Bandura, Dorothea Ross and Sheila A. Ross conducted a research on incidental learning by imitating behaviors exhibited by adult model in 1961. The research involved 72 children; 36 boys and 36 girls, aged between 37 to 69 months, with two adults a man and a woman serving as the model, one female experimenter conducting the study for all the children. The children were divided into eight experimental groups of six children each and a control group consisting of 24 children. They were then subdivided to two groups, with one exposed to aggressive models and the other exposed to subdued and nonaggressive models. They were further subdivided to group by their gender, male and female. The control group had no prior exposure to adult models and was tested only in generalization situation (Bandura, Ross and Ross, 1961).
The experimenter and a nursery school teacher rated the children on four five-point rating scales, both were well acquainted with the children. The scales measures the extent of aggression displayed from physical, verbal, aggression towards inanimate objects, and aggressive inhibition by the children. To permit assessment of inter-rater agreement, fifty-one children were rated independently by both observers (Bandura, Ross and Ross, 1961).
The mean score estimated from Pearson product-moment correlation was .89 towards reliability of composite aggression. The score obtained by summing the ratings on the four aggression scales, were later used as basis to arrange the children in triplets. They were then randomly assigned to one of two models or the control group (Bandura, Ross and Ross, 1961).
The model was invited in after the children were being individually brought to a corner in the experimental room; the corner of the room was structured as the play area for the children. Varies activities that was rated as high interest value for children by previous studies in the nursery were demonstrated how it is being used and left it for the children to play with. The model was then escorted to the opposite corner where a small table, chair, a tinker toy set, and a 5-foot inflated Bobo doll was placed, those were the materials for the model to play with, the experimenter left the room and started observing the situation in the room through a one way mirror from the next room (Bandura, Ross and Ross, 1961).
In non-aggressive condition, the model displayed a quiet subdued manner while assembling the tinker toy ignoring the Bobo doll. Vice versa, in aggressive condition, model aggresses towards the Bobo doll after assembling the tinker toys (Bandura, Ross and Ross, 1961).
A sequence of physical and verbal aggression was demonstrated repetitively a few times by the model, with sufficient pattern of responses to be scored as aggressive behavior. The children were provided with a task in which diverts their attention while insuring observation of the model's behavior was done without noticing the absence of the experimenter (Bandura, Ross and Ross, 1961).
All members in the experiment were exposed to mild aggression arousal as to insure every member were under some degree of instigation to aggression. As so observation of aggression displayed by others would not reduce the aggression probability on the part of the observer. Children in non-aggressive condition expressed little aggression towards appropriate instigation indicates inhibitory process is present (Bandura, Ross and Ross, 1961).
- Aggression Transmission through Imitation of Aggressive Models (Albert Bandura, Dorothea Ross and Sheila A. Ross, 1961)