Bandura 61

• Aim: Albert Bandura aimed to see whether aggressive behaviour could be acquires through observation of aggressive models. 
• • study was a laboratory experiment using a matched pairs design.
• • 36 boys and 36 girls aged 3-6 were selected from the Stanford university nursery in California
• • The dependant variable was the level of agression the children displayed
  • The independent variables included; modelling of aggression, sex of model and same/different sex model and observer.
• • Aggressiveness (. Potential confounding variable) was controlled by ensuring that each group contained equally aggressive children
  • Ratings of the children were determined before hand by an experimentor who knew the children well and one of the children’s teachers, and used to match the children
• • Children were placed in threes 3 boys or three girls
  • Randomly allocated to one of the three conditions; aggression group; observed an aggressive adult model male or female punching kicking and shouting at a large inflatable bobo doll, Non agression group; non aggressive model assembling mechanicals toys, Control group: no model was present while children were playing
• Step 1 - the children were brought into a room with an observation window and allowed to play with toys (with or without model present)
• Step 2- children were then taken to another room where the children were deliberately frustrated shown new shiny toys and being told they were for other children (if they were angry they would have the urge to take it out later
• Step 3- finally the children were taken into a playroom containing a range of toys including an inflatable bobo doll. The behaviour was observed through a one way mirror by the male model ie covert. A second observer was present for half of the participants to determine inter-rater reliability.
• Behaviours were categorised as

  • imitative aggression - sitting on bobo and behaving aggressively or saying pow
  • Partially imitative- using mallet aggressively on toys other than bobo
  • Non imitative aggressive- punching bobo doll saying hostile things not said by model
• • children who witnessed aggressive model were more likely to mimic aggression either partially or fully.
  • Children who did not observe aggressive model or adult at all displayed much less aggressive behaviour 
  • Children in aggressive group were slightly more likely to engage in non imitative aggression 
  • Boys more likely to imitate aggression shown by same sex model then opposite and more likely overall to imitate physical aggression 
• Conclusion 

  Social behaviour such as aggression can be learned by the imitation of models. Imitation is more likely when the modelled behaviour in gender typical and when the model and observer are of the same gender.
• Generalisability strength: Banduras 1961 study has high generalisability due to the large sample size of 72 with an equal number of boys and girls, therefore Banduras study is less likely to risk having anomalies his results and that the study is representative of both genders of children
• reliability strength: Reliability was strong due to the standardisation of the procedure with the same amount of time in each phase for each child, same scripted information and same toys eliminating risks in consistency and making the results more reliable 
• Validity: Banduras study has high environmental validity due to construction of a familiar playroom setting built in the laboratory, this setting would’ve increased the validity due to the elimination of the risk that the children would act abnormally in an unusual setting. By making the labs more familiar to the children they are more likely to feel more comfortable and act as they would in a real life situation
• Application strength: A strength in the results is that the study was used to change social policy understanding how viewing aggressive impacted children and their development. Banduras findings on imitative aggression from model have been beneficially applicated in real life with restrictions on tv content before 9pm to reduce exposure of aggressive, inappropriate or negative models so it is less likely to be imitated by children,
• Conclusion: Bandura’s study had a very good level of experimental control and the researchers built in some clever design features. It is also a good example of a study with important policy implications, such as the effects of parental violence on children’s behaviour.  Like all laboratory experiments, however, Bandura et al’s study is artificial and may not generalise entirely to everyday life.
• Weakness in generalisability: Cultural bias due to lack of cultural diversity in study done with only American students from Stanford University nursery California means that there is a weakness in generalisability as the study may not be representative of children from multiple cultures
• reliability weakness: A weakness in reliability is that because of ethical issues informed by the British psychology society we would be unable to replicate it today which invalidates Bandura study due to lack of supporting research 
• validity weakness: A weakness in validity was the use of the popular bobo doll toy in the study looking at imitative aggression, this is because the bobo doll was advertised as a doll meant to be punched and hit which invalidates Banduras result as it is not clear if the children imitated the aggressive modelling or the tv advertisement