4.1.8 – Bandura (1961)



The aim of Bandura (1961) was:

To investigate whether children would imitate aggression if shown an aggressive aggressive role model, compared to children not shown this aggressive role model.

To investigate whether the gender of this role model (compared to the child) had an effect.


Experimental Design – Matched pairs.

Experimental Method – Lab experiment and a naturalistic experiment (because the IV of child’s gender is naturally occurring)

IV – Whether the child is shown an aggressive role model or a non-aggressive role model.
– The gender of the role model in comparison to their gender

DV – The aggression shown by the children.

Sample – 72 children from Stanford University Nursery.

(1) 24 – Aggressive model (Of which, 6 boys and 6 girls saw a male model, and 6 boys and 6 girls saw a female model)

(2) 24 – Non-aggressive model (Of which, 6 boys and 6 girls saw a male model, and 6 boys and 6 girls saw a female model)

(3) 24 – Control group (All 24 saw no model)

Bandura wanted to control for base levels of aggression and therefore they were all rated on a 5-point scale for all of these characteristics and distributed evenly into the groups, so one group wasn’t full of particularly violent children.

Firstly, the children were put in the model room. In this room, there were toys such as finger paints.
The experimental conditions then saw a model enter the room with a Bobo doll.
The non-aggressive model ignored the Bobo doll and played quietly next to the children.
The aggressive model pushed the doll over, punched it, hit it with a plastic mallet and shouted aggressive phrases such as “Hit him down”

Then, the children were placed in the arousal room, where there were some appealing toys. They were told after a few minutes that they weren’t allowed to play with them toys. This was meant to make the children feel frustrated.

Finally, they were brought into the observation room. 
This room contained a mixture of aggressive (Such as a plastic mallet, and a 3 foot Bobo doll) and non-aggressive toys (Such as cars, trucks, crayons and paper).

The control group saw no role model.

For 20 minutes they were observed by 2 experimenters who recorded the child’s behaviour every 5 seconds.
They looked for some pre-determined behaviours such as: imitative verbal aggression, imitative physical aggression, imitative non-aggressive verbal statements and also acts of non-imitative physical or verbal aggression.


Those who observed an aggressive role model were more aggressive.

Male role models seem to have more of an impact.

Boys watching a male aggressive role model showed an average of 25.8 acts of violence, compared to girls watching a female aggressive role model who showed 5.5 acts of violence on average.


Behaviours can be learnt by imitation even if not reinforced.
Male role model more influential
Boys more likely to be physically aggressive.
Verbal aggression –> boys imitated the male role model more, girls imitate the female model more. (The model they identify more with)



(+/-) 72 children, large amount. But only 6 in each experimental condition
(-) All children of people who work at/go to Stanford University –> Representative? Children of academics might be brought up differently.
(-) Can children be generalised to adults? Children would probably be more quick to pick up new behaviours due to them still learning.


(+) Very standardised procedure (same rooms for everybody, same model for everybody in that condition, etc)
(+) Structured observation –> Decided on what to look out for beforehand and decided on behaviours
(+) Inter-rater reliability –> 2 observers.

(-) Each time the model performed the action, they might have done things slightly differently which might have affected what the children saw.
(-) There is no guarantee that all the children saw all of the actions.


(+) Can be applied to real life in terms of bringing up children. It shows that if a child’s role models are not aggressive, they are also less likely to be aggressive. Especially shows that models of the same gender can be more effective at times, for things like verbal aggression.


(-) Ecological validity – unusual environment watching an adult stranger playing with toys.
(-) Due to this unusual environment, there may have been demand characteristics because they may have done what they thought the experimenters wanted them to do, like hit the Bobo doll.
(-) Doesn’t consider the biological approach – only the learning theories.


(+) Could argue that the benefits to society could outweigh the distress caused because it taught us how to reduce aggression.
(-) Harm –> May have caused distress to the participants by seeing the violent behaviour.
(-) “Normalising unhelpful behaviours” (BPS Guidelines) because the children were shown violence and became violent, this may have stayed with them.
(-) No valid consent gained. Presumptive consent given by their teachers.
(-) Could not withdraw.