Analyzing Roles And Relationships

We analyzed the data from the CRC cards and from notes taken during the role plays in various ways. We analyzed the data on the CRC cards quantitatively. Over time, three meaningful quantitative measures emerged. They are:

Number of Roles: This is simply the number of roles in the organization. It is not the number of people.
Communication Saturation: Each role has the potential to communicate with every other role. The communication saturation is the percentage of the potential communication paths that are actually used.
Communication Intensity Ratio: Not every role has the same number of communication paths to other roles. In many organizations, one role has the lion's share of the communication paths. The communication intensity ratio is the ratio of the number of communication paths of the 'busiest" role to the average number of communication paths in an organization. It measures how much the communication is concentrated in a single role.

These measures helped form some of the patterns; you will see them mentioned later in the patterns themselves. The patterns will explain the implications of the measures.

Visual Analysis
Besides the quantitative analysis, we found that looking at the data from the CRC cards yielded some interesting — and sometimes surprising — insights.

The first "picture" of the data we use is called a sociogram. It is best explained by a metaphor. At the conclusion of a session with an organization, we have a deck of cards, one for each role. We take the cards back home to our laboratory. We begin by rubbing each card through our hair, thus imparting a weak positive charge to the cards (and a negative charge to our hair.) We then deal the cards out on a frictionless table (we have one of those in our lab.) Then we reach for our jar of protons, and very carefully place one proton in the middle of the table (we have a jar of protons.) This causes all the cards to move away from the center, and away from each other. But then we hook the cards that communicate with each other together with rubber bands. In fact, we have three strengths of rubber bands, representing the strength of communication. They show up as lines of different thickness. We have noticed that the more rubber bands that a card gets, the darker it is shaded.

Finally, we step back and let the cards settle down, and view the resulting patterns. As you can imagine, cards that communicate strongly with each other clump together. And some cards are left pretty much alone. The cards with the most communication end up in the middle. In the following sociodiagram, for example, you can see that a role called "SPM" is central to the entire organization. On the other hand, the factory role is isolated from the rest of the organization.

With practice, we were able to learn many things about an organization, literally at a glance. These diagrams led to other patterns; the above pattern is representative of the HubSpokeAndRim pattern.

The second diagram we use is called an interaction grid. Here is a simple interaction grid for the same organization as depicted in the sociogram above:


This diagram shows both communication and direction -- which roles initiate communication to other roles. The roles shown along the left of the grid (the y axis) initiate communication to the (same) roles along the bottom of the grid (the x axis). The shading represents the intensity of the communication, just as in the sociogram.

These visualizations have their roots in social network theory. For those of you who are interested, we explore their origins in SocialNetworkTheoryFoundations, as well as the tools used to create the pictures, after we present the patterns.