A brief history of balance through time

We present methods for establishing the amount of reciprocity, transitivity and group balance (a generalization of structural balance) in sociometric structures. These methods are applied to the second time series of sociometric data provided by the Newcomb (1961) study. The amount of reciprocity was above chance levels at the outset and showed no systematic variation thereafter. Transitivity has a very different time scale. It climbed steadily through the first nine weeks and remained stable thereafter. While consistent with chance at the beginning of the study period, it grew to above chance levels at week 3. Group imbalance declined throughout the entire period. The reasons for these different time scales are discussed.