Method and Procedure
Bandura's experiment was brilliantly designed with 72 children aged 3-5 from Stanford University's nursery. The researchers cleverly controlled for existing aggressiveness levels by ensuring each group had equally aggressive children to start with.
The experiment had three phases that built up the tension perfectly. In Phase 1 (modelling), children watched either an aggressive adult attack a Bobo doll with punches and shouts of "Pow!", a non-aggressive adult, or no model at all. Phase 2 (aggression arousal) was genius - kids were allowed to play with amazing toys, then suddenly told they couldn't, creating frustration.
Finally, Phase 3 (delayed imitation) was the real test. Children were taken to a new room with toys including a Bobo doll, observed through a one-way mirror. This setup tested whether they'd copy the aggressive behaviour in a completely different setting without the original model present.
Key Point: The three-phase design was crucial - it showed that children didn't just mindlessly copy behaviour immediately, but actually learned and applied it later in new situations.