiple disciplines (see the awesome article by my fave peeps, Borgatti, Mehra, Brass, & Labianca, 2009).

Yes, the subject is old and seemingly boring but, over the past generation there has been considerable growth in business and academic interest in social networks. This is mainly due to high tech communication and the proliferation of ideas such as the 6 degrees of separation, the force that is THE Globalization (yep…THE), and the increase in interconnectedness of the world.

You know that cool Google algorithm that was developed by Sergey and Brinn way back (which still forms the foundations of the web—and therefore world commerce)…yep.. you guessed it..the PageRank algorithm is nothing more than social network analysis (SNA) modelling and formulas in smart disguise. It sucks that they operationalized it first, right?

Even network analysis software has experienced a recent increase in demand, and there are tens of network analysis applications freely and widely available online. Many of which combine statistical analysis with the traditional methods of network analysis to provide a highly quantitative method of analyzing organizational dynamics (see Otter, 2010).

So let’s just glean over some of the fundamentals so that you can get a sense of things.

The two integral parts of any social network are its actors and its actors’ relations with others. An actor is considered to be any entity ranging from a person, to the team, to an entire organization or even a nation-state.

A relation is defined as a particular connection between one actor and another (could also be called a dyad or a link). That connection is not necessarily a function only of one of the actors, but can be subordinate to both actors simultaneously. Both actors that share a connection control the relationship in some dynamic that never truly allows one party to dominate the relationship in most cases. In other words, it is not a property or an attribute of only one entity in an organizational or social system that only one person controls. To compare, an attribute like the color of your hair has nothing to do with the color of hair of the person sitting next to you in class—its just an attribute.

Network analysis is all about consequences and prediction—that’s why we study it. I like this quote from Knoke and Yangs teeny tiny book — “The central objectives of network analysis are to measure and represent the structural relations accurately, and to explain both why they occur and what are the consequences.”

When it comes to every organization on the planet we all know that they have a hidden structure of communication, collaboration, and authority that is invisible to all possible individual and organizational perspectives—especially human resources. Even though we of the HR profession think that we are the all-knowing people of the firm and that we can somehow sense and perceive all, when it comes to people. But boy is it powerful!! The most important accomplishment of organizational network analysis is that it allows leaders to align individual and collective action with the organization’s strategic objectives.

Of course, formal structure determines in part who is sought out in a network (for information for example), but informal relationships tend to be more crucial, which is why SNA is so important. Because it is through informal relationships that technical expertise and organizational knowledge are provided and sought.

As I mentioned before, the network perspective is distinct from the dominating social science perspectives in that it looks on the relations between actors of a system, rather than the attributes of the actors. Of course, that is not to say that attributes of a given actor do not affect its (his/her) relation to other actors in a given system. For example, it could be that smart people (an attribute) like to build relationships with other smart people. However, the traits of an actor in a given system exemplify the type and nature of the relations to other actors of any given system.

For example, patterns and rates of the diffusion of technological innovations are better explained by taking into account the structures of communication and advice among actors than by their education, age, class, gender, or race ( that’s from Cowan & Jonard, 2004 with a touch of Valente, 1995).

That’s enough for now..look up the articles I mentioned and make sure to let me know if you have any questions..

About The Author
Dr. Joseph A.E. Shaheen
Computational Social Scientist with a twist of network science, social network analysis, data science, and random thoughts.
Leave a response

Leave a Response

This site uses Akismet to reduce spam. Learn how your comment data is processed.