Many of the patterns in this book came from insights offered by a tool called Dot, a public domain tool authored by Steve North [is this true?]. We used scatterplots to find patterns in the data. For example, we could plot the Communication Intensity Ratio against the number of roles in the corresponding organization, against the number of communication links, against other sociometric quantities--anything and everything. We created dozens of these plots for many data sets and then looked at them to find patterns. Some scatterplots showed a roughly linear correlation, as in this plot of the number of roles as a function of Communication Intensity ratio:
Other plots showed polynomial trends; many others showed linear trends.