Bandura et al. (Evaluation Essay)

P – Point (Strength/Weakness) E – Explain the point E – Example from the study R – Relate to the point on why is it a strength / weakness in the study

Low on ecological validity

The extent to which the findings of research in one situation would generalise to other situations. This is either the situation represents the real world effectively or the task is relevant to real life.

For example, the study was conducted in a laboratory with several controls. The toys in room 1 and room 3 were always the same and always in the same position when the children entered.

OR

The actions of the aggressive models were always the same, in the same order and for the same length of time.

OR

The study was conducted in a specially designed environment with three different rooms specially equipped with toys, games etc: 

    • Children are not normally escorted into a room, sat in a corner to play with stickers and potato prints whilst a strange adult bashes a bobo doll in the opposite corner
    • Children are not normally given nice toys to play with and then told by an experimenter that they could no longer play with the toys because they had to be reserved for other children

The laboratory and controlled setting is not a natural environment for the children that were being observed.

OR

The behaviours that the children were tasked to watch were lacking in mundane realism as it is not true to real life.

OR

Therefore, it could be argued that the findings of this study lacked in ecological validity.

High in validity The extent a researcher is testing what they claim to be testing.

For example, the children were matched for pre-existing levels of aggression .

OR

Another example would be the actions of the aggressive models were always the same, in the same order for the same length of time

Using a matched pair design allows the researcher to be more confident in concluding that any differences in the results were due to learning and not due to participant variable.

OR

The standardised behaviours of the models in each condition ensured that all participants were exposed to the same behaviours. This helped to further minimise the effect of confounding variables on the dependent variables, therefore this study is high in validity.

Ethical issue (psychological harm) Protection of participants means that participants should not be harmed in any way from the experiment regardless if it is mentally or physically Children were exposed to the models that behaved in an aggressive way with actions such as, sitting on the bobo doll and punching the nose saying ‘sock him on the nose’ and this resulted in children displaying aggressive behavior. Observational learning: If imitation leads to learning, display of aggressive behavior by children may have continued even after the study had ended. The children did not leave the study in the same physical or psychological state in which they entered.
Quantitative data

Describing human behaviour and experience using numbers and statistical analysis.

OR

Data that focus on numbers and frequencies rather than on meaning or experience.

For example, the results showed that female group that were exposed to the aggressive female model had more instances of aggression (m = 5.5) than the female group that were exposed to the non- aggression female model (m = 2.5). The use of quantitative data enables clear comparisons to be made between all groups to see the effect the model was having on behaviour of the children. However, we do not know why the children were acting in the ways they did as no further details were obtained.
Use of children (strength)

Children aged under 16 must get parental permission to participate in studies or a guardian (e.g. from someone who looks after them in a nursery).

For example, Bandura studied how aggression is learned using children aged between 37 & 69 months as participants. Consent was given by the classroom teacher.

The use of children in this study allows researchers to understand children’s thoughts and behaviour which in turn might help us understand adult’s thoughts and behaviour.

OR

Additionally, children can be better participants than adults as they are naïve and can be more open and truthful.

Use of children (weakness) Some other issues to consider when working with children Include their language capabilities and whether the study involves children concentrating more than they would usually do.

For example, Bandura studied how aggression is learned using children aged between 37 & 69 months as participants. Consent was given by the classroom teacher.

Children might not understand the task or the complex language of an experimenter. This might therefore affect the way the children interprets the experimenter’s instructions.

OR

Additionally, while consent was given by the teacher, children under 16 years can never give full inform consent and if debriefed, they might be too young to understand hence raising ethical concerns.

OR

Children might not understand the meaning behind the behaviours shown by the model and if it is a desirable behaviour, demonstrating a limitation in the learning process. Especially if compared to different age groups who have the ability to implement cognition and end up not imitating the behaviour shown. This limits the applicability of the results to children only.

Other analysis:

Useful because:

  • helps to guide parents / teachers in the way they model behaviour to children.
  • helps to guide society e.g. in protecting children from harmful models, e.g.: – film / game certificates. – TV watershed. Not useful because:
  • situation was limited, the behaviours were very standardised, with an unfamiliar Bobo doll and were ‘accepted’ so may not apply to more realistic situations, e.g. where children are free to play with any toys, including familiar ones, or are punished for aggression.
  • children may have believed they were supposed to copy the adult, so the effects may not generalise to other situations where they know that the behaviour is wrong.

A number of controls were in place in this study. For example, the time they watched a model for, the layout of the room and which toys were available were the same for all the children. Therefore, other researchers could easily replicate this study to test it for reliability.

Ethics: the issue is this study is protection. The children displayed aggressive behaviour and this may have continued after the study had ended. The children did not leave the study in the same physical or psychological state in which they entered.

The child would play with the toys with a stranger (model) on the other corner. The natural world typically does not provide a quiet, and distraction-reduced environment. Children don’t usually engage in play alone.

The set up was artificial because the child (especially the first stage) was in a setting not really familiar to children. As a result the findings could be argued to be low in ecological validity.

Some of the tasks expected of the child were not usual (e.g. simply to sit and watch an adult play with some toys and not get involved in the play). Therefore, aspects of the study could be low in mundane realism.

The set-up was artificial because the children (especially the first stage) were in a setting not really familiar to them. As a result, the findings could be argued to be low in ecological validity. Some of the tasks expected or the child were not usual (e.g. simply sitting there and watching an adult play with some toys and not get involved in the play). Therefore, aspects of the study could be low in mundane realism.

As the controls were high for both parts of the study (time watching the model, priming before entering the observation room, allocating an equal balance of preexisting level of aggression in each manipulated condition), the researchers could be confident that it was the actions of the model that caused the children to show aggressive and nonaggressive behaviour.

Quantitative data: this enabled clear comparisons to happen between all groups to see the effect the model was having on behaviour. However, we do not know why the children were acting in they ways they did as no qualitative data were collected to explore this.