Albert Bandura, Dorothea Ross & Sheila Ross 1961
AIMS: This study is a laboratory experiment investigating the effects of observing
aggression and was led by Albert Bandura who was the principal developer of Social
Learning Theory.
The aim of Bandura's study was to demonstrate that if children were passive witnesses
to an aggressive display by an adult they would imitate this aggressive behaviour
when given the opportunity.
The researchers made the following 5 predictions:
- Children exposed to aggressive models will reproduce aggressive acts resembling those
of the models
- Children exposed to non-aggressive models will reproduce less aggressive acts
- The control group, who saw no modelled behaviour, would show more aggression than
the group that saw non-aggressive behaviour.
- Children will imitate the behaviour of a same-sex model to a greater degree than
a model of the opposite sex
- Boys will be more predisposed than girls towards imitating aggression
PROCEDURE (METHOD): Bandura et al tested 36 boys and 36 girls aged between 37 to
69 months (mean = 52 months) who were enrolled in the Stanford University Nursery
School.
The design of the experiment has three major conditions; the control group, the group
exposed to the aggressive model, and the group exposed to the passive model.
The child was kept in this room for 20 minutes during which time their behaviour
was observed by 2 judges through a one-way mirror. Observations were made at 5-second
intervals - therefore, giving 240 response units for each child.
3 measures of imitation were obtained. The observers looked for responses from the
child that were very similar to the display by the adult model. These were:
- Imitation of physical aggression (for example, punching the doll in the nose)
- Imitative verbal aggression (for example, repeating the phrases "Pow!" or "Sock him
in the nose".
- Imitative non-aggressive verbal responses (for example, child repeats “He keeps coming
back for more”)
They also looked at two types of behaviours that were not complete imitations of
the adult model:
- Mallet aggression (for example, child strikes toy with mallet rather than Bobo.)
- Sits on Bo-Bo (for example, child sits on Bobo but is not aggressive towards it)
They also recorded 3 aggressive behaviours that were not imitations of the adult
model. These were all aggressive behaviours which were not carried out by the model.
- Punches Bo-Bo
- Non-imitative physical and verbal aggression
- Aggressive gun play
The results enabled the researchers to consider
- Which children imitate the models,
- Which models the children imitate
- Whether the children showed a general increase in aggressive behaviour or a specific
imitation of the adult behaviours.
The two experimental groups of children, who were to be exposed to the adult role
models, were further sub divided by their sex, and by the sex of the model they were
exposed to.
This complicated design therefore has 3 independent variables. The condition the
children were exposed to, the sex of the role model and the sex of the child.
However, the number of children in each group is quite small (six) and the results
could be distorted if one group contained, say, 3 children who are normally quite
aggressive. For example, if the researchers found that a particular group, such as
the 6 boys who were witness to an aggressive display by a male, were the most aggressive
this could have resulted because this small group of 6 boys were already the most
aggressive children.
The researchers attempted to reduce this problem by pre-testing the children for
how aggressive they were. They did this by having an experimenter and a nursery school
teacher observe the children in the nursery and judge their aggressive behaviour
on 4 x 5-point rating scales. It was then possible to match the children in each
group so that they had similar levels of aggression in their everyday behaviour -
ie: matched pairs. The rating scales were;
- physical aggression
- verbal aggression
- aggression towards inanimate objects
- aggressive inhibition
Each child’s score was obtained by adding the result of the 4 ratings.
In stage 1 of the experiment children were brought individually to the experimental
room by the experimenter and shown how to design a picture in one corner of the room,
equipped with a small table and chair, potato prints and picture stickers. The model,
who was in the hallway outside the room, was invited to come in and taken to the
opposite corner where there was a small table, chair, tinker toy, mallet and 5-foot
inflatable Bo-Bo doll. After the model was seated the experimenter left the experimental
room.
In the non-aggressive condition, the model ignored Bo-Bo and assembled the tinker-toys
in a quiet, gentle manner.
In the aggressive condition the model began by assembling the tinker-toys, but after
one minute turned to Bo-Bo and was aggressive to the doll in a very stylised and
distinctive way.
An example of physical aggression was: "raised the Bo-Bo doll and pommeled it on
the head with a mallet".
Example of verbal aggression were: "Pow!" and "Sock him in the nose".
The control group underwent the same procedures for play but no model was present.
After 10 minutes the experimenter entered and took the child to a new room which
the child was told was another games room.
In stage 2 the child was subjected to 'mild aggression arousal'. The child was taken
to a room with relatively attractive toys. As soon as the child started to play with
the toys the experimenter told the child that these were the experimenter's very
best toys and she had decided to reserve them for the other children.
The frustrated child was then taken to a third room for stage 3 of the study - the
child had to walk some distance to the third room to increase their sense of frustration.
The child was told they could play with any of the toys in the third room. The experimenter
stayed in the room "otherwise a number of children would either refuse to stay alone
or would leave before termination of the session".
In this room there was a variety of both non-aggressive and aggressive toys.
The non-aggressive toys included a tea set, crayons, three bears and plastic farm
animals.
The aggressive toys included a mallet and peg board, dart guns and a 3-foot Bo-Bo
doll.
FINDINGS (RESULTS):
- The children in the aggressive model condition made more aggressive responses than
the children in the non-aggressive model condition and the control group - around
70% of the participants in the latter 2 groups had a zero score for aggression
- About a third of those in the aggressive model condition also imitated the the models’
non-aggressive responses - no one in either of the other 2 groups made such remarks
- The mallet was used aggressively on objects other than Bo-Bo to a greater extent
by those in the aggressive model and control conditions than those in the non-aggressive
model group - this was particularly true for girls, with those in the non-aggressive
model condition performing a mean of 0.5 mallet responses compared with 18 for those
in the aggressive model condition and 13.1 for girls in the control group.
- Boys made more aggressive responses than girls
- The boys in the aggressive model condition showed more aggressive responses if the
model was male than if the model was female;
- The girls in the aggressive model condition also showed more physical aggressive
responses if the model was male but more verbal aggressive responses if the model
was female; (However, the exception to this general pattern was the observation of
how often they punched Bo-Bo, and in this case the effects of gender were reversed).
Interestingly Bandura reported that the aggression of the female model had a confusing
effect on the children, perhaps because it did not fit in with their prior learning
about what is culturally appropriate behaviour. For example, one of the children
said, "Who is that lady? That's not the way for a lady to behave. Ladies are supposed
to act like ladies...". Another child said, "You should have seen what that girl
did in there. She was punching and fighting but no swearing".
However the aggressive behaviour of the male model fitted more comfortably into a
cultural stereotype of appropriate behaviour. For example, one boy said, "Al's a
good sucker, he beat up Bo-Bo. I want to sock like Al." One of the girls said, "That
man is a strong fighter, he punched and punched and he could hit Bobo right down
to the floor and if Bo-Bo got up he said, 'Punch your nose'. He’s a good fighter,
like Daddy."
The girls spent more time playing with the tea set and the crayons while the boys
played more with the guns. No sex differences were found with regard to farm animals,
cars or other toys. Children from the non-aggressive model condition spent more time
sitting quietly and not playing.
CONCLUSIONS: The findings support Bandura's Social Learning Theory and show that
not all behaviour is shaped by reward or punishment. The model were not reinforced
for their behaviour and yet it was imitated. This supports the idea that children
learn social behaviour such as aggression through the process of observation learning
- through watching the behaviour of another person.
Central to Social Learning Theory is the identification of which types of models
are more likely to be imitated.
In the study it was found that aggressive male models were more likely to be imitated
than aggressive female models. One probable reason for this is to do with sex roles:
perhaps it is more acceptable in Western culture for men to be aggressive than women,
and even by 3 or 4 years of age children are learning the dominant stereotypes that
relate to sex-role differences. So aggressive male models are more likely to be imitated
since this is seen by the child as more fitting or appropriate for men (in general)
than for women (in general).
Bandura found that boys were more likely to imitate the aggressive male model than
the female role model. Perhaps the greater relevance of the male model's behaviour
for boys lies in the fact that boys perceive the similarity between themselves and
the model.
CRITICISMS (EVALUATION): As already noted Bandura believed that his findings supported
his Social Learning Theory.
The study was carefully set up and controlled, using measurable acts that could be
recorded. This enables cause-and-effect conclusions could be drawn because the variables
were isolated and operationalised. The dependent variable - the various physically
and verbally violent actions - were observed by 2 judges and those observations
checked for reliability. One judge did not know to which condition the child under
observation had been allocated - ie: a ‘blind’ procedure was used to avoid bias when
recording the child’s behaviour. Therefore, the results can be considered reliable.
However, the experiment can be criticised for having low ecological validity - how
often will children observe adults being either violent towards a Bo-Bo doll or playing
quietly with tinker toys? As Kevin Durkin (1995) pointed out: “Where else in life
does a 5-year-old find a powerful adult actually showing you how to knock hell out
of a dummy and then giving you the opportunity to try it out yourself?”
Durkin also accused Bandura of failing to distinguish between play fighting and real
violence, arguing that the children would not have been so violent towards another
child.
The children, after seeing the adult’s behaviour, may simply have been doing what
they were meant to - demand characteristics.
G Cumberbatch raised the issue of whether familiarity with Bo-Bo doll influenced
behaviour towards it? Cumberbatch found that children who were unfamiliar with the
doll were 5x more likely to imitate aggressive behaviour against it than children
who had played with it before.
As the children were all from a California university nursery, the study may be subject
to culture bias - making it dangerous to generalise from it.
The study can also be criticised on ethical grounds. The children were shown adults
demonstrating violent behaviour and tacitly encouraged to copy their behaviour. While
it is reasonably safe to assume Stanford’s ethics committee approved the research,
Bandura et al omit any mention in their documentation of gaining parental consent
for their children to participate.
With Richard Walters (1963) and then on his own (1965), Bandura investigated the
role of punishment and reward, using similar Bo-Bo doll scenarios, and found that
they had a powerful effect on observational learning.
Developed initially from separate work by Mark Hollah and Christine Brain
